Open session of the standing technical committee of the EUFMD- 2008 by EuFMD

AGA: EUFMD/RG/08

REPORT OF THE

SESSION OF THE RESEARCH GROUP OF THE STANDING TECHNICAL COMMITTEE OF THE

EUROPEAN COMMISSION FOR THE CONTROL OF FOOT-AND-MOUTH DISEASE (EUFMD)

HELD AT

ERICE (ITALY) 14-16 OCTOBER 2008 (OPEN SESSION) 17 OCTOBER 2008 (CLOSED SESSION)

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS ROME, 2008

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

TABLE OF CONTENTS REPORT ON DISCUSSIONS HELD IN THE OPEN SESSION - 14-17 OCTOBER 2008 OPENING CEREMONY –OPEN SESSION

RECOMMENDATIONS OF THE OPEN SESSION

SESSION 1: GLOBAL FMD CONTROL THROUGH REGIONAL CO-ORDINATED ACTIONS: OPPORTUNITIES AND CONSTRAINTS

SESSION 2A: OPTIMIZING PROGRAMS IN A SITUATION OF LIMITED RESOURCES

SESSION 2B: VACCINATION: OVERCOMING ANTIGENIC DIVERSITY IN ORDER TO SIMPLIFY PREVENTIVE PROGRAMS

SESSION 2C: NOVEL VACCINE DELIVERY APPROACHES: PROGRESS UPDATES

SESSION 2D: FMD VACCINE STANDARDS REQUIRED FOR GLOBAL CONTROL

SESSION 3: BIOSECURITY AND BUY-IN

SESSION 4: MEASURE PROGRESS IN GLOBAL AND REGIONAL FMD CONTROL, AND EARLY WARNING OF FMDV EMERGENCE

SESSION 5: DIAGNOSTICS: MAKING QUALITY SERVICE AVAILABLE WHERE NEEDED

SESSION 6: COMPONENTS AND CAPACITY FOR EFFECTIVE CONTROL

SESSION 7: PROSPECTS FOR INTEGRATING ANTI-VIRAL APPROACHES

SESSION 8: STRATEGY FOR REGIONAL AND GLOBAL CONTROL

REPORT OF THE CLOSED SESSION OF THE STANDING TECHNICAL COMMITTEE OF THE EUFMD COMMISSION ITEM 1: AGENDA OF THE SESSION

ITEM 2: FMD LAB MINIMUM CONTAINMENT STANDARDS

ITEM 3: MINIMUM DIAGNOSTIC CAPACITY IN EUFMD MEMBER STATES FOR THE LABORATORY CONFIRMATION OF FMD

ITEM 4: SAMPLING INSTRUCTIONS FOR FIELD VETERINARIANS TO COLLECT SAMPLES WITH THE PRIMARY BIOCONTAINMENT AT THE POINT OF SAMPLING FOR USE IN

RT-PCR.

ITEM 5: POSITION PAPER ON THE OPTIONS OF DECENTRALIZED TESTING.

15 15

ITEM 6: SERO-SURVEILLANCE IN TURKEY: THE QUESTION OF HARMONISING THE PERFORMANCE/INTERPRETATION OF SP ANTIBODY DATA WITH/BETWEEN

RG MEMBER LABORATORIES.

ITEM 7: VACCINE SELECTION AND POTENCY OF ANTIGENS IN THE EUVB FOR USE AGAINST THE CURRENT SAT2 FMDV CIRCULATING IN

BOTSWANA/NAMIBIA

ITEM 8: REMAINING BUSINESS; WORKPLANS AND TASKS

16 17

ITEM 9: LIST OF ACTIONS

CHAIRMAN’S AGENDA LIST - FROM 2007 SESSION

III

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

LIST OF APPENDICES APPENDIX 1 AGENDA OF THE OPEN SESSION..............................................................................................................19 APPENDIX 1B POSTER SESSION...............................................................................................................................25 APPENDIX 2 LIST OF PARTICIPANTS.........................................................................................................................27 APPENDIX 3 DR. GAETANA FERRI FULL ADDRESS..........................................................................................................39 APPENDIX 4 KEYNOTE THE GLOBAL CONTROL OF FMD; CHALLENGES AND OPPORTUNITIES

K. SUMPTION AND J. LUBROTH .................................................................................. ..........41 APPENDIX 5 GLOBAL FMDV DISTRIBUTION AND REGIONAL VIRUS RESERVOIRS: AN OPPORTUNITY TO DIVIDE AND CONTROL?

J. M. HAMMOND, D. P. KING, N. J. KNOWLES, J. WADSWORTH, K. G. SWABEY, B. STATHAM, Y. LI, P. KEEL, P. HAMBLIN, G. H. HUTCHINGS, S. M. REID, K. EBERT, J. M. STIRLING, N. P. FERRIS AND D. J. PATON ......... 47 APPENDIX 6 THE HEMISPHERIC PROGRAM FOR THE ERADICATION OF FOOT-AND-MOUTH DISEASE

V. SARAIVA AND D. GEALE ................................................................................................ 55 APPENDIX 7 FOOT-AND-MOUTH DISEASE VIRUS TYPE C SITUATION: THE FIRST TARGET FOR ERADICATION?

P.L. ROEDER AND N.J. KNOWLES ......................................................................................... 65 APPENDIX 8 PROGRESS IN FMD CONTROL IN IRAN AND PERSPECTIVE ON LONGER TERM CONTROL IN THE REGION

F. GEIGER, V.OTAROD ..................................................................................................... 67 APPENDIX 9 PHYLOGENETIC STUDY OF FMDV ISOLATES COLLECTED IN RUSSIA, CIS-COUNTRIES AND MONGOLIA IN

2000-2007

A. SCHERBAKOV, A. TIMINA AND V. BORISOV........................................................................... 69 APPENDIX 10 CONSTRAINTS AND OPPORTUNITIES FOR THE CONTROL OF FOOT AND MOUTH DISEASE IN INDIA

D. HEMADRI, A. SANYAL, RP. TAMIL SELVAN, S. SARAVANAN, JK. MOHAPATRA, LALKRISHNA, KM. BUJARBARUAH B. PATTNAIK ........................................................................................................... 73

AND

APPENDIX 11 THE GLOBAL FMD RESEARCH ALLIANCE (GFRA)-AN R&D RESPONSE TO THE MAJOR CONSTRAINTS TO GLOBAL FMD CONTROL

L. RODRIGUEZ, C.G. GAY, M. JEGGO, S. ALEXANDERSEN, P. KITCHING, B. CHARLESTON, D. PATON, K. MǿLLER AND T.K. NIELSEN .......................................................................................................... 79 APPENDIX 12 KEYNOTE: VACCINATION: OVERCOMING THE CONSTRAINTS TO ACHIEVING EFFECTIVE IMMUNITY RATES

A. DEKKER................................................................................................................... 81 APPENDIX 13

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

INTRADERMAL VACCINATION WITH

1/10 DOSE AGAINST FMDV PROTECTS PIGS AS WELL AGAINST CLINICAL DISEASE AND

SUBCLINICAL VIRUS SHEDDING AS INTRAMUSCOLAR VACCINATION WITH A FULL DOSE

P. EBLÉ, K. WEERDMEESTER, F. VAN HEMERT-KLUITENBERG AND A. DEKKER .......................................87 APPENDIX 14 FOOT-AND-MOUTH DISEASE (FMD) VACCINATION STRATEGIES IN SHEEP AND LAMBS BY USING COMMERCIAL OIL VACCINE

E.J.A. SPÄTH, B. ROBIOLO, E.A. LEÓN, J.A. MANAZZA, S.J. DUFFY, J. FILIPPI, A. HAM, J. LA TORRE AND E. SMITSAART ...................................................................................................................95 APPENDIX 15 LONGEVITY OF PROTECTION IN CATTLE FOLLOWING VACCINATION WITH EMERGENCY FMD FROM THE UK STRATEGIC RESERVE

–

PRELIMINARY RESULTS

S. COX , S. PARIDA, P. HAMBLIN, B. BANKOWSKI, B VERONICA CARR, D. PATON AND P. BARNETT............ 105 APPENDIX 16 HIGH POTENCY VACCINES INDUCE PROTECTION AGAINST HETEROLOGOUS CHALLENGE WITH FOOT-AND-MOUTH DISEASE VIRUS

K. BREHM, N. KUMAR, H.-H. THULKE AND B. HAAS .................................................................. 111 APPENDIX 17 PREDICTING ANTIGENIC SITES ON THE FMDV CAPSID FROM CROSS-REACTIVITY DATA

F. F. MAREE, R. REEVE, B. BLIGNAUT, J. J. ESTERHUYSEN, E. FRY, T. DE BEER, E. RIEDER AND D. HAYDON 113 APPENDIX 18 VACCINE MATCHING: RELIABILITY OF FOOT-AND-MOUTH DISEASE (FMD) R-VALUES DETERMINATION

N. MATTION, N. GORIS, T. WILLEMS , B. ROBIOLO, E. MARADEI, C. PEREZ,, A. PEREZ, E. SMITSAART, N. FONDEVILA, E. PALMA, K. DE CLERCQ AND J. LA TORRE............................................................. 123 APPENDIX 19 TOWARDS VACCINE SELECTION GUIDELINES FOR EACH REGIONAL VIRUS POOL OF FOOT-AND-MOUTH DISEASE

D. J. PATON ...............................................................................................................131 APPENDIX 20 EXPRESSION OF FMDV CAPSID PROTEINS IN SILKWORM-BACULOVIRUS EXPRESSION SYSTEM AND ITS UTILIZATION AS AN EMPTY CAPSID VACCINE

Z. LI, Y. YI, X. YIN, ZHIDONG. ZHANG, Y. LI, ZHIFANG ZHANG AND J. LIU ....................................... 133 APPENDIX 21 POTENTIAL OF MODIFIED VACCINIA ANKARA (MVA) AS A VACCINE DELIVERY VECTOR FOR FOOT-AND-MOUTH DISEASE VIRUS

(FMDV).

J. CASTILLO-OLIVARES, D. PATON, B. CHARLESTON AND S. PARIDA ............................................... 141 APPENDIX 22 MARKER POTENTIAL OF A VACCINE PREPARED FROM A FOOT-AND-MOUTH DISEASE VIRUS WITH A PARTIAL VP1 G-H LOOP DELETION

V.L. FOWLER, N.J. KNOWLES, D.J. PATON, E. RIEDER

AND

P.V. BARNETT. ...................................... 149

APPENDIX 23 FRENKEL LECTURE: FMD VACCINE DEVELOPMENT

- PAST AND FUTURE

L. ROBINSON, M. WINDSOR, J. HOPE, G. MACPHERSON, B. CHARLESTON ........................................ 157 APPENDIX 24 KEYNOTE:VACCINE QUALITY TESTS: THE VALUE OF ALTERNATIVE METHODS?

N. GORIS

AND

K. DE CLERCQ ........................................................................................... 159

APPENDIX 25 A STANDARDIZED DIRECT CONTACT CHALLENGE METHOD FOR FMDV IN SWINE

J. M. PACHECO, M. TUCKER, E. HARTWIG, L. RODRIGUEZ ........................................................... 169

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

APPENDIX 26 CAN IN VITRO METHODS RELIABLY REPLACE EXISTING VACCINE CHALLENGE TESTS?

R. REEVE, P. BARNETT AND D. HAYDON ............................................................................... 171 APPENDIX 27 CORRELATING VACCINE INDUCED PROTECTION WITH HUMORAL AND CELLULAR IMMUNE RESPONSES TO FMDV: POTENTIAL IN VITRO ASSAYS FOR REPLACEMENT OF POTENCY TEST

Y. OH, P. HAMBLIN, B. STATHAM, B. CHARLESTON, D. J. PATON, J.H. PARK, Y.S. JOO, S. PARIDA .......... 179 APPENDIX 28 DEVELOPMENT OF A FOOT-AND-MOUTH DISEASE VACCINE POTENCY TEST WITHOUT CONDUCTING ANIMAL CHALLENGE EXPERIMENT

M. ALKAN, S. GURCAN, M. F. SARAC, Y. GULTEKIN, A. ARSLAN, E. UZUNLU, S. AKYUZ AND G. AYNAGOZ .. 181 APPENDIX 29 KEYNOTE BIOSECURITY: WHAT’S IN IT FOR ME?

N. HONHOLD, P. ANKERS, A. MCLEOD................................................................................. 187 APPENDIX 30 FOOT-AND-MOUTH DISEASE CONTROL IN ENDEMIC SETTINGS: COMBINING EPIDEMIOLOGY, RISK ASSESSMENT AND VALUE CHAINS ANALYSIS TO IDENTIFY RISK CONTROL POINTS

J.O PINTO, G. FERRARI; N. TAYLOR, JN RUSHTON, K. SUMPTION AND J. LUBROTH .............................. 189 APPENDIX 31 BRIDGING THE DIVIDE BETWEEN TWO BONA FIDE GUARDIANS OF SOCIETY’S INTERESTS: THE DISEASE EXTERMINATORS AND THE SUSTAINABLE GROWTH ADVOCATES

B. PERRY, K. RICH ....................................................................................................... 195 APPENDIX 32 KEYNOTE: GLOBAL SURVEILLANCE FOR FMD

– WHAT ARE WE DOING AND WHAT COULD BE DONE

D. J. PATON, N. P. FERRIS, Y. LI, D. P. KING, N. J. KNOWLES, J.M. HAMMOND AND J. BASHIRUDDIN ...... 197 APPENDIX 33 KEYNOTE: GLOBAL SURVEILLANCE OF FOOT-AND-MOUTH DISEASE: CONCEPTS, CONSIDERATIONS, AND REQUIREMENTS

A.PEREZ

AND

M. THURMOND ........................................................................................... 199

APPENDIX 34 SAMPLING, SHIPPING, ORGANISATION AND TESTING OF INACTIVATED SAMPLES TO TAKE ADVANTAGE OF MOLECULAR ANALYSIS METHODS

S. ALEXANDERSEN, J. KLEIN, T. FREDERIKSEN AND M. HUSSAIN. .................................................. 203 APPENDIX 35 USE OF INFRARED THERMOGRAPHY TO DETECT SIGNS OF FOOT–AND-MOUTH DISEASE IN WILD AND DOMESTIC UNGULATES

M. R. DUNBAR, S. R. JOHNSON, J. C. RHYAN, M. MCCOLLUM .................................................... 209 APPENDIX 36 SEQUENCE IDENTIFICATION AND GENETIC PROFILE OF FMDV IN A

2007 DISEASE OUTBREAK IN ISRAEL USING FULL LENGTH

GENOMIC ANALYSIS

L. XU, J. BIEKER, J. ROWLAND, W. HURTLE, C. CARRILLO, H. YADIN, A. PEREZ, T. BECKHAM, M. MCINTOSH S. METWALLY ........................................................................................................ 211

AND

APPENDIX 37 IDENTIFICATION OF VIRULENCE DETERMINANTS IN FMDV: POTENTIAL USE OF FUNCTIONAL GENOMICS TO PREDICTING VIRAL PATHOTYPES

J. ARZT, M.E. PICCONE, J. PACHECO, M. BORCA, E. RIEDER, J. ZHU, L. RODRIGUEZ .......................... 217 APPENDIX 38

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

THE DISTRIBUTION OF NSP POSITIVE ANIMALS IN REGULAR VACCINATED HERDS AS CRITERION FOR EFFECTIVENESS OF VACCINE AND VACCINATION DURING ENDEMIC YEAR

H. YADIN ...................................................................................................................219 APPENDIX 39 ESTIMATING THE VACCINATION COVERAGE IN AN FMD FREE ZONE-BRAZIL

2005/2006

H.V. BARBOSA JR., G.M. DE MORAES, A. MENDES DA SILVA, J. L. NARANJO ..................................... 223 APPENDIX 40 MANAGEMENT OF SAT-TYPE FMD IN SOUTHERN AFRICA: PRESENT CONTROL AND TRADE STANDARDS ARE INAPPROPRIATE

FRED BROWN DEBATE: G R THOMSON .................................................................................. 225 APPENDIX 41 KEYNOTE DIAGNOSTICS: A NEW ERA IN ACCESS TO HIGH PERFORMANCE DIAGNOSTICS

D.P. KING .................................................................................................................229 APPENDIX 42 FMD AND SVD COMBINED PROFICIENCY TEST STUDIES

2008 – VIROLOGICAL ASSAYS

N. P FERRIS, G. WILSDEN, D. P KING, G. H HUTCHINGS, S. M REID, K. EBERT, Y. LI AND D. J PATON ..... 231 APPENDIX 43 FMD AND SVD COMBINED PROFICIENCY TEST SCHEME STUDIES

2008 (PHASE XXI) – SEROLOGY

Y. LI, P. KEEL, G. WILSDEN, P. HAMBLIN, N. FERRIS, K. SWABEY, B. STATHAM, J. HAMMOND AND D. PATON ..............................................................................................................................237 APPENDIX 44 KEYNOTE: DRIVING UP GLOBAL STANDARDS FOR FMD DIAGNOSTIC: A KEY ROLE FOR PROFICIENCY TESTING AND INTERNATIONAL ORGANISATIONS

K. DE CLERCQ, K. LUYTEN, D. PATON, D. KING, K. SUMPTION AND N. GORIS ................................... 245 APPENDIX 45 AN ADVANCED FIELD DEPLOYABLE “PEN SIDE” SAMPLE PREPARATION AND PCR SYSTEM

DOUG GREEN, C. VOLPE, JOHN CZAJKA, JASON BETLEY AND JAY LEWINGTON...................................... 249 APPENDIX 46 VALIDATION OF REAL-TIME RT-PCR: MATRIX EFFECT, UNCERTAINTY OF MEASUREMENT AND PRECISION

N. GORIS, F. VANDENBUSSCHE, J. VILLERS, C. HERR, Y. VAN DER STED

AND

K. DE CLERCQ .................. 251

APPENDIX 47 DEVELOPMENT OF SOLID PHASE COMPETITIVE ELISAS BASED ON MONOCLONAL ANTIBODIES FOR THE SEROLOGY OF FMDV SEROTYPES SAT1 AND SAT2

S. GRAZIOLI, E. BROCCHI, V. TRANQUILLO, S. PARIDA AND D. PATON ............................................ 259 APPENDIX 48 FULL GENOME SEQUENCING TO SUPPORT TRACING OF UK OUTBREAKS OF FMD

N. J. KNOWLES, E. M. COTTAM, J. WADSWORTH, K. EBERT, D. P. KING AND D. J. PATON ..................... 269 APPENDIX 49 SERO-SURVEILLANCE AGAINST FOOT-AND-MOUTH DISEASE VIRUS (FMDV) NON-STRUCTURAL PROTEIN ANTIBODIES IN SHEEP, GOATS AND CATTLE IN JORDAN AFTER

2006 OUTBREAK

S. AMAREEN, P. GRAINGER, L. FLEMING, M. MAHAPATRA, H. KHALIL, I. BANI YOUNIS, A. TAHAINEH, D. PATON, F. ALDOMY AND S. PARIDA ............................................................................................... 271 APPENDIX 50 ANTICIPATORY CONTROL MEASURES: GEOGRAPHICAL INFORMATION SYSTEMS-BASED IDENTIFICATION OF TOPOGRAPHIC FACTORS ACTING AS OBSTACLES OR DISSEMINATORS IN THE

2001 URUGUAYAN FMD EPIDEMICS

A. L. RIVAS, A. L. HOOGESTEYN, S. J. SCHWAGER

AND

K.L. ANDERSON ......................................... 275

VII

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

APPENDIX 51 NETWORK ANALYSIS OF LIVESTOCK MOVEMENTS TO ESTIMATE POTENTIAL SILENT SPREAD OF FOOT-AND-MOUTH DISEASE

C. DUBÉ, C. RIBBLE , D. KELTON, B. MCNAB, S. JAVIER AND A. RIVERA ......................................... 279 APPENDIX 52 MODELLING OF FMD OUTBREAKS IN THE NETHERLANDS: VACCINATION AND REGAINING THE STATUS ‘FREEDOM OF INFECTION’

J. A. BACKER, T. J. HAGENAARS, G. A. NODELIJK AND H.J.W. VAN ROERMUND ................................. 289 APPENDIX 53 DEFINING THE PERIOD OF INFECTIOUSNESS IN CATTLE NATURALLY INFECTED WITH FOOT-AND-MOUTH DISEASE VIRUS

B. M. BANKOWSKI, R. HOWEY, N. JULEFF, D. GIBSON, S. J. COX, P. V. BARNETT, M. E.J. WOOLHOUSE AND B. CHARLESTON .............................................................................................................. 293 APPENDIX 54 UK

2001 FOOT-AND-MOUTH DISEASE EPIDEMIC: SEQUENCE DATA AND POSSIBLE AIRBORNE SPREAD

G. A. KÖNIG, E. M. COTTAM, S. UPADHYAYA, J. GLOSTER, L. M. MANSLEY, D. T. HAYDON AND D. P. KING ............................................................................................................................. 301 APPENDIX 55 LOCALISATION OF FOOT-AND-MOUTH DISEASE VIRUS AFTER ACUTE INFECTION IN CATTLE; A NOVEL, IMMUNOLOGICALLY SIGNIFICANT SITE.

N. JULEFF, M. WINDSOR, E. REID, J. SEAGO, Z. ZHANG, P. MONAGHAN, I. W. MORRISON AND B. CHARLESTON ............................................................................................................................. 305 APPENDIX 56 INVESTIGATIONS INTO FINDINGS OF FMD SEROPOSITIVE SHEEP AND GOATS IN CYPRUS, AN FMD-FREE COUNTRY

D. J PATON, N. FERRIS, G. HUTCHINGS, Y. LI, K. SWABEY, P. KEEL, P. HAMBLIN, D.P. KING, S. REID, K. EBERT, S. SAVVA, K.GEORGIOU AND C. KAKOYIANNIS............................................................... 311 APPENDIX 57 RESULTS OF MOLECULAR TESTING OF SAMPLES FROM PAKISTAN

S. ALEXANDERSEN, J. KLEIN AND M. HUSSAIN ........................................................................ 313 APPENDIX 58 TRACING

2007-2008 EMERGENCY EPISODES OF FOOT-AND-MOUTH DISEASE VIRUS IN SOUTH AMERICA: PHYLOGENETIC

ANALYSIS

V. MALIRAT, I. E. BERGMANN, R. DE MENDONÇA CAMPOS, E. NEITZERT, M. VILLAMIL, J. L. QUIROGA CIVERA, F. CONDE AND G. SALGADO JIJÓN ......................................................................................... 329 APPENDIX 59 SERO-PREVALENCE OF FOOT-AND-MOUTH DISEASE IN SMALL RUMINANTS UNDER CONTRASTING HUSBANDRY PRACTICES IN UGANDA

S. N. BALINDA, C. MASEMBE, K. TJØRNEHØJ, A. SANGULA, F. MWIINE, C. AYEBAZIBWE, R. ADEMUN, S. ALEXANDERSEN, H. SIEGISMUND, V. MUWANIKA ..................................................................... 335 APPENDIX 60 THE STATUS OF FOOT-AND-MOUTH DISEASE (FMD) IN ETHIOPIA

G. AYELET, E. GELAYE, J. GUITIAN, M. SAHLE, N. J. KNOWLES AND M. MAHAPATRA ............................ 341 APPENDIX 61 FACTORS INFLUENCING GLOBAL FMD REPORTING AND RISK

R. GARABED, W. JOHNSON, A. PEREZ AND M. THURMOND .......................................................... 347 APPENDIX 62 PARTICIPATORY EPIDEMIOLOGY AS COMPARED TO CONVENTIONAL FOOT AND MOUTH DISEASE SURVEILLANCE TOOL

T. RUFAEL, A. CATLEY, A. BOGALE, M. SAHLE AND Y. SHIFERAW .................................................. 353 APPENDIX 63

VIII

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

DEVELOPMENT OF AN IMPROVED CAPABILITY IN SUPPORT OF NATIONAL BIO-SECURITY FOR THE SURVEILLANCE AND CONTROL OF FOOT-AND-MOUTH DISEASE IN CATTLE AND PIGS IN VIET NAM

C. MORRISSY, N.T. LONG, L. WRIGHT, D.M. HOA, P.P. VU, N.T. PHONG, L.H. VU, I. PRITCHARD, S. RIDDELL, D. SCHAFER, D. EAGLES, W. HA, W. GOFF, J. HAMMOND, S. JUZVA, M. JOHNSON AND P. DANIELS......... 361 APPENDIX 64 DETECTION OF FMDV SEROTYPES O, A AND ASIA

1 BY REAL-TIME RT-PCR

S. M REID, N. J KNOWLES, M. H. N.SHIRAZI, D. P KING........................................................... 363 APPENDIX 65 LINEAR-AFTER-THE-EXPONENTIAL (LATE) PCR: NEW DETECTION TECHNOLOGIES FOR PAN-FMDV AND SEROTYPE-SPECIFIC ASSAYS IN THE FIELD

K.E. PIERCE, R. MISTRY, S. BHARYA, S.M. REID, K. EBERT, D.P. KING, L.J. WANGH ......................... 369 APPENDIX 66 EVALUATION OF A LATERAL FLOW DEVICE FOR THE PEN-SIDE DIAGNOSIS OF FOOT-AND-MOUTH DISEASE

N. P FERRIS, A. NORDENGRAHN, G. H HUTCHINGS, S. M REID, D. P KING, K. EBERT, D. J PATON, T. KRISTERSSON, E. BROCCHI, S. GRAZIOLI AND M. MERZA............................................................ 371 APPENDIX 67 DETECTION OF PERSISTENTLY FOOT-AND-MOUTH DISEASE INFECTED CATTLE BY SALIVARY IGA TEST

J. K BISWAL, D. PATON, G. TAYLOR AND S. PARIDA ................................................................. 377 APPENDIX 68 DEVELOPMENT AND EVALUATION OF IGM ELISA FOR THE DETECTION OF FMDV SPECIFIC IGM ANTIBODIES IN BOVINE AND OVINE SERA

U. WAHEED, D. GIBSON, D. J PATON, Q. M KHAN2 AND S. PARIDA................................................ 383 APPENDIX 69 ENHANCED PROSPECTS FOR FMDV ANTI-VIRALS TARGETED TO THE

3C PROTEASE

S. CURRY, P. ZUNSZAIN, T. SWEENEY, N. ROQUÉ-ROSELL, S. KNOX, A. JAULENT AND R. LEATHERBARROW .. 387 APPENDIX 70 AN ANTIVIRAL AGENT, T-1105 PREVENTS FROM VIRUS EXCRETION FROM PIGS INFECTED WITH PORCINOPHILIC FOOT-AND-MOUTH DISEASE VIRUS

S. OHASHI, K. SAKAMOTO, K. FUKAI , K. MORIOKA, R. YAMAZOE, K. TAKAHASHI AND Y.FURUTA ............. 393 APPENDIX 71 ELEMENTS OF A GLOBAL STRATEGY

J. DOMENECH, JUAN LUBROTH AND K. SUMPTION, IN CONSULTATION WITH OIE ................................... 399 POSTER SESSION........................................................................................................................... 409 APPENDIX 72 A FOETAL GOAT TONGUE CELL LINE FOUND HIGHLY SENSITIVE FOR FOOT-AND-MOUTH DISEASE VIRUS

K. BREHM, M. LENK, R. RIEBE AND B. HAAS .......................................................................... 409 APPENDIX 73 EPIZOOTIOLOGICAL STUDY OF FOOT-AND-MOUTH DISEASE IN THE SUDAN

M. HABIELA, M. A. GAFFAR, Y. A. RAOUF, Y. H. ALI ................................................................ 411 APPENDIX 74 STATUS OF FOOT-AND-MOUTH DISEASE IN PAKISTAN

S.M. JAMAL, S. AHMED, M. HUSSAIN AND Q. ALI .................................................................... 421 APPENDIX 75 THE RELATION ANTIBODY AND PROTECTION AFTER FOOT-AND-MOUTH DISEASE VACCINATION CANNOT BE STANDARDISED

A. DEKKER, N. GORIS, S.M. JAMAL, Y. LI ............................................................................. 425

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

APPENDIX 76 STUDY OF IMMNUNOBIOLOGIC CHARACTERISTICS

A. TIMINA, S.R KREMENCHUGSKAYA., V.V. MIKHALISHIN, V.I. DIEV, V.V. BORISOV, V.M. ZAKHAROV. .... 429 APPENDIX 77 PIGLETS WITH MATERNALLY DERIVED ANTIBODIES CAN BE VACCINATED AT

2 WEEKS OF AGE

G. CHÉNARD, P. SELMAN, P. EBLÉ, N. STOCKHOFE AND A. DEKKER ............................................... 431 APPENDIX 78 IN VIVO AND IN VITRO TESTS TO DETECT NON-STRUCTURAL PROTEINS IN FOOT AND MOUTH DISEASE VACCINES.

M. TROTTA, D. COMPAIRED, O. ZABAL, M. PÉREZ-FILGUEIRA, J. LA TORRE AND N. FONDEVILA ............... 435 APPENDIX 79 TARGETING FMDV MINIGENES TO SLA II POSITIVE CELLS ENHANCES THE INDUCTION OF CELLULAR RESPONSES IN SWINE AND CONFERS PROTECTION AGAINST VIRAL CHALLENGE

B. BORREGO, J. M. ARGILAGUET, E. PÉREZ-MARTÍN, A. EZQUERRA, M. PÉREZ-FILGUEIRA, J. M. ESCRIBANO, F. SOBRINO AND F. RODRÍGUEZ ............................................................................................ 441 APPENDIX 80 CONTROL OF FOOT AND MOUTH DISEASE UNDER PUBLIC-PRIVATE PARTNERSHIP (PPP)

S.N. SINGH ............................................................................................................... 451 CLOSED SESSION............................................................................................................................459 APPENDIX 81 LIST OF PARTICIPANTS........................................................................................................................459

APPENDIX 82 PROVISIONAL AGENDA

- STANDING TECHNICAL COMMITTEE OF THE EUFMD COMMISSION...........................................461

APPENDIX 83 ITEM

2: MINIMUM CONTAINMENT STANDARDS FOR FMD LABORATORIES

B. HAAS ................................................................................................................... 463 APPENDIX 84 ITEM2.1: MINIMUM STANDARDS OF BIORISK MANAGEMENT FOR LABORATORIES UNDERTAKING DIAGNOSTIC INVESTIGATIONS OF LOW-RISK SAMPLES DURING AN OUTBREAK OF FMD

B. HAAS ................................................................................................................... 479 APPENDIX 85 ITEM

3 PROPOSED MINIMUM REQUIREMENTS FOR ADOPTION AT THE 38TH EUFMD GENERAL SESSION (2009) AS A MINIMUM FOR

MEMBER STATES...............................................................................................................................483

APPENDIX 86 OPTIONS FOR DECENTRALISED DIAGNOSIS OF SECONDARY CASES OF FOOT-AND-MOUTH DISEASE IN ANY FUTURE OUTBREAK.

D. SAMMIN, N. FERRIS, D. KING, E. RYAN, S. ZIENTARA, B. HAAS, H. YADIN, AND D. PATON................ 487 APPENDIX 87 TRANSPORT OF FMDV RNA RATHER THAN LIVE VIRUS; OPTIMISATION OF VIRAL RECOVERY BY TRANSFECTION OF INFECTIOUS RNA

G. BELSHAM ............................................................................................................... 495 APPENDIX 88 ITEM

6. SERO-SURVEILLANCE IN TURKEY: THE QUESTION OF HARMONISING THE PERFORMANCE/INTERPRETATION OF SP ANTIBODY

DATA WITH/BETWEEN RG MEMBER LABORATORIES

N. BULUT AND C. POTZSCH .............................................................................................. 497

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

REPORT ON DISCUSSIONS HELD IN THE OPEN SESSION - 14-17 October 2008 An Open Session of the EuFMD Research Group of the Standing Technical Committee was held in Erice, Italy, between 14th and 17th October 2008, with 11 members of the Committee and over 170 observers from across the world. Eight Technical Items were considered, relating to technical constraints to improved regional and global control of FMD. The Session considered seven keynote papers and sixty-six presentations, relating to the eight items. Two debates were held, and forty two posters presented. A panel discussion, with representatives of OIE, European Commission (EC), of the President of the EuFMD Commission, and the Global FMD Research Alliance (GFRA), followed the final presentation on development of a global strategy for progressive control of FMD. The Agenda of the Session is found at [Appendix 1]; the list of participants is found in [Appendix 2]. OPENING CEREMONY –OPEN SESSION The Session was opened by Dr. Gaetana Ferri, Deputy Chief Veterinary Officer, Ministero della Salute, (Ministry of Health), Government of Italy. Her full address is given in [Appendix 3]. She emphasised that foot-and-mouth disease is still one of the diseases that calls for the greatest efforts in prevention and control of the national and international health agencies, because of its extreme contagiousness and the direct damage it causes to animal production and the severe trading restrictions connected to outbreaks. She drew attention to the international institutions that have made collaboration at all levels a pillar of their action, with indisputable results, and gave a special thanks to the EuFMD Commission, for organizing this important multidisciplinary congress, as well as for its continual activity in supporting and enhancing prevention in its 35 European member states, in close collaboration with the European Commission (DG SANCO). She drew attention to the importance of the OIE in promoting a more transparent exchange of information on the occurrence of this disease in countries all around the world and, in the framework of the SPS agreements, for attaining the full harmonization of the health measures and diagnostic standards. In regard to international actions, she drew attention to the importance of European support for projects in developing countries, including those funded by Italy in central Asia and elsewhere, and the significance of the solid commitment of the European Union in funding research. Of central importance to FMD control is the collection of epidemiological data and the funding of surveillance plans should be the priority, especially in those countries where there is still scarce knowledge concerning the presence of the disease and the kind of serotypes circulating. Epidemiological surveillance is a valid instrument to assess the risks entailed by the disease, and therefore provide for an optimization of resources and enhance the effectiveness of vaccination plans. Finally, she stressed the need to reinforce veterinary services, for they play a vital role in all the aspects related to the implementation of the control measures for FMD. She expressed the hope that the Session would offer a relevant contribution to solving various problems that are faced in different parts of the world, and would enable improve international strategies for effective control. Dr Christianne Bruschke, on behalf of the President of the EuFMD Commission, Dr Peter de Leeuw, thanked Dr Ferri for opening the Session and emphasised the importance of scientific research and technical developments to make possible changes in policy and control measures against FMD. The global nature of the FMD threat makes the subject of control of infection in endemic regions of importance to free countries in Europe. On behalf of the OIE, Professor Willeberg, Secretary General of the OIE Scientific Commission for Animal Diseases, indicated that the OIE viewed the Session as a valuable preparatory meeting ahead of the international conference on FMD to be held in Paraguay on 24-26th June, 2009. The outcome and recommendations should assist in developing documents and positions for that meeting, and he thanked the organisers for their efforts to ensure the major technical issues were addressed. Dr. Alf Fuessel, for DG-SANCO of the European Commission, supported the previous comments and emphasised the importance of scientific decisions reached in the Research Group Sessions to the European Community. The EC strongly supported the work of the EuFMD Commission to reduce the risk of FMD to the European member states, and wished the Session a great success.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

RECOMMENDATIONS OF THE OPEN SESSION Session 1: Global FMD control through regional co-ordinated actions: opportunities and constraints The keynote paper presented by Keith Sumption [Appendix 4] and seven papers [Appendix 5, 6, 7, 8, 9, 10 and 11] were discussed, relating to regional FMD control, eradication of type C, and co-ordination of research. Considering that: 1. FMD is a “one world” problem, requiring preventive actions both in countries that are endemic and at-risk; 2. There is an enormous variation in risk and capacity to control disease within and between geographical regions; 3. FMD serotypes and antigenic strains show a regional specificity and distribution; 4. Many countries have no coherent long term strategy for FMD control, and that national efforts are at risk if the neighbouring countries or regions do not take effective action in parallel; 5. Expertise on FMD is scattered across the continents and there is a need for sharing of ideas, technical information and expertise between FMD experts and to build effective regional networks to improve technical information available to disease control agencies; 6. FMD Type C incidence has markedly declined over the past 20 years to a point that circulation in the wild may have ceased. Recommends that: 1. The approach of developing regional roadmaps in each of the seven major virus pools for the control of FMD is adopted as a central part of the global strategy for progressive FMD risk reduction by FAO and OIE, and should involve a greater effort to foster effective laboratory and epidemiology networks within each region; 2. Regional roadmaps should utilise, to the greatest extent, public-private partnerships to overcome financial, social and technical barriers to delivery of FMD control programs in endemic and at risk regions; 3. Effort is continued by the EuFMD Commission and FAO (EMPRES) after the current Session to ensure that FMD expertise in each region is able to benefit from the global and regional discussions at the Session, and they should support or develop a global network of FMD expertise that fosters greater participation in future in the regional and global efforts against FMD; 4. A pathway towards verified international freedom from circulating Type C infection is developed by FAO and OIE, that should include the cessation of vaccination against Type C, surveillance efforts in the last known reservoirs, and emergency planning for possible escape or re-appearance of infection. Session 2A: Optimizing programs in a situation of limited resources The keynote paper presented by Aldo Dekker [Appendix 12] and three papers [Appendix 13, 14 and 15] were discussed relating to timing of vaccination and dose requirements. Considering that: 1. Vaccine coverage is the major constraint in control of FMD by vaccination mainly due to insufficient vaccination, and partly due to decreased quality of the vaccine at time of vaccination or poorly timed vaccination programmes. A vaccination coverage lower than 100% will hardly be sufficient, because current vaccines are not 100% effective (3 PD50 vaccines only protect 75-85% of the cattle); 2. Global FMD vaccine production is constrained by a lack of demand in parts of the most affected regions in Africa and Asia, and significant public and private investment in vaccine production would be needed to increase supply using current production methods, over many years; 3. Financial resources to fund vaccination are limited in most countries, and optimization of programs e.g. the protection of young animals before they enter the animal movement chain, could have positive financial and epidemiological impacts; 4. Intra-dermal application of FMD vaccine is very promising and could be a simple way to expand the number of doses available given current antigen production methods; 5. Good quality vaccines can give a long-lasting protection from clinical disease for at least 6 months, but good vaccines need to be applied correctly in the field. Special attention

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

should be given to the cold chain as that can affect the antigen stability, and good quality vaccines can induce an antibody response in presence of maternally derived antibodies, but it is important to revaccinate in time to increase the vaccination coverage in young animals. Recommends that: 1. The FAO/OIE Global Strategy should place emphasis on vaccination policy guidance and technical support to Veterinary Services (VS) to assist them to rationalize and optimize programs to achieve the greatest reduction in transmission and/or reduction in disease impact, through targeting vaccine to critical populations and ages (being mainly young animals). The support should be channelled through the existing regional specialised agencies and animal health centres of FAO/OIE, and make effective use of the technical expertise in the EuFMD Research Group and associated technical network; 2. Countries buying vaccine should check the vaccine, for both the antibody inducing capacity at the time of delivery and also half-way through the shelf-life; 3. Countries using prophylactic vaccination should on a regular basis test for the optimal moment to immunise young animals, because this could be influenced by the quality of the vaccine used in both dams and offspring, and carry out field studies to determine the duration of immunity following a single application of an FMD vaccine, when using the selected vaccine in their own population; 4. New vaccines should be developed that induce a higher protection level in the population and that are less sensitive to problems in the cold-chain; 5. Further studies should be carried out to validate the finding that intradermal inoculation could enable reduction in the dose and cost of vaccination programs.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 2b: Vaccination: overcoming antigenic diversity in order to simplify preventive programs The keynote paper presented by Bernd Haas [Appendix.16], and two papers [Appendix. 17 and 18], were discussed relating to vaccine matching methods and prediction of antigenic sites from cross-reactivity data. The papers and discussion can be summarized as follows: Considering that: 1. Harmonization of vaccination programs, in terms of vaccine performance and antigenic spectrum, has provided positive benefits to regional FMD control in South America and in Europe; 2. There is a lack of guidance to decision makers on vaccine suitability in much of subSaharan Africa and parts of Eurasia; 3. High payload vaccines against type A FMDV have been found to confer a significant protective effect even against FMDV where the r-value suggests a poor level of protection. Recommends that: 1. In order to promote and develop regional control efforts, the OIE/FAO FMD Refece laboratory network are encouraged to produce an annual guidance paper on vaccine strain selection for each region/virus pool, and on the availability and properties of vaccine strains worldwide; 2. To assist the above, efforts to identify the prevalence and continually monitor the characteristics of FMD strains in the seven regional “virus pools” should be internationally supported, and the international agencies, working with the Ref Centres, co-ordinate efforts to determine whether established vaccine strains offer protection against new field strains; 3. Members of the OIE/FAO FMD Ref lab network as well national reference laboratories (NRLs) and laboratories of vaccine producers should exchange protocols, data and sera in order to standardize methods for r-value determination; 4. Further research should be funded on the correlation of heterologous protection with parameters that can be determined in-vitro, in particular serum titres, and on epitopes relevant for protection (e.g. cross-reactivity, structural and sequence data).

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 2C: Novel vaccine delivery approaches: progress updates Three papers [Appendix. 20, 21 and 22], were presented and discussed in this section, on trails of a Baculovirus expressed empty capsid FMD vaccine, on use of modified vaccinia virus, and on marker vaccine using an FMDV with a VP1 G-H loop deletion. Considering that: 1. Enhancing the duration of immunity after vaccination will have a profound effect on the practical implementation of control programmes; 2. Improvements in vaccine efficacy will significantly increase stockholder’s confidence in vaccination programmes; 3. Research into vaccine stabilization, targeting of antigen to antigen presenting cells and providing the appropriate “danger signal” in vaccines is likely to improve vaccine efficacy; Recommends that: 1. Research and Development on new vaccines is supported long term as a central pillar of the global strategy against FMD; 2. Research into In vitro expression of empty capsids to substitute for conventional vaccine antigen production should be pursued. However, studies to address whether a ‘single sequence’ capsid provides equivalent protection to a quasi species of capsid (conventional vaccines) should be explored; 3. Research into the use of GH Loop negative viruses as vaccine antigen shows promise. The applicability for other serotypes should be explored.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 2D: FMD vaccine standards required for global control The Keynote paper of Nesya Goris [Appendix 24] and four presented papers [Appendix. 25, 26, 27 and 28], were discussed and can be summarized: Considering that: 1. There is a need for buyers and users of FMD vaccines to have confidence in the claims of the producer and to have sufficient information on duration of protection, stability, and cross-protection to enable rational design of programs; 2. There is a need for buyers to be able to check independently the performance of vaccine without requiring a full vaccine challenge under high containment conditions. Recommends that: 1. FMD laboratories that have performed quality assessments should share the data on vaccine quality assessment to improve analyses through the use of larger datasets; 2. Laboratories performing challenge tests should standardize their methodologies for assessment of immune responses, as well as the analytical techniques, on a statistical sound basis; 3. Based on statistical evaluations the OIE and European Pharmacopoeia should choose a measure of vaccine quality which matches previous potency tests and the epidemiological needs; 4. OIE and FAO should support the monitoring and communication of vaccine quality based on serological responses in target population using standardized laboratory tests or international accredited laboratories.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 3: Biosecurity and buy-in The keynote paper presented by Nick Honhold [Appendix 29], the two presented papers [Appendix 30 and 31], were discussed and can be summarized: Concludes that: 1. Risk based disease surveillance using analysis of livestock production and marketing chains and livestock production systems must be a guiding element in decision-making and control interventions; 2. Governments have a responsibility to: a. b. c. d.

3. 4. 5.

Craft FMD control strategies and plans based on objective assessment of the risks in different epidemiological, economic and social settings and within different production and marketing chains; Facilitate and sustainably finance processes that will require key stakeholders to listen to each other and take each other’s concerns into account in planning FMD control, and to look for areas of joint interest and potential gain; Ensure that up to date information is available on all prevention and control measures through a range of media; Work in partnership with animal keepers, veterinary staff, intermediaries and owners of gathering points to develop biosecurity measures that are feasible and agreed for peacetime, raised risk periods and outbreaks;

Researchers and those who finance research have a responsibility to provide more documented evidence on the impacts of applying biosecurity; Farmers, intermediaries and operators of gathering sites have a responsibility to consider the impact of their actions on the livestock sector and to apply good management including preventive biosecurity; Veterinarians have a responsibility to ensure that their knowledge of FMD prevention and control is up to date, to encourage farmers to adopt a range of control measures including preventive biosecurity. Others who communicate with farmers e.g. vaccine providers can also be a valuable source of information.

Recommends that: 1. The global strategy being developed by FAO/OIE should address the issue of how public and private investment in FMD prevention and control can be encouraged, and in particular the promotion and safeguarding of stakeholder investment in prevention and control efforts in endemic regions; 2. In developing the global strategy, the potential impact upon investment and FMD risk of changes to the standards affecting compartmentalisation, commodity based trade, and criteria for gaining or retaining FMD freedom should be examined; 3. FAO/OIE should encourage countries to develop National FMD risk reduction plans, based on realistic assessment, and stakeholder consultation and engagement, of the risk of introduction and circulation of FMD virus and an assessment of the policies, capacity, incentives and opportunities for risk reduction; 4. FAO/OIE should assist in implementing and update risk assessment in FMD endemic settings using critical risk control point (CCP) identification and livestock market chain analysis to assist countries to identify how FMD is maintained, spread and the possible control interventions for reducing or breaking FMD transmission; 5. The research and development communities should work together over the next few years to develop projects and guidance that will enable CCP to be identified in major regions/countries within each FMD virus pool, to support the progressive control of FMD in endemic settings. Session 4: Measure progress in global and regional fmd control, and early warning of fmdv emergence Two keynote papers were presented by David Paton [Appendix.32] and Andres Perez [Appendix 33], and seven papers [Appendix. 34, 35, 36, 37, 38, 39 and 40], relating to improving surveillance for emergent virus threats through use of inactivated samples in shipment to reference laboratories, use of thermal imaging, full genome sequencing for epidemic virus characterisation, and monitoring of vaccination programs through measurement of NSP positive rates and through surveys for population coverage.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Considering that: 1. Very little attention has been placed in international reporting systems on the use of measurements of incidence, force of infection and risk of new FMD infections, which constrains the rational design of preventive measures in at-risk and affected countries; 2. Monitoring of progress against FMD requires a set of comparative indicators, to measure change in risk, incidence and capacity to detect and control FMD epidemics; 3. The submission of live virus samples to reference laboratories is constrained by high transport costs, and carries some risk that may be reduced by use of alternative sample handling arrangements and use of services within the affected region. Recommends that: 1. The FAO/OIE and their reference laboratories should develop a multilateral, multiagency, international strategy and infrastructure to survey FMD globally for FMD capture, sample collection, reporting, modelling, and development of indicators of FMD prevalence and risk at global and regional scales, and in selected countries and regions specifically targeted; 2. International and regional activities should be developed that promote team-working between disease control agencies and reference laboratories, and the application of stateof-the-art epidemiological methods, and field, theoretical/informatics-based and molecular epidemiology techniques; 3. Greatly increased support is required for strategic centres and epidemiological projects to better understand and sample ecosystems in each of the virus pools, where local capacity/incentives are insufficient. This should address the requirement for better and more complete field epidemiological information submitted together with samples for analysis by national and international reference laboratories. The existing Network of Reference Laboratories should be sustained and twinning projects should be established to support the development of laboratories and associated surveillance projects in the Middle East, West Africa and East Africa. New opportunities for simplified sample submission should be explored; 4. FAO/OIE are encouraged to produce, at least twice a year, a report on the Global FMD risk situation incorporating the OIE/FAO laboratory network report and an analysis of key epidemiological indicators that provide insight into factors contributing to increasing and to decreasing risk. Key epidemiological indicators should include, for example, quantitative estimates of risk, vaccine-induced immunity, attack and transmission rates, reproductive ratio, spatial distribution of population at risk, and patterns and frequency of animal movements and trade; 5. The global strategy being developed by FAO/OIE is encouraged to promote greater monitoring of the evolution and efficacy of control programs, using indicators such as vaccine coverage, incidence of FMDV exposure, and estimates of the reproductive ratio in countries or regions that are not free of infection.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 5: Diagnostics: making quality service available where needed Two keynote papers were presented, by Dr Donald King [Appendix 41], on development and potential application of new diagnostic test systems relevant to endemic and at risk areas, and Kris de Clercq [Appendix 44], on raising the performance standards of national and regional reference laboratories through their involvement in a global system of proficiency testing. A further seven papers [Appendix. 42, 43, 45, 46, 47, 48 and 49], were presented and discussed, and a debate was held on the priorities for new diagnostic kits for use in endemic regions. Considering that: 1. Progress has been made to develop and validate rapid and simple penside test systems for FMD; 2. Virus type information is usually vital to planning of immediate follow-up control actions in endemic as well as normally free countries; 3. On behalf of FAO, and supported by EuFMD, the WRL at Pirbright distributes panels of samples as part of a proficiency testing service (PTS), but participation of non-European laboratories is limited; 4. Currently most national reference laboratories in endemic regions currently do not participate in proficiency testing schemes, and therefore confidence is lacking in their services at national and regional level. Concludes that: 1. Assay validation should exploit statistical methods in order to define uncertainties of measurement. A clear strategy for dealing with samples that generate weak values is required. Recommends that: 1. As part of a co-ordinated global effort, the international agencies promote the continued development and application of the new rapid and simple FMDV test systems to assist every endemic and at risk country to establish a sufficient level of diagnostic capacity for the early confirmation of FMD; 2. International Reference standards (IRS) for assay validation and calibration should be developed and made available to NRLs. These materials are required for all FMDV serotypes – including SATs; 3. Support for the organization of proficiency testing schemes (PTSs) for FMDV should be continued: resources required to undertake this work should be prioritized; 4. Guidelines and optimized protocols for effectively transporting inactivated FMDV between the field and NRLs (and WRL) should be developed and the potential of replacing live virus transport in international surveillance programs further developed by the international agencies.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 6: Components and capacity for effective control Three papers [Appendix. 61, 62 and 63], were presented, followed by a wide discussion on level of capacity required to implement effective control programs in endemic regions. Considering that: 1. There is a lack of reliable surveillance data from many countries; 2. Participatory epidemiology has shown an ability to provide detailed indications of the perceived importance, occurrence and epidemiology of the disease at farmer level; 3. The provision of improved laboratory facilities is important but highlights weaknesses in control programs because of weak capacity of field services and political will; 4. It is important to ensure that improvements in capacity are sustainable. Concludes that: 1. There is a need to ensure that required capacity is available at all levels of the key actors in control such as farmers, private veterinary services, government veterinary services, diagnostic laboratories and research institutes; 2. The reasons for low levels of disease surveillance and reporting varies between different countries; 3. Financing of vaccination schemes should be considered to examine if farmers can pay for all or some of the vaccination to release resources for surveillance; 4. The EuFMD meeting organizers are to be congratulated on fostering the broader scope of the meeting to address capacity and biosecurity issues. Recommends that: 1. The global FMD control strategy being developed by FAO and OIE should address the issue of capacity for surveillance in every FMD affected country and promote actions at all levels of the key actors in control such as farmers, private veterinary services, government veterinary services, diagnostic laboratories and research institutes. There must be a focus on ensuring the sustainability of these capacities and on capacities for disease surveillance and reporting as well as post-vaccination sero-surveillance; 2. The global strategy should also address the financing of vaccination schemes and examine if farmers can pay for all or some of the vaccination to release resources for surveillance. Public-private sector co-operation should be encouraged to contribute towards these; 3. At regional and national level, studies of the costs and benefits of different control strategies (including the impact of FMD at farmer level) should be undertaken to indicate if refined control strategies may be as or more effective than broad control strategies such as blanket vaccination; 4. The EuFMD Commission should expand subject areas such as biosecurity, capacity building and disease surveillance and reporting systems in the EuFMD research group Sessions and efforts should be made to increase attendance by representatives of veterinary field services.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 7: Prospects for integrating anti-viral approaches The two papers [Appendix. 69 and 70], and discussion can be summarised as follows: Considering that: 1. Promising studies have provided information on the structure of the 3C protease of FMDV, which is required for the generation of infectious virus output, and new tools to easily screen potential inhibitors; 2. Studies indicate that anti-viral compounds reduced the level of clinical disease and virus excretion in pigs challenged with the porcinophilic O TAW/97 but administered the previously reported potential anti-viral compound T-1105 only 1 hour before challenge followed by administration of the same compound twice a day for 7 days. Concludes that: 1. Very promising specific approaches and tools are available for anti-viral studies targeting the important 3C protease of FMDV; 2. Certain compounds already show significant promise for potential use in a control strategy including administration of anti-virals. Recommends that: 1. Further studies should be funded at a level enabling focussed screening of potential antiviral compounds directed against the 3C protease of FMDV; 2. Additional studies are needed to look at already established potential anti-viral compounds showing promise in reducing disease and in particular excretion and transmission and such studies should be aimed producing evidence supporting the efficacy and safety, including the potential for generation of escape mutants, during simulated field conditions.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Session 8: Strategy for regional and global control The keynote paper of FAO, presented by Dr Joseph Domenech [Appendix n. 71], was discussed by a panel comprising Dr. Alf Fuessel, DG-SANCO (EC), Professor Preben Willeberg (OIE Scientific Committee on Animal Diseases), Dr. Christianne Bruschke (Deputy CVO, the Netherlands) and Dr Cyril Gay (Global FMD Research Alliance). The paper and discussion can be summarized as follows: Considering that: 1. FAO and OIE are in the preparation phase of a global initiative for the progressive control of FMD, to be developed and presented at an International Conference in Paraguay in June 2009; 2. There is a need for buy-in at national Government level to FMD control and to progressive and co-ordinated control at regional level, and for advocacy and convincing arguments of the benefits of FMD control at all levels; 3. There is a need to fund international surveillance for FMD in endemic regions, and Research and Development of new tools, if change is to occur in the capacity at national level to control FMD. Concludes that: 1. The current Open Session addressed some of the most major issues affecting long term FMD control in endemic regions, and the recommendations of the Session should therefore be taken up by FAO and OIE in preparing the Global FMD Control Strategy. Recommends that: 1. The regional approach based on the concept of actions within the 7 major virus type ecosystems/"pools" is promoted in the FAO-OIE International Initiative for the progressive control of FMD, with co-ordination and progress monitoring involving a global platform and a Secretariat; 2. The FAO and OIE continue to develop the regional and global strategy papers, based on a progressive risk reduction approach, making full use of expertise available to the EuFMD Commission, the OIE/FAO Reference Lab network, and in the global FMD research and development community; 3. A series of regional workshops is organized by FAO and OIE to develop the long term vision and regional FMD control strategies (road maps). The EuFMD Commission should support these to the extent possible, contributing to the efforts of FAO EMPRES and in close collaboration with Regional Organizations; 4. The FAO and OIE should address, in the global strategy, how increased investment and effort in FMD control by stakeholders in the most affected and at-risk regions can be encouraged and sustained, and this element should be a major component of a future Conference on controlling FMD in the worst affected regions; 5. Socio-economic issues affecting investment and uptake of FMD control measures by stakeholders in endemic and at-risk countries be addressed in the strategy; 6. FAO with OIE produce an annual global FMD report to indicate the progress made in the seven major virus pools, that will bring together virological, epidemiological and programmatic progress in each region; 7. The next Open Session of the EuFMD in 2010 is utilised as a major technical forum that will assist technical progress to the global FAO OIE initiative.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

REPORT OF THE CLOSED SESSION OF THE STANDING TECHNICAL COMMITTEE OF THE EUFMD COMMISSION Friday 17th October: Location-Palazzo Sales, “Board Room”-1st Floor. A Closed Session of the Research Group was held on Friday 17th October, 2008, at Erice, Italy. The Session was chaired by Dr Aldo Dekker, with Keith Sumption as Secretary. Nine members of the group (Aldo Dekker (AD), Naci Bulut (NB), Georgi Georgiev (GG), Bernd Haas (BH), Hagai Yadin (HY), Kris de Clercq (KdC), Emiliana Brocchi (EB), David Paton (DP), Soren Alexandersen (SA) were present. Apologies were received from Mark Bronsvoort, Stefan Zientara and Donal Sammin; the work of the latter for the group was represented by Eoin Ryan (ER). One member (Hakan Vigre) had resigned from the group as his professional position had changed. Additional participants for technical items were Dr Jef Hammond (JH), Head of the WRL Pirbright, Dr Graham Belsham (GB) Lindholm, Dr Gavin Thomson (GT), SADC FMD project. Observers included Christianne Bruschke (representing the Chairman, EuFMD), and Dr Alf Fuessel (AF; DG-SANCO, EC), and Susanne Munstermann (SM), FAO Gaborone, Botswana. The list of participants is found in [Appendix. 81] Item 1. Agenda of the Session [Appendix 82] Two additional items were added to the Agenda: 1) EB requested the issue of technical guidance on testing of animal products coming from an affected European country for the presence of FMDV be discussed. The issue was clarified by AF, who agreed that some guidance to Chief Veterinary Officers (CVOs) was needed. The unanimous opinion of the group was that it is not feasible to rule out the presence of FMDV in consignments of meat through sampling and lab testing. Such an answer would require a very high level of sampling and sensitivity of diagnostic procedures. DP drew attention to a previous opinion (of the SCAHW, 2003, on FMD diagnostic tests) which supported the above. Conclusion: 1. The testing of consignments of meat for presence of FMDV cannot be used to rule out the presence of levels of FMDV that could pose a risk of infection if fed to pigs. 2) Concept Notes presented to the 76th Executive Committee of the EuFMD, June 2008 Three concept notes had been presented to the Executive, and each were recommended as important studies; the EC had indicated their financial support in principle. One proposal (on the role of sheep in transmission of FMDV, and relating to the question of benefit of vaccinating sheep to prevent transmission to cattle) arose from the Closed Session of 2007. The other two arose from the Final Workshop in March 2008, of the EC funded FMD surveillance support program in Turkey operated by EuFMD (2007-3/2008). These were: a study on the use of full length sequencing, to address issues of type A epidemiology (focus on the new antigenic variant) but also to better identify if Full length sequencing can be used to predict the number of unreported outbreaks; and a study on the stability of FMDV 146s in oil adjuvanted FMD vaccines. Implementation of each study had not commenced, but letters of agreement had been developed with IAH Pirbright and Lelystad, respectively. AD answered questions on the stability study; the intention was in addition to the in vitro results to compare with results with longitudinal serological studies in Turkey. Dr Bulut explained the initial response of the SAP Institute to the proposed study, and was confident that within a month their support would be given to providing the vaccine batches required. AF made clear that the support of the official veterinary service (OVS) for such a study would be helpful and that FAO should also liase with them if difficulty to gain the required study materials, and use of the results, remained an issue. Item 2. FMD Laboratory Minimum Containment Standards 2.1 Minimum Containment Standards (MCS) for FMD Laboratories [Appendix 83] The Secretary indicated that the RG was formally asked to give their technical clearance on the final version of the revised Minimum Containment Standards (MCS) for FMD Laboratories. This document had been prepared by a working group which had invested over 300 hours of time in

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

revision and review; the document had been circulated to all FMD laboratories in the member states that were listed as working with live FMDV, and comments reviewed by the Chairman of the working group, Bernd Haas. The final stage was therefore the clearance of the document after the above consultation. BH indicated that he had taken each comment from consultation, and considered only two areas of concern, the first being to clarify the risk categories in the risk table, and the issue of inclusion of infectious RNA as a risk on page 3. It was agreed that the latter should be mentioned in the hazard identification section. Conclusion: 1. The document, with final editorial changes agreed from the Session, was technically cleared by the Group; 2. The final Standard (Appendix would be provided within a week of the Closed Session, for the Report and for utilization by interested parties including DG-SANCO, with the expectation that the document would be formally approved by the General Session of the EuFMD in April 2009). 2.2 Minimum standards of biorisk management for laboratories undertaking diagnostic investigations of low-risk samples during an outbreak of FMD [Appendix 84] BH presented a second document, relating to emergency FMDV labs for serology and for testing inactivated samples. This document was proposed to the group to supersede the previous two biocontainment guidelines relating to FMDV sero-diagnostic facilities and confirmation of FMDV through RNA or antigenic detection procedures not involving non-live virus manipulation. Conclusion: 1. The Group supported the consolidation of the previous Guidelines into a single document, with updating to bring into line with the new Minimum Containment Standards for FMDV laboratories; 2. The Group agreed that the document should be referenced in the Minimum Containment Standards for FMDV laboratories, but could also be added to the FMD diagnostic manual being prepared by the CRL for use by European NRLs; 3. The biorisk assessment and management sections of the guideline should be harmonised with the MCS document, and should include a statement on avoiding RNA in contact with cell culture. Action: Secretariat to arrange consultation on the revised Guidelines, with the NRLs following the RG Session. Item 3. Minimum Diagnostic Capacity in EuFMD Member States for the laboratory confirmation of FMD [Appendix 85] KS introduced this Item; the issue was raised because of the low participation in the annual proficiency panel exercises organized for EuFMD and EC by the WRL/CRL of the NRLs of EuFMD member states that are outside of the EU, for example in the western Balkan countries. Following the 75th Executive Committee, the Secretariat produced a paper on Minimum Diagnostic Capacity that has the aim of harmonizing non-EU countries towards the expected diagnostic capacity stipulated for the EU member states (Council Directive 2003/85/EC). The RG were asked to review the paper, which would affect mainly non-EU countries if adopted by the EuFMD General Session in April 2009. Conclusion: 1. The Minimum Diagnostic capacity paper should assist to clarify expectations for countries outside of the EU in respect of FMD diagnostic services and was technically cleared by the group. Action: 1. Feedback on the paper is expected from Dr Yadin and Dr Bulut, representing countries that are EuFMD members but not in the EU; 2. The Secretariat should organise a consultation with NRLs in the non-EU European countries before the 77th Executive Committee.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Item 4: Sampling instructions for field veterinarians to collect samples with the primary biocontainment at the point of sampling for use in RT-PCR. Item 5: Position paper on the options of decentralized testing. The RG were requested to give clearance to this position paper [Appendix 86] whose development had been led by Donal Sammin. The paper was briefly presented by Eoin Ryan. The Chairman thanked the authors for their excellent work, and opened the item for discussion. It was agreed that: • The paper should continue to retain the focus on free countries but could mention issues or options for decentralized testing relevant to non-free countries; • Company or trade names must be avoided; • Use of swabs must be added; • If possible, a summary table to give decision makers the options and issues relevant to different situations in which DCT is applicable should be made; • The need for a system to deliver sufficient and correct materials for DCT to the field operatives should be mentioned; • The example decision tree is relevant and should be retained as an example, and emphasis given to each MS adapting a decision tree for its own circumstances. Action: finalization of the paper within one month (Dr Eoin Ryan and Dr Donal Sammin). Item 5.1 Transport of FMDV RNA rather than live virus; Optimization of viral recovery by transfection of infectious RNA KS presented the background to this item; the 76th Executive Committee had recommended that studies be conducted to reduce or replace live virus shipments while maintaining the possibility of recovery of live virus from RNA. Graham Belsham (Lindholm) outlined a proposed study for the optimization of methods to recover live virus from RNA [Appendix 87]. The variables to be studied included electroporation conditions, cell type (BHK, BTY), and time of harvesting. The proposed study would be conducted within the next 6 months and at a cost of circa 20,000 US$ (laboratory costs only, the personnel would be provided through a visiting scientist funded through an FAO project). In discussion, it was agreed that the safety issue also needed to be studied, since viral RNA retained infectivity, although of a quite different level of risk to live virus. Conclusion: 1. The group fully supported the proposed study, and requested a report to the next Session; 2. bio-risk studies should be conducted, at least in mice, to better identify the level of infectivity of FMDV RNA Action: 1. Secretariat to follow up with EC following the approval in principle given at the Session to fund the study, and the Letter of Agreement with Lindholm.

Item 6. Sero-surveillance in Turkey: the question of harmonizing the performance/interpretation of SP antibody data with/between RG member laboratories [Appendix 88] Presenters: Naci Bulut, Carsten Potzsch (PC) NB summarised the situation: to end of September 2008, Turkey had reported 30 type O, and 130 type A outbreaks; a relatively big upsurge in type A compared to 2007. As a result of vaccine matching, the vaccine to be applied in 2009 should contain or cover the A TUR 06 antigenic type. Sero-surveillance in Thrace and Anatolia; this is being implemented from September 08, after assistance in design from EuFMD RG members; the NSP and SP serological testing is not expected to be finalised until after March 2009. For sero-prevalence of FMD exposure, some 33,900 samples will be collected, using a 2 stage sampling, 565 villages selected, prevalence of 2% infected villages and 10% intra-unit prevalence. Sixty cattle per village will be sampled aged between 4-24 months.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Villages (20,000) with less than 100 cattle were excluded from the survey (in total 166,000 cattle in total, average 8 per village). For vaccine efficacy, 3 provinces (and 3 villages in each) within each of 7 regions were selected, out of those selected for sampling for NSP sero-prevalence, in total the sera of 60 cattle in each of 63 villages will be tested, in total 9780 sera. In Thrace region, 9780 samples between 4 and 24 months of age, from 152 villages will be sampled. The issues for discussion 1) SP surveys: validity of thresholds for protection in ELISA tests conducted by the SAP Institute NB explained the basis for the threshold currently applied by the SAP Institute to interpret test results. Action: AD include data of the Turkish laboratory in the international comparative study and to review the current threshold applied. 2) Design of surveys in 2009: CP proposed that future surveys are examined and commented upon by the RG, before implementation in Turkey. Dr A. Fuessel drew attention to the fact that as EuFMD are invited to the National FMD task force meetings, then this can be used to introduce the proposed surveillance design, but the end result has to be agreed by the National FMD project SC, involving the CVO, and must meet the requirement in the project document for minimum number of samples. Action: 1. EuFMD Secretariat, in liason with the CVO of Turkey and NB, should propose a meeting of RG experts next May or June for analysis of the results of the 2008 survey, and to design the 2009 survey, to occur ahead of the national TF and SC meetings. 3) Inter-laboratory proficiency testing scheme: For NRLs in countries where the EuFMD is implementing capacity building programs under the support of the EC TF (GEO/ARM/AZB/TUR/IRAN). CP proposed that a more frequent ILPT is implemented for this set of countries to expedite raising of the performance, given the importance placed on sero-monitoring in the programs in each country and the very limited experience in the Caucasus and Iran in FMD sero-diagnosis. It was agreed that each laboratory should participate in future in the WRL/CRL annual scheme, but the current requirements for these labs are different from the majority of FMD free countries and specific panels could assist with harmonisation of SP serology. Action: 1. WRL to select from available sera for a pool relevant to west Eurasia, and would further identify sera that could be of similar use in the other virus pools (e.g. African pools, south and east Asia). Item 7. Vaccine selection and potency of antigens in the EuVB for use against the current SAT2 FMDV circulating in Botswana/Namibia The issues raised by the ongoing SAT2 epidemic in Botswana/Namibia were reviewed by Kris de Clercq; David Paton reviewed the vaccine matching results for African recent African SAT2 viruses. Major issues include: • Potency of the SAT2 vaccine formulated from antigens in the EU vaccine bank against the SAT2 epidemic type or other relevant SAT2 viruses; • The lack of a comprehensive set of cross-matching and heterologous potency studies on SAT2 vaccines against the range of genotypes of SATs that exist in Africa. Discussion points: • • •

The EU vaccine bank contains both SAT2 Zimbabwe and SAT2 Eritrea antigens; SAT2 Eritrea has been used in vaccine matching by WRL following potency tests, not the SAT2 Zimbabwe; On which body or how to influence producers to develop new /adapted SAT vaccines;

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

•

On a heterologous potency test to determine if the EuVB antigen would give at least least 3 PD50 in challenge.

Conclusions: 1. 2.

A heterologous challenge test was recommended by the Chairman, with a representative challenge virus from the ongoing epidemic of SAT2; A subgroup of the RG should follow keep in close contact to follow up the discussions and identify studies or actions required (members: JH (WRL), DP, KdC, AD, HY, KS)

Item 8. Remaining business, workplans and tasks 1.

Guidelines on the Clinical examination of cattle and sheep.

Action: 1. Guidelines need finalization with comments of BH and SA. 2. Membership of the Group; SA would remain a member of the group, until the next EuFMD General Session (April 2009). 3. A replacement for Hakan Vigre should be identified that could bring epidemiology experience; possible replacements to be approached for interest include Jean-Francois Valarcher (Sweden) and Nick Honhold (UK); 4. Workplan: AD would revise the workplan and send for comments within 4 weeks, after liason with the Vice Chairman and Secretariat on priorities; 5. Next meeting: the date and location of the 2009 Session were not fixed but would be expected to occur in September/October. ITEM 9: LIST OF ACTIONS

Agreement with CVO Turkey and SAP Institute to supply vaccine batches for stability analysis at Lelystad;

Minimum Standards for emergency labs;

Minimum Diagnostic Capacity;

Position paper on decentralized testing;

Study: optimization of live virus recovery from RNA;

SP survey - thresholds values for protection;

Workshop to design 2009 survey;

Identify panel of standard sera for use in SP serology for EuFMD beneficiary countries in West EurAsia;

Guidelines on clinical examination –cattle and sheep;

10. Workplan.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

CHAIRMAN’S AGENDA LIST - FROM 2007 SESSION: Priority list for Research Group HIGHEST PRIORITY 1. 2. 3.

Position paper on the options of decentralized testing (DS 1 December 2007); Sampling instructions for field veterinarians to collect samples with the primary biocontainment at the point of sampling for use in RT-PCR; Minimum Diagnostic capacity in EuFMD Member States for the laboratory confirmation of FMD: a. Requirement for ISO17025; b. Every year a proficiency test.

MEDIUM PRIORITY 1.

Definition and reporting of epidemiologically significant events: guidance paper, based on criteria developed under the EuFMD/EC/IVO program in Iran [Action point: delivery 12/07]; 2. Paper on the risk of spread from wild boar in the Israeli circumstances; 3. Report on the use of thermal imaging camera for use in extensively kept cattle for the purpose of detection of febrile animals (HY, to be reviewed by DP); 4. Complete the review of duration of immunity after type O vaccination, to include the antibody decline by VNT, and the data should be analyzed with linear mixed effect model. (by 12/07); 5. Draft contract or form of agreement whereby a network of FMD labs in Europe could agree to provide services to NRLs that are temporarily incapacitated; 6. Revised position paper, as an addition or addendum to the document on live virus facilities, to include the requirements for applying non-live FMD techniques in serology and for RT-PCR (BH 31 December 2008); 7. Finalize paper on standardization of information collection and information output (Carsten Potzsch); 8. Discuss the protocol for RNA stabilisation study in Pakistan before referring it to the Exec committee; 9. In vitro stability study for FMD vaccines; 10. Role of sheep in maintaining or spreading FD in a temperate European climate with mixed sheep and cattle farms (UK, 2001); 11. Application of tools for high-resolution FMDV molecular epidemiology in Western EurAsia.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 1 AGENDA OF THE OPEN SESSION

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 1b POSTER SESSION

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 2 LIST OF PARTICIPANTS

Research Group members are listed together. Remaining participants are listed under “Others” in alphabetical order (by surname) and by “Country” at the end Open Session of the Research Group of the Standing Technical Committee Erice, Italy 14 - 17 October 2008

RESEARCH GROUP

Dr Aldo DEKKER (Chairman) Senior Scientist Laboratory Vesicular Diseases Central Institute for Animal Disease Control PO Box 2004, Lelystad 8203 AA, The Netherlands Tel/Fax: +31-320-238603 / +31-320-238668 e-mail: Aldo.Dekker@wur.nl Dr Kris DE CLERCQ (Vice Chairman) Head Department of Virology Section Epizootic Diseases CODA-CERVA-VAR Groeselenberg 99 B-1180 Ukkel, Belgium Tel/Fax: +32-2-3790400 / +32-2-3790666 e-mail: kris.de.clercq@var.fgov.be Dr Søren ALEXANDERSEN Research Professor Danish Institute for Food and Veterinary Research Department of Virology, Lindholm DK-4771, Kalvehave, Denmark Tel/Fax: +45-72-347833 / +45-72-347883 (direct) or 347901 e-mail: AlexandersenS@inspection.gc.ca Dr Emiliana BROCCHI Head National Reference Laboratory for Vesicular Diseases Istituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia Romagna Via A. Bianchi, 7/9 25124 Brescia, Italy Tel/Fax: +39-30-2290310 / +39-30-2290369 e-mail: ebrocchi@bs.izs.it Dr Naci BULUT FMD Expert Head of the Diagnosis Department

FMD Institute Ankara, Turkey

Tel: +90 312 2873600 / Fax: +90 312 2873606 Mobile: +90 533 3571484 nacib@sap.gov.tr

Dr Georgi Kirilov GEORGIEV Head of Exotic and Emerging Diseases National Diagnostic and Research Veterinary Medical Institute 1606 Sofia, Bulgaria Tel/Fax: +359-2-8341004 e-mail: georgivet2@yahoo.com Dr Bernd HAAS Head of National FMD Reference Laboratory Friedrich-Loeffler-Institut Federal Research Institute for Animal Health Boddenblick 5 a 17493 Greifswald, Insel Riems, Germany Tel/Fax: +49-(0)3835170 / +49(0)383517151 e-mail: bernd.haas@fli.bund.de Dr Stephan ZIENTARA AFSSA – Lerpaz - BP 67 94703 Maisons-Alfort Cedex, France Tel/Fax: +33-1-49-771333 / +33-1-43689762 s.zientara@afssa.fr Dr Hagai YADIN Head of Virology Division and FMD Laboratory Kimron Veterinary Institute c/o Ministry of Agriculture PO Box 12 Beit-Dagan 50250, Israel Tel/Fax: +972-3-968-1619 / +972-3-9681788 mob: +972-50-6241382 e-mail: hagaiy@moag.gov.il FAO WRL Dr David PATON (ex-officio) Pirbright Laboratory Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-231012 / +44-1483232621

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

e-mail: david.paton@bbsrc.ac.uk

e-mail: paul.barnett@bbsrc.ac.uk

OTHERS

Dr John BASHIRUDDIN IAH Pirbright Laboratory Ash Road, Pirbright Surrey GU24 ONF, UK Tel/fax: 44-1483-231127 / 44-1483-232448 email: john.bashiruddin@bbsrc.ac.uk

A Dr. Ronello ABILA Regional Coordinator OIE SE Asia, Regionaql Coordination Unit, 09/1, PhayaThai Rd, Bangkok, Thailand. Email: r.abila@oie.int Dr Gelagay AYELET National Veterinary Institute Debre Zeit, Orumia 19, Ethiopia Tel/Fax: +251-114-338411 / +251-114339300 e-mail: gelagayayelet@yahoo.com Dr Gulhan AYNAGOZ Deputy Director SAP institute PO Box 714 Ankara, Turkey Tel: +90 312 2873600 e-mail: gulhana@sap.gov.tr B Dr Labib BAKKALI-KASSIMI Agence française de sécurité sanitaire des aliments, UMR 1161. AFSSA/INRA/ENVA, 23 Avenue du Général de Gaulle, 964703 Maisons-Alfort, France e-mail: lbakkali@vet-alfort.fr Dr Sheila Nina BALINDA Makerere University, Institute of Environment and Natural Resources, Molecular Biology Laboratory, P.O. Box 7298, Kampala, Uganda e-mail:sbalinda@muienr.mak.ac.ug Dr Helio BARBOSA Jr Esplanada dos Ministérios, Bloco D, Anexo A, 3° Andar, Departamento de Saúde Animal Ministério da Agricoltura Brasilia D F 70043-900, Brazil Tel/Fax: +55-61-3218-2236 / +55-61-32244180 e-mail: helio.vilela@agricultura.gov.br Dr Simon J BARTELING Nieuwe Keizersgracht 438 Amsterdam 1018, VG The Netherlands Tel: +31-206207688 e-mail: simon.barteling@orange.fr Dr Paul BARNETT Institute for Animal Health Ash Road, Pirbright Surrey GU24 ONF, UK Tel/fax: 44-1483-232441 / 44-1483-232448

Dr Yasser BASYOUNI Dr Graham BELSHAM National Veterinary Institute Technical University of Denmark Lindholm 4771 Kalvehave, Denmark Tel +45 72347985 e-mail: grbe@vet.dtu.dk Dr Ingrid BERGMAN Head of laboratory Pan American FMD Center PAHO/WHO Tel: +55 21 3559056 e-mail: Ingrid.bergmann@hotmail.com Dr Jitendra Kumal BISWAL Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey, GU240NF, UK e-mail:jitendra.biswal@bbsrc.ac.uk Dr Ester BLANCO Centro de investigación en Sanidad Animal, INIA; Valdeolmos, 28130 Madrid, Spain e-mail:blanco@inia.es Dr Belinda BLIGNAUT Onderstepoort Veterinary Institute, Transboundary Animal Diseases Programme, Private Bag X05, Onderstepoort, 0110, South Africa e-mail: bohmerB@arc.agric.za Dr Katrina BREHM Friedrich-Loeffler-Institute, Südufer 10, 17493 Greifswald-Insel Riems, Germany e-mail: katharina.brehm@fli.bund.de Dr Belen BORREGO CISA-INIA Valedeolmos 28130 Madrid, Spain Tel: +34 91 6202300 e-mail: Borrego@inia.es Dr Anette BOTNER Department of Virology, Danish Institute for Food & Veterinary Research Lindholm, Kalvehave 4771, Denmark Tel/Fax: +45-72-34-78-58 / +45-72-34-7901 e-mail: bot@dfvf.dk

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Dr Christianne BRUSCHKE Deputy Chief Veterinary Officer Ministry of Agriculture, Nature and Food Quality, PO Box 20401, 2500 EK The Hague, The Netherlands Tel/Fax: + 31 70 3784683 / + 31 70 3786134 Cellphone +31 6 21689854 e-mail:c.bruschke@minlnv.nl Dr Marco BUGNETTI Istituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia Romagna, Brescia, Italy Email: marco.bugnetti@bs.izs.it C Dr. Alejandra CAPOZZO ICT Milstein - cevan - Conicet Saladillo 2468, Buenos Aires, Argentina Tel/fax: +5491168631789 Email: acapozzocevan@centromilstein.org.ar Dr Veronica CARR Institute for Animal Health Compton Nr Newbury Berkshire, RG20 7NN, UK e-mail: Veronica.Carr@bbsrc.ac.uk Dr Javier CASTILLO-OLIVARES Institute for Animal Health, Pirbright Laboratories, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, UK e-mail:javier.castillo-olivares@bbsrc.ac.uk Dr. Byung-Sik CHANG Department of Research and Development Laboratory, Jenobiotech Inc. Chuncheon, Gangwon, Korea e-mail: bschang@jenobiotech.com Dr Bryan CHARLESTON Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: bryan.charleston@bbsrc.ac.uk Mr Gilles CHENARD Animal Sciences Group, Products Division PO Box 65, Lelystad 8200 AB, The Netherlands Tel/Fax: +31-320-238006 / +31-320-238237 e-mail: gilles.chenard@wur.nl Dr. Grant CLARKE Veterinary Adviser Veterinary Exotic Notifiable Diseases Unit Department of Environment, Food and Rural Affairs Area 5 D, Nobel House, 17 Smith Square, London, SW1P 3JR, UK Tel/ Fax 0207 238 5489, 0207 238 5051 e-mail: Grant.Clarke@defra.gsi.gov.uk

Dr Sarah COX Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: sarah.cox@bbsrc.ac.uk Dr. Carolina CUBILLOS Centro de investigación en Sanidad Animal, INIA; Valdeolmos, 28130 Madrid, Spain Tel/Fax:+34-94-6202300 e-mail: Cubillos@inia.es Dr Stephen CURRY

PhD, Professor of Structural Biology Biophysics Section, Blackett Laboratory, Imperial College, London SW7 2AZ, UK Tel/ Fax: +44-(0)20-7594-7632/+44(0)7986-201707/+44-(0)20-7589-0191 e-mail: s.curry@imperial.ac.uk D Dr Mihail Claudiu DIACONU 63, Dr Staicovici Street, Sect. V Bucharest 050 557, Romania Tel/Fax: +40-21-410-0945 / +40-21-4113397 e-mail: diaconu.claudiu@idah.ro Dr Vyacheslav DIEV OIE Regional Reference laboratory for FMD, Federal Centre for Animal Health, Vladimir, Russia Dr Timothy DOEL Merial Animal Health, Ash Road Pirbright, Woking, Surrey GU24 0NQ, UK Tel/Fax: +44-01483-238111 / +44-01483238102 e-mail: tim.doel@merial.com Dr. Caroline DUBÉ MV MS PhD Candidate office/bureau, Epidemiologist, Animal Health Division, Disease Control Section, Canadian Food Inspection Agency, 59 Camelot, Ottawa, ON, K1A 0Y9, Government of Canada Tel/Fax: (450) 424-0549 / cell (514) 2610236 / (613) 228-6144 e-mail: dubecm@inspection.gc.ca Dr Philippe DUBOURGET 29 Avenue Tony Garnier, Lyon 69007, France Tel/Fax: +33-472-723000 / +33-472-723181 e-mail: philippe.dubourget@merial.com Dr M.R. DUNBAR Michael R. Dunbar, MS, DVM Project Leader for Rabies and Bovine Tuberculosis Research,

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

USDA/APHIS/WS/National Wildlife Research Center 4101 LaPorte Avenue Fort Collins, CO 80521, USA Tel/Fax: (970) 266-6360/ (970) 266-6138 Email: Mike.R.Dunbar@aphis.usda.gov Dr Rahana Mohan DWARKA Dr Rahana M Dwarka (PhD), Senior Researcher, ARC-OVI- Transboundary Animal Diseases Programme Private Bag X05 Onderstepoort, 0110, UK Tel/Fax: +27 12 529 9529 (office)/529 9589 (lab)/+27 12 529 9543/Cell: 083 555 2747 Email: dwarkar@arc.agric.za E Ms. Katja EBERT Institute for Animal Health, Ash Road, Pirbright, Surrey, GU24 0NF, UK Ms Phaedra EBLE Central Institute for Animal Disease Control (CIDC-Lelystad), PO Box 2004, Lelystad 8203 AA, The Netherlands Tel/Fax: +31-320-238680 / +31-320-238668 e-mail: phaedra.eble@wur.nl Prof. Marianne ELVANDER National Veterinary Institute – SVA Uppsala SE – 75189, Sweden Tel/Fax: +46-18674000 / +46-18674445 e-mail: marianne.elvander@sva.se F Dr Nigel FERRIS Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: nigel.ferris@bbsrc.ac.uk G Dr Rebecca GARABED Department of veterinary Preventive medicine, Ohaio State University, 1900 Coffey road, Columbus, 0H43210, USA e-mail: rbgarabed@ucdavis.edu Dr. Ciryl GAY DVM, PhD,National Program Leader, Animal Health, National Program Staff, Animal Production and Protection USDA -Agricultural Research Service, 5601 Sunnyside Avenue, Beltsville, Maryland, 20705-5148, USA Tel: (301) 504-4786 Email cyril.gay@ars.usda.gov Dr Dorothy GEALE 59 Camelot Drive Ottawa, Ontario KIA O49, Canada

Tel: +1-613-221-4234 e-mail: dgeale@inspection.gc.ca Dr Kyriacos GEORGIOU Veterinary Services, 1417 Nicosia, Republic of Cyprus e-mail: Kgeorgiou@vs.moa.gov.cy Dr Deborah GIBSON BSc, Institute for Animal Health Ash Road, Pirbright Woking, Surrey GU24 0NF, UK Tel: 01483 231153 E-mail: debi.gibson@bbsrc.ac.uk Mr John GLOSTER Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-231023 / +44-1483232448 e-mail: john.gloster@bbsrc.ac.uk Dr. Danny GOOVAERTS Ms Nesya GORIS Groeselenberg 99, Ukkel 1180, Belgium Tel/Fax: +32-23790514 / +32-23790666 e-mail: negor@var.fgov.be Dr Santina GRAZIOLI Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna Via Bianchi 9, Brescia 25124, Italy Tel/Fax: +39-030-229-0370 / +39-030-2290369 e-mail: sgrazioli@bs.izs.it Dr. Yasemin GULTEKIN Sap Institute, P.O. Box 714 Ulus, Ankara, Turkey Tel/fax: +90312287/3600 e-mail: ygultekin2@gmail.com

Dr Jianhong GUO Xujiaping No. 1, Yanchangpu Lanzhou Gansu 730046, China Tel/Fax: +86-931-8342585 / +86-9318342052 e-mail: gregjh@126.com H Dr. Mohammed Ahmed HABIELA Senior Assistant Research Ministry of Science and Technology. Animal Resources Research Laboratories (ARRC), Central Veterinary Research Laboratories (CVRL), Unit of Foot and mouth disease- Soba- Khartoum, Box: 8067 Amarat- Khartoum- Sudan Tel/Fax: 00249 922 846856 e-mail: mhabiela979@hotmail.com Dr Jeffrey HAMMOND

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

CSIRO, AAHL Private Bag 24, Geelong, VIC 3220, Australia Tel/Fax: +61-03-522-75767 / +61-03-52275555 e-mail: jef.hammond@csiro.au

Dr Nicholas JULEFF Centre for Tropical Veterinary Medicine, University of Edinburgh, Easter Bush Veterinary Centre, Roslin, Midlothian EH25 9RG, UK email: nicholas.juleff@bbsrc.ac.uk

Dr Michiel HARMSEN Edelhertweg 15, Lelystad Flevoland 8219 PH, The Netherlands Tel/Fax: +31-32-238436 / +31-32-238961 e-mail: michiel.harmsen@wur.nl

Dr Jijung HE Dr. Divakar HEMADRI Senior Scientist, Project Directorate on FMD IVRI campus, Mukteswar, Nainital-263 138, India, Uttarakhand, Tel/Fax: 0091-5942-286122 e-mail: divakar.hemadri@gmail.com Dr Eileen HERRERA Office of International Research Programs ARS-USDA 5601 Sunnyside Avenue, Beltsville, MD 20705-5141, USA Tel +1 (301) 504-4521 FAX +1-301-504-4528 email: Eileen.Herrera@ars.usda.gov Dr Helen HONDROKOUKI Neapolfos 25, Ag. Paroskevi, Athens 15310, Greece Tel/Fax: +30-2106007016 e-mail: fmdi@otenet.gr Dr Nick HONHOLD AGA, FAO, Viale delle Terme di Caracalla 00100 Rome, Italy e-mail: Nick.Honhold@fao.org

Dr Andrzej KESY Department of Foot & Mouth Disease National Veterinary Research Institute, Wodna Str. 7, Zdunska Wola 98-220, Poland Tel/Fax: +48-43-823-51-34 / +48-43-82352-75 e-mail: andrzejke@piwzp.invar.net.pl Dr Yong Joo KIM Scientific and Technical Department OIE, Rue de Prony, 2, 75015, Parisa, France e-mail: yjokim@oie.int Dr Donald KING Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: donald.king@bbsrc.ac.uk Mr Nick KNOWLES Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: nick.knowles@bbsrc.ac.uk Dr Josef KOEFER Austrian Agency for Health and Food Spargelfeldstrasse 191 1, 1210, Vienna, Austria Tel/Fax: 33 14415863/33 142670987 e-mail: josef.koefer@ages.at

J Dr Liesbeth JACOBS Cedi-Diagnostics, Edelhertweg 15 Lelystad 8219 PH, The Netherlands Tel: +31-622909642 e-mail: liesbeth.jacobs@wur.nl Dr Peter JADUD State Veterinary Institute Pod drahami 918, Zvolen 960 86, Slovak Republic Tel/Fax: +421-45-5320803 / +421-455332486 e-mail: jadud@svuzv.sk Mr Syed M JAMAL National Veterinary laboratory (NVL) Park Road, 45500. Islamabad, Pakistan Tel: +92-302-2409599 e-mail: jamal115@yahoo.com

Dr GUIDO KÖNIG Instituto de Biotecnología, INTA, N. Repetto y de los Reseros, Hurlingham, Buenos Aires, 1686, Argentina Institute for Animal Health, Pirbright, Surrey, GU24 0NF, UK e-mail: gkonig@cnia.inta.gov.ar Dr Vlastimil KRIVDA L Mr Kwang-Nyeong LEE National Veterinary Research & Quarantine Service 480 An-Yang 6 Man-An, An-Yang Gyeong-Gi 430 824, Republic of Korea Tel/Fax: + 82-31-469-1918 / + 82-31-4495882 e-mail: leekwn@nvrqs.go.kr

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Dr Yanmin LI Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: yanmin.li@bbsrc.ac.uk Dr Zhidong LI Research Assistant, Lanzhou Veterinary Research Institute, Chinese Academy of Agriculture Sciences, 1 Xujiaping, Post Code: 730046, Lanzhou , China Tel/Fax: +86-931-8342685/+86-9318340977 E-mail: li_zhiyong56@yahoo.com.cn Dr Wilai LINCHONGSBONGKOCH Dr Eric LEFEVRE Institute for Animal Health, Compton Laboratory, Compton, Newbury, Berkshire, RG20 7NN, UK e-mail: eric.lefevre@bbsrc.ac.uk Dr Yves LEOFORBAN Inspectour Général de la santé publique vétérinaire, President de la commission science technique et société, 251, rue de Vaugirard 75732, Paris cedex 15, France Tel/Fax: +33 149558129/149558169 Email: yves.leoforban@agricolture.gouv.fr Dr Linda LOGAN USDA/APHIS Embassy of the United States 8 Kamal El Din Salah Street Garden City, Cairo, Egypt Tel/Fax: +1-2-797-3011 / +1-2-792-4812 e-mail: Linda.L.Logan@aphis.usda.gov Dr Angelica LOITSCH Austrian Agency for Health and Food Safety (AGES) Institute for Veterinary Disease Control Moedling, Robert Kochgasse 17 A-2340 Moedling, Austria Tel./Fax 43 1 802 12 12 40/+43 664 966 83 26/43 1 802 12 12 11 e-mail: angelika.loitsch@ages.at Dr Zengjun LU Xujiaping No. 1, Yanchangpu Lanzhou Gansu 730046, China Tel/Fax: +86-931-8342706 / +86-9318342052 e-mail: luzengjun920@126.com M Dr Mana MAHAPATRA

Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: mana.mohapatra@bbsrc.ac.uk Dr Francois MAREE (PhD) Exotic Diseases Division, OVI, ARC Private Bag X05, Onderstepoort 0110, South Africa Tel/Fax: +27 12 5299584/94/+27 12 5299595/05 e-mail: maree@arc.agric.za Dr Caterina MARINOU BVMS, MVM, Foot-and-Mouth Disease Institute, Greek Ministry of Rural Development and Food, 25 Neapoleos Str., 15310 Agia Paraskevi, Athens, GREECE Tel/ Fax: +30210 6007016 E-mail: kmarinou@gmail.com Dr. Nora MATTION Vicedirectora Centro de Virología Animal CEVAN – CONICET, Saladillo 2468 C1440FFX Buenos Aires, Argentina Tel/Fax: +54 11 4686 6225 - +54 11 46876735/7886 e-mail: nmattioncevan@centromilstein.org.ar

Prof. Malik MERZA Svanova Biotech AB Uppsala Science Park, Uppsala SE-75183, Sweden Tel/Fax: +46-18-654901 / +46-18-654999 e-mail: malik.merza@svanova.com Dr Samia METWALLY USDA/APHIS PO Box 848, Greenport, New York 11944, USA Tel/Fax: +1-631-323-3063 / +1-631-3233366 e-mail: joan.m.sawicki@aphis.usda.gov Dr Miroslav MOJZIS Dr Mokganedi MOKOPASETSO Dr Kazuki MORIOKA 6-20-1 Josuihoncho, Kodaira Tokyo 187-0022, Japan Tel/Fax: +81-42-321-1441 / +81-42-3255122 e-mail: morioka@affrc.go.jp Dr. Chris MORRISSY Diagnostic Mammalian Virology Australian Animal Health Laboratory PMB 24 Geelong 3213 Australia Tel/Fax: +61 3 5227 5555 email: chris.morrissy@csiro.au

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

e-mail: votarod@hotmail.com Dr Otto MOZZER Odorico P. Santos, 526 Montes Claros, MG 39401810, Brazil Tel/Fax: +55-38-3229-7107 / +55-38-32297089 e-mail: mozzer@vallee.com.br Dr Alcan MUSA SAP institute PO Box 714 Ankara, Turkey Tel: +90 312 2873600 e-mail:musaa@sap.gov.tr N Dr Singanllur Nagendra KUMAR India Immunological Limited, Hiderabad, 500032, India email: Nagu@indimmune.com Dr Than Long NGO Center for Veterinary Diagnostics – Regional Animal Health office No.6 Department of Animal Health of VIET NAM Regional Animal Health Office No.6 124 Pham The Hien st.,Dist.8, Ho Chi Minh city - Viet Nam Tel/Fax: 84 8 8518454/+84 913 894 891/ + 84 8 8569050 email: n.t.long@hcm.vnn.vn Dr Wieslaw NIEDBALSKI Department of Foot & Mouth Disease National Veterinary Research Institute Wodna Str. 7, Zdunska Wola 98-220, Poland Tel/Fax: +48-43-823-51-34 / +48-43-82352-75 e-mail: wieslaw@piwzp.invar.net.pl O Dr Jarlath O’CONNOR Ms Yooni OH Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: Yooni.oh@bbsrc.ac.uk Dr Hiroyouki ONOZATO Exotic disease research station, National institute of animal health, Kodaira, Tokyo, 187-0022, Japan Dr Vahid OTAROD Iran Veterinary Organization 23, Seyed Asad Abadi St. PO Box 14155-6349, Tehran, Iran Tel/Fax: +98-982-188806407 / +98-982188902712

P Dr Dorothee PAEFFGEN Intervet International Wim de Koerverstraat 35, PO Box 31 Boxmeer Brabant 5830 AA, The Netherlands Tel/Fax: +31-485-587219 / +31-485-587491 e-mail: dorothee.paeffgen@intervet.com Dr Satya PARIDA Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: satya.parida@bbsrc.ac.uk Dr Jong-Hyeon PARK national Veterinary research and Quarantine Service (NVRQ), 480 Anyang, 6 Dong, Anyang-City, Korea tel/fax; +82314611719/+82315882 email: park@nvrqs.go.kr Dr Aravindh Babu PARTHIBAN Institute for Animal Health Pirbright laboratory, Institute House- 4 Bridgemead Pirbright, Woking ,Surrey, GU24 0ND,UK Tel/Fax: 01483 231111/01483 232448 email: aravindh.babu@bbsrc.ac.uk Dr Emma PAUL DEFRA, Nobelhouse, 175 Mith Square, London, 51018, UK Email: emma.paul@defra.gsi.gov.uk Dr Guntram PAUL Intervert International GmbH Osterather Str. 1a, Cologne 50739, Germany Tel/Fax: +49-221-1778-233 / +49-2211778-299 e-mail: untram.paul@intervet.com Dr Andres PEREZ 279 Cousteau Place Suite 500, Davis, California 95618, USA Tel/Fax: +1-530-297-4621 / +1-530-2974618 e-mail: amperez@ucdavis.edu Dr Mariano PEREZ FILGUEIRA Researcher INTA-CONICET, Instituto Virologia-CICVyA, INTA, Las Cabañas y Los Reseros s/Castelar-Buenos Aires (1712), Argentina Tel/Fax: +54 11 4621 1676/1127 ext. 159/+54 11 4621 1743 email: mperez@cnia.inta.gov.ar Prof Brian PERRY OBE, BVM&S, DTVM, MSc, DVM&S, FRCVS

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Visiting Professor of Tropical Veterinary Medicine, Nuffield Department of Clinical Medicine, University of Oxford, UK and Department of Veterinary Tropical Diseases, University of Pretoria, South Africa, P.O. Box 437, Gilgil, Kenya 20116 Tel/Fax: +254-20-2136090/+254-734600250/+254-20-4223001 email: b.perry@cgiar.org Dr Kenneth PIERCE Senior Scientist, Department of Biology, MS008, Brandeis University Waltham, MA 02454-9110, USA email: pierce@brandeis.edu Ms Liliyana POLIHRONOVA P. Slaveikov Blvd. No. 15 Sofia 1606, Bulgaria Tel/Fax: +359-2-8341004 e-mail: c/o georgivet@yahoo.com

USDA/ARS Plum Island Animal Disease Center, P.O. Box 848 Greenport NY 11944, USA Tel/Fax: 631 323 3223/631 323 3006 Email: luis.rodriguez@ars.usda.gov Dr Peter ROEDER Dr Ir. Herman ROERMUND Quantitative Veterinary Epidemiology and Risk Analysis (QVERA), Central Veterinary Institute (CVI) of Wageningen UR. P.O.Box 65, 8200 AB Lelystad, The Netherlands. Visiting address: Houtribweg 39, 8221 RA, Lelystad, the Netherlands. Tel/Fax: +31 (0)320 238392/+31 (0) 320 238668. Internet: www.cvi.wur.nl e-mail: herman.vanroermund@wur.nl Dr Michael ROYALS

Dr Helen PRENTICE R Mr Scott REID Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: scott.reid@bbsrc.ac.uk Ms Michelle REMOND AFSSA-LERPAZ 23 Avenue du General De Gaulle BP 67 Maisons Alfort F 94703, France Tel/Fax: +33-1-49-77-13-17 / +33-1-43-6897-62 e-mail: m.remond@afssa.fr Dr Richard REEVE Boyd Orr Centre for Population and Ecosystem Health, Graham Kerr Building, University of Glasgow, Glasgow G12 8QQ, UK e-mail: R.Reeve@bio.gla.ac.uk Dr Aida Elizabeth RIEDER Foreign Animal Disease Research Unit, United States Department of Agriculture, Agricultural Research Service, Plum Island Animal Disease Center, Greenport, NY 11944. USA Mr Ariel RIVAS Department of Public Health & Pathobiology College of Veterinary Medicine, NCSU Raleigh, North Carolina 14850, USA e-mail: alr4@cornell.edu Dr Luis RODRIGUEZ Ph.D. Research Leader, Foreign Animal Disease Research Unit

Dr Tesfaye RUFAEL National Animal Health Diagnostic and Investigation Center, Sebleta, P.O.Box 04, Ethiopia Tel/Fax: +251911 764972 email: rufaelc@yahoo.com Mr Eoin RYAN Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: eoin-d.ryan@bbsrc.ac.uk S Dr Kenichi SAKAMOTO 6-20-1 Josui-honcho Kodaria, Tokyo 167-0022, Japan Tel/Fax: +81-423211441 / +81-423255122 e-mail: skenichi@affrc.go.jp Dr Abraham SANGULA Foot & Mouth Disease Laboratory Embakasi, PO Box 18021, Nairobi 00500, Kenya Tel/Fax: +254-20-553-633 e-mail: abrahamsangu@yahoo.com Dr Victor SARAIVA Dr Christian SCHELP Dr Bommeli AG Stationsstr 12, Liebefeld CH-3097, Switzerland Tel/Fax: +41-319706265 / +41-319706279 e-mail: christian-schelp@idexx.com Dr Kamil SEDLÁK State Veterinary Institute Prague Sidlistni 24/136, Prague 16503, Czech Republic

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Tel/Fax: +420-776-189-049 / +420-220920-655 e-mail: kamil.sedlak@svupraha.cz Dr Moshe SHALEV PO Box 848, Greenport, New York 11944, USA Tel/Fax: +1-631-323-3338 / +1-631-3233295 e-mail: moshe.shalev@dhs.gov Dr Alexey SHERBAKOV e-mail: a.sherbakov@arriah.ru Dr Lidiya SHERBAKOVA Dr Michael SHERIDAN Deputy Chief Veterinary Officer, Department of Agriculture, Fisheries and Food Agriculture House Kildare Street, Dublin 2, Ireland Tel/Fax: +353-1-6072645 / +353-1-6762989 email: michael.sheridan@agriculture.gov.ie Dr. Wim SCHIELEN Business Manager FMD Prionics Lelystad BV Platinastraat 33, NL-8211 AR Lelystad, The Netherlands Tel/Fax +31 320 714 003/+31 320 714 029 e-mail: wim.schielen@prionics.com Prof. Liisa SIHVONEN Evira, Mustialankatu 3, Helsinki 00790, Finland Tel/Fax: +358-50-5539226 / +358-207724363 e-mail: liisa.sihvonen@evira.fi Dr Roland SILBER AGES, IVET-MOE Robert Kochgasse 17, Modling A-2340, Austria Tel/Fax: +43-1-802-121210 / +43-1-802121211 e-mail: roland.silber@ages.at Dr Shree Narayan SINGH Biovet Private Limited 560, "C" Block 1st Main, Aecs Layout Kundalahalli, Bangalore, Karnataka 560037, India Tel/Fax: +91-080-28495915 - 17 / +91-08028495512 e-mail: singhsn@biovet.in Dr Emiliana SMITSAART DVM PhD, Research & Development, Biogénesis Bagó S.A. Ruta Panamericana km 38.5 Garín B1619 IEA Prov. Buenos Aires, Argentina Tel/Fax: 54 (0) 3327 44 8355 / 54 (0) 3327 44 8347, www.biogenesisbago.com

e-mail: eliana.smitsaart@biogenesisbago.com Dr Villuppanor SRINIVASAN e-mail: vas@indimmune.com Dr Arthur SUMMERFIELD T Dr Lazare TANO Merial 13B, Avenue Albert Einstein Villeurbanne 69623, France Tel/Fax: +33-472-72-34-20 / +33-472-7259-90 e-mail: lazare.tano@merial.com Dr Barbara THUER Institute for Virology and Immunoprophilaxy, 3147, Mittelhaeusen, Switzerland mail: barbara.thuer@ivi.admin.ch Dr Gavin THOMSON Technical Specialist: SADC FMD Project Food, Agriculture & Natural Resources Directorate, SADC Secretariat Plot 116, Kgale Mews, Gaborone, Botswana Tel/Fax: +267 391 3357/ +267 74201231S e-mail: gthomson@sadc.int Dr Tore TOLLERSRUD U V Dr Paul VAN AARLE PO Box 31, Boxmeer 5830 AA, The Netherlands Tel/Fax: +31-485-585-228 / +31-485-587418 e-mail: paul.vanaarle@intervet.com Dr Jean Francois VALARCHER Senior Lecturer, Department of Clinical Sciences, Swedish University of Agricultural Sciences, P.O. Box 7054, SE-750 07 Uppsala, Sweden Tel/ Fax: +46 (0) 18 671 391 / +46 (0) 18 673 545 e-mail: jean-francois.valarcher@kv.slu.se Dr VENKATARAMAN e-mail: aniket.sanyal@gmail.com Dr Carmelo VOLPE Smiths Detection, 459 Park Avenue, Bushey Watford Hertfordshire WD23 2BW, UK Tel/Fax: +44-7880-784879 / +44-1923236407 e-mail: carmelo.volpe@smithsdetection.com Dr Vilna VOSLOO e-mail: VoslooW@arc.agric.za

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

W Dr Ryan WATERS e-mail: ryan.waters@bbsrc.ac.uk Dr Preben WILLEBERG CADMS VET:MED, UCDAVIS, Davis, ca 95616 e-mail: pwillerberg@ucdavis.edu. Dr. Tom WILLEMS Dr Miriam WINDSOR CHAPMAN email: Miriam.windsor@bbsrc.ac.uk Mr John WOOD Kritis 35, Athens, Greece Tel: +30-2109913835 Dr Caroline WRIGHT email: caroline.wright@bbsrc.ac.uk X Prof. Jingshan XUE No. 16 Building, 8 Block, No. 188 South 4th Ring West Road, Beijing 100070, China Tel/Fax: +86-10-637-02600 e-mail: xuejs@tom.com

Tel/Fax: +32-2-2950870 / +32-2-2953144 e-mail: alf-eckbert.fuessel@ec.europa.eu EFSA Dr Hubert DELUYKER Department of Scientific Expert Services European Food Safety Authority Largo Natale Palli, 5/A, Parma 43100, Italy Tel/Fax: +39-0521-036 448 / +39-0521036-548 e-mail: hubert.deluyker@efsa.europa.eu OIE World Organization for Animal Health (OIE) 12, rue de Prony 75017 Paris, France Tel/Fax: +33-1-44151888 / +33-1-42670987 e-mail: oie@oie.int Dr Ghazi YEHIA World Organization for Animal Health (OIE) Beirut, Lebanon Tel/Fax: +961-5-430741 / +961 5 430742 e-mail: g.yehia@oieme.org OTHERS EUFMD SECRETARIAT

Z Prof Valery ZAKHAROV FGI ARRIAH Jur'evets, Vladimir 600901, Russia Tel/Fax: +7-4922-261755 e-mail: zaharov@arriah.ru Dr Zhifang ZHANG Professor, Biotechnology Research Institute, ChineseAcademy of Agricultural Sciences, 12 Zhongguancun South Street, Beijing , Post Code: 100081, China Tel/Fax: +86-10-68919854/ +86-1062136981, E-mail: zhifangzhang@yahoo.com Dr Zhidong ZHANG Institute for Animal Health Ash Road, Pirbright, Surrey GU24 0NF, UK Tel/Fax: +44-1483-232441 / +44-1483232448 e-mail: zhidong.zhang@bbsrc.ac.uk EC Dr Alf-Eckbert FÜSSEL Head of Sector DG-SANCO/D1 European Commission Rue Froissart 101 – 3/67 1049 Brussels, Belgium

Dr Keith SUMPTION Secretary, EUFMD Animal Health Service Animal Production and Health Division FAO – Viale delle Terme di Caracalla 00100 Rome, Italy Tel/Fax: +39-065705-5528 / +39-0657055749 e-mail: keith.sumption@fao.org Ms Nadia RUMICH EUFMD Secretariat Animal Health Service Animal Production and Health Division FAO – Viale delle Terme di Caracalla 00100 Rome, Italy Tel/Fax: +39-065705-2637 / +39-0657055749 e-mail: nadia.rumich@fao.org Dr Nick HONHOLD FAO, Turkey e-mail: nick.honhold@fao.org Dr Carsten PÖTZSCH FAO Regional FMD Office 15a Tamarashvili Str. Tbilisi 0177, Georgia Tel: +995-99-626546 e-mail: Carsten.Potzsch@fao.org DELEGATES BY COUNTRY

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

-Ayelet, Gelagay -Rufael, Tesfaye

Argentina -Capozzo, Alejandra -Filgueira, Mariano, Perez -König, Guido -Mattion, Nora -Smitsaart, Eliana

Finland -Sihvonen, Liisa France -Bakkali-Kassimi, Labib -Dubourget, Philippe -Kim, Yong Joo -Leoforban, Yves -Remond, Michelle -Tano, Lazare -Stephen, Zientara

Australia -Hammond, Jeffrey -Morrissy, Chris Austria -Koefer, Josef -Loitsch, Angelika -Silber, Roland

Georgia -Pötzsch, Carsten

Belgium -Kris De Clercq, (Vice Chairman

of the Research

Group) -Füssel, Alf-Eckbert -Nesya Goris

Brazil -Helio Barbosa Jr -Mozzer, Otto

Germany -Haas, Bernd (Member of the Research Group) -Brhem, Katrina -Paul, Guntram Greece -Hondrokouki, Helen -Marinou, Caterina -Wood, John

Botswana -Thomson, Gavin Bulgaria -Georgiev, Georgi Kirilov

India -Hemadri, Divakar -Singh, Shree Narayan (Member of the Research

Group)

-Polihronova, Liliyana

Iran -Otarod, Vahid

Canada -Dubé, Caroline -Geale, Dorothy

Ireland -Sheridan, Michael Israel -Yadin, Hagai

China -Guo, Jianhong -Li, Zhidong -Lu, Zengjun -Zhidong, Zhang -Zhifang, Zhang -Xue, Jingshan

Italy -Brocchi, Emiliana -Bugnetti, marco -Deluyker, Hubert -Grazioli, Santina -Honhold, Nick -Rumich, Nadia -Sumption, Keith

Cyprus -Georgiou, kyriacos Czech Republic -Sedlák, Kamil Denmark -Alexandersen, Søren Group) -Belsham, Graham -Botner, Anette -Willeberg, Preben

Egypt -Basyouni, Yasser -Logan, Linda Ethiopia

(Member of the Research Group)

(Member of the Research

(Member of the Research Group)

Japan -Morioka, Kazuki -Ohashi, Seiichi -Onozato, Hiroyouki -Sakamoto, Kenichi Kenya -Sangula, Abraham Korea, Republic of -Lee, Kwang-Nyeong -Chang, Byung-Sik -Park, Jong-Hyeon

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Lebanon -Yehia, Ghazi Netherlands -Barteling, Simon J -Bruschke, Marianne -Chenard, Gilles -Dekker, Aldo (Member of the Research Group) -Eble, Phaedra -Harmsen, Michiel -Jacobs, Liesbeth -Paeffgen, Dorothee -Roermund, Herman Ir -Shielen, Wim -Van Aarle, Paul Pakistan -Hussein, Manzoor -Jamal, Syed M Poland -Kesy, Andrzej -Niedbalski, Wieslaw Romania -Diaconu, Mihail Claudiu Russia -Borisov, Vladimir -Diev, Vyacheslav -Zakharov,Valery Slovak Republic -Jadud, Peter Spain -Blanco, Esther -Borrego, Belen -Cubillos, Carolina South Africa -Blignaut, Belinda -Maree, Francois

Uganda -Balinda, Sheila Nina United Kingdom -Barnett, Paul -Bashiruddin, John -Biswal, Jitendra Kumal -Carr, Veronica -Castillo-Olivarres, Javier -Charleston, Bryan -Clarke, Grant -Cox, Sarah -Curry, Stephen -Doel, Timothy -Dwarka, Rahana Mohan -Ebert, Katia -Ferris, Nigel -Gibson, Deborah -Gloster, John -Juleff, Nicholas -King, Donald -Knowles, Nick -Lefevre, Eric -Li, Yanmin -Mahapatra, Mana -Oh, Yooni -Parida, Satya -Partiban, Aravindh Babu -Paton, David (Member of the Research Group) -Paul, Emma -Perry, Brian -Reid, Scott -Reeve, Richard -Ryan, Eoin -Volpe, Carmelo -Zhang, Zhidong

Switzerland -Schelp, Christian -Thuer, barbara

United States of America -Dunbar, M. R. -Gay, Ciril -Garabed, Rebecca -Herrrera, Eileen -Metwally, Samia -Perez, Andres -Pierce, Kenneth -Rieder, Elizabeth Aida -Rodriguez, Louis -Rivas, Ariel -Shalev, Moshe

Thailand -Abila, Ronello

Viet Nam -Ngo, Than Long

Sudan -Habiela, Mohammed Ahmed Sweden -Elvander, Marianne -Valarcher, Francois -Merza, Malik

Turkey -Aynagoz, Gulhan -Bulut Naci -Gultekin, Yasemin -Musa, Alkan

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 3

DR. GAETANA FERRI FULL ADDRESS

Ministero del Lavoro, della Salute e delle Politiche Sociali. Dipartimento per la Sanità Pubblica Veterinaria, la Nutrizione e la Sicurezza degli Alimenti, Direzione Generale della Sanità Animale e del Farmaco Veterinario OFFICE III.

Dear Colleagues, Ladies and Gentlemen, First of all allow me to convey you the greetings of the Ministry of Health, and our gratitude for your participation to this Congress. It is an exceptional opportunity, for all the authorities and researchers involved, to express new ideas and approaches, and take the international community one step ahead in fighting against FMD. Foot and mouth disease today is still one of the diseases that calls for the greatest efforts of national and international health agencies alike, both in terms of prevention and care, because of its high diffusibility, for the direct damage it causes to animal production and the severe trading restrictions connected to its outbreak, which can seriously jeopardise a country’s economy. There have been remarkable breakthroughs in the fight against this disease, thanks to the consolidation of a number of international institutions that have made collaboration at all levels a pillar of their action, attaining undeniable and indisputable results. In this respect, we would like to give our special thanks to EU-FMD, an organism of FAO, for organising this important multidisciplinary congress, as well as for its tireless activity in supporting and enhancing prevention in its 37 European member states, of course in close collaboration with the European Commission DG SANCO. A number of initiatives conducted by EU-FMD have led to a greater sharing of information among its member states, to the implementation of surveillance plans for risk analysis and to the execution of plans to combat the disease in the various countries that adhere to this organism. Furthermore, the possibility to intervene in case of emergency with short term operations based on the ready availability of vaccines, diagnostic reactors and experts, represents an additional service this international organisation renders to countries that request it. In this same regard, the activity of OIE stands out: for promoting a more transparent exchange of information on the occurrence of this disease in countries all around the world and, in the framework of the SPS agreements, for attaining the full harmonisation of the health measures and diagnostic standards that superintend to international exchanges of animals and products. Still, with regard to OIE, we need to underscore its efforts to enhance collaboration between countries in the fields of research and diagnosis, by establishing six OIE reference laboratories in different parts of the world, and appointing a group of international experts dealing with research and training activities within the network of national laboratories. There are two further elements I would like to stress in this regard, that involve both member states and extra European partners in achieving important goals: a number of projects in developing countries that involve also Italy, and the solid commitment of the European Union in funding research under its frame programmes. The activities and experiences mentioned above highlight the need, even today, to make further progress toward full collaboration between national governments and international institutions, in order to promote alliances that may develop every aspect involved in the control of this disease. The struggle against FMD indeed affects a wide range of elements and I will therefore only underscore a few, that I believe need to be the object of specific attention in the near future. In particular, with regard to diagnostics, all the laboratories active in this field need to meet the international standards required to guarantee a rapid identification of the disease. And considering the crucial role played by vaccination in FMD control, there are great expectations concerning the availability of validated confirmation tests to confirm the screening tests that help distinguish infected and vaccinated animals. Further collaboration is necessary, between laboratories on different continents, in order to exchange material and diagnose those serotypes that once were thought to be present only in one specific territory but today, in this increasingly globalized world, may pose a threat to new countries. Finally, in this same regard, and giving due consideration to events that have occurred several times in the past, I believe we need an overall laboratory update, especially for those employing live viruses, in accordance with international safety requirements.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

With regard to the collection of epidemiological data, the increase and funding of surveillance plans is a priority, especially in those countries where there is still scarce knowledge concerning the presence of the disease and the kind of serotypes circulating. In addition, epidemiological surveillance is a valid instrument to assess the risks entailed by the disease, and therefore provide for an optimal use of resources and enhance the effectiveness of vaccination plans. In this respect, Italy is giving its contribution by funding a three-year FAO project in central Asian countries, some in a state of warfare, founded on surveillance and training activities in their laboratories. Finally, I would like to recall some general aspects concerning vaccination. Vaccinal prophylaxis is one of the most effective means of fighting this disease, but as I mentioned before, it needs to be associated with surveillance in order to be founded on sound information concerning the risks involved and the identity of the serotypes circulating. We should also bear in mind the costs of vaccination plans, which call for further funding of projects, especially in developing countries, and for collaboration with the Institutions in charge of vaccine production, so that they may produce according to internationally recognised standards of quality. Several countries are setting up vaccine banks, to intervene in case of emergency. In this process it would be good to seek a higher degree of collaboration at a regional level, so that States may decide together, before purchasing them, which serotypes need to be stored in the different banks. This would allow to cover a higher number of serotypes and avoid superfluous purchases in the various countries, of course on condition that a specific serotype is supplied to the country that requests it. Finally, the implementation of vaccination plans is often the reason for importing countries to set up unjustified trade barriers. This calls me to remind all the countries of the need to meet OIE rules. They are, and I would like to emphasise this, founded on scientifically sound information. Last, but not least, allow me to stress the need to reinforce the veterinary services, for they play a vital role in all the aspects related to the implementation of the control measures for this disease. I believe this congress, for the wide range of topics it covers, may offer a relevant contribution to solving various problems that are faced in different parts of the world, and allow to attain even greater results in the struggle against this formidable disease. But I do not want to take away time from your contributions, and I end here: I wish you a fruitful exchange of ideas, and I hope you will get a chance also to enjoy the beauty of this city and this precious and ancient land. I thank you for your attention.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 4

KEYNOTE: THE GLOBAL CONTROL OF FMD; CHALLENGES AND OPPORTUNITIES K. Sumption1 and J. Lubroth2 1

Secretary, European Commission for the Control of Foot-and-Mouth Disease (EuFMD) Head, Emergency Prevention System-Livestock component, Animal Health Service, Food-andAgriculture Organization of the UN, Rome, Italy.

SUMMARY At global scale, FAO estimates that food production needs to rise by 50% by 2030 to meet targets for hunger reduction. Foot-and-mouth disease (FMD) is a “One world problem”, severely affecting livestock production efficiency where it is endemic, preventing free and safe movement in livestock, and requiring a high and continuous level of public and private investment to protect animals in affected and free regions. Long term FMD control should have benefits at every level, and contribute to global efficiency in food production. This paper summarises the challenges and opportunities for working at global to local scale to bring FMD under control and enable eradication, and has been prepared as part of the development process between FAO and the OIE (World Organisation for Animal Health) of a global approach to long term FMD control. OBJECTIVES The objective is to present a perspective on how efforts against FMD viruses in the principal ecological and livestock production systems at regional level might be organised, under a global coordinated programme. Constraints to control include limited access to quality vaccines, the lack of incentives to drive public and private investments in high burden countries, limited capacity to deliver public veterinary services, and risk of progress being undermined by lack of commitment. Opportunities include global services to assist each region and nation to adopt approaches tailored to their risk /epidemiological region, together with the driver of high livestock and commodity prices which could be used to create incentives for producers and countries to invest in protecting livestock and livestock marketing chains. Through regional approaches and global support services, the global community can expect that progress will be measured using verifiable data and indicators, and improved early warning of risks of reversions and epidemic incursions to free regions. GLOBAL FMD CONTROL; CHALLENGES AND OPPORTUNITIES FMD is an extremely infectious disease of cloven hooved livestock, caused by seven serotypes of FMD virus (O, A, C, Asia-1, SAT1, SAT2 and SAT3). Multiple antigenic types exist within each serotype, and vaccine selection for types A and SAT viruses must in particular be adapted to local epidemic strains to provide effective protection. Although very widely distributed, occurring in all continents (except Australasia), through national and regional co-ordinated control efforts, several large regions (north and central America, most of Europe (including the EU-27 and EFTA states), Australasia and parts of South America and southern Africa, and off shore islands of East Asia (plus Republic of Korea) are recognised as free of FMD. These co-ordinated efforts have required more than 50 years of effort, and remain at risk because the majority of Africa and the Eurasian livestock populations are not free of FMD. FMD is thus a classical “One world problem”, since all countries suffer economic consequences, be it though costs of maintaining freedom, or direct losses from disease and costs of preventive measures. These main “reservoirs” of FMD have seen little improvement in disease control over the past 30 years, and therefore “parallel worlds” currently exist, with on one side major investment in maintaining freedom, and on the other, high recurrent costs to maintain routine vaccination (minority of countries) and a relatively uncontrolled situation (the majority of endemic countries) where sweeping epidemics cause widespread losses. FAO has been aware of these asymmetries in investment in FMD control and the impact of incursions of FMD into free areas; as a result it has championed the “One world” approach of

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

“tackling disease at source” through globally supported and regional co-ordinated efforts. To this end it signed an agreement with the OIE in 2004 (Global framework for the progressive control of transboundary animal diseases; GF-TADS), which has seen the establishment of governance structures, active regional committees and support units. It was agreed at the establishment of the GF-TADs agreement, and ratified at the first Global Steering Committee for GF-TADs, that a global initiative on FMD should be developed by FAO and OIE, with a view to launching the initiative after scientific and other consultations planned for 2009. Global distribution is characterized by regional reservoirs; a challenge and an opportunity The global burden of FMD virus infection is maintained by seven major reservoirs (pools) of infection; each pool has at least 3 serotypes of virus, and since virus circulation is mainly within these regional reservoirs, strains have evolved which are specific to the region and which often (in the case of type A and SAT viruses) require tailored vaccines. Progress in the last 50 years has probably not changed the prevalence of infection in 5 or 6 of the 7 major reservoirs. In the case of the western Eurasian “virus reservoir”, the area directly involved straddles the “traditional ethnic, political or economic boundaries”, involving countries in the European, Middle East and western and central Asian economic groupings. This has the effect that the problem is always too big for one region and responsibility is often seen to belong to a third party, and national investment to improve control is hindered by risk from neighbouring countries/zones. The opportunity that can be seen is to address each regional virus reservoir with its own long term program: a “regional roadmap approach to FMD control”. On the basis of the 7 major reservoirs, 7 such regional programs are needed; three (Europe, South America and South-East Asia) are already underway. Continuation of the current programs is vital, as is developing the plans for the other four. This approach fits well with the joint FAO and OIE initiative on trans-boundary animal diseases (GF-TADS), which has Regional Steering committees. Vaccination against FMD as major tool in regional control; challenges and opportunities FMD vaccination has been applied since 1938, and has been used in large national programs for circa 50 years. Some large regions (e.g. western Europe, parts of South America), using policies of national full bovine population immunization have been able to discontinue vaccination and rely on sanitary measures and emergency response to maintain freedom. In other regions, premature cessation of vaccination has resulted in overwhelming epidemics and therefore return to preventive vaccination (certain countries of the Southern Cone of South America). Global use of FMD vaccination is asymmetric, with very high amount of the global total (of circa 2.3 billion doses per year) being used by only a few countries (China, Argentina, Brazil); with most regions at direct risk in the majority (endemic infection or incursions > once per 5 years), the gap between required immunity (usually considered >80% of the target population) to prevent virus circulation, and current coverage rates (mostly <5%) is enormous. Very little of these vaccination programs are externally funded; the majority are fully or partially state funded. In addition, where vaccination has been used for >10 years, the losses may be seen as acceptable by stakeholders; which reduces one of the drivers to progress to eradication. In the majority of African countries, and south Asia, vaccination coverage is very low; raising economic prosperity in Asia may create an improved tax base on which to plan long term control. Since India can provide a virus reservoir for the wider region, national plans to build up coverage towards national levels by 2015 or 2020 will be crucial to wider success. A further problem is that vaccine quality varies enormously; too much vaccine that is used remains of inadequate quality to maintain a durable immunity (6-12 months). Delivery systems also often fail to vaccinate animals before they are first traded (as calves), exposing these animals to infection in transit or market place, and spreading infection through the marketing chain to new regions. Therefore even where vaccines are used, a combination of poor quality, timing and coverage result in gaps that allows infection to circulate. Confidence is therefore eroded, among veterinarians as well as stakeholders. Vaccine supply is definitely an issue, as the inactivated vaccine production cannot easily be scaled –up; however, the sector has shown it can respond over years to increasing demand. What are the opportunities? First, the work of surveillance and laboratory networks (e.g. FAO/OIE FMD networks) should provide rapid access to selection of suitable vaccines for each virus pool; if no suitable vaccine exists, market studies should assist the private sector, or possibly international agencies, to decide on producing tailored vaccines. Second, the pharmaceutical sector has shown itself capable of responding to demand, but scale of demand may be limited for “niche vaccines” ,

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

likely to be needed in parts of Africa. Third, the demand–side could be stimulated at the level of the livestock producer, if vaccination became a requirement for movement to market/trade (internal and regional trade). Fourth, through regional “roadmaps”” countries could be expected to agree to vaccination coverages, which will be required for at least 3-5 years (and often much longer because of risk from neighbouring countries) to knock-down infection to manageable levels, generating additional demand. Fifth, a renewed effort in research is likely to bring vaccines to market with longer effect (reducing need for repeat vaccination) within the next 15 years. The impact of such new vaccines could be very positive in achieving the shelf life needed for use in remote areas, and the duration required to break transmission cycles, which could allow redirection of the limited vaccine resource to better effect. Without livestock producers support, can anything be achieved? Challenges and opportunity Success in FMD control in Europe and South America has occurred under conditions (including OIE and bilateral standards) where freedom from FMD at national, or zonal level, offered livestock trade opportunities. These economic drivers operated at the level of national strategic planning, and with strong support – or even leadership - from farmers associations. In most of the developing world, the prospect of achieving country-wide or zonal freedom is very distant, as the national livestock supply and demand conditions may favour import rather than export, or where export volumes could be produced, the required attainment of disease control cannot be achieved because of a weakness in civil society drivers (limited farm lobby) and/or in regulation (public sector unable to deliver policy, especially in movement control). Given the above, significant debate on how to achieve incentives for investment in animal health is welcome and necessary, including the possibility of exporting commodities under conditions where the product, not the originating location, would be considered safe in terms of FMD risk. If such commodity treatment can be safe, the argument goes, then the FMD status of the wider region, or even the herd or animal is no longer important, enabling wider access to higher value trade. Linking animal movement control, at any level, to vaccination status could create an important driver for improving vaccination rates, and for overall success, since non-immune traded animals are often implicated in short and long distance spread. Where direct losses are usually small (e.g. under vaccination), such linkages could address gaps in vaccination coverage. Since vaccination status cannot be seen by the buyer, visual or other forms of rapid verification will be needed where regulatory controls are weak, as in most endemic countries. If an animal product can be traded without need for vaccination guarantees, the driver for vaccination will rely mainly on fear of disease, which is unlikely to motivate owners to use vaccination at the required 80% to have the required impact on decreased local viral circulation. Under these circumstances the driver for national co-ordinated control may be weakened, and those countries that have already achieved free status may themselves consider that maintaining this status is not economically viable. One option that could both create opportunities without losing incentives towards wider control would seem to be to adopt policy of development of commodity based trade providing that the FMD vaccination status of the animal and herd is satisfactory (to the regional virus risk). Incentives for involvement in FMD control are also needed in currently free and low risk areas. Biosecurity measures cost and human behaviour is often to take risks if no negative consequence is perceived. Identifying incentives, that operate at level of the livestock trader, will be important to achieve their participation in FMD control, as the main lesson of the recent past is that capacity of countries to control animal movement is weak, and regulatory measures without popular support are unlikely to succeed. Incentives for trading safe (immune) animals could reduce the need for systems based on geographical or regulated animal separation. Stimulating regional efforts as part of global progress; challenges and opportunities Each of the seven major FMD virus reservoirs have within their regions countries with striking intra-regional variations in FMD control capacity, together with risk factors for transboundary spread, such as long distance animal movement patterns between animal breeding areas and terminal fattening /consumption areas, differentials in wealth and meat price, and long land borders with or without transboundary rangelands that mix animals from more than one country. Such conditions allow for periodic epidemics that may touch all countries in the region, which generally respond after the event by sanitary efforts and preventive vaccination programs. The flourishing transboundary animal movements, often informal/illegal, are a feature of many pathogen reservoirs, and cannot be wished away; could they present an critical point/opportunity for regional control? Development of regional roadmaps to FMD control could help to clarify

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

responsibilities and expectations among neighbours, to stimulate political will, to provide early warning of likely failure and to channel and co-ordinate the efforts of donors. To be appreciated, the regional efforts must also serve national interests; FAO and OIE can assist to provide the neutral forum, to assist the region to respond to new risks, including those of countries which fail to control epidemics, and to develop the regional monitoring effort and expertise. The Roadmap concept is not new; Europe, through the EUFMD Commission, had a strategy for FMD control involving national pledges to maintain freedom through vaccination or through stampingout; rapid progress was made in a decade in the 1960s, and possibly vaccination could have ceased earlier than it did (in 1992). The current EUFMD Strategic Plan involves risk reduction measures in areas in parts of west Asia/Middle East that pose a threat to European countries, assisting Turkey to progress towards FMD control. A West Eurasian Roadmap makes sense, since epidemics usually emerge within the region and spread to the margins of Central Asia, to the European borders and to parts of west Asia (as far as Syria/Jordan/Saudi Arabia and even to Egypt in 2007). To the east, there is a case for a South Asia regional Roadmap (SAARC cluster), as well as the current South East Asia (SEAFMD) 2020 Roadmap. China is so large, and with borders that touch Ccentral Asia, east and south Asia, that they have a unique situation and there is a case for them contributing to all 3 Eurasian Roadmaps. From virus typing, there is a case for Regional Roadmaps in West/Central Africa, East Africa, and Southern Africa (SADC region). The virus-type “watersheds”” require mapping, and political boundaries will also influence the inclusion of countries in one or more Roadmaps. Only the SADC region currently has an FAO or OIE reference laboratory within its boundaries, and significant virus typing is required to build up the information base for selection of vaccines, for the time when significant investment will occur. Monitoring progress and keeping ahead of the epidemic wave Since building prevention and response capacity will be essential to progress, stimulating regional prevention and containment effort will be central. How can this be achieved? Regional donor funded projects can be catalytic, but almost all investment to the scale required will need to come from national funds, including private funds. It follows that regional programs will require national parties to argue consistently and effectively in their national and regional economic and political fora to maintain consensus and effort; and FAO and OIE to provide advocacy and supportive information to keep the will to proceed. Therefore monitoring of progress must go beyond the traditional analysis of outbreaks and FMD strains; it needs to follow the political commitment indicators (adoption and implementation of national action plans), objective and comparative indicators of success and effort, and apply greater effort to measure prevention (vaccination performance), capacity of veterinary services to perform key tasks, as well as indicators of impact on infection parameters. At the global level, the use of comparative and objective indicators will be needed to monitor the progress of the regions. Some new information collect and analysis effort will be needed, which could be managed by FAO and OIE, adapting the current international information systems to the new requirements. Further, regional networks will be crucial to each Roadmap and regional effort; they should interface and be supported under a global information effort under GF-TADS, using the strengths of the FAO and OIE FMD reference laboratory network. The global community could expect a yearly appraisal of global progress, based on evidence of commitment, effort and indicators made under the regional programs. At present the global community invests heavily in prevention measures but no one institution undertakes serious measures to determine global progress, and a Secretariat will be needed to produce the progress reports. Since no single organization has the capacity, competence or reach to change the situation on the ground, the overall responsibility to progress the global progress will need inputs of many partners, with suitable governance and advisory structures. If the global efficiency of livestock production is to rise, tools for combating the major transboundary diseases will need to be available and affordable for the majority of livestock at risk, which is not currently the case. Global, national and private sector partners will all need to contribute, and the time is right for setting out a global framework that is likely to speed up the benefits from regional and global control. ACKNOWLEDGEMENTS

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

This paper reflects the views of the authors, should not be considered the formal view point of EuFMD Commission or of FAO. The author is grateful to many persons in FAO for helpful discussions, in the scientific networks on FMD discussions (World Reference laboratory at Pirbright, and OIE/FAO lab network, in research consortia including GFRA) and in the regional organizations with which EUFMD/FAO co-operates.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 5

GLOBAL FMDV DISTRIBUTION AND REGIONAL VIRUS RESERVOIRS: AN OPPORTUNITY TO DIVIDE AND CONTROL? J. M. Hammond*, D. P. King, N. J. Knowles, J. Wadsworth, K. G. Swabey, B. Statham, Y. Li, P. Keel, P. Hamblin, G. H. Hutchings, S. M. Reid, K. Ebert, J. M. Stirling, N. P. Ferris and D. J. Paton Institute for Animal Health, Ash Road, Pirbright, Surrey, GU24 0NF, UNITED KINGDOM

ABSTRACT For Europe, FMD is a continuous threat to FMD-free areas. The activities and information provided by the FAO World Reference Laboratory for Foot-and-Mouth Disease (WRLFMD) to monitor FMD around the globe during the period from October 2006- September 2008 are described. During this period 1250 clinical samples were received to the WRLFMD from 38 countries outside of the United Kingdom including Africa, Asia, Europe and the Middle East. This material was submitted for primary and referral diagnosis (to confirm the presence of FMD virus) and for antigenic characterisation and epidemiological monitoring of circulating virus strains globally. Additionally a further 3774 samples were received from the UK. Collation of such global based knowledge enables the visualisation of FMDV reservoirs as distinct pools providing a real time virus map for the implementation of better informed and improved control measures for FMD. 1. INTRODUCTION Foot-and-mouth disease (FMD) is a highly contagious vesicular disease affecting up to 70 species of domesticated and wild cloven-hoofed mammals including cattle, sheep, goats, swine and wild ruminants. Infected animals exhibit blisters and ulcers in the mouth, feet and udder, lose weight and stop producing milk, and although rarely fatal in adult animals, high mortality can result in the young. On recovery, at least 50% of ruminants become ‘carriers’ with persistent sub-clinical infection. FMD is caused by a non-enveloped RNA virus of the family picornaviridae (FMDV). The virus exists as seven immunologically distinct serotypes (O, A, C, Asia 1, SAT 1, SAT 2 and SAT 3) that do not induce protection against each other after infection. In addition, a large number of subtypes have evolved within serotypes. FMD is endemic in large areas of Africa, Asia and South America and does not respect international boundaries causing epidemics in previously free areas. The distribution of different FMD serotypes and subtypes varies according to countries and continents which can result in several serotypes and subtypes occurring at the same time. However, the global pattern of serotypes and subtypes is not static and is influenced by the type of disease control practised. This highly contagious virus is mainly transmitted by direct contact between naïve and infected animals or their products, but can also be spread by airborne transmission or fomites. Control of disease is also further complicated by the emergence of new strains. In order to develop and implement appropriate and effective control measures including improved or new generation vaccines, diagnostics and practical contingency plans, it is important to determine the origin and monitor spread of known FMDV strains causing outbreaks and screen for any new subtypes that might arise. 2. MATERIALS AND METHODS The WRLFMD at IAH Pirbright receives both samples for FMD diagnosis and virus isolates made in reference laboratories from all around the world. Submitted material was tested by diagnostic assays (virus isolation [using primary bovine thyroid cultures and the IB-RS-2 cell line], antigen ELISA [Ferris and Dawson, 1988] and automated real-time RT-PCR [rRT-PCR: Reid et al., 2003]) to detect the presence of FMDV. In addition, genetic sequencing of these viruses was performed to allow us to precisely characterise new isolates of FMDV and trace their origin by comparing them to those characterised viruses present in the WRLFMD collection (Knowles and Samuel, 2003). This phylogenetic analysis enables us to monitor the emergence and spread of FMD virus globally. The

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

in-vitro antigenic properties of selected isolates were assessed by serological cross-reactivity (virus neutralisation test: VNT and liquid phase blocking ELISA) using bovine vaccinal sera (BVS) raised against reference vaccine strains (Hamblin et al., 1986; Kitching et al., 1989). 3. RESULTS 3.1 Samples received to WRLFMD During this period, 1250 clinical samples were received to the WRLFMD from 38 countries outside of the United Kingdom including Africa, Asia, Europe and the Middle East (figure 1). This material was submitted for primary and referral diagnosis (to confirm the presence of FMD virus) and for antigenic characterisation and epidemiological monitoring of circulating virus strains globally. Additionally, a further 3774 samples were received from the UK. From these samples 523 FMD viruses were isolated and initial characterisation showed that 5/7 FMD virus serotypes were represented with O being the most frequently recovered (figure 2). The WRLFMD website also contains quarterly and annual reference laboratory reports and can be found at http://www.wrlfmd.org/ref_labs/fmd_ref_lab_reports.htm 3.2 Summary of Global FMD situation During 2007 an issue of great concern was the emergence of a highly transmissible lineage of the Pan Asia strain of serotype O (O-PanAsia-2) which spread from India to the east, north and west causing epidemics in a number of countries in the Middle East. This was similar to that seen prior to 2000-2002 when another O Pan Asia strain spread into several normally FMD-free countries including Taiwan, Japan, South Africa, UK, France, Netherlands and South Korea. Although, the O Manisa vaccine is predicted to provide protection against this new Pan Asia variant, the vaccine has a slightly poorer serological match compared to that which was found against the O UKG 2001 Pan Asia virus. During 2007, this lineage spread west through Turkey to cause outbreaks in Thrace and through Jordan, Lebanon, Israel, Palestinian Territory and into Egypt. In addition to Serotype O, there were reported outbreaks due to serotype A (Iran 05 lineage) in Turkey and Jordan. In the Middle East, FMD outbreaks were reported (February 2008) in cattle in northern Israel (due to serotype O) and Lebanon (no serotype reported). In Lebanon, movement of infected animals via a local market (in Bekkaa province) was proposed to be the likely route of infection. An un-expected increase in mortality due to FMD (serotype O) in Bahrain was reported and there were reports of FMD cases in Kuwait. Phylogenetic analysis showed a close relationship between FMD viruses recovered from these outbreaks in Bahrain and Kuwait and other members of the O-Pan Asia 2 lineage. The O-PanAsia-2 strain (ME-SA topotype) continues to dominate in the Middle East region (Pakistan, Iran, Turkey, and Saudi Arabia), while in Southeast Asia (Thailand, Laos) most outbreaks were of the O Mya-98 strain (SEA topotype). In the Middle East (Pakistan, Afghanistan, Iran, Turkey, Jordan, Saudi Arabia), the A-Iran-05 (ASIA topotype) has dominated for the last three years. However, since August 2007, a new sub lineage of this strain (named A-Iran-05ARD07) has been found in Turkey. In Southeast Asia, a local unnamed strain of type A has been circulating for a number of years, without any introductions from outside the region. There continue to be further outbreaks of FMD (Serotype Asia 1) in China. As India is an important reservoir of FMDV, the relationship between O, A and Asia 1 viruses circulating in India (and monitored by the Project Directorate on FMD in Mukteshwar) and those catalogued from neighbouring countries by WRLFMD needs clarification. In Vietnam, new cases of FMD were reported in March 2008 affecting 2 central provinces (Nghe An and Ha Tinh). In order to attempt to control FMD, 100,000 animals in the central province of Quang Tri province were vaccinated with trivalent (serotype O, A and Asia 1) vaccine. Notably, in 2007, two countries within Europe (previously FMD-free without vaccination) reported outbreaks of FMD. In the United Kingdom, FMD was initially confirmed on 3rd August in beef cattle in Surrey, England: the first outbreak in the country since 2001. Subsequently, a total of 8 premises (11 holdings) were found to have animals that were infected by FMDV (Ryan et al., 2008 and Cottam et al., 2008). Trade restrictions with the EU were lifted in December following 3 months without any subsequent outbreaks of disease. The United Kingdom’s status of FMD freewithout vaccination was restored by OIE on 19 February 2008. In Cyprus, serological evidence of FMD infection was uncovered in small ruminants. The initial case was identified in October following investigation of a flock of 25 sheep which were exhibiting suspect clinical signs. Although conclusive evidence of FMDV circulation was not obtained for this first farm, testing of further samples collected from neighbouring farms (near Larnaca on the south coast of the island) revealed 3 further flocks with serological evidence of FMD infection including antibodies against FMDV structural proteins of serotype O. Based upon these serological data and clinical evidence of vesicular lesions in some of the animals, an FMD outbreak was declared to the

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

OIE on the 5th November 2007. Further surveillance in the affected area and other parts of the island failed to demonstrate active infection and serologically positive farms were not identified outside of the 10 km surveillance zone surrounding the culled Larnaca flocks. As from 21 February 2008 Cyprus regained its FMD-free status without vaccination. Further information is presented elsewhere in these proceedings. Throughout 2007-2008 there continued to be sporadic reports of FMD due to Asia 1 in Central Asia. In addition to reports from China (Qinghai, Gansu and Xinjiang Provinces), in January 2007 this serotype also caused the first outbreak in North Korea since 1960. Although initial analysis indicated that serotype O, had caused this outbreak, subsequent investigation of clinical material collected from affected animals recovered FMDV serotype Asia 1. This outbreak was located in P`yongyan-Si close to the capital. Outbreaks of FMD due to serotype O were reported in Kazakhstan, and due to serotype O and serotype A in Kyrgyzstan. Elsewhere in Asia, there were continued reports of FMD in endemic areas including Vietnam, Malaysia, Myanmar, Bhutan, India and Laos. In Africa, types O, A, SAT 1 and SAT 2 were isolated from Kenya. In Somalia, type O viruses (EA-3 topotype) were linked with those occurring in the Yemen Arab Republic, although ultimately viruses belonging to EA-3 probably originate in the horn of Africa. In Nigeria, outbreaks of type O and SAT 2 were suspected to be linked with viruses occurring in Sudan. New outbreaks of FMD SAT 2 were reported in Botswana and Namibia and are continuing to cause problems in northern Botswana, eastern Namibia (Caprivi Strip) and southern Zambia. It would appear that there have been multiple introductions into Botswana during 2007-2008. But it is not clear if the origin is wildlife (African buffalo) within Botswana or from cattle/wildlife in neighbouring countries. In September 2008, a suspected outbreak of FMD was reported on Kaombe Ranch, Nsanje, southern Malawi and this is the first since 2003. Tracing the origin of the infected animals indicated that some were brought in from an area close to Lengwe National Park harbouring buffaloes that were a source of the 2003 outbreak. In South America, outbreaks of FMD (serotype O) were reported to the OIE from Bolivia and Ecuador during 2007. In addition, FMDV serotypes O and A continued to cause outbreaks in Venezuela. In many countries of the continent, mass vaccination programmes are regularly being employed to control FMD. A 15 km high surveillance zone (HSZ), not considered FMD-free, was created in the common borders of Argentina, Bolivia, Brazil and Paraguay, which is being closely monitored. Except for this HSZ zone, all the Argentine territory was considered FMD-free either with or without vaccination The FMD-free without vaccination region has been extended to include Northern Patagonia and the area of FMD-free status with vaccination (suspended due to the 2006 emergency) restored by OIE. In Brazil, despite a large zone with status suspended, the state of Santa Catarina was established FMD-free without vaccination. Part of the state of Para was recognized FMD-free with vaccination. Further north, in Colombia, the border with Ecuador, part of the Valley and Caqueta, and Western Cundinamarca were declared FMD-free by the OIE. Peru added the central-eastern zone to the already recognized FMD-free without vaccination zone, reaching 85% of its territory with a free status. An outbreak of FMD type A was reported on three farms in Sardinata, Norte de Santander, Colombia in July-August 2008 which was the first since Feb 2005. During this period WRL have received samples from Gabon for the first time, received samples from Somalia for the first time since 1983 and received samples from Nigeria for the first time since 1984-85. This is a very welcome improvement in the geographical range of samples sent to WRL and may reflect the efforts of the EUFMD in encouraging sample submission from these regions. This is a very positive move for the future with regard to obtaining regular ‘real time’ information on virus populations circulating in particular regions and will greatly aid informed disease management and control. 3.3 Current Threats to Europe The current threats to Europe are considered to be, the new O-Pan Asia 2-lineage and A-Iran-05 from the Middle and near East, O, A and Asia 1 from South and South-East Asia and Asia 1from China, O, A, SAT1 and SAT2 from North Africa, SAT1 and SAT2 from Southern Africa and O and A from non-free regions of South America. 3.4 Vaccine Recommendations WRL vaccine recommendations have been changed to reflect the variation in FMDV serotype A activity in the Middle East and western Asia (Table 1). A Iran 96 has been reduced from high to medium priority reflecting continued dominance of the A Iran 05 strain. A22 Iraq vaccine remains

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

at high priority to cover against A Iran 05, although it has been noted that recent Turkish isolates of the A Iran 05 strain (named A-Iran-05ARD-07) showed a poor antigenic match to A22 Iraq vaccine. Of note also is that the SAT2 vaccine used in Botswana showed limited cross-reactivity to some recent isolates from Botswana and neighbouring countries. Although O Manisa and A22 Iraq remain the most important vaccine strains for protection against viruses circulating in the Middle East, not all of the virus isolates that have been studied show a strong match to these vaccines. Therefore, emergency vaccination would require the use of high potency vaccines to guarantee protection and there may be a case for developing new vaccine strains with greater antigenic homology. Asia 1 Shamir remains the vaccine strain of choice for this serotype. In Africa, there is a great diversity of viruses circulating and in some cases, vaccines that provide good matches do not seem to be readily available. In South America, circulating viruses are all of the type that has been indigenous to the continent and are reasonably well matched by O Campos and A24 Cruzeiro, although supplementary strains of serotype A are also used in some countries to improve vaccine match. 4. DISCUSSION 4.1 Global Control - Regional Knowledge It can be argued that broader and more in depth regional knowledge of FMD outbreaks and identification of these within particular reservoirs or pools of FMD activity can provide an increased potential for globally informed regional FMD control programmes. Such programmes would utilise the most up to date information and would allow control strategies to be developed based upon greater regional knowledge. The suggested global locations of 7 FMD virus pools, comprising 3 pools covering Europe, the Middle-East and Asia, 3 pools covering Africa and 1 pool covering the Americas are shown in figure 3. These 7 pools can be considered as containing multiple serotypes but with subtypes/topotypes mainly confined to that pool. This then allows for better informed control measures to be implemented within that pool, but within the context of a real time map of global FMD status. It also follows that if vaccination is to be a major tool for control then each pool could benefit from tailored or more specific vaccines relevant to the topotypes present in that pool and not necessarily be reliant on the generically available vaccines. It should also be noted that currently there is an asymmetry of effort to identify improved vaccines for use in Africa compared with those used to equip the developed world’s vaccine banks. Such a global approach would require that each region or pool would need laboratory support for typing and vaccine matching work. Obviously regional reference laboratories and WRL can provide major support but certain regions will need development of their own regional laboratory perhaps utilising support from twinning programmes encouraged by both FAO and OIE. 4.2 Future Issues Issues for consideration in the future include how to generate an increase in the knowledge of circulating viruses in Africa in order to improve the current sparse information on effective vaccine strains? To develop ways to improve interactions and information sharing with other reference laboratories. Identify twinning opportunities with laboratories in Middle East and Africa. Ensure that strong links are maintained with FMD initiatives already underway, such as involvement in SEAFMD and GFRA. And finally, to provide mechanisms to ensure that WRLFMD continues to provide the best information possible to the global FMD network. 5. OVERALL CONCLUSIONS AND RECOMMENDATIONS

Material submitted to WRL provides confirmation of the presence of FMD virus, antigenic characterisation and epidemiological monitoring of circulating virus strains globally Such global based knowledge enables the visualisation of FMDV reservoirs as distinct pools providing a real time virus map for the implementation of better informed and improved control measures for FMD Networks facilitating the improvement of this information generation should be fully supported by decision makers and funding bodies

6. ACKNOWLEDGEMENTS The funding support of Defra, FAO and DG SANCO is gratefully acknowledged. The efforts of EUFMD, the IAH staff, SEAFMD, GFRA, and OIE are also much appreciated. Also the ongoing collaborations with other FMD reference laboratories play an important part in WRLFMD activities.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 1: WRLFMD: 1250 samples from 37 countries October 2006 to June 2008

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

O A SAT 1 SAT 2 Asia 1

382

11 2

Figure 2: 523 FMD Viruses Isolated Between October 2006 and June 2008

Pool 3 O,A,

Pool 2 O,A,

Pool 1 O,A, Po ol

Pool 5 O,A,S

Pool 4 O,A, Pool 6

OIE Laboratories

Reference

FAO

Additional

Reference

Regional/Natio nal

Figure 3: Visualization of Regional FMD Virus Pools as an Aid to Global FMD Control

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

HIGH PRIORITY

MEDIUM PRIORITY

LOW PRIORITY

O Manisa O BFS or Campos A24 Cruzeiro A22 Iraq Asia 1 Shamir SAT 2 Saudi Arabia (or equivalent) A Argentina 01 A Iran 96 A Iran 99 A Eritrea A Iran 87 or A Saudi Arabia 23/86 (or equivalent) A Malaysia 97 (or Thai equivalent such as A/NPT/TAI/86) O Taiwan 97 (pig-adapted strain or Philippine equivalent) SAT 1 South Africa SAT 2 Zimbabwe

A15 Bangkok related strain A Kenya A87 Argentina related strain SAT 1 Kenya SAT 2 Kenya SAT 3 Zimbabwe C Noville Within category: not in order of importance

Table 1: Vaccine Bank Recommendations 7. REFERENCES [1] Cottam E.M., Wadsworth J., Shaw A.E., Rowlands R.J., Goatley L., Maan S., Maan N.S., Mertens P.P., Ebert K., Li Y., Ryan E.D., Juleff N., Ferris N.P., Wilesmith J.W., Haydon D.T., King D.P., Paton D.J. and Knowles N.J. 2008. Transmission Pathways of Foot-and-Mouth Disease Virus in the United Kingdom in 2007. PLoS Pathog 4(4): e1000050. doi:10.1371/journal.ppat.1000050 [2] Ferris N.P. and Dawson M. 1988. Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Vet Microbiol. 16(3):201-9. [3] Hamblin, C., Barnett, I.T.R. & Hedger R.S. 1986. A new enzyme-linked immunosorbent assay for the detection of antibodies against foot-and-mouth disease virus. J. Immun. Meth. 93: 115-121. [4] Kitching, R.P., Knowles, N.J., Samuel, A.R. & Donaldson, A.I. 1989. Development of footand-mouth disease virus strain characterisation – a review. Trop. Anim. Health Prod. 21: 153-166. [5] Knowles, N.J. & Samuel, A.R. 2003. Molecular epidemiology of foot-and-mouth disease virus. Virus Res. 91: 65-80. [6] Reid, S.M., Grierson, S.S., Ferris, N.P., Hutchings, G.H. & Alexandersen, S. 2003. Evaluation of automated RT-PCR to accelerate the laboratory diagnosis of foot-and-mouth disease virus. J Virol Methods 107:129-39. [7] Ryan E., Gloster J., Reid S.M., Li Y., Ferris N.P., Waters R., Juleff N., Charleston B., Bankowski B., Gubbins S., Wilesmith J.W., King D.P. and Paton D.J. 2008. Clinical and laboratory investigations of the outbreaks of foot-and-mouth disease in southern England in 2007. Veterinary Record (2008) 163, 139-147.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 6

THE HEMISPHERIC PROGRAM FOR THE ERADICATION OF FOOT-AND-MOUTH DISEASE [PLAN HEMISFÉRICO DE ERRADICACIÓN DE FIEBRE AFTOSA (PHEFA)]: SUCCESSES AND CHALLENGES

V. Saraiva,

1,*

and D. Geale 2

Pan-American Foot-and-Mouth Disease Center –VP-PANAFTOSA-PAHO/WHO, Av. Presidente Kennedy, 7778, Duque de Caxias, RJ, Brazil, CEP 25040-004 2 Terrestrial Animal Health Division, Canadian Food Inspection Agency, 59 Camelot, Ottawa, Ontario Canada, K1A 0Y9

ABSTRACT The Pan American Foot-and-Mouth Disease Centre (PANAFTOSA) is one of seven specialized Pan American Health Organisation (PAHO) centres. It was established in 1951 to coordinate FMD programs throughout the Americas, as well as functioning as an OIE/FAO reference centre. It functions as the secretariat for the Hemispheric Committee for the Eradication of FMD [COHEFA] and for Inter-American Group for the Eradication of Foot-and-Mouth Disease [GIEFA]. It also annually organizes the South American Commission for the Fight against FMD [COSALFA], which is the principal hemispheric event for FMD in the Americas. The aim of the COSALFA is to review progress in FMD eradication in member countries and provide up to date information. In reviewing the progress of the Hemispheric Program for the Eradication of FMD (PHEFA) at the 15th Inter-American Meeting at the Ministerial Level on Health and Agriculture [RIMSA] held in Rio de Janeiro in June 2008, it was noted that progress was inconsistent among countries, whereas some achieved FMD eradication, others still struggle with endemism. The factors contributing to FMD persistence are well known. The PHEFA goal of an America FMD-free for 2009 needs to be reconsidered. Transparency is critical in the final stages of FMD eradication. All countries must cooperate to achieve the final goal. 1. INTRODUCTION Foot-and-mouth Disease (FMD) was introduced to Argentina and the United States in 1870. The first outbreak in Brazil occurred in the same year. FMD continued to spread to Bolivia and Peru in 1910 and 1912. Chile experienced its first outbreak in 1920 with animals introduced from Argentina. In 1950/51 FMD was identified in Colombia and Venezuela and ten years later in Ecuador (1961). Hemispheric eradication was achieved first in North America—the USA in 1929; Canada in 1952 and Mexico in 1954. Central America has never been affected by FMD. In South America, Chile achieved eradication in 1981 with two re-introductions in the 1980s. From the early to late 1990s, the number of FMD cases in South America fell from an average of 766 cases per year to 130 cases per year. By the year 2000, four countries, Argentina, Chile, Guyana, and Uruguay, were recognized as FMD free without vaccination. In the spring of 2001, there was a widespread reoccurrence of FMD into Argentina, Uruguay and the state of Rio Grande do Sul in Brazil and the number of outbreaks reached 4,318 (Figure 3) (Correa Melo et al 2002). The re-emergence was attributed to (i) reduced implementation of preventive measures by participating nations and (ii) decreased investment in infrastructure for animal health and surveillance, after declarations of freedom (Correa Melo et al 2002). Following these outbreaks, most South American countries reinstated vaccination and improved movement and border controls. The Pan American Foot-and-Mouth Disease Centre, of the Veterinary Public Health Project (VP/PANAFTOSA/PAHO/WHO) was established in 1951 as the FMD reference network coordinator for the Americas. As an OIE/FAO reference centre, it harmonizes and ensures the quality of diagnostic methodologies in the region by distributing international reference reagents as well as

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

organizing inter-laboratory testing. PANAFTOSA also coordinates planning, research, technology transfer, training and evaluation of sanitary strategies for the control and eradication of FMD. PANAFTOSA first entered into projects with Argentina in 1961; Brazil in 1965; Paraguay and Uruguay in 1967; Chile in 1969 and Colombia in 1972. The development of oil adjuvanted vaccines for use in FMD eradication in 1972 spurred vaccination coverage through the 1970s and 1980s. However during the 1980s, South America experienced a weakening of the National FMD programs which were revitalized in 1987 at the Inter-American Meeting at the Ministerial Level on Health and Agriculture [RIMSA 5]. 2. MATERIALS AND METHODS 2.1 Plan hemisférico de erradicación de fiebre aftosa (PHEPA) This underlying plan for the Hemispheric Eradication of FMD [PHEFA] was signed in 1987 at InterAmerican Meeting at the Ministerial Level on Health and Agriculture [RIMSA 5]. It divided the Continent into sub regional projects, namely: North America, Central America, Caribbean, Andean, Amazonian and Southern Cone, and developed for each one different strategies for control, eradication and prevention with defined goals (Figure 1). Early in the 1980s a methodology to classify production and disease ecosystems types in South America was developed by VPPANAFTOSA/PAHO/WHO Four FMD ecosystems were described: (i) primary endemic; (ii) secondary endemic; (iii) para-endemic/sporadic and (iv)free areas. The basic strategies of the PHEFA were to Gain knowledge of the nature of livestock production and trade Apply a veterinary model of service that incorporates the producer Characterize the risk and spread of FMD in different zones according to defined macroregionalization and ecosystems. Administer technical coordination of the FMD control and eradication programs Ten specific programs underlie PHEFA: Epidemiological Surveillance. Prevention Systems Sanitary control: Customs, Airports, Ports. Borders etc. Laboratory Diagnosis and Vaccine Control. Vaccine production. Massive and Systematic vaccination. Outbreaks Detection and Control Continental Surveillance (Sistema Continental de Informação e Vigilancia - SIVCONT ) International Border Agreements Community/Public Participation Since 1973, PANAFTOSA hosts the annual meeting of the South American Commission for the Fight against FMD [Comisión Sudamericana para la Lucha contra la Fiebre Aftosa-COSALFA], the principal hemispheric event for FMD in the Americas. VP-PANAFTOSA/PAHO/WHO is secretariat for several oversight bodies for the PHEFA plan. These include the Hemispheric Committee for the Eradication of Foot-and-Mouth Disease [Comité – Hemisférico para la Erradicación de la Fiebre Aftosa - COHEFA] also established at RIMSA 5 in 1987. The role of COHEFA is to ensure the political will of governments and private sectors to eradicate FMD in the hemisphere; to coordinate and evaluate the progress of the PHEFA; and to obtain the economic and financial resources for national and subregional programs covered by PHEFA. In 2004, RIMSA 14 created the Inter-American Group for Foot-and-Mouth Disease Eradication [Grupo Interamericano para la Erradicación de la Fiebre Aftosa -GIEFA}, as a working group, with the participation of the public and private sectors. The objective of GIEFA was to reorient the PHEFA, supporting the final stages of FMD eradication in South America targeted for 2009. The revisions included specific action plans designed to complement national programs that were already in place under the original PHEFA. The overall objective was to control and eradicate the disease from endemic areas and preventing it from entering free areas in the continent, as well as strengthening the zoo sanitary infrastructure of all countries in the region. Endemic zones which required detailed actions include: The Chaco region including parts of Bolivia, Paraguay and the North of Argentina,

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Zones in Ecuador and Venezuela, The north zone of Bolivia, the north-eastern part of Paraguay and its border with Brazil, and Zones of undetermined risk in the northern states of Brazil. addition, there were bi-or tri national border zones of significant concern including Argentina – Brazil – Uruguay, The north-eastern part of Paraguay and the State of Paraná in Brazil, The border of south east Paraguay and the north of Formosa in Argentina, Bolivia – Chile – Peru, Peru – Ecuador, Ecuador – Colombia, Colombia – Venezuela (Llanos and Atlantic coast) Brazil, Venezuela – Guyana, and Brazil - Colombia

Subordinate to the PHEFA (2005-2009) plan are regional plans such as the PAMA plan in the Southern Cone [Programa de Acción MERCOSUR Libre de Fiebre Aftosa.-PAMA] overseen by the Permanent Veterinary Committee [Comité Veterinario Permanente del Cono Sur –CVP] and the Andean Technical Committee on Animal Health [Comité Técnico Andino de Sanidad Animal – COTASA]. The CVP also oversees the High Surveillance Zone [Zona de Alta Vigilancia (ZAV)] that brings together the efforts of Argentina, Brazil, Bolivia and Paraguay to strengthen sanitary actions at their common borders. Notwithstanding this detailed classification, efforts are being directed nowadays to the three main problem areas. (Figure 2): 3. RESULTS In spite of all the resolutions, committees, meetings, reviews, action plans and international as well as national support, the PHEFA goal of an FMD free Americas by 2009 may not be realized. There have been considerable successes but unfortunately also a number of challenges. 3.1 Successes Since 1987 the number of cases of FMD in South America has decreased from 825 to 212 in 1998 to 45 in 2007, with 43 to date in 2008 (Figure 3). Advances in the declaration of FMD free zones, with and without vaccination, in accordance with OIE statutes are illustrated in Figure 4. Since the adoption of the concept FMD free with vaccination by the OIE, this status was achieved in 1994 by Uruguay, Brazil in 1998, Argentina and Paraguay in 1997, Colombia in 2001 and Bolivia in 2003. Zones free without vaccination were recognized in Argentina in 2002, Peru in 2005, Brazil in 2007 and Colombia in 2008.Throughout these achievements, reoccurrences of FMD, starting in 2000, have caused the temporal loss of status. Geographically the area, number of herds and number of bovines vaccinated have shown a significant increase in the area now FMD free with vaccination and FMD free without vaccination (Table 1). Vaccine production has recently been standardized and controlled in South America. Vaccines comply with the OIE Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, but a few production facilities still struggle with purity standards, which are not yet harmonized in some countries. Most of the advanced countries have developed effective strategies for emergency FMD outbreak control, for example in Uruguay, Argentina and Brazil, in 2001, Brazil in 2005 and in Argentina in 2006. On the other hand, FMD endemism persists in Venezuela and Ecuador, posing a threat to other countries 3.2 Generic Challenges Inconsistencies in the development and implementation of national FMD eradication strategies are largely due to three factors, (i) lack of resources, (ii) lack of political will and (iii) poor publicprivate coordination. These clearly define the current challenges. There is a need for an epidemiological characterization of livestock production systems as a foundation for FMD risk determination which has not been carried out in all South American countries Use of automated information sharing technologies such as the SIVCONT provided in seminars and operational training by VP-PANAFTOSA/PAHO/WHO has only been adopted by Brazil, Colombia and Peru while the remaining countries still use the traditional Continental System.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Although bovine movement control systems exist in all countries, they have been developed with different technologies and purposes so that transit control except in the High Surveillance Zone [Zona de Alta Vigilancia-ZAV] is not harmonized. VP-PANAFTOSA/PAHO/WHO has been stressing that the national country programs need to enhance submissions of field material from premises with vesicular clinical signs, for complementary diagnosis. Although recently improved such submissions still only represent a small percentage of a potential effective passive surveillance system (Table 2). Of the samples submitted from premises with vesicular clinical signs, 93, 9% result in confirmed diagnoses with 5.4% of these FMD. Unfortunately, 6.1% remain undiagnosed, which reflects a deficiency in the final eradication phase of FMD (Table 3). Information on vaccination coverage is based on the number of vaccines sold per known bovine population and does not reflect the actual coverage determined by the regional policy which may vaccinate once, twice or more times during the year, depending on the level of risk established; the production system involved; and the type of eradication program adopted by the country. Immunity coverage studies have been carried out by some countries, but such studies should be applied in all vaccinating countries. 3.3 Specific Challenges: Residual Endemic FMD In 2007, outbreaks of FMD types A and O have occurred in Venezuela (26) and Ecuador (11) and Bolivia (15) and these have continued in 2008 with 29 outbreaks in Venezuela, 7 in Colombia and 7 in Ecuador. Such outbreaks highlight the challenges faced by South America as final continental eradication is approached. In August and September 2008 in response to GIEFA and COHEFA recommendations, VP/PANAFTOSA-PAHO/WHO conducted missions to these countries with the participation of OIE, FAO, IICA and FEDEGAN/Colombia (livestock owners) to coordinate technical cooperation and recommend adjustment of the PHEFA plan (2005-2009), if necessary. Ecuador is still in an FMD endemic situation. The Ecuadorian Livestock Health Service (SESA) has developed a national eradication program but lacks legislative authority to enforce it and build up veterinary infrastructure. Private sector participation is divided between the National FMD Eradication Commission (CONEFA), and a few livestock owners associations. The country is dependent on support of international agencies such as the FAO and IICA. FMD vaccination is compulsory in Venezuela under two schemes. The first, officially coordinated and operated by vaccination brigades is directed to the small farm holders. The second scheme is carried out by the larger producers themselves, under the supervision of the State. Whereas the first scheme covers 16 % of the animal population that of the private vaccination scheme lacks coverage Supporting legislation for a national plan is under discussion; Veterinary Services (SASA) is being reformed and the involvement of the private sector through the Organizaciones de Ganaderos which suffers from political polarization Bolivia could not reach the PHEFA goals anticipated due to deficiencies in the formal zoosanitary structure. Although recognition of two FMD free zones reflect partial achievement, nevertheless, current politico-social differences between the official service and the production sector put at risk the achievements and continuity of the program. Finally, the northern Amazonian region was not initially a priority for the National Brazilian Program for FMD Eradication. Now authorities and local producers have a proposal to develop and implement surveillance and respond to vesicular events building on public education and awareness. Outbreaks in 2004 tested the efficiency of the response capacity. In 2008, the southcentral part of the State of Pará was recognized as FMD free with vaccination, reflecting the improved veterinary service. 3.4 Specific Challenges: International Border Zones Combined bi- and tri- national programs were created at risk borders in an effort to strengthen and coordinate actions under PHEFA (2005-2009). This concept builds on strengths of bordering countries to bring stability to common areas. The creation of the High Surveillance Zone on the borders between Argentina, Bolivia, Brazil and Paraguay of the Southern Cone Region (Figure 4) was recognized by the OIE in 2007 as a separate border zone to encourage reporting through reduced consequences for the individual countries. Trade is permitted between this zone and the rest of the countries under a special surveillance scheme. Similar coordination exists at the Argentina-Brazil-Uruguay; northeast Paraguay and states of Parana and Mato Grosso do Sul, Brazil; and southeast of Paraguay and northern Argentina with high compliance, based on the

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

common interests of the participant countries. Characterization of properties at risk and coordination of actions and are managed by the CVP as part of the PAMA. In the Andean Region, recent meetings have been held between Bolivia, Chile and Peru. The PeruEcuador border has seen rapid coordinated response to FMD type O through movement control and vaccination. A formal cooperative agreement exists between the Ministries of Agriculture of Ecuador and Colombia with the involvement of the private sector for both FMD control programs and response through strategically located animal health out-posts. Colombia is developing a cooperative agreement with Venezuela including a system of active surveillance and border control points. The FMD border control between Brazil and Colombia is aimed at the prevention of animal entry from endemic areas of the Amazon. There is little activity between Brazil-Venezuela-Guyana but efforts are scheduled for 2008. 4. CONCLUSIONS Is it possible to eradicate FMD from South America? The answer is yes, but first it should be understood that the hemispheric eradication of FMD requires a comprehensive integrated plan supported nationally and regionally by all stakeholders including private and public, rigorous coordination and political commitment, all driven by economics. Transparency is critical in the final stages of FMD eradication Lack of resources is not the only reason for the pitfalls and the residual vestiges of FMD require concentrated effort by all countries so that the PHEFA goal of an Americas FMD-free even if not by 2009 can be achieved. 5. REFERENCES [1] Annual Laboratory Report, 2005. PANAFTOSA-PAHO/WHO. [2] Annual Laboratory Report, 2006. PANAFTOSA-PAHO/WHO. [3] Annual Laboratory Report, 2007. PANAFTOSA-PAHO/WHO. [4] COHEFA http://www.panaftosa.org.br/Comp/Eventos/cohefa_11_novo/english/default_i.html [5] Correa Melo, E., Saraiva, V. & Astudillo, V. 2002 Review of the status of foot-and-mouth disease in countries of South America and approaches to control and eradication. Rev. sci. tech. Off. int. Epiz. 2002, 21 (3), 429-436. [6] GIEFA (2005) Executive Summary Hemispheric Plan for the Eradication of Foot-and-Mouth Disease (PHEFA_ PHEFA’s Action Plan for the final step in the eradication of FMD from the Americas 2005-2009 (versión del 30 de marzo del 2005 - Lima - Peru). Accessed Sept 16 2008 at http://www.panaftosa.org.br/Phefa/ [7] PHEFA Action Plan (2005-2009). PANAFTOSA-PAHO/WHO. [8] Report on the XXXIII COSALFA Meeting. PANAFTOSA-PAHO/WHO. [9] Report on the XXXIV COSALFA Meeting. PANAFTOSA-PAHO/WHO. [10] Report on the XXXV COSALFA Meeting. PANAFTOSA-PAHO/WHO. [11] Report on the Situation of FMD Programs, 1988. PANAFTOSA-PAHO/WHO [12] Report on the Situation of FMD Programs, 2005. PANAFTOSA-PAHO/WHO. [13] Report on the Situation of FMD Programs, 2006. PANAFTOSA-PAHO/WHO. [14] Report on the Situation of FMD Programs, 2007. PANAFTOSA-PAHO/WHO. [15] Saraiva, V 2008, ANÁLISE DOS AVANÇOS E PROBLEMAS OBSERVADOS NA EXECUÇÃO DO PLANO DE AÇÃO DO PHEFA - 2005-2009 and PHEFA PLAN OF ACTION 2005-2009: WHERE ARE WE?, presentations at COHEFA 10, Rio de Janeiro, June 10 2008-09-30 Table 1: South America Sanitary Situation, 1988 and 2008 (September, 2008). Years

1988

Unit

Bovine/Bubaline Herds Area

Free without vaccination

Free with vaccination

Not Free

Total

n°

3.371.100

243.426.300

246.797.400

% n° % n° %

1,4 108.557 2,7 757.820 4,4

0 0 0 0 0

98,6 3.963.537 97,3 16.560.725 95,6

100 4.072.094 100 17.318.545 100

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2008

Bovine/bubaline Herds Area

n°

10.604.822

176.206.874

50.163.000

336.974.696

% n° % n° %

3,1 916.748 16,3 2.739.520 15,8

81,9 2.761.907 49,2 8.450.557 48,8

15,0 1.940.451 34,5 6.125.514 35,4

100 5.619.106 100 17.315.591 100

Source: PANAFTOSA-PAHO/WHO.

Table 2: Samples received at PANAFTOSA-PAHO/WHO and number of farms with signs compatible with vesicular disease. South America, 2005-2007 Year

Samples received at PANAFTOSA

Number of farms with clinical signs compatible with vesicular disease.

2005

1673

2006

1238

2007

187

1210

Source: Annual Laboratory Report, 2005, 2006 y 2007; Report on the Situation of FMD Programs, 2005, 2006 y 2007.

Table 3: Number of farms* with notification of clinical symptoms compatible with vesicular diseases, according to diagnosis and countries. South America, 2007

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 1: PHEFA Macro Regions in the Americas

Source: Data sent to COSALFA by the countries, 2007

Figure 2: Priority areas, 2008.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 3: FMD Outbreaks in South America, 1972-2008 (September 2008).

FMD Outbreaks 6000 Outbreaks

5000 4000 3000 2000 1000 0 72 19

76 19

80 19

84 19

88 19

92 19

Year

96 19

00 20

04 20

08 20

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 4: FMD Status recognized by the OIE September, 2008

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 7

FOOT-AND-MOUTH DISEASE VIRUS TYPE C SITUATION: THE FIRST TARGET FOR ERADICATION? P.L. Roeder1* and N.J. Knowles2 2

1 Taurus Animal Health, Hampshire GU35 8SY, UK Institute for Animal Health, Pirbright Laboratory, Surrey GU24 0NF, UK

After its first description in Europe in 1926, the Waldmann C foot-and-mouth disease (FMD) serotype, later referred to simply as type C, was for many years a significant component of the FMD complex in Europe, South America, South Asia, the Philippines and parts of East Africa and was included in preventive vaccination programmes. The outbreak areas in different continents were often linked by trade in meat products and vaccines. However, for reasons which are obscure, this FMD virus type, never perhaps as robust as other FMD virus types, has been in decline for some 30 years and its distribution has become severely limited in the last decade. This situation has led to an understanding that FMD virus type C could be considered as a candidate for eradication. Drawing on molecular epidemiological data this report briefly describes the recent history of FMD type C occurrence and presents the epidemiological determinants to be understood if eradication is to be attempted. Recommendations are made with respect to the actions which need to be taken. 1. INTRODUCTION What became known by international agreement as Waldmann type C FMD virus was first described in 1926 (Waldmann and Trautwein) in Germany. Historically the type C FMD virus has had a relatively narrow distribution compared to the other European types O and A and type Asia 1, being found principally in Europe and South America with reported occurrence in South Asia and Africa. Although it became relatively widespread, and was an important component of the global FMD situation, it had always a relatively restricted distribution and was never perhaps as robust as other types of FMD. In Africa it has been reported, and then only infrequently, from Kenya (from 1957 almost every year until 1988), Ethiopia (initially in 1957, then from 1971 to 1983) and Uganda (after 1970 and 1971 and possibly until 1979). In North Africa outbreaks were confirmed in Tunisia in 1965, 1967 and 1969, however, other anecdotal reports from Algeria in 1971 and 1972 were never confirmed. The recent history of type C foot-and-mouth disease (FMD) virus occurrence indicates a greatly reduced incidence of outbreaks and geographic distribution compared to the mid-twentieth century situation. This has led some workers to suggest that the current situation lends itself to contemplating global eradication of this virus type in the near future. However, this enthusiasm needs to be tempered by a realisation that the areas where FMD virus type C outbreaks were last detected in Asia and Africa and surrounding countries are not well covered by surveillance. Here, and in Latin America, where surveillance is considered to be relatively effective, the epidemiology of FMD virus type C occurrence is enigmatic. This paper explains the elements of the enigma and suggests that without a knowledge of the determinants of virus persistence between outbreaks in the foci where the virus has occurred in recent years it is difficult to assess with certainty the actual status of these foci and the likelihood that the virus could be eliminated from each of them. 2. MATERIALS AND METHODS RNA was extracted from virus-containing preparations using the RNeasy Kit (Quiagen) and a onestep amplification of the VP1-coding region was performed (primer sequences available on request to the authors) using Ready-To-Go™ RT-PCR beads (GE Healthcare) with amplicon clean-up using ExoSAP-IT (USB Corp.). Cycle sequencing used the CEQ Quick Start Kit and the CEQ8000 automated sequencer (Beckman Coulter). The region amplified is depicted in Figure 1. Phylogenetic analysis using MEGA 4.0 (Tamura et al., 2007) enabled the construction of mid-point rooted Neighbor-joining trees.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 1: An indication of the region amplified and sequenced for viral phylogeny 3. RESULTS 3.1 A brief history of recent FMD virus type C occurrence In the last 20 years the type C virus has been restricted to four countries or geographic clusters of countries, namely: South America, The Philippines, South Asia and Eastern Africa (records of the FAO World reference Laboratory for FMD, Institute of Animal Health, Pirbright Laboratory, UK). Several factors constrain identification of the epidemiology of FMDV type C. These relate to the insensitivity of surveillance programmes combined with the dubious quality of laboratory diagnostics in some countries especially in the relatively remote areas which constitute foci of type C occurrence and related to this the paucity of type C viruses or samples presented in a timely manner for confirmation and characterisation. The comments which follow must be viewed in the light of this constraint. 3.2 The South American Focus Type C FMD viruses were rapidly identified from South America after their identification in Europe having spread through livestock movements and through the use of vaccines. Two serogroups were relatively widespread and prevalent in South America in the 1950s to 1970s. The most persistent was the sero-group designated C3 but related viruses were last detected in Argentina in 1993. An indication of the results of phylogenetic analysis is given in Figure 2. It is important to note the clustering of South American type C FMD viruses around two strains extensively used as vaccines i.e. C3/Resende/BRA/55 [AY593807] and C3/Indaial/BRA/71 (78) [M90376] and the close relationship between C3 Resende and C3 Philippines. Notwithstanding the fact that the formerly overtly-persisting type C FMD viruses were not seen after 1993, an outbreak occurred in September 2004 in buffaloes and cattle on an island in the Amazon River in the Municipality of Careiro da Várzea, Amazonas, Brazil (FAO, 2005). The area concerned was within a routinely-vaccinated (trivalent A, O and C vaccine) zone of the country. According to PANAFTOSA, VP1 gene sequencing led to the conclusion that the virus was “historically endogenous to the continent”, “... it was not possible to establish a close relationship with any of the isolates of the PANAFTOSA-PAHO/WHO data bank (maximum homology of 89 per cent)” and “... comparison between the isolates C3/Careiro and C3/Indaial/BRA/71 (vaccines strain) showed a genetic difference of 13 per cent in the region studied. This result precludes the hypothesis that the virus resulted from an escape from the vaccine industry.” Thus, although the virus groups within the C3 clade, it is a distinct virus and its provenance remains unknown. One must ask not only “Where did the virus originate?” but also “Where has it been until now?” This is clearly an enigma which unless it is resolved casts doubt on any claims that the type C virus has been eliminated. Clearly the answer will relate to the statements “Amazon region is in the northern livestock crescent where the animal health and veterinary delivery systems are not as effective as those in other parts of the country. The area as a whole depends on the import of animals and meat from other parts of the country.” (FAO 2005).

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

C1 Europe C2 Europe/SA C/Leticia/COL/67 C/Leticia/COL/70 C/General Roca/Cordoba/ARG/02/93 (AJ3062 C/General Villegas/BA/ARG/93 (AJ306217) C/San Cristobal/SF/ARG/94 (AJ306218) C3/Chaco/PAR/74 C3/Sao Jose dos Campos/BRA/72 C3/Goias/88 C3/Indaial/BRA/71 (K01202) C3/Indaial/BRA/71 (AY593806) C3/Santa Fe/ARG/75 C3/Alegrete/BRA/82 C3 C3/Indaial/BRA/71(78) (M90376) C3/Indaial/BRA/71 (J02184) C/ANG/3/73 C3/San Antonio de Giles/ARG/92 (AJ308704 C/Salto/BA/ARG/91 (AJ308703) C/Rivadavia/BA/ARG/93 (AJ306216) C3/ARG/85a (M19762) C/General Lamadrid/ARG/93 (AJ306213) C3/Cordoba/ARG/85 (L29062) C3/ARG/85b (AJ007347) C3/ARG/83a C3/Resende/55 C3/ARG/83b C3/Resende/BRA/55 (AY593807) C/ISR/4/70 C/LEB/3/69 C/PHI/7/76 C3 Resende-like & C Phil C/PHI/1/79 C3/ARG/84 (M19761) C5/ARG/69a C5/ARG/69 (AY593809) C5/BEL/1/69 C/GRE/1/70 C3/PAR/69

I ndaial-like

C Phil C/UKG/149/34 (AY593810)

AFRI CA ASIA C/GER/c.26 (CGC) C1/GER/c.26 (CGC) (M90368) 0.02

Figure 2: Phylogeny of Type C FMD viruses found in South America. 3.2 The Philippines Focus The close relationship which existed earlier between the swine and animal health industries in South America and The Philippines, with transfer of animals and vaccines, are believed to have provided the means for the movement of C3 Resende-like type C viruses to The Philippines in the 1970s or earlier. The viruses display an evolutionary continuum over more than 10 years from 1984 to 1994 explicable as the evolution of a single clade of virus (Figure 3). This is quite remarkable because from 1991 to 1994 FMD (not just type C FMD) was not detected in The Philippines despite intensive active surveillance including sero-surveillance being conducted in 1993/4 (P.L. Roeder, personal observation). When FMD was again detected in Quezon City of Manila in September 1994 it was identified as belonging to type C. However, it was only the first case detected which yielded type C FMD virus, the epidemic which evolved to engulf the island of Luzon for the next 10 years was, with three exceptions, always type O belonging to the “CATHAY” topotype. The three exceptions were in Cavite, close to Manila, in June 1995 and Bulacan, far south of Manila in January and March 1995. The relationship, if any, between these outbreaks could not be determined; unfortunately these viruses were not subjected to phylogenetic characterisation. The Philippines veterinary authorities have pursued FMD elimination vigorously for more than 14 years and in recent years have conducted extensive serological studies to support an application for accreditation of freedom from FMD without finding any evidence of FMD virus circulation. Once again we are faced with an enigma – that of where this virus had been in the period between its last sighting in 1991 and its reappearance in 1994; it clearly was not re-introduced from vaccine.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

C 1 E u ro pe C 2 E u ro pe / S A C C o lom bia C 3 I nd a ia l-lik e C 3/AR G /8 3a C3 /R es en de/ 55 C 3/AR G /83 b C 3/R es ende /BR A/5 5 (AY593 807) C /I S R/ 4/70 C /LEB/ 3/69 C/ PH I/7/ 76 C 3 R e s e n de -lik e & C /PH I/1/ 79 C 3/AR G /84 (M197 61) C 5/ARG /6 9a C 5/ARG / 69 (AY593 809) C 5/BEL /1/6 9 C /G R E/1/7 0 C 3/PAR/ 69 C /PH I/ 4/85 C /P HI /6/8 9 C /PH I/ 7/8 4 VS C /PH I/ 7/84 a C /P H I/ 3/9 4 C /PH I/4/ 94 C /P H I/ 1/90 C/ PHI /11/ 89 C P hil C /PH I/3 /88 C /PH I/ 10/8 9 C /PH I/3 /87b C /P H I /3/8 7a C /P H I/ 1/88 C /P H I/ 4/88 C /P H I/5 /88 C/ UKG / 149/ 34 (AY593 810 )

C P h il

AF RI CA AS I A C /G ER /c. 26 (C G C ) C 1/ GER /c .26 (C G C) (M9 0368 ) 0 .0 2

Figure 3: phylogeny of the Philippines type C FMD viruses and their relationship to other C3 Resende-like viruses. 3.3 The Eastern African Focus Type C FMD virus was first recorded in sub-Saharan Africa in Kenya and Ethiopia in 1957. With the exception of an Angolan outbreak in 1973, clearly linked to South America phylogenetically, (see Figure 2; C/ANG/3/73), the early isolates show independent lineages with a common origin possibly representing a single introduction. All other isolates emanate from East African countries, primarily Kenya, and the Kenyan isolates are closely related to the vaccine strain K267/67. The virus was last detected in Ethiopia in 1983 and was clearly endemic in Kenya until 1988 but the later situation is uncertain. 3.4 The South Asia Focus The type C FMD virus was probably introduced into the sub-continent in 1950s through the use of vaccines prepared in Europe. All the later viruses from India are closely clustered but outbreaks have not been detected since 1996 (see Figure 5). Reports from Pakistan and Afghanistan of type C viruses are highly unlikely to be correct. Viruses from the other South Asian countries (Nepal, Bangladesh, Pakistan, Sri Lanka and Bhutan) reflect the situation in India. The observed situation is suggestive of a relationship between outbreaks and vaccine use but vaccine strains in use are needed to confirm this and the information is not available.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

E U R O-S A

C/UKG/149/34 (AY593810) C/KEN/1/2004 (K6/04) C/KEN/5/96 (K14/96) C/K221/83 (Kenya) C/KEN/32/70 (K267/67) C/ETH/1/71 C/ETH/6/2005 (1983) C/ETH/7/2005 (1983) C/UGA/1/71 C/UGA/35/71A C/UGA/18/70 C/UGA/19/70 C/UGA/20/70

AF R I CA

AS I A C/GER/c.26 (CGC) C1/GER/c.26 (CGC) (M90368) 0.02

Figure 4: Phylogeny of eastern African isolates of Type C FMD virus.

E U R O-S A

C/UKG/149/34 (AY593810)

AF R I CA C/N65/Tadjikis tan/USSR/67 (vac s tr) C/CEY/1/74 C/SRL/4/78 C/CEY/2/74 C/CEY/4/71 C/SRL/1/84 C/SRL/4/84 C/IND/3/83 C/KUW/2/82 C/IND/9/82 C/IND/7/76 C/IND/14/80 C/IND/12/82 C/IND/51/79 (IND/42/77*) C/IND/1/82 C/SAU/1/84 C/SAU/12/84 C/NEP/35/96 C/IND/63/96* (1991 IVRI) C/IND/147/93* (IVRI) C/IND/26/93* (IVRI) C/IND/146/93* (IVRI) C/NEP/21/93 C/NEP/1/94 C/NEP/10/93 C/IND/66/96* (1991 IVRI) C/NEP/124/90 C/IND/67/96* (1991 IVRI) C/IND/64/96* (1991 IVRI) C/IND/65/96* (1991 IVRI) C/IND/8/93* (1992 IVRI) C/BAN/1/92 C/BAN/2/92 C/BHU/10/91 C/IND/89/92* (1991 IVRI) C/BHU/7/91 C/IND/9/93* (1992 IVRI) C/IND/7/92* (1991 IVRI) C/IND/136/92* (IVRI)

AS I A

C/GER/c.26 (CGC) C1/GER/c.26 (CGC) (M90368) 0.02

Figure 5: The phylogeny of South Asian viruses related to the European/South American groups.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

4. CONCLUSIONS AND RECOMMENDATIONS 4.1 Feasibility of eradication There is no reason to suspect the presence of type C FMD virus outside the foci described. The virus could well have been eliminated from The Philippines, where it was last seen more than 13 years ago, and might have ceased to circulate in South Asia and East Africa, where it was last seen more than 10 years ago. The proximate analysis presented here suggests that latterly the persistence of virus in the reservoirs where it was occurring was related largely to vaccine use and less to livestock trade than it had been earlier. If type C vaccine use is discontinued in East Africa and South Asia it seems likely that the disease will not return to those areas. Similarly, the best way to determine what is happening in South America is to cease using type C FMD vaccine and intensify surveillance. Bearing in mind the analysis presented above, however tentative it might be at this stage, it is possible to examine the feasibility of eradication of type C FMD virus. It is axiomatic that an eradication programme is unlikely to succeed unless it is founded on a sound understanding of the epidemiology of the disease targeted. In the case of FMD type C and the areas in which it has occurred until recently, there are serious deficits in understanding which could compromise eradication. These are the issues relating to where the virus had been in Brazil and The Philippines prior to the disease reappearing in 1994 and 2004, respectively – and where it is now. Without significant improvements in surveillance and intensive efforts focussed at disclosing any remaining occult virus transmission uncertainty will remain. The global eradication of FMD virus type C is probably feasible but it requires a coordinated effort with a strong active surveillance and epidemiological component and economic justification. 4.2 Recommendations If a serious attempt is to be made to achieve verified eradication of type C FMD virus - or at least its elimination from three of its former reservoirs in South Asia, The Philippines and eastern Africa – the following actions appear to be essential: 1. Elaborate a sound justification for undertaking the exercise and a coordination mechanism. 2. Cease all use of type C FMD vaccine. 3. Maintain type C FMD vaccine bank(s) within an emergency preparedness programme to be deployed if type C FMD re-emerges. 4. Strengthen FMD surveillance in the foci of attention where the virus was last seen and include an active virological and serological search for type C FMD virus. 5. Strengthen the availability of laboratory diagnostic capability (antigen trapping and typing ELISA, solid phase blocking ELISAs, PCR) and access to Reference Laboratory services. 5. ACKNOWLEDGEMENTS We gratefully acknowledge the assistance of Kate Swabey, Beck Midgley, Paul Davies and Jemma Wadsworth (IAH Pirbright Laboratory) with the sequencing analysis and Ingrid Bergmann (PANAFTOSA) and Blesilda Verin (Philippine Animal Health Center, BAI) for valuable contributions of data. 6. REFERENCES [1] FAO 2005. EMPRES Bulletin 26 of 2004: Outbreaks of FMD type C in Amazonas, Brazil. [ftp://ftp.fao.org/docrep/fao/007/y5754e/y5754e02.pdf accessed October 2008] [2] Tamura, K., J. Dudley, M. Nei, and S. Kumar, 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24: 1596-1599. [3] Waldmann, O. and Trautwein, K, 1926. Experimentelle Untersuchungen über die Pluralität des Maul-und Klauenseuchevirus. Berl. Tierärztl. Wochenschrift. 42: 569-571.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 8

PROGRESS IN FMD CONTROL IN IRAN AND PERSPECTIVE ON LONGER TERM CONTROL IN THE REGION F. Geiger1, V.Otarod2* 2

1 FAO Consultant in charge of EuFMD project in Iran. National Coordinator of Project, Iran Veterinary Organization.

INTRODUCTION Comparison of FMD outbreaks in 2007 and 2006 shows a significant decrease in the number of outbreaks in 2007 although there was an epidemic because of Type O Pan Asia. (1626 outbreaks in 2006 and 1114 0utbreaks in 2007) The Condition continued in 2008 with stability in Type O Pan Asia but more decrease in the number of outbreaks in March and April. Meanwhile Type A seems to be more sporadic although need to be under special surveillance. There is no record of circulation of Type Asia1. RESULTS Data show that most of the outbreaks occurred in villages (monthly reports, active village clinical surveillance). Occurrence of the disease in other Epidemiological units is not too high, but is the matter of attention for control of disease. In the year 2007 the pattern of the disease is rather same as 2006 with concentration of disease mostly in the central provinces. Overall disease density analysis show better situation in 2007. Implementation of regular clinical investigation and circulating virus identification with a rapid outbreak reports through electronic reporting system linked to Georeferenced database (GIS_VET) helped Iran to improve the knowledge of disease by outbreak mapping, vaccination mapping, and field studies and have led to different prevention and response for better decision about the control of disease. In this respect decrease in the number of outbreaks may be due to: Implementation of more regular vaccination, biosecurity measures, and better knowledge of animal movements. In the case of an outbreak in units, provinces determine the control zone and vaccinate all the susceptible animals in 3 kilometer zone. (targeted vaccination) Provinces are allowed to use the vaccine only in the specified phases. Guidelines are prepared for provinces and recommend them regularly to implement the work plan precisely. Ongoing training. CONCLUSION Movement of the animals and animal products between the countries around Iran (export and import of animals which may be legal or illegal), and circulation of the different type of the FMD virus in the Central Asia and Middle East which are evidently in contact with the other countries in the Africa and Asia, and the geographical location of Iran which is critical in the region, required an effective gathering and sharing of information among those responsible for disease control in the region, and mutual co-operation in FMD surveillance should assist each of the countries involved to reduce the risk of FMD by earlier action following the alerts and early warning messages. Global roadmap for control and eradication of disease is a need which is well understood.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 9

PHYLOGENETIC STUDY OF FMDV ISOLATES COLLECTED IN RUSSIA, CIS-COUNTRIES AND MONGOLIA IN 2000-2007 A. Scherbakov*, A. Timina and V. Borisov Federal Centre for Animal Health, 600901, Yur’evets, Vladimir, Russia e-mail: ascherbakov@arriah.ru

1. INTRODUCTION In the last 8 years FMD outbreaks caused by the O, A and Asia 1 serotypes were detected in the post-USSR territory. FMDV-O was responsible for the outbreaks in Russia in 2000 and 2004, Armenia, Georgia and Uzbekistan in 2000, Tajikistan in 2001-2003, Kyrgyzstan in 2001 and 2007, Kazakhstan and Nagorno-Karabakh Republic in 2007. Serotype A virus was detected in Tajikistan in 2003, Armenia in 2006 and Kyrgyzstan in 2007. Asia 1 virus caused outbreaks in Georgia and Armenia in 2000-2001, Tajikistan, Kyrgyzstan and Uzbekistan in 2003-2004 and Russia in 20052006. In Mongolia FMD was reported in 2000, 2001 and 2004 (O serotype) and 2005 (Asia 1 serotype). To understand the putative origin of these emergencies phylogenetic study of the viruses isolated during the outbreaks was undertaken. 2. MATERIALS AND METHODS Complete VP1 sequences were determined following viral RNA amplification by RT-PCR and phylogenetic analysis was carried out using the Neighbor-joining algorithm. 3. RESULTS The complete VP1 gene sequences of 19 FMD type O, 3 type A and 16 type Asia1 viruses were determined and compared with the GenBank database. The phylogenetic analysis placed all but one studied FMDV-O isolates in the PanAsia genetic lineage (Middle East–South Asia topotype). Isolates collected in 2000-2004 belonged to PanAsia1 strain and those collected in 2007 to PanAsia2. O/Mongolia/2004 isolate belonged to the South-East Asia topotype and was closely related to the isolate HLJOC12/03 (DQ119643) collected in China in 2003 (fig.1). All three isolates of serotype A (A/Tajikistan/2003, A/Armenia/2006 and A/Kyrgyzstan/2007) presented the A/Iran/2005 genetic lineage (fig.2). Three different sublineages of Asia 1 serotype caused FMD outbreaks in the post-USSR territory in 2000-2006 (fig.3). The strain detected in Iran, Turkey and Greece in 1999-2000 was found in Armenia and Georgia in 2000-2001. Viruses collected in Tajikistan, Uzbekistan and Kyrgyzstan in 2003-2004 were closely related to Asia 1 isolates from Pakistan. Isolates responsible for Asia 1 outbreaks in the Far East of Russia and Mongolia in 2005-2006 had 99-100% VP1 gene homology with the virus spread in the mainland China in 2005. 4. DISCUSSION Phylogenetic studies showed that the FMDV isolates responsible for the outbreaks in Russia, CIScountries and Mongolia were very closely related to the viruses previously detected in China, Pakistan, Iran and Turkey. These findings prove that the reappearance of FMDV in the post-USSR territory and Mongolia resulted from the virus introduction from neighboring Asian countries. 6. ACKNOWLEDGEMENTS

The work was supported by ISTC project No 2538.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Euro-SA

O1/BFS [J02185] O/Russia/1995

Cathay

TAW/81/97 [AJ296321] O/HKN/12/91 [AJ294921] O/PHI/5/95 [DQ164946] O/MYA/2/2000 [DQ164927]

SEA

HLJOC12/03 [DQ119643] O/Mongolia/2004 O1/Manisa [AJ251477]. O/Kazakhstan/1/2007 O/Kazakhstan/2/2007 O/Kazakhstan/3/2007 Pakistan [EF494481]

ME-SA

PanAsia-2

O/Kyrgyzstan/1/2007 O/Kyrgyzstan/2/2007 O/Nagorno-Karabakh/1/2007 O/Nagorno-Karabakh/2/2007

PanAsia

O/Russia/2004 O/Tajikistan/2003 O/Uzbekistan/2000

PanAsia-1

O/Kyrgyzstan/2001 O/Tajikistan/2001 O/Tajikistan/2002 O/Armenia/2000 O/Georgia/2000 O/Mongolia/2000

0.01

O/Mongolia/2001 O/CHA/2/99 [AJ318831] O/Russia/2000 O/JPN/2000 [AB050978]. UKG/10/2001 [AJ311722]

Figure 1: Neighbour-joining tree of foot-and-mouth disease virus serotype O based on a comparison of the complete VP1 gene.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Africa

A/KEN/29/2005 [EF208773] A/ARG/2001 [AM180023]

Euro-SA A24/Cruzeiro/BRA/55 [AJ251476] A/IRQ/24/64 [AJ251474]

Irq64 A/Uzbekistan/1989 A/Kyrgyzstan/1999 A22 550/USSR/65 (vaccine strain)

Asia

A/Georgia/1999

Irn96 A/IRN/1/96 [EF208771] A/Armenia/1998 A/TUR/45/01/1998 [DQ296542]

Irn99

A/IRN/22/99 [EF208772] A/TUR/382/06/1999 [DQ296550] A/IRN/7/2004 [EU414530]

Irn05

A/Kyrgyzstan/2007 A/Tadjikistan/2003 A/IRN/1/2005 [EF208769]

0.01

A/Armenia/2006 A/Turkey/1/2006

Figure 2: Neighbour-joining tree of foot-and-mouth disease virus serotype A based on a comparison of the complete VP1 gene.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Asia1-48/RUS/vaccine strain As1/Amursky/RUS/2006 As1/Jiangsu/China/2005 DQ156527 As1/Amursky/RUS/2005 As1/Chita/RUS/2006 As1/Khabarovsk/RUS/2005 As1/ Prymorsky/RUS/2005 As1/Mongolia/2005 As1/QH/China/2005 EF187272 As1/Beijing/China/2005 EF185303 As1/Gansu/China/2005 EF185304 As1/IND/15/1981 DQ121117 As1/IND/18/1980 DQ121116 As1/Shamir/ vaccine strain As1/IND114-04/2003 DQ101239 As1/IND175-04/2004 DQ101237 As1/IRN/58/99 DQ121122 As1/Armenia/2000 As1/TUR/10/99 DQ121131 As1/Georgia/2000 As1/Georgia/2001 As1/GRE/2/2000 DQ121113 As1/PAK/30/2002 DQ121124 As1/PAK/1/2004 DQ121128 As1/PAK/69/2003 DQ121127 As1/Uzbekistan/2003 As1/Tajikistan/1/2004

0.01

As1/Tajikistan/2/2004 As1/Kyrgyzstan/2/2004 As1/Kyrgyzstan/1/2004 As1/Tajikistan/1/2003 As1/Tajikistan/2/2003

Figure 3: Neighbour-joining tree of foot-and-mouth disease virus Asia 1 serotype based on a comparison of the complete VP1 gene.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 10

CONSTRAINTS AND OPPORTUNITIES FOR THE CONTROL OF FOOT AND MOUTH DISEASE IN INDIA D. Hemadri1, A. Sanyal1, RP. Tamil Selvan1, S. Saravanan1, JK. Mohapatra1, Lalkrishna2, KM. Bujarbaruah2 and B. Pattnaik1 1

Project Directorate on Foot and Mouth Disease, Mukteswar-Kumaon, Nainital-263 138 India 2 Indian Council of Agricultural Research, Krishi Bhavan, New Delhi-110 001, India

SUMMARY Foot-and-mouth disease (FMD) is endemic in India and occurs round the year with higher incidences being recorded in the winter months and early monsoon. About 4000-5000 outbreaks are reported every year, which an underestimate. Serotypes, O, A and Asia 1 are responsible for the outbreaks and add to the already complicated disease situation due to free movement of animals within the country. The economic losses caused by the disease are huge and according to some studies, the direct losses alone could be more than 2 billion US dollars per year. Agriculture, which was major contributor to the Gross Domestic Product (GDP) in the fifties, occupies a less dominant position now, due to the increased growth of industrial and service sectors. At the same time, contribution of livestock sector to the total GDP has remained steady and its contribution to the agriculture GDP has gone up from 13.8 % in 1981 to 23.8% in 2002 -2003 and thus has potential for further growth. In order to realize this potential, India needs to conquer many infectious diseases of livestock including FMD. Results of the pilot scale control programme involving 54 districts initiated in the 10th plan period (2002-2007) showed a lower incidence of FMD in the areas where it was operational. Though desirable, taking up country wide FMD control programme may not be feasible at the moment due to many constraints; however, there exists a prospect to launch it in a phased manner. This paper discusses those constraints and opportunities for the control of FMD in India. 1. INTRODUCTION Foot-and-mouth disease (FMD) is a major concern to both FMD-free countries as well as to countries not free from it. Apart from its impact on animal health, FMD has been described as the most important constraint to international trade in animals and animal products, which restricts trade in a south-to-north direction (1). These authors have shown that the world distribution of this one disease is almost a mirror image of the world-wide global economic structure with the highincome, industrialised countries being generally free from FMD, while the disease is persistently endemic in low-income countries suffering from food deficits. A recent Department for International Development (DFID)-ILRI study has ranked FMD as the disease that most affects livestock productivity of small-scale farmers, especially in Asia (2) while a consultative document by the FAO has identified FMD as one of the key impediments to the access of poor livestock farmers to formal markets (3). In India, FMD is endemic and occurs round the year with higher incidences being recorded in the winter months and early monsoon. About 4000-5000 outbreaks are reported every year, which could be an underestimate. Serotypes, O, A and Asia 1 are responsible for the outbreaks and add to the already complicated disease situation due to free movement of animals within the country. The economic losses caused by the disease are huge and according to some studies, the direct losses alone could be more than 2 billion US dollars per year. Agriculture, which was major contributor to the Gross Domestic Product (GDP) in the fifties, occupies a less dominant position now, due to the increased growth of industrial and service sectors. At the same time, contribution of livestock sector to the total GDP has been increasing and its contribution to the agriculture GDP has gone up from 13.8 % in 1981 to 23.8% in 2002 -2003 (source: CSO, MoSPI, GOI) and thus has potential for further growth. In order to realize this potential, India needs to conquer many infectious diseases of livestock including FMD. Results of the pilot scale control programme involving 54 districts initiated in the 10th plan period (2002-2007) showed a lower incidence of FMD

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

in the areas where it was operational. Though desirable, taking up country wide FMD control programme may not be feasible at the moment due to many constraints; however, there exists a prospect to launch it in a phased manner. This paper looks at the constraints and opportunities for the control of FMD in India. 2. OPPORTUNITIES 2.1 Availability of large network of laboratories Project Directorate on FMD, which lends technical support to Government of India in decision and policy making, has a network of laboratories spread across India. These laboratories totalling 23, located in almost every state, are primarily involved in collection of epidemiological data, clinical samples and laboratory diagnosis besides advising on the control measures to be undertaken. Availability of such laboratories is a boon considering that any suspected outbreaks can be attended and reported to as quickly as possible and control measures can be put in place. Fig 1 shows the location and distribution of network of FMD laboratories in India. 2.2 Large peninsular region India has the advantage of being a large peninsula and it is worth mentioning that India's coast is 7,517 kilometres (4,671 miles) long; of this distance, 5,423 kilometres (3,370 mi) belong to peninsular India. Considering that FMD is trans-boundary in nature, Southern India (Deccan Peninsula), covered by sea/ocean in all directions except north, is an ideal geographical region to initiate the control programme. 2.3 Availability of large base line data The Project Directorate on FMD, which was set up in 1968, as All India Co-ordinated Research Project for FMD virus serotyping, has over the years generated a large pool of data on disease prevalence and distribution of serotypes etc. This information along with antigenic, genetic and vaccine matching analysis has added to our knowledge on the antigenic and genetic spectrum/nature of the circulating filed strains and the suitability/relevance of the current vaccine strains. 2.4 One country, one strain Currently, trivalent vaccine consisting of serotypes O, A and Asia 1 is manufactured in the country. Molecular epidemiology studies carried out at PD-FMD indicates that field strains (within each serotype) belonging to a single genotype are circulating in the recent years. Interestingly, cocirculation of two different genotypes, particularly in serotype A, which was a common feature just a few years ago, has been not observed in the past 4-5 years. Antigenic analysis shows that, barring in type A, most of the field strains are related to the vaccine strains. In serotype A, a change of vaccine strain has been recommended recently and will replace the old strain in any vaccine batch prepared from December 2008. Use of single strain per serotype is advantageous as it helps in cutting down the cost of the vaccine as well as the dose volume. Further it saves labour on the production of vaccine containing different strains for different regions. 2.5 Availability of companion diagnostic tests When it comes to diagnosing the disease, foot-and-mouth disease in large animals, seldom poses confusion in the minds of veterinary clinicians in India. However, clinical diagnosis in sheep and goats is difficult because of the transient appearance of lesions and their similarity to those caused by other common diseases of small ruminants. In this context, one has to rely on laboratory diagnosis. Recently, one step pen-side diagnostic test has been developed and validated with some success during the 2007 outbreak in UK. A similar test is under development at PD-FMD besides a sandwich ELISA that is routinely used for serotyping (4). Availability of molecular diagnostics such as PCR could be useful in situations where the clinical material is insufficient or the amount of antigen present is not sufficient to be detectable by routine tests such as ELISA. Recently a multiplex PCR (5) for the serotyping of Indian FMDV strains has been developed and is being used as a back up test for ELISA. Seromonitoring of the vaccinated animals is done by indigenously developed liquid phase blocking ELISA (LPBE). Differentiating infected from vaccinated animals becomes imminent once the disease incidence comes down. Options in this area include detecting antibodies to non structural proteins, particularly 3ABC and 2C. An indirect ELISA and electro immuno transfer blot (EITB) are being used world over for the detection NSP antibodies, and at PD-FMD, an indirect ELISA employing recombinant 3AB3 is being validated for the said purpose. Besides vaccine matching exercises are routinely carried out at PD-FMD in order to assess the relevance of the vaccine strains to the field situation.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.6 Large pool of qualified manpower Veterinary services in India are primarily are state funded and there are about 9000 veterinary hospitals/polyclinics and 18000 veterinary dispensaries with more than 43,000 veterinarians and other supporting workers who provide the Veterinary services to the huge livestock population of the country (6). The field level veterinarians are responsible for the reporting of the diseases, collection of samples, data and executing the required control measures. Besides there are mobile disease surveillance/investigation teams established under veterinary colleges/state agricultural universities and also PD-FMD trained-manpower to undertake laboratory diagnosis. 2.7 Fairly strong political will As seen from the European and South American experience control of foot-and-mouth disease leading to ultimate eradication is a long drawn process with huge investments. Obviously a political will to sustain and support a control programme is a must for the fruitful outcome. In these lines, to prevent economic losses due to FMD and to build herd immunity in large ruminants, foot-andmouth disease control programme (FMD-CP) was initiated in 54 specified districts of the country in the tenth five year plan by Government of India (GOI). The programme involved 100% central funding towards the cost of vaccine, maintenance of cold chain and other logistic support to undertake vaccination. The State Governments provided other infrastructure and manpower. Encouraged by the results of this pilot scale programme, GOI is planning to expand this activity in a phase wise manner in the upcoming five year plans. 3. CONSTRAINTS 3.1 Lack of public awareness Though farmers in India recognize FMD quiet easily, the disease has never been taken seriously by them. The reason being FMD being non fatal and has not much perceivable impact on the milk production in low milk producing animals. Besides illiteracy and poverty make the situation much more complicated. Further, the inability of FMD vaccines to produce sterilizing immunity adds up the pressure on any vaccination programme being taken up, as the vets needs to answer some difficult questions. 3.2 Less than required quantity of vaccine production At present, India has three commercial producers of foot-and-mouth disease vaccine, namely, Indian Immunologicals Ltd., Intervet India Ltd. and Indian Veterinary Research Institute, Bengaluru. Products of M/S Brilliant Industries and Biovet Pvt. Ltd. are in the process of entering the market. Indian Immunologicals are the biggest producer of FMD vaccine in India and they produce about 120-150 million trivalent doses annually. Intervet and Indian Veterinary Research Institute together produce about 60-70 million trivalent doses. In all, there is about 180-220 million trivalent doses of FMD vaccine produced. This is less considering that the country has about 483 millions of FMD susceptible livestock population and the animals have to be vaccinated two times a year. In India, large ruminants (cattle and buffaloes) constitute nearly half of the total domestic livestock population and FMD is recorded mostly in these two species. Obviously, the present control strategy in the country has involved vaccinating these two species with the hopes to bring down disease incidences in small ruminants automatically. 3.3 Lack of strong legislation for animal movements and disease reporting For any disease control programme to be successful, movement of animals within and at the international borders needs to be strictly monitored and controlled. Although check posts exist at state borders in most of the states in India, they need to be strengthened with man power and quarantine facilities. Then a strong legislation requiring any animal entering the state to have vaccination certificates as well as to undergo quarantine may help restricting the disease at the entry level. The legislation should also encourage farming community report the disease immediately to the veterinary authorities, who in turn may be responsible for ordering movement restrictions to prevent further spread of the disease, quarantine, cordoning, disinfecting, ring vaccination and informing neighbouring districts and provinces immediately. It may be recalled that a farmer was convicted of failing to inform the authorities of a notifiable disease, and later of feeding his pigs "untreated waste" during 2001 UK outbreak. In a country like India, reporting can be encouraged by offering incentives rather than penalising. Recently, Government of Kerala has issued a notification amending the Kerala prevention and control of animal diseases rules 1968 providing compulsory vaccination against FMD and identification of vaccinated animals by ear tagging on cost basis (GO No.176/04/AD dt.24/8/2004). Further, the state also has issued the

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

orders to strengthen surveillance and checking FMD at State borders (GO (MS) No. 36/2004/AD dated 19/3/2004) and compulsory disease reporting. (Source: www. ahdkerala.gov.in) 3.4 Ban on cow slaughter Although FMD has been controlled in many countries in Europe and South America by systematic and repeated rounds of vaccinations, none have been successful in eradicating the disease without a stamping out policy, particularly when re-introductions occurred. This is mainly because, international trade regulation impose a heavy penalty on the use of vaccine against FMD in the form of import/export restrictions of animals and animal products and by practicing stamping out policy, one can hope to regain disease free status quickly, there by reducing the trade and other direct losses due to the disease. It can be recalled that in 1981, Chile was the first country in South America to be declared officially free of FMD. The eradication strategy was based on the gradual elimination of the disease working in a south-to-north direction; applying quarantine measures across different regions of the country and following a vaccination campaign using aqueous vaccines of guaranteed quality, mainly produced in Uruguay. The country suffered two reintroductions of FMD caused by illegal animal movements: one in March 1984 and the second one in March 1987. Both episodes were eradicated by stamping-out and quarantine measures. Chile has been recognized by OIE as free of FMD since 1988 (7). In India, at present, Kerala, Arunachal Pradesh, Meghalaya, Mizoram, Nagaland and Tripura are the only states that permit cow slaughter. This is due to the fact that cow is considered sacred in Hinduism and worshipped for the various graces she bestows on humanity. Although, not very pertinent at the present day FMD situation in India, ban on cow slaughter may become an issue, once the disease incidence comes down to zero level. 3.5 Porous international borders Illegal trafficking of livestock and livestock products has been the biggest threat for maintaining a disease free status. Similarly, 2001 FMD outbreak in UK has been attributed to feeding of infected swill. Chile suffered two re-introductions of FMD caused by illegal animal movements: one in March 1984 and the second one in March 1987(7). There was one outbreak in Botswana in 2002 in cattle in a communal area near Francistown in the east of the country, apparently resulting from the illegal movement of carrier or subclinically infected cattle from Zimbabwe. Similarly, illegal trafficking of livestock and livestock products along India’s international borders takes place and may pose a serious threat in the years to come. In view of the above and considering that FMD is transboundary in nature concerted efforts by all the neighbouring countries may be required to attain full freedom from the disease. 4. CONCLUSIONS Foot-and-mouth disease is endemic in India and despite some constraints, efforts are being made to control this dreaded disease in a phased manner. However, it would be ideal if the countries within the Indian sub-continent take up the FMD control programme in unison to attain full freedom from the disease. 5. ACKNOWLEDGMENTS Authors wish to thank ICAR for finance and facilities. DH’s participation in EUFMD research group meet is supported by FAO. 6. REFERENCES [1] Rweyemamu M.M. & Leforban Y. (1999). – Foot-and-mouth disease and international development. Adv. Virus Res., 53, 111-126. [2] Perry B.D., Randolph T.F., McDermott J.J., Sones K.R. & Thornton P.K. (2002). – Investing in animal health research to alleviate poverty. International Livestock Research Institute (ILRI), Nairobi, 140 pp. [3] Food and Agriculture Organization (FAO) (2002). – Improved animal health for poverty reduction and sustainable livelihoods. FAO, Rome [4] Bhattacharya, S., Pattnaik, B., Venkataramanan, R., 1996. Development and application of a sandwich enzyme-linked immunosorbent assay (ELISA) for type identification of foot-andmouth disease (FMD) virus in direct field materials. Indian J. Anim. Sci. 66, 1201–1209.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[5] Giridharan P, Hemadri D, Tosh C, Sanyal A, Bandyopadhyay SK. (2005). Development and evaluation of a multiplex PCR for differentiation of foot-and-mouth disease virus strains native to India. J Virol Methods. 126:1-11. [6] Venkataramanan R, Hemadri D, Bandyopadhyay SK and Taneja VK (2006) Foot-andmouth Disease in India: Present status. Paper presented at a workshop on Global Roadmap for improving the tools to control foot-and-mouth disease in endemic settings. 29 Nov-1 Dec 2006, Agra, India. [7] Sutmoller P., Barteling S, Casas Olascoaga R, Sumption KJ (2003) Control and eradication of foot-and-mouth disease. Virus Research 91: 101-144.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 11

THE GLOBAL FMD RESEARCH ALLIANCE (GFRA)-AN R&D RESPONSE TO THE MAJOR CONSTRAINTS TO GLOBAL FMD CONTROL L. Rodriguez1, C.G. Gay1*, M. Jeggo2, S. Alexandersen3, P. Kitching3, B. Charleston4, D. Paton4, K. Mǿller5 and T.K. Nielsen5 1

The Agricultural Research Service, United States Department of Agriculture, United States. 2 Commonwealth Scientific and Industrial Research, Australia. 3 National Centre for Foreign Animal Diseases, Canada. 4 Institute for Animal Health, United Kingdom. 5 National Veterinary Institute of the Technical University of Denmark, Denmark.

INTRODUCTION A group of international animal health scientists1 met on Plum Island May 2008 to define the purpose and goals of the Global Foot-and-Mouth Disease Research Alliance (GFRA). The group agreed that the purpose of the GFRA should be to establish a coordinated global alliance of scientists to produce evidence and innovation that will enable the progressive control and eradication of Foot-and-mouth Disease (FMD). The group also agreed that the following five strategic goals should drive the work of the GFRA: 1) Facilitate research collaborations; 2) Conduct strategic research to better understand FMD; 3) Development of the next generation of control measures and strategies for their application; 4) Determine social and economic impacts of new generation of improved FMD control and 5) Provide evidence to inform development of policies for safe trade of animals and animal products in FMD endemic areas. DISCUSSION There are currently no research laboratories with the necessary critical mass and support structures to achieve the GFRA strategic goals. It is therefore imperative that laboratories worldwide with active FMD research programs work together to reach the critical mass needed to achieve the GFRA goals. Critical will be to establish research programs that will meet the needs of countries that are endemic for FMD and that are the most affected by the devastating economic impact of this disease. The current members of the GFRA have therefore agreed to the following action plan: 1) Identify partnership opportunities and promote funding of collaborative research projects; 2) Expand and coordinate the alliance; 3) Promote mechanisms and bring together the necessary experts to do gap analysis and set research priorities; 4) Organize and manage GFRA and related meetings including issues of sponsorship and 5) Seek funding for GFRA coordination activities. NEXT STEPS The GFRA will hold a workshop on the last day of the FAO/EU-FMD Erice conference to recruit new members, review current FMD research projects, solicit input on research priorities, and establish new strategic research collaborations to advance the progressive control and eradication of FMD.

C. Gay, L. Rodriguez, R. Moore, USDA/ARS-USA; J. Hammond, M. Jeggo, CSIRO-Australia; B. Charleston, D. Paton-Pirbright-United Kingdom; B. Perry, Nuffield Department of Clinical Medicine, University of Oxford, UNITED KINGDOM; P. Kitching-CFIA-Canada; A. Dekker-Central Veterinary Institute of WageningenNetherlands; K. DeClerq, CODA-CERVA-VAR-Belgium; S. Andersen, National Veterinary Institute-Denmark; R. Drummond, Food & Rural Affairs- United Kingdom.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 12 KEYNOTE: VACCINATION: OVERCOMING THE CONSTRAINTS TO ACHIEVING EFFECTIVE IMMUNITY RATES A. Dekker* Central Veterinary Institute of Wageningen University and Research Centre, P.O. Box 65, 8200 AB Lelystad, The Netherlands

1. INTRODUCTION Foot-and-mouth (FMD) vaccination has been applied world-wide to control FMD. Although vaccination alone is not capable of eliminating the disease, along with other control measures it has been shown that FMD can be eradicated. The history of FMD control in the Netherlands (Figure 1) shows that the Netherlands was only successful when our main trading partner, Germany, in 1966 also introduced mass vaccination of their cattle. This history also shows that eradication takes a long time.

100000

Number of cases

Start mass vaccination cattle

10000 1000 End vaccination campaign

100 10 1 2004 1999

1994 1989 1984

1979 1974 1969

1964 1959 1954

1949 1944 1939

1934 1929 1924

1919 1914 1909

Year Figure 1: Number of FMD cases in the Netherlands since 1909 The first vaccines were formaldehyde inactivated virus suspensions derived from infected cattle, but when it was possible to culture the virus in vitro and to analyse the virus for its immunogenicity it became clear that the whole virus was more immunogenic than subunits of the virus (T.R. Doel & W.K.T. Chong, 1996). This information is very essential for the development of new vaccines which are discussed in another session (2C). One of the biggest constraints in vaccination is the antigenic variation. Antigenic variation can be shown using cross-protection studies, serological tests with polyclonal sera, by testing isolates with monoclonal antibodies and by genome analysis. Different tests have different advantages and disadvantages, but this will be discussed in a separate session (2B). In this paper I will focus on the use of vaccines in the field the problems with duration of immunity, use of different adjuvants and the needs from an epidemiological point of view. 2. DURATION OF IMMUNITY

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

In many countries FMD vaccines are applied twice a year. Studies in countries that apply vaccination on a regular basis support the need for repeated vaccination (A.K. Sen & S.P. Nair, 1994; M.E.J. Woolhouse et al., 1996). There are however several reports that vaccines in the past induced a longer lasting immune response (T.R. Doel, 2005). Studies in the Netherlands showed that cattle were protected to transmission with homologous FMD virus for three years after three annual vaccinations (C. Terpstra et al., 1990). In fact we have come across one cow still positive for antibodies against type A and O, and negative for type Asia-1, 15 years after a single vaccination with trivalent FMD Frenkel vaccine. Figure 2 shows studies on Oil Emulsion vaccines that also show longer durations of immunity (A. Dekker, Terpstra, C., Barteling, S.J., 1992).

2.5

1.5

titre ( log)

Cattle

Pig Sheep 0.5

0 0

100

120

140

160

180

200

Days after vaccination Figure 2; Neutralising antibody titres after a single vaccination with O type FMD double oil emulsion vaccine in cattle (solid line), pigs (dashed line) and sheep (dotted line) (A. Dekker, Terpstra, C., Barteling, S.J., 1992). The previous observations were confirmed in later studies (P. Selman et al., 2006). The question remains, why does vaccine not always work that well in field situations. Doel (2005) suggests that the quality of the vaccine might be responsible. A high quality vaccine induces a higher immune response after initial vaccination which will consequently last longer. The quality of vaccines is difficult to measure, although recently a test has been developed that can measure antigen quality in formulated vaccines (see poster session). The quality of the vaccine depends on the amount of antigen in the vaccine, the stability of the antigen in the formulation and most important the quality of the adjuvant. First most vaccines were adjuvanted with aluminium hydroxide, but later saponin was added to aluminium hydroxide vaccine (A. Martins, 1971), or purified quil-A to oil vaccine (C. Xiao et al., 2007). Recent studies in pigs (P.L. Eblé et al., 2007) show that the immune response can be enhanced by using a four-fold dose, which indicates that improvement to adjuvants is still possible and should be studied more extensively. We also studied intradermal vaccination in pigs, which is reported later in this session. For consumers (mostly governments) of FMD vaccine it is difficult to assess the quality of the vaccine, therefore every batch of vaccine should be tested in at least 5 calves free of maternally derived antibodies in which serum samples are taken before vaccination and preferably at a weekly basis until 6 – 8 weeks after vaccination. Sufficient serum should be stored, that it can be used in comparative serological studies when new batches are procured or batches have to be tested for stability. It should however be acknowledged that some vaccines protect cattle with a lower amount of neutralising antibodies than other vaccines. We have indications that 12S from degraded 146S can induce neutralising antibodies to a good level, without protecting the animals. So serology should be used carefully in the interpretation of cross producer differences, but can be used to study stability of vaccines when results of one vaccine batch are compared after storage. In this context studies on in vitro assays to measure immunity that relates to protection are essential, but this specific issue is dealt with in session 2D.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

3. INTERFERENCE WITH MATERNALLY DERIVED ANTIBODIES Young animals receive FMD specific antibodies from their dams via the colostrum. These antibodies can interfere with immune response. Earlier studies have shown that the height of the maternally derived antibody titre is crucial for young animals to be able to respond to vaccination (I. Gomes, 1984). Maternally derived antibodies decline very quickly. Recent studies in calves and piglets show a half-life of 16 respectively 15 days. In piglets, however, the antibody titre not only declines due to degradation but mainly due to the growth of the animal. When compensation for the growth of the piglets and the dilution due to increase in blood volume the chemical and physical stability of maternal derived antibodies in piglets is much higher than in calves, in our experiments estimated half-life of 27 – 40 days depending on the growth data and method used for compensating. In old experiments even a longer half-life has been calculated (M.J. Francis & L. Black, 1984a, b). On farms where prophylactic vaccination is practised it is essential to re-vaccinate in time to have sufficient vaccination coverage on the farm. The right timing will depend on the height of the antibody titre at birth and that is correlated to the antibody titre of the dam. The antibody titre of the dam will depends on the quality of the vaccine, e.g. the type of adjuvant type of antigen (intratype heterologous antigen can more easily induce antibodies), and amount of antigen.

10log

VNT titre

2.0

1.5

1.0

0.5

Weeks post vaccination

Figure 3: Antibodies in calves vaccinated with Al(OH)3 saponine adjuvanted Frenkel vaccine (solid squares) and calves vaccinated with a similar vaccine using double oil emulsion formulation (S.J. Barteling et al., 1990) In calves born from cows vaccinated with Frenkel vaccine routinely used in the Netherlands calves were selected with different levels of maternally derived antibodies (Figure 3). In calves with titres below 1.0 (10log) a good antibody response is seen with both Al (OH) 3 saponin adjuvanted vaccine as well as with double oil emulsion (DOE) adjuvanted vaccine. However, in the calves with titres close to 1.0 (10log) only the DOE adjuvanted vaccine induced an antibody response (S.J. Barteling et al., 1990). Later studies in South-America confirmed this finding (E.J.A. Spath et al., 1995). In 2006 an outbreak in Turkey occurred with a strain that was better covered with old A22Iraq antigen. The dams had been vaccinated with A Iran 96 like vaccine and the question rose whether the maternally derived antibodies against A Iran 96 like vaccine would interfere with an A22Iraq vaccination. A similar situation was studied in our lab and it was shown that homologous A Turkey 14/98 vaccine could not induce antibodies in calves born from A Turkey 14/98 vaccinated dams, but A22Iraq vaccine could (see poster session). This shows that intratype heterologous vaccination can induce an antibody response and is a possible solution when an outbreak occurs in a area with prophylactic vaccination. It has been shown that the amount of antigen in a vaccine is well correlated with the probability of protection (T.W.F. Pay & P.J. Hingley, 1987). Increasing the dose of antigen in the vaccine will undoubtedly increase the efficacy, but antigen is often the most expensive part of the vaccine and not always stable. It has been shown that the dose-effect relation studies that the effect of adjuvant is much bigger than the effect of antigen (C. Stellmann et al., 1977). Improving the

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

adjuvant is often cheaper. The study in the Netherlands mentioned above also shows that changing the adjuvant can increase the response in cattle with maternally derived antibodies. It is essential to study the effect of maternally derived antibodies on the antibody response in newborn animals. If newborn animals are vaccinated to soon, the vaccine will not have an effect (E.J.A. Spath et al., 1995). But if the newborn animals are vaccinated to late the gap in immunity might leave the farm susceptible for introduction of FMD virus. In the poster session there will be an example of a study in piglets and during the oral session an example in sheep. Countries relying on prophylactic vaccination should study the immunity in piglets and calves on a regular basis. Vaccination strategies should be adapted if new vaccines induce higher or lower antibody titres. 4. EPIDEMIOLOGICAL NEEDS Epidemiological models will be discussed in session 5. But when looking at the needs for vaccination the OIE in 1972 defined "At least 70% immediate protection of cattle vaccinated with one dose of vaccine, with a probability of 95%". This definition has been the basis of vaccine potency tests, although the current definitions never followed the OIE definition exactly. Vaccine testing is part of a separate session, but 70% protection in a population would mean that if an infected individual is capable of infecting on average 3.3 susceptible individuals. If the 70% of those individuals are protected, then 30% of 3.3 susceptible animals become infected which is only 1. That means that in all cases transmission will most likely die out. So by setting this standard the OIE has assumed a reproduction ration Ro of 3.3. In groups of unvaccinated animals the estimates of reproduction ratio's in cattle (Ro = 6) and pigs (Ro = 40) are higher (P.L. Eblé et al., 2008; K. Orsel et al., 2007). So the potency of vaccines should even be higher. But in case of an outbreak the transmission between farms is important and not the within pen transmission, therefore 3 PD50/dose which always has been used in Europe which is estimated to protect 70 – 80% of the cattle was sufficient if other control measures were in place. When setting a goal of FMD eradication world-wide means that also transmission within groups of animals should be blocked. The stringent control measures that were applied after outbreaks in Europe are difficult to implement in other countries. We need therefore more potent vaccines, as mentioned earlier studies with a four-fold dose shows that there is still room for improvement also with inactivated vaccines, but knowing the constraints of the stability of these inactivated products, live vector vaccines are probably the only way to achieve world-wide eradication of FMD. 5. LITERATURE [1] Barteling, S. J., Van Maanen, C., Yadin, H. & Anemaet, D. A. J., 1990, A foot-and-mouth disease vaccine bank: purified inactivated antigen stored at ultra-low temperatures for the rapid preparation of double oil emulsion (D.O.E.) vaccines, in: European commission for the control of FMD, session of the research group of the standing technical committee, Lindholm. Denmark. [2] Dekker, A., Terpstra, C., Barteling, S.J., 1992, Antibody kinetics and protection against challenge of FMD vaccines formulated from stored antigen, in: European commission for the control of FMD, session of the research group of the standing technical committee, Berne, Switserland. [3] Doel, T. R., 2005, Natural and vaccine induced immunity to FMD, Foot-and-mouth disease virus Vol 288:103-131. [4] Doel, T. R. & Chong, W. K. T., 1996, Comparative immunogenicity of 146S, 75S and 12S particles of foot-and-mouth disease virus, Archives of Virology 73(2):185-191. [5] Eblé, P. L., Bouma, A., Weerdmeester, K., Stegeman, J. A. & Dekker, A., 2007, Serological and mucosal immune responses after vaccination and infection with FMDV in pigs, Vaccine 25(6):1043-1054. [6] Eblé, P. L., de Koeijer, A. A., de Jong, M. C. M., Engel, B. & Dekker, A., 2008, A metaanalysis quantifying transmission parameters of FMDV strain O Taiwan among non-vaccinated and vaccinated pigs, Preventive Veterinary Medicine 83(1):98-106. [7] Francis, M. J. & Black, L., 1984a, Effect of the sow vaccination regimen on the decay rate of maternally derived foot-and-mouth disease antibodies in piglets, Research in Veterinary Science 37(1):72-76. [8] Francis, M. J. & Black, L., 1984b, The effect of vaccination regimen on the transfer of footand-mouth disease antibodies from the sow to her piglets, Journal of Hygiene 93(1):123-131. [7] Gomes, I., 1984, Observations on the influence of colostral antibodies on the anamnestic response of calves revaccinated against foot-and-mouth disease, Boletín del Centro Panamericano de Fiebre Aftosa 49-50:19-27. [10] Martins, A., 1971, Vacunas antiaftosas hidroxido-sapinonadas inactivadas por el formol, Boletín del Centro Panamericano de Fiebre Aftosa 1:1-19.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[11] Orsel, K., de Jong, M. C. M., Bouma, A., Stegeman, J. A. & Dekker, A., 2007, The effect of vaccination on foot-and-mouth disease virus transmission among dairy cows, Vaccine 25(2):327335. [12] Pay, T. W. F. & Hingley, P. J., 1987, Correlation of 140S antigen dose with the serum neutralizing antibody response and the level of protection induced in cattle by foot-and-mouth disease vaccines, Vaccine 5(1):60-64 [13] Selman, P., Chénard, G. & Dekker, A., 2006, Cedivac-FMD; Duration of Immunity in cattle, sheep and pigs, in: European commission for the control of FMD, session of the research group of the standing technical committee, Paphos, Cyprus. [14] Sen, A. K. & Nair, S. P., 1994, Demonstration of antibodies against foot-and-mouth disease virus (FMDV) type O and Asia-1 in non-descriptive crossbred calves, Acta Virologica 38(3):169171. [15] Spath, E. J. A., Smitsaart, E., Casaro, A. P. E., Fondevila, N., Fernandez, F., Leunda, M. R., Compaired, D., Buffarini, M. & Pessi, H., 1995, Immune response of calves to foot-and-mouth disease virus vaccine emulsified with oil adjuvant. Strategies of vaccination, Vaccine 13(10):909914. [16] Stellmann, C., Terré, J., Favre, H., Brun, A. & Fontaine, J., 1977, Comparison of footand-mouth disease vaccine potency testing on cattle in terms of the nature of the diluent, Archives of Virology 54:61-74. [17] Terpstra, C., van Maanen, C. & van Bekkum, J. G., 1990, Endurance of immunity against foot-and-mouth disease in cattle after three consecutive annual vaccinations, Research in Veterinary Science 49(2):236-242. [18] Woolhouse, M. E. J., Haydon, D. T., Pearson, A. & Kitching, R. P., 1996, Failure of vaccination to prevent outbreaks of foot-and-mouth disease, Epidemiology and Infection 116(3):363-371. [19] Xiao, C., Rajput, Z. I. & Hu, S. H., 2007, Improvement of a commercial foot-and-mouth disease vaccine by supplement of Quil A, Vaccine 25(25):4795-4800.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 13 INTRADERMAL VACCINATION WITH 1/10 DOSE AGAINST FMDV PROTECTS PIGS AS WELL AGAINST CLINICAL DISEASE AND SUBCLINICAL VIRUS SHEDDING AS INTRAMUSCOLAR VACCINATION WITH A FULL DOSE. P. Eblé*, K. Weerdmeester, F. van Hemert-Kluitenberg and A. Dekker Central Veterinary Institute of Wageningen UR (CVI), P.O. Box 65, 8200 AB Lelystad, The Netherlands

ABSTRACT The aim of this study was to investigate whether intradermal (ID) vaccination against foot-andmouth disease (FMD) is suitable as an alternative for the usually used intramuscular (IM) route. We used vaccines containing a normal or 10-fold dose antigen and compared groups of pigs that were vaccinated ID with either 0.2ml or 4x0.2ml with groups of pigs that were vaccinated IM with a standard 2ml, 0.2ml or 4x0.2ml dose. As compared with the 2ml IM vaccinated pigs, the pigs vaccinated ID with 0.2ml of the same vaccine were equally protected against clinical disease and subclinical virus shedding. Moreover, although VN-titres at 28 days post vaccination (dpv) of the 0.2ml ID vaccinated group were lower as compared to the IM vaccinated 2ml group, logistic regression analysis of the virus neutralising antibody (VN) titres at 28 dpv, the number of virus shedding pigs and application method showed that the ID vaccinated pigs needed significantly lower VN-titres as compared to the IM vaccinated pigs in order to be protected against virus shedding. We conclude that the ID route might be used as an alternative for IM application of FMD vaccine. ID application might induce more efficient immunity against FMD and, moreover, because the dose required in the ID route is lower compared to the IM route, ID application may reduce the cost of FMD vaccination markedly. 1. INTRODUCTION Despite all effort to eradicate the disease, FMD is still present in large parts of the world. The disease is endemic in large parts of Africa, Asia and South America. In some endemic areas, vaccination is applied as tool to eradicate the disease, but costs of vaccination are high. After a successful vaccination program in Western Europe, the European Union adopted a non-prophylactic vaccination strategy in 1992. Since then, the control policy for outbreaks in the EU has been primarily based on ‘stamping out’, in combination with movement restrictions and hygienic measures. Emergency vaccination during an outbreak was usually not applied because of the adverse economic consequences of vaccination as compared to slaughter of infected and in-contact animals. Since the European FMD outbreak in 2001, in which huge numbers of animals were killed, the OIE and the EU (Anonymous, 2005; Anonymous, 2003) have amended their regulations, and the option to use emergency vaccination (as a vaccinate-to-live policy) during an outbreak of FMD, has become more favourable (Anonymous, 2005). All registered FMD emergency vaccines are based on inactivated virus particles in an adjuvant and usually are administered intramuscularly in doses of 2ml. If the dose used per animal could be reduced, the total number of animals that can be vaccinated from one formulated (emergency-) vaccine batch is higher which would be an advantage in an outbreak situation in the EU. Not only for emergency vaccination, but also for vaccination in endemic regions reduction of vaccine dose would have benefit because the costs per vaccine dose could then be reduced. Reduction of vaccine dose might be accomplished by using another application method such as intra-dermal (ID) vaccination. Studies with hepatitis B, rabies and influenza vaccines suggest that intradermal vaccination has potential greater immunogenicity, because the skin is populated with dendritic cells, which are efficient and potent antigen-presenting cells for induction of protective immunity, and can result in dose-sparing (Glenn and Kenney, 2006). Moreover, intradermal vaccination has also other advantages compared to intramuscular vaccination. Intradermal application of a vaccine is less painful than IM application, ID application reduces lesions of edible tissue and, when a needle-less device is used, iatrogenic transfer of blood related agents cannot take place. In the present study we used a needle-less device for intradermal vaccination of pigs against FMD. This

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

device is already applied e.g. for vaccination against Aujezsky's disease (Visser et al., 1994) and PRRS (Martelli et al., 2007). In the presented study, we compared the efficacy of ID vaccination of pigs against FMD with that of the usually used intramuscular route. 2. MATERIALS AND METHODS 2.1 Animals and experimental design Three animal experiments were performed using conventionally reared 6-week old piglets. In the first experiment, 12 piglets were randomly allocated to 4 groups of 3 pigs each. Pigs in group 1 were not vaccinated, pigs in group 2 were vaccinated intramuscularly (IM) at 28 days before inoculation (-28 dpi) with a single dose double-oil-in-water emulsion (DOE) vaccine that contained 3µg of O Taiwan (O TAW 3/97) 146S antigen per 2 ml dose (vaccine A). Pigs in group 3 were vaccinated with the same vaccine, applied intradermally (ID) with 0.2ml (0.3µg, 1/10 dose) at – 28 dpi using a needle-less device especially designed for this purpose (Intradermal Application Liquids (IDAL) Injector, Intervet International B.V., Boxmeer , The Netherlands). Pigs in group 4 were also vaccinated ID with 0.2ml vaccine, but with a vaccine that contained 10 times more antigen than usual (vaccine B) and thus received 3µg antigen, analogous to the IM vaccinated group. All vaccinations were given in the neck of the left-hand side of the pig. Before challenge, the pigs were moved to the high-containment unit and the groups were housed in separate pens. After an acclimatization period, challenge of the pigs was performed at 28 days post vaccination by intradermal inoculation in the bulb of the heel of the left hind-foot with 0.1 ml of FMD virus type O TAW 3/97 containing 105 TCID50/ml. In the second experiment, 25 pigs were randomly allocated to 5 groups of 5 pigs each. Pigs in group 1 were not vaccinated, pigs in group 2 were vaccinated IM with 2 ml of DOE vaccine that contained 3µg of FMDV O Taiwan 146S antigen per 2ml dose (vaccine C), pigs in group 3 were vaccinated IM with 2 ml of DOE vaccine that contained 30 µg of FMDV O Taiwan 146S antigen per 2ml dose (vaccine D), pigs in group 4 were vaccinated ID with 0.2 ml of vaccine C and thus received 0.3µg 146S antigen and pigs in group 5 were vaccinated ID with 0.2 ml of vaccine D and thus received 3µg 146S antigen. Housing, inoculation etc. were identical as in the first experiment. The experiment was ended at 7dpi. In the third experiment, 48 pigs were randomly allocated to 6 groups of 8 pigs each. Pigs in group 1 received 0.2ml PBS ID (non-vaccinated group), pigs in group 2 were vaccinated IM with 2 ml of DOE vaccine that contained 3 µg of O Taiwan 146S antigen per 2ml dose (vaccine E); pigs in group 3 were vaccinated IM with 0.2 ml of vaccine E and thus received 0.3µg 146S antigen; pigs in group 4 were vaccinated IM with 4x0.2 ml of vaccine E and thus received 1.2µg 146S antigen; pigs in group 5 were vaccinated ID with 0.2 ml of vaccine E and thus received 0.3µg 146S antigen and pigs in group 6 were vaccinated ID with 4x0.2 ml of the vaccine and thus received 1.2µg 146S antigen. The pigs that were vaccinated four times were vaccinated four times in the neck of the left-hand side of the pig. Of each group, 3 pigs were removed for other research purposes at 1, 3 and 7 dpv respectively. At 28 dpv, the remaining 5 pigs of each group were challenged with FMDV. Housing, inoculation etc. were identical as in the first experiment. The experiment was ended at 14dpi. 2.2 Data collection and sample preparation After vaccination, the pigs were inspected daily (exp. 1, 2 and 3) and the location at which the vaccination was given was examined every day for local reactions (exp. 3 only). Serum samples were collected twice a week. After challenge, clinical signs (rectal temperature and vesicle score for feet, nose and mouth) of the pigs were recorded daily. OPF was collected daily after challenge from 0-7 dpi (exp. 1 and 2) or 0-14 dpi (exp. 3) using cotton mouth swabs. In the laboratory, the swabs were incubated for 30 minutes in 4 ml EMEM and then centrifuged. Half of each sample was stored at –20°C until ELISAs were performed. To the other part of each sample 5% FBS and 10% antibiotics was added and stored at -70°C for virus isolation and RT-PCR. Serum samples were centrifuged and serum was stored at –20°C.

At the end of the experiment 3, examination of the location of vaccination was performed. 2.3 Laboratory tests Challenge virus and OPF samples were assayed for the presence of virus by plaque titration on monolayers of secondary pig-kidney cells (De Leeuw et al., 1979). Virus titres were expressed as 10 log plaque forming units (pfu) per ml.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Neutralising antibody titres (VN-titres) against FMDV O Taiwan in serum samples were measured using the neutralisation assay as described previously (De Leeuw et al., 1979). End-point titres were calculated as the reciprocal of the final serum dilution that neutralised 100 TCID50 of FMDV in 50% of the wells. The titres after infection where compared with the titres at the day of challenge. Antibodies against non-structural proteins of FMDV in serum samples were determined using a commercially available ELISA (Ceditest® FMDV-NS) used according to the instructions of the manufacturer. 2.4 Statistical methods The results of the experimental groups with the same treatment in experiments 1, 2 and 3 were pooled. Differences in number of pigs with generalized FMD and virus excretion between the nonvaccinated and all the vaccinated groups, the IM 3 µg vaccinated and the other vaccinated groups and between the ID 0.3 µg vaccinated and the other vaccinated groups were statistically analysed using the Fisher Exact test (StatXact®-5). Differences in VN-titres at 28 days post vaccination between the non-vaccinated and all vaccinated groups, the IM 3 µg vaccinated and the other vaccinated groups and the ID 0.3 µg and the other vaccinated groups were statistically analysed using the non-parametric Kruskal Wallis test (StatXact®-5). A non-parametric permutation test (StatXact®-5) was used for pair-wise comparison between groups if the Kruskal-Wallis test gave a significant result. Using logistic regression (R for Windows, version 2.6.2) we examined whether or not application method (IM versus ID) had influence on the relation between VN-titre and protection against virus shedding. For this, of all vaccinated pigs, VN-titres at 28 dpv, application method (IM or ID) and virus shedding (+ or -) were included in the model. First, the relation between VN-titre and protection against virus shedding was modelled. Then, the models with and without inclusion of application method were compared using a likelihood ratio test. All significance levels were set at p<0.05. 3. RESULTS 3.1 Clinical signs and virus shedding After vaccination, no systemic reactions were observed in any of the vaccinated pigs. In the IM vaccinated pigs no local reactions could be seen at the location of vaccination. In the ID vaccinated pigs, a swelling at the location of vaccination could be observed with a diameter that differed from approximately 2-5cm. After challenge, all pigs of the non-vaccinated groups showed signs of generalized FMD (vesicles at another site than the inoculated foot), all but one showed fever and all shed virus from day 1-2 post inoculation. None of the pigs vaccinated IM with the standard 3 µg dose developed generalized FMD after challenge, although in some of the pigs vesicles at the inoculated foot were observed. Two pigs shed virus, although virus shedding was brief and titres were low as compared to the non-vaccinated groups. All pigs IM vaccinated with the vaccine that contained 10 times more antigen than usual were clinically protected against challenge and did not shed virus. In the group vaccinated IM with a 1/10 dose, also in none of the pigs generalized FMD was observed after challenge, although in one pig vesicles at the inoculated foot and fever was observed. Three pigs of this group shed virus after challenge. The pigs that were vaccinated IM four times with a 1/10 dose all were protected against challenge and no virus shedding was observed in this group. Of the intradermally vaccinated pigs with the 0.3 µg dose, in three pigs lesions on the inoculated foot were observed and one pig shed virus subclinically. In one pig, generalized FMD was observed although no virus was detected in OPF samples from this pig. Also in the group intradermally vaccinated with the vaccine that contained 10 times more antigen than usual (ID 3 µg vaccinated pigs) one pig showed generalized FMD without virus shedding. In this group, in five pigs lesions at the inoculated foot were observed of which one pig also showed fever. Three pigs shed virus subclinically from which two only briefly and at low titres. Of the ID vaccinated group that was vaccinated four times with the 0.3µg, all pigs were protected against clinical disease and no virus shedding was observed in this group (Table 1). Table 1; Results of all experiments

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Non-vaccinated

IM vaccinated (3µg)

exp 1

Clinical signs 3/3

VI OPF 3/3

exp 2

5/5

<0.3

0/5a

exp 3 total

5/5 13/13

<0.3 <0.3

0/5b

exp 1

0/3

2.5

0/3

exp 2 exp 3 total

0/5 (2LH ) 0/5 0/13

2/5 0/5 2/13

2.1 2.3 2.2

0/5a 0/5

IM vaccinated (30µg)

exp 2

0/5

2.5

0/5a

IM vaccinated (0.3µg)

exp 3

0/5 (1LH)

3/5

1.8

1/5

IM vaccinated (4x0.3µg) exp 3

0/5

2.1

0/5

ID vaccinated (0.3µg)

exp 1

0/3

1.7

0/3

exp 2 exp 3 total

1/5 (2LH) 0/5 (1LH) 1/13

1/5 0/5 1/13

1.8 2.2 1.9

0/5 0/5

ID vaccinated (3µg)

experiment ended at 7 dpi;

exp 1

1/3 (2LH)

1/3

1.3

3/3

exp 2 total

0/5 (3LH) 1/8

2/5 3/8

1.4 1.4

0/5a

0/5

2.2

0/5

ID vaccinated (4x0.3µg) exp 3 a

VN-titer 28dpv NS-ELISA <0.3 3/3

early euthanasia;

left hindfoot

All vaccinated groups differed significantly from the non-vaccinated group (p<0.01) with regard to generalization of FMD. Between the IM 3 µg vaccinated groups and the other vaccinated groups, no significant differences could be detected (p>0.05). Also between the ID 0.3 µg vaccinated and the other vaccinated groups, no significant differences could be detected (p>0.05). Regarding the number of pigs that shed virus, all vaccinated groups differed significantly from the non-vaccinated group (p<0.01), except the IM 0.3µg vaccinated group (p=0.07). Between the IM 3 µg vaccinated groups and the other vaccinated groups, no significant differences could be detected (p>0.05). Between the ID 0.3 µg vaccinated and the other vaccinated groups, only the IM 0.3µg vaccinated group (p=0.04) differed significantly in the number of pigs shedding virus, but the other vaccinated groups did not (p>0.05). In the 3 µg IM vaccinated pigs, at the location of vaccination, intramuscularly, reactions to the vaccine of on average 5x4x3 cm that consisted of granulation tissue and necrosis were observed. Also the other IM vaccinated groups had similar reactions to the vaccine. In the ID vaccinated pigs, intradermally, reactions to the vaccine of on average 1-2 cm diameter consisting of granulation tissue and necrosis were observed. 3.2 Serological responses All vaccinated pigs developed neutralising antibodies against FMDV after vaccination. The mean virus neutralising antibody titre at 28 dpv of the vaccinated pigs is shown in Table 1. Increase of the antigen-payload (vaccines B and D) had a positive effect on the VN titres of the IM vaccinated groups, but resulted in no to almost the opposite effect for the ID vaccinated groups. Variation of the administered volume gave a dose-effect response in both the IM vaccinated groups (2ml>4x0.2ml>0.2ml) as the ID vaccinated groups (4x0.2ml>0.2ml). The titres of the IM vaccinated pigs were in general higher than those of the ID vaccinated pigs, except for the 4x0.3 µg ID vaccinated group that had a very good VN response (Figure 1). Statistically, at 28 days post vaccination the VN-titres of the pigs of all vaccinated groups differed significantly from the non-vaccinated group. The IM 3 µg vaccinated group differed significantly from the ID 3 µg (p=0.02) and the IM 0.3 µg vaccinated groups (p=0.01), but not from the other vaccinated groups. The ID 0.3 µg vaccinated group differed significantly from the IM 30 µg vaccinated group (p=0.04) and the ID 3 µg vaccinated group (p=0.02), but not from the other vaccinated groups.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

At a group level, a clear boost in VN-titre post challenge could be detected in the non-vaccinated and the ID 3 µg group of exp. 1 (Figure 1). VN-titre at time of challenge had a significant relation with protection against virus shedding (p<0.001). Moreover, the application method plus interaction between application method and VNtitre fitted significantly better than the model without inclusion of application method (p=0.002). Based on this statistical model it can be concluded that the IM vaccinated pigs needed significantly higher VN-titres as compared to the ID vaccinated pigs in order to be protected against virus shedding. Positive reactions in the NS-ELISA were seen in all pigs of the non-vaccinated and the ID 3 µg vaccinated group of exp.1. Of exp. 2, all pigs tested negative, probably due to early euthanasia of these pigs. In exp. 3 only pig 9761 of the IM 0.3 µg tested positive in the NS-ELISA (Table 1). Figure 1; VNT responses

VNT responses Exp. 1

VNT responses Exp. 2

3.5 3

2.5

1.5

non-vac ID 0.3µg

0.5

IM 3µg IM 30µg

1.5

ID 0.3µg ID 3µg

IM 3µg

non-vac

log VN-titre

3.5

0.5

ID 3µg 0

0 0

days post vaccination

VNT responses Exp. 3 3.5 3

non-vac 2

IM 3 µg IM 0.3 µg

1.5

IM 4x0.3 µg

log VN-titre

2.5

ID 0.3 µg ID 4x0.3 µg

0.5 0 0

days post vaccination

Group means of virus neutralising antibody responses 4. DISCUSSION We compared the efficacy of ID vaccination of pigs against FMD with the normally used intramuscular route at varying antigen doses. In the study presented in this paper, ID vaccination with 0.2ml vaccine was as effective as IM vaccination with the full 2ml dose. Effectiveness was established both as the number of pigs that were protected against generalization of FMD after challenge but also as number of pigs that were protected against (subclinical) virus shedding. ID inoculation has been investigated before as a means of vaccinating humans, laboratory animals and domestic farm animals (Hunsaker et al., 2001). For humans, effective ID vaccination is described for e.g. influenza (Auewarakul et al., 2007; Kenney et al., 2004; Belshe et al., 2004), rabies (Warrell et al., 1985; Warrell et al., 2008; Chutivongse et al., 1990) and hepatitis B (Karahocagil et al., 2006, Sangfelt et al., 2008) vaccination. Successful ID vaccination of pigs has been reported for Aujeszky's disease (Visser et al., 1994; Vanderpooten et al., 1997; Vannier and Cariolet, 1991; Mikulska-Skupien et al., 2004) and PRRS (Martelli et al., 2007). To our knowledge, our study is the first in which intra-dermal vaccination of pigs against FMD was assessed. ID

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

vaccination was effective even when a 1/10 vaccine dose was used. Using a lower vaccine dose will not only reduce the costs of vaccination but also could be essential to achieve a high vaccination coverage during an emergency vaccination in an FMD epidemic. Moreover, ID vaccination is less painful and reduces lesions in tissue that later might be used for consumption and, if delivered by a needle-less device, also avoids unintentional transfer of blood related agents. Therefore, ID vaccination might be a good alternative for the routinely used IM application of FMD vaccine. Even though we used a vaccine that is optimised for IM administration, ID vaccination with 1/10 vaccine dose resulted in a similar (slightly higher) VN-titre and was more successful for protection against virus shedding after challenge with FMDV as compared to IM vaccination with 1/10 vaccine dose. Therefore, ID application of FMD vaccine may result in a better immunological response as compared to IM vaccination with the same vaccine dose. Moreover, although VN-titres of the 0.2ml ID vaccinated group were lower as compared to the IM vaccinated 2ml group at time of challenge, protection against clinical disease and subclinical virus shedding of both groups were comparable and our results show that the ID vaccinated pigs needed significantly lower VN-titres as compared to the IM vaccinated pigs in order to be protected against virus shedding. In cattle, it is well established that protection against challenge at 3 weeks after vaccination is correlated with induced VN-antibody response caused by vaccination (Pay and Hingley, 1987). For pigs (Haas, 1999) this correlation is less clear but this is probably due to paucity of data. Recently, we demonstrated that in pigs’ reduction of virus shedding after challenge was correlated with vaccine induced VN-titres (Eble et al., 2007). However, the results of the present study show that after ID vaccination protection against challenge is accomplished at a lower VN-titre and thus suggest that ID application, as compared to IM application, might induce other immune mechanisms that contribute to protection. In the present study, we used several control groups in order to compare the vaccination methods with respect to total administered volume and/or antigen dose. For both IM and ID administration, for the vaccines that contained a normal dose antigen (vaccines A, C and E), a dose effect response was observed. Thus, the more total volume was administered, the better the neutralising antibody response. The vaccines with a 10-fold antigen payload induced a higher neutralising antibody response than the 'normal' vaccine when administered IM. Similar findings have been reported for cattle (Cox et al., 2006), sheep (Barnett et al., 2004) and pigs (Eble et al., 2007). However, if the vaccines that contained a 10-fold dose antigen were administered ID the results were relatively poor. This was probably due to the properties of the device that we used, which was optimised for use with adjuvated vaccine of the manufacturer. The vaccines (B, D) which we prepared with a 10-fold antigen load became very viscous because of the high protein content it contained. The IDAL device had difficulty in coping with these vaccines and probably delivered less than 0.2ml per dose, which is probably the cause of the disappointing results of these vaccines delivered ID. Although the pigs that were ID vaccinated with 1/10 of the normally used vaccine dose were equally protected against clinical disease and subclinical virus shedding as the pigs that were vaccinated IM with a 2 ml dose, still some of them showed clinical signs of FMDV (albeit merely locally at the site of challenge) and (subclinical) virus shedding after challenge. Therefore, the vaccination regime that we used could be further optimised. An improvement was already found in the group that was vaccinated ID 4x with 0.2ml. In this group, all pigs were completely protected against clinical signs and virus shedding and the mean VN-titre post vaccination was remarkably higher as compared to the 1x 0.2ml ID vaccinated group. As described previously, the height of VN-titre is correlated with protection against infection (Eble et al., 2007). In the 4x vaccinated groups, all vaccinations were given at the left side of the neck. Maybe less vaccination spots but separated from each other at locations with different draining lymph nodes would be as effective, as has been described for rabies vaccination (World Health Organisation, 2007). We conclude that ID vaccination with FMD vaccine seems to work very well in pigs. However, before ID vaccination against FMD can be used in field circumstances, more research should be carried out with respect to optimization of the used device, vaccine composition, administration strategy and efficacy in other species. 5. ACKNOWLEDGMENTS The authors wish to thank the laboratory assistants and staff of the animal isolation units that participated in the described research for their assistance. We thank Intervet International BV, Boxmeer, and The Netherlands for providing the IDAL device. This work was supported financially by the Netherlands Ministry of Agriculture, Nature and Food Quality. 6. REFERENCES

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[1] Anonymous. (2005) Chapter 2.2.10: Foot-and-mouth Disease. In: OIE Terrestrial Animal Health Code 15th edition 2005; Article 2.2.10.7. www.oie.int [2] Anonymous. 2003. Council Directive 2003/85/EC. In: Official Journal of the European Union; 46: L306. europa.eu.int [3] Auewarakul P, Kositanont U, Sornsathapornkul P, Tothong P, Kanyok R, Thongcharoen P. 2007. Antibody responses after dose-sparing intradermal influenza vaccination. Vaccine, 25(4):659-63. [4] Barnett PV, Keel P, Reid S, Armstrong RM, Statham RJ, Voyce C, Aggarwal N, Cox SJ. 2004. Evidence that high potency foot-and-mouth disease vaccine inhibits local virus replication and prevents the "carrier" state in sheep. Vaccine, 22(9-10): 1221-1232. [5] Belshe RB, Newman FK, Cannon J, Duane C, Treanor J, Van Hoecke C, et al. 2004. Serum antibody responses after intradermal vaccination against influenza. N Engl J Med, 351(22):2286-94. [6] Chutivongse S, Wilde H, Supich C, Baer GM, Fishbein DB. 1990. Postexposure prophylaxis for rabies with antiserum and intradermal vaccination. Lancet, 335(8694):896-8. [7] Cox SJ, Voyce C, Parida S, Reid SM, Hamblin PA, Hutchings G, Paton DJ, Barnett PV. 2006. Effect of emergency FMD vaccine antigen payload on protection, sub-clinical infection and persistence following direct contact challenge of cattle. Vaccine, 24(16):3184-90. [8] De Leeuw PW, Tiessink JWA, Frenkel S. 1979. Vaccination of pigs with formaldehydeinactivated aluminium hydroxide foot-and-mouth disease vaccines, potentiated with diethylaminoethyldextran (DEAE-D). Zentralbl Veterinarmed B, 26: 85-97. [9] Eble PL, Bouma A, Weerdmeester K, Stegeman JA, Dekker A. 2007. Serological and mucosal immune responses after vaccination and infection with FMDV in pigs. Vaccine, 25(6):104354. [10] Glenn GM, Kenney RT. 2006. Mass vaccination: solutions in the skin. Current topics in microbiology and immunology, 304:247-68. [11] Haas B. 1999. In vivo and in vitro testing of FMD vaccines for pigs. Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of FMD, Maison-Alfort, France, 29 September- 1 October 1999; 94-99. [12] Hunsaker BD, Perino LJ. 2001. Efficacy of intradermal vaccination. Vet Immunol Immunopathol, 79(1-2):1-13. [13] Karahocagil MK, Buzgan T, Irmak H, Karsen H, Akdeniz H, Akman N. 2006. Comparison of intramuscular and intradermal applications of hepatitis B vaccine in hemodialysis patients. Renal failure, 28(7):561-5. [14] Kenney RT, Frech SA, Muenz LR, Villar CP, Glenn GM. 2004. Dose sparing with intradermal injection of influenza vaccine. N Engl J Med, 351(22): 2295-301. [15] Martelli P, Cordioli P, Alborali LG, Gozio S, De Angelis E, Ferrari L, et al. 2007. Protection and immune response in pigs intradermally vaccinated against porcine reproductive and respiratory syndrome (PRRS) and subsequently exposed to a heterologous European (Italian cluster) field strain. Vaccine, 25(17):3400-8. [16] Mikulska-Skupien E, Szweda W, Procajlo Z, Platt-Samoraj A. 2004. Indices of nonspecific cellular immune response in pigs after intradermal vaccination with deleted Aujeszky's disease vaccine and after experimental infection. Bull Vet Inst Pulawy, 48(4):347-54. [17] Pay TW, Hingley PJ. 1987. Correlation of 140S antigen dose with the serum neutralising antibody response and the level of protection induced in cattle by foot-and-mouth disease vaccines. Vaccine, 5(1): 60-64. [18] Sangfelt P, Uhnoo I, Reichard O, Weiland O. 2008. A low-dose intradermal hepatitis B vaccine programme in health-care workers and students is highly effective and cost saving: a retrospective follow-up survey in the clinical setting. Scandinavian journal of gastroenterology, 43(4):465-72. [19] Vanderpooten A, Goddeeris B, De Roose P, Hendrickx L, Biront P, Desmettre P. 1997. Evaluation of parenteral vaccination methods with glycoproteins against Aujeszky's disease in pigs. Vet Microbiol, 55(1-4):81-9. [20] Vannier P, Cariolet R. 1991. Vaccination of pigs against Aujeszky's disease by the intradermal route using live attenuated and inactivated virus vaccines. Vet Microbiol, 26(1-2):1123. [21] Visser N, Egger W, Lutticken D. 1994. Intradermal application of Aujeszky's disease virus strain Begonia with tocopherol-based adjuvant and a novel design injection device. Acta veterinaria Hungarica, 42(2-3):413-8. [22] Warrell MJ, Nicholson KG, Warrell DA, Suntharasamai P, Chanthavanich P, Viravan C, et al. 1985. Economical multiple-site intradermal immunisation with human diploid-cell-strain vaccine is effective for post-exposure rabies prophylaxis. Lancet, 1(8437):1059-62.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[23] Warrell MJ, Riddell A, Yu LM, Phipps J, Diggle L, Bourhy H, et al. 2008. A simplified 4site economical intradermal post-exposure rabies vaccine regimen: a randomised controlled comparison with standard methods. PLoS neglected tropical diseases, 2(4):e224. [24] World Health Organization. 2007. Weekly epidemiological record 7 DECEMBER 2007. Rabies vaccines WHO position paper. http://www.who.int/wer/2007/wer8249_50.pdf

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 14 FOOT-AND-MOUTH DISEASE (FMD) VACCINATION STRATEGIES IN SHEEP AND LAMBS BY USING COMMERCIAL OIL VACCINE E.J.A. Späth1, B. Robiolo2, E.A. León3, J.A. Manazza1, S.J. Duffy3, J. Filippi4, A. Ham4, J. La Torre2 and E. Smitsaart4* 1

Animal Health Group. INTA. Ruta 226 km 73, 5; 7620 Balcarce, Prov. Buenos Aires, Argentina 2 CEVAN. Instituto de Ciencia y Tecnología Dr. César Milstein, CONICET. 3 CICVyA. INTA 4 Biogénesis Bagó S.A. Argentina

ABSTRACT In most countries in South America, FMD routine control and eradication programs require compulsory vaccination of the entire cattle population be vaccinated compulsory. However, in an emergency situation or when a buffer zone is being established, the vaccination of all susceptible species is essential. In this context, appropriate vaccination strategies in sheep and lambs with maternal derived antibodies (MDA) are needed in order to reach and maintain a satisfactory protection level in the herd. A commercial oil adjuvanted FMD vaccine (Biogénesis-Bagó, Argentina) was administered to seronegative adult sheep during breeding season. Three groups were revaccinated at 10, 20 and 30 days after primary vaccination (dpv), respectively. After lambing, lambs born from vaccinated sheep were grouped according to age and vaccinated at 30, 60 or 90 days of age using the same FMD vaccine. FMDV specific antibodies were examined up to one year after vaccination in adults and up to five months of age in lambs. One dose induced satisfactory antibody levels in adult sheep, which persisted up to 330 dpv. However, revaccinated groups had significantly higher antibody titres (P<0.05) than the singly vaccinated group from 40 to 120 dpv, regardless of the time the second dose was administered. In the absence of MDA, lambs responded as adults to FMD vaccination. In spite of the MDA interference, vaccination at young age induced a good immune response in the three vaccinated groups and protective antibody levels remained above log10 2.5, until the end of the study. Taken together these results indicate that a vaccination scheme consisting of one or two doses to adult sheep and a single dose to lambs at any age (with or without MDA) should be encouraged in order to achieve and maintain sufficient herd protection levels and prevent further spread of the disease. 1. INTRODUCTION In most countries in South America, FMD routine control and eradication programs require compulsory vaccination of the entire cattle population. However, in an emergency situation or when a buffer zone is being established, the vaccination of all susceptible species is mandatory. In this context, and in regions with high density of sheep, appropriate vaccination strategies are necessary in order to reach and maintain a satisfactory protection level in the herd. In addition, there is a need to demonstrate the efficacy and safety of vaccination in newborn ruminants in the presence of maternally derived antibodies (MDA) as they interfere to varying degrees with active immunization (Kitching and Salt, 1995, EMEA, 2007). Previous work had demonstrated that lambs are immunologically competent at an early age and no evidence of a blocking effect of MDA on vaccinal response was found (Cunliffe and Graves, 1970, Terpstra and Dekker, 1996). The efficacy of mineral oil adjuvanted vaccines in inducing protective immune responses in sheep has been well documented (Cox et al., 1999, Parida et. al, 2008, Selman et. al., 2006). Most of the experiments previously reported used either few animals, experimental vaccines specifically prepared and/or high potency vaccines (Patil et al., 2002, Nair & Sen, 1993, Cox et al., 2003). The objective of this study was to provide information to allow determination of the optimum vaccination strategy for sheep and lambs that are required in different risk situations and using an oil vaccine commercially available for systematic vaccination in cattle. 2. MATERIALS AND METHODS

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.1. Vaccines Two industrial batches of commercial polyvalent inactivated water-in-oil single emulsion vaccine containing FMD virus (FMDV) strains O1 Campos, A Arg 2000, A Arg 2001 and A24 Cruzeiro manufactured by Biogénesis Bagó S.A., Argentina (each batch size was 2500 000 cattle doses, Bioaftogen®) were used. In trial 1, a vaccine batch manufactured in 2004 was used, and in trial 2, one manufactured in 2005. The vaccine batches were controlled by the Argentine National Animal Health Authorities and complied with requirements for efficacy, safety, purity related to non structural protein (NSP) antibodies and stability before released to the market (SENASA, Resolution 351/06). 2.2. Animals and experimental design A Corriedale sheep flock of the Experimental Station from INTA localized in Balcarce, Buenos Aires province, Argentina was used. The last FMD outbreak reported in this area was 4 years prior the commencement of trial 1 (2000). Animals were identified individually by ear tags. The vaccine was administered by the intramuscular route on the side of the neck, using a 1 ml dose. General and local reactions at the site of vaccination were recorded. Two experiments were carried out, one in adult sheep and the other in lambs (Trials 1 and 2, respectively). Figure 1 shows a scheme of reproductive time and vaccination/bleeding schedule. The flock was kept under grazing conditions, in a 7 hectare paddock with a consociated pasture (about 15 sheep per hectare). No nutritional supplement was given during the winter months. Two rams were introduced in the flock during the two breeding months (April-May). Ewes were pregnancy tested the 14th of June using ultrasonic equipment, and marked according to pregnancy status (empty, single, double). Afterwards, lambing ewes were moved to a second paddock of the same size. All sheep were de-wormed prior to the trial beginning and fecal egg counts were performed routinely to monitor internal parasite burden. Only one treatment was necessary during the trial. Blood samples were taken before and at regular intervals after vaccination (up to 1 year in adults and 5 months in lambs), and the serum was separated, aliquoted and stored at –20ºC for further serological determinations. 2.2.1. Trial 1 One-hundred and two female sheep aged 1-4 years, (free of structural and NSP antibodies to FMDV) were allotted to 4 vaccinated groups (n=23 each) homogeneously distributed with regard to age, weight and body condition. Ten non-vaccinated sheep served as control animals. FMD vaccination was performed during breeding season. Three groups were revaccinated at 10 (Group 2), 20 (Group 3) and 30 (Group 4) days after primary vaccination (dpv), respectively; Group 1 was not revaccinated and received only a single dose. 2.2.2. Trial 2 After lambing, lambs born from vaccinated sheep were grouped according to age and vaccinated at approximately 35±1 (n=28), 61±3 (n=28) and 94±2 (n=30) days old (abbreviated as G-30, G-60 and G-90, respectively). Non-vaccinated lambs born to vaccinated sheep were sampled to evaluate the decay of levels of MDA (n=13). Four (4) and 2 lambs born to non-vaccinated ewes free from FMDV antibodies (from control group of Trial 1) were vaccinated at 30 and 90 days of age, respectively. Body weight was measured at birth, at 32-49 and at 122-140 days of age, and daily weight gain calculated for each group. 2.3. Serology Specific antibody responses were determined by liquid-phase blocking sandwich ELISA (LPBE) against FMDV strains O1 Campos and A Arg 2001 (Robiolo et al., 1995). Neutralizing antibodies against O1 Campos were also determined in serum samples of sheep of Trial 1 by the virus neutralization test (VNT, OIE, 2004) using baby hamster kidney (BHK21 clone 13) cell monolayers. Antibodies to the 3ABC NSP were determined by 3ABC ELISA (Robiolo et al., 2006) as a screening test, and the positive reactors determined using the Ceditest FMDV-NS, (Cedi-Diagnostics). The Ceditest FMDV-NS was also used for serum samples of Group 4, Trial 1 (before vaccination, 30 days after the first dose, and again 30 days after the second dose). Serum samples of nonvaccinated sheep were also tested at those intervals. 2.4. Statistical analysis Analysis of variance was used to compare the means of the LPBE and VNT antibody titres, and mean daily weight gain of the experimental groups. A nonparametric Kruskal-Wallis test was used to compare VNT antibodies at 330 days of age due to lack of homogeneity of variance of data at this bleeding time. Statistical analyses were performed using SAS version 9.1, Mixed and Glimmix procedures were used (SAS Institute, 2002-2005). Differences were considered significant at

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

P<0.05. Linear Regression was used to analyze the MDA decay according to age to estimate mean half-lives, and also to analyze the interference of MDA to vaccination at 30 days of age. A Student´s t-test was used to compare LPBE antibody titres at 30 days of age between lambs born to sheep receiving one or two doses of vaccine. 3. RESULTS 3.1. Tolerance Trial 1 and 2 No general adverse side effects were observed in adult sheep or lambs of any age after vaccination or revaccination. No visible and palpable local reactions were detected at the site of inoculation after vaccination/s in any adult or lamb. Lambs had at birth an average body weight of 4.41 kg (SD 0.83) and was balanced for all groups. At 32-49 days of age the mean body weight reached 15.3 kg (SD 3.15) and at 122-140 days of age it reached 29.17 kg (SD 4.36). The mean for daily weight gain in the vaccinated groups was 0.182 g, whereas in the non-vaccinated group was 0.191 g (P> 0, 05) showing no effect of the vaccination on the body weight gain during the study. 3.2. Trial 1. Immune response in adult sheep after single or double vaccination at different intervals Non antibodies to FMDV were detected in the control group during the whole trial (either by LPBE or 3 ABC ELISA). One dose of vaccine induced satisfactory LPBE antibody levels with regards to O1 Campos and A Arg 2001 in adult sheep, reaching log10 2.80 and log10 2.71, respectively at 20 dpv. These levels persisted up to the end of the trial (330 dpv). However, revaccinated groups had significantly higher LPBE antibody titres (P<0.05) than the single vaccinated group from 40 to 120 dpv, regardless of the time the second dose was administered (Figure 2). Neutralizing antibody kinetics of O1 Campos followed a similar pattern to that of the LPBE antibody profile (Figure 3), and revaccinated groups had higher VNT antibody titres at 60 dpv (P<0.05). At 180 and 330 dpv, the mean VNT antibody titres ranged between log101.35 and 1.61, and non significant differences were detected between vaccinated and revaccinated groups. 3.3. Trial 2. Decay of MDA in non vaccinated lambs born to vaccinated ewes Four out of the 13 lambs did not have detectable antibodies to both FMDV strains in the LPBE presumably due to either inadequate ingestion or very poor absorption of colostrum, and were consequently excluded from this analysis. High MDA titres were observed in the 9 remaining lambs at the first sampling date corresponding with an average age of 17.4 days (SD 1.4, range 15-19): log10 3.47 for O1 Campos and log10 3.41 for A Arg 2001. Figure 4 shows the decay of the MDA antibodies determined by LPBE to both FMDV strains up to 153 days of age. The regression lines calculated with 83 observations were: O1Campos: y = 3.8061-0.0191x; r2=0.82 A Arg 2001: y = 3.6487-0.0177x; r2=0.78 From these equations, the half-lives of MDA for A Arg 2001 and O1 Campos were estimated in 17 and 15.2 days, respectively. 3.4. Trial 2. Immune response of lambs with MDA vaccinated at 30, 60 and 90 days of age High MDA titres were recorded in lambs at 30 days of age (log10>3.00 for both FMDV strains). Moreover, lambs born to revaccinated sheep (Group 2, 3 and 4, Trial 1) had significantly higher MDA titres than those born to singly vaccinated sheep (Group 1, Trial 1) (P<0.01). Vaccination at this age induced seroconversion in 39 and 49% of the lambs (G-30 group) at 30 dpv (Figure 5). High titres were also detected before vaccination in G-60 group lambs (mean titre log10 2.42 for O1 Campos and log10 2.48 for A Arg 2001) and seroconversion after vaccination was evident in 27 out of 28 lambs. At 90 days of age, lambs had medium to low levels of MDA (mean titre log10 1.94 for A Arg 2001, and log10 2.08 for O1 Campos) and vaccination at this age induced high immune response with seroconversion in the 100% of the lambs. High antibody titres were detected at 30 dpv (LPBE titres log10 3.04-3.29) with non significant differences among G-30, G-60 or G-90 lamb groups. In addition, similar antibody kinetics was observed in all three groups after vaccination. Lambs showed high levels of LPBE antibodies (mean titre group log10>2.70) at 5 months of age regardless of age the vaccination was administered (Figure 5). Individual responses were related to the antibody level at the time of vaccination and a negative postvaccinal response above log10 3.00 was observed. Regression of postvaccinal antibody increase or decrease at 32 dpv (y) to the antibody titres at vaccination (x) at 30 days of age (G-30 group)

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

produced a significant fit for both FMDV strains with a negative slope (Figure 6). Regression equations are: y=2.8208-0.9270x; r2= 0.8515 for A Arg 2001 and y=3.1787-1.0193x; r2=0.8658 for O1 Campos. 3.5. Trial 2. Immune response of lambs without MDA vaccinated at 30 or 90 days of age Vaccination of lambs born to non vaccinated ewes either at 30 or 90 days of age induced good immune responses similar to those of G-30 or G-90 in presence of MDA and high antibody levels persisted to 5 months of age (Figure 5). The mean group antibody titres at 60 dpv were equivalent in lambs and sheep (Trial 1) regardless of the presence of MDA in lambs. 3.6. Serology to NSPs NSP antibody reactors (by 3ABC ELISA-CEVAN) were detected in 7/1684 serum samples (0.41%), with a low percentage of positive reactivity (% reactivity= 22 to 32%) and 17/1684 serum samples (1 %) were classified as suspicious (% reactivity= 15 to 19%). No association was established between positive or suspicious reactors and age group, interval post vaccination, vaccinated group, or to samplings of an individual animal. Results were verified by applying the Ceditest NSP FMDV to all the serum samples of animals that were scored as suspicious or positive at any bleeding time by 3ABC ELISA CEVAN; no reactors were detected. In addition, no reactors were detected by Ceditest NSP FMDV in serum samples 30 dpv of group 4 and in the non vaccinated group. 4. DISCUSSION FMD control in most South American countries relies on systematic vaccination of cattle -once or twice a year- movement control, differential diagnosis with other vesicular diseases and serosurveillance to monitor antibodies to NSP in the susceptible population. In South America, specific strategies are being implemented in areas with a high level of risk in which vaccination of all susceptible domestic livestock is practiced. This study provides information regarding vaccination strategies in sheep and lambs with MDA gained by using a commercial polyvalent oil adjuvanted vaccine currently used in the vaccination campaign in Argentina (120 million doses per year during the last 5 years). The safety of the vaccine was demonstrated in vivo in lambs and sheep as no side effects - local or systemic - were observed after vaccination in any animal. With regards to the purity of antigens formulated in the vaccine, previous work has demonstrated a sufficient level of purity based on the lack of seroconversion to NSP in cattle after repetitive vaccinations and under field conditions (Mattion et al., 2004, Espinoza et al., 2004). In this study, no detectable NSP antibodies were found in sera from vaccinated and revaccinated sheep. Therefore, the use of this immunogen in sheep in addition to cattle will not interfere with the surveillance programs for the detection on NSP and recognition of FMD free status. Previous challenge studies performed in vaccinated sheep have shown protection against clinical disease and viremia when sheep had an antibody titre ≥ log101.3 (VNT) at 10 dpv (Parida et al., 2008); approximately ≥ log101.3 (VNT) at 14 dpv (Barnett et al., 2004); ≥ log102.45 (LPBE) at 90 dpv (Fondevila et al., 1993) and ≥ log10 2.3 (LPBE) at 21 dpv (Deghaidy et al., 2002). However, it is difficult to directly compare titres obtained in different locations and this is particularly true for the VNT when different cells are used. In our study, protective levels were observed in adult sheep one year after a single dose (mean group VNT titre of log10 1.35 and mean group LPBE titre log10 2.8), suggesting a long lasting duration of immunity which is of particular relevance in areas with difficulties in moving animals for booster doses. However, when a higher protection level is required in high risk situations, a second dose given 10 to 30 days after primary vaccination assures a high flock immune status is achieved within a short period. Further studies are necessary to determine the antibody levels required in sheep to confer protection against clinical signs, reduce viremia and prevent transmission of relevant FMDV strains. Lambs born to non vaccinated sheep responded as adults against FMD vaccination. This observation was in line with previous reports (Terpstra and Dekker, 1996, Cunliffe and Graves, 1970). Half-lives of MDA of 15-17 days found in this study were slightly shorter than the values calculated by other authors. Cunliffe and Graves (1970) reported a mean value 22,4 days (range of 12.2 to 29.7 days) for different FMDV strains. Although vaccination of lambs with high MDA titres produced a high antibody response on average, individual differences were observed. In lambs with high MDA titres (>log10 3.00), there was a slightly impaired response at 30 dpv; however, the biological impact of this suppression is negligible as protective levels of antibodies persisted during the whole trial. MDA in lambs stayed at or above protective levels (log10>2.30) for up to 80 days of age. Vaccination of lambs at 30, 60 or 90 days of age in the presence of MDA induced a protective response that persisted at least up to 150 days of age.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

In this regard, an increase of mean titre was observed in all lambs with MDA after vaccination. In contrast, a previous study in calves with MDA vaccinated at 20, 30 or 40 days of age showed a decline in the mean group titres up to 90 dpv, even though half of calves had protective antibody levels (Späth et al., 1994). Differences between immune responses of calves and lambs with MDA could provide insights in the role of colostrum in the regulation of the immune system and the mechanisms of active immunisation in young ruminants. Interestingly, both lambs (with or without MDA) and adult sheep reached high antibody levels at 60 dpv. Moreover, these levels were higher against both FMDV strains than those detected in cattle at the time of the official batch release potency test of the two vaccine batches used in this study (differences ranged from log10 0.45 to 0.89; data not shown). Other correlates of protection such as gamma interferon, IgA in probang samples or saliva need to be examined in further vaccination/challenge studies to characterize the protective immune response induced after vaccination of sheep. In conclusion, these results indicate that a vaccination scheme consisting of one or two doses for adult sheep and a single dose to lambs at any age (with or without MDA) should be encouraged in order to achieve and maintain sufficient herd protection levels and prevent further spread of the disease. 5. CONCLUSIONS One dose of oil adjuvanted vaccine induces a duration of immunity in adult sheep of at least one year A booster effect is observed shortly after revaccination at 10, 20 or 30 days post primo vaccination. Lambs 30-90 days old with MDA respond to FMD oil vaccine and produce high levels of antibodies that persist at least for 4 months 6. RECOMMENDATIONS One or two doses annually in adult sheep is recommended for systematic vaccination programs Two doses of oil vaccine in adult sheep, 10-30 days apart, is encouraged in order to enhance herd protection in a high risk situation Vaccination of young lambs with high levels of MDA is beneficial as this induces an effective immune response 7. ACKNOWLEDGEMENTS The authors thank Mariela Guinzburg for technical assistance, Marcelo Spitteler, Adriana Cano and Claudia Faverin, for statistical analysis and Gilles Chénard for reviewing the manuscript. The assistance of field personnel at the INTA Balcarce is greatly appreciated. 8. REFERENCES [1] Barnett, P.V., Keel, P., Reid, S., Armstrong, R.M., Statham, R.J., Voyce, C., Aggarwal, N. & Cox, S.J. 2004. Evidence that high potency foot-and-mouth disease vaccine inhibits local virus replication and prevents the “carrier” state in sheep. Vaccine 22: 1221-1232 [2] Cox, S.J., Aggarwal, N., Statham, R.J., Barnett, P.V. 2003. Longevity of antibody and cytokine responses following vaccination with high potency emergency FMD vaccines. Vaccine. 21:1336-47. [3] Cunliffe, H.R. & Graves, J.H. 1970. Immunologic response of lambs to emulsified foot-andmouth disease vaccine. Arch Ges Virusforsch 32: 261-268 [4] Deghaidy, W., Daoud, A. & El-Molla, A. 2002. Immune response of sheep to foot-and-mouth disease vaccines containing different adjuvants. Small ruminant research 45:185-192 [5] EMEA Concept paper on the need for requiring data to demonstrate the influence of maternally derived antibodies on the vaccination of very young animals. 31 July 2007. European Medicines Agency. Veterinary Medicines Inspection. Doc Ref. EMEA/CVMP/IWP/501304/2006CONSULTATION. (Available at http:/www.emea.europa.eu) [6] Espinoza, A.M., Maradei, E., Mattion, N., Cadenazzi, G., Maddonni, G., Robiolo, B., La Torre, J., Bellinzoni & Smitsaart, E. 2004 Foot-and-mouth disease polyvalent oil vaccines inoculated repeatedly in cattle do not induce detectable antibodies to non structural proteins when evaluated by various assays. Vaccine 22:69-74. [7] Fondevila, N., Smitsaart, E., Marcovecchio, F., O´ Donnell, V., Frick, E., López, A.G., Schudel, A. 1993. An oil adjuvanted vaccine against foot-and-mouth disease virus in sheep. Study of the immune response. Rev. Med. Vet. Bs.As. 74, 2. 89-95.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[8] Kitching, P. & Salt, J. 1995. The interference by maternally-derived antibody with active immunization of farm animals against foot-and-mouth disease. Br. vet. J.151: 379-389 [9] Mattion, N., König, G., Seki, C, Smitsaart, E., Maradei, E., Robiolo, B., Duffy, S., León, E., Piccone, M., Sadir, A., Bottini, R., Cosentino, B., Dotta, J., Falczuk, A., Maresca, R., Funes, G., Periolo, O., Bellinzoni, R., Espinoza, A., La Torre, J., Palma, E. 2004. Reintroduction of Foot-and-mouth Disease in Argentina: Characterisation of the Isolates and Development of Tools for the Control and Eradication of the Disease. Vaccine 22:4149-4162. [10] Nair, S.P. & Sen, A.K. 1993. A comparative study of the immune responses of sheep against foot-and-mouth disease virus types Asia-1 and O PEG-concentrated aluminum hydroxide gel and oil adjuvanted vaccines. Vaccine 11: 782-786 [11] OIE (World Organisation for Animal Health). Foot-and-mouth disease. In: Manual of diagnostic tests and vaccines for terrestrial animals. 5th ed. Paris, France: OIE Standards Commission, Office International des Epizooties; 2004 [chapter 2.1.1]. [12] Parida, S., Fleming, L., Oh, Y., Mahapatra, M., Hamblin, P. Gloster, J., Paton, D.J. 2008. Emergency vaccination of sheep against foot-and-mouth disease: Significance and detection of subsequent sub-clinical infection. Vaccine 26: 3469-3479 [13] Patil, P.K., Bayry, J., Ramakrishna, C., Hugar, B., Misra, L.D., Prabhudas, K. & Natarajan, C. 2002. Immune responses of sheep to quadrivalent double emulsion Foot-and-mouth disease vaccines: rate of development of immunity and variations among other ruminants. J.Clin.Microbiol. 40: 4367-4371 [14] Robiolo, B., Grigera, P., Periolo, O., Seki, C., Bianchi, T., Maradei, E. & La Torre, J. 1995. Assessment of foot-and-mouth disease vaccine potency by liquid-phase blocking ELISA: a proposal for an alternative to the challenge procedure in Argentina. Vaccine. 14:1346-1352. [15] Robiolo, B., Seki, C. Fondevila, N., Grigera, P., Scodeller,E., Periolo, O., Jose La Torre, J., Mattion, N. 2006. Analysis of the immune response to FMDV structural and non-structural proteins in cattle in Argentina by the combined use of liquid phase and 3ABC-ELISA tests. Vaccine 24: 997–1008 [16] SAS Institute Inc., SAS Online Doc 9.1.3, Cary, NC: SAS Institute Inc., 2002-2005. [17] Selman, P., Chénard, G. & Dekker, A. 2006. Cedivac-FMD; Duration of immunity in cattle, sheep and pigs. .Session of the Res. Group of the Standing Technical Committee. EUFMD. Paphos, Cyprus. Appendix 31 215-219. [18] SENASA Resolución 351/2006 Servicio Nacional de Sanidad y Calidad Agroalimentaria Sanidad Animal (SENASA) 2006 Argentina - Boletin Oficial nº 30.940 - 5/7/06 (available at http://www.infoleg.gov.ar/infolegInternet/anexos/115000-119999/117636/norma.htm) [19] Späth, E.J.A., Smitsaart, E., Casaro, A.P.E., Fernández, F., Leunda, M. R., Compaired, D., Buffarini, M., Pessi, H. 1995. Immune response of calves to foot-and-mouth disease virus vaccine emulsified with oil adjuvant. Strategies of vaccination. Vaccine. 13: 909-914. [20] Terpstra, C. & Dekker, A. Tolerance and efficacy of FMD-double oil emulsion vaccine in 2week-old-calves, lambs and piglets. 1996. Session of the Res. Group of the Standing Technical Committee. EUFMD. Ma´ale Ha´amisha, Israel. Appendix 26. pp 183-190.

BREEDING

2004

2005

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

JAN FEB MAR

VACCINATION ADULTS

BLEEDINGS ADULTS

VACCINATION LAMBS

BLEEDINGS LAMBS

LAMBING

WEANING

Figure 1: Breeding cycle and vaccination/bleeding schedule in adult sheep and lambs

100

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Control Group 1-1 dose Group 2-rv 10

mean group LPBE titres log 10 (O1 Campos)

4,50

Group 3-rv 20

4,00

Group 4-rv 30

3,50 3,00 2,50 2,00 1,50 1,00 0,50 0,00 0

100 150 200 250 days after primary vaccination

300

350

Figure 2: Mean group LPBE antibody titres (log 10) following vaccination (Group 1-1 dose) and revaccination at 10 days (Group 2-rv 10), 20 days (Group 3-rv 20) or 30 days (Group 4-rv 30) after primary vaccination. Open circles indicate non vaccinated control group.

101

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Control mean group VNT titres to O1 Campos (log 10)

Group 1-1 dose Group 2-rv 10 Group 3-rv 20

2,5

Group 4-rv 30 2

1,5

0,5 0

100 150 200 days after primary vaccination

250

300

350

Figure 3: Neutralizing antibody titres (log 10) following vaccination (Group 1-1 dose) and revaccination at 10 days (Group 2-rv 10), 20 days (Group 3-rv 20) or 30 days (Group 4-rv 30) after primary vaccination. Open circles indicate non vaccinated control group.

4 A Arg 2001 Maternally derived antibodies in lambs by LPBE (log 10)

3,5

O1 Campos

3 2,5 2 1,5 1 0,5 0 0

100

120

140

160

days of age

Figure 4: Decay of maternally derived antibodies in lambs born to vaccinated ewes, as measured by LPBE to A Arg 2001 (open circles) and O1 Campos (solid squares) FMDV strains.

102

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

LPBE titres log 10 (A Arg A2001)

4 3,5 3 2,5 2 1,5 1 0,5 0 30

90 days of age

120

control G-30

150

non vacc-MDA w/o MDA vacc 30

LPBE titres log 10 (A Arg 2001)

B 4 3,5 3 2,5 2 1,5 1 0,5 0 30

120

150

days of age control

non vacc-MDA

G-60

LPBE titres log10 (A Arg 2001)

C 4 3,5 3 2,5 2 1,5 1 0,5 0 30 control G-90

90 days of age

120

150

non vacc-MDA w/o MDA vacc 90

Figure 5: Immune response in lambs following vaccination as measured by LPBE (mean group titres log 10+SD). Arrows indicate vaccination. Open circles indicate naïve control sheep. Open squares indicate decay of colostral antibodies. A. Lambs born to vaccinated sheep (G-30) and non vaccinated sheep (w/o MDA vacc 30) vaccinated at 30 days of age. B. Lambs born to vaccinated sheep vaccinated at 60 days of age (G-60) C. Lambs born to vaccinated sheep (G-90) and non vaccinated sheep (w/o MDA vacc 90) vaccinated at 90 days of age.

103

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.0 1.5

DIFA2001

1.0 0.5 0.0 -0.5 -1.0 -1.5 1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

A2001

Figure 6: Effect of maternally derived antibodies (MDA) on primary response. Y axis: Difference between LPBE titres (log10, A Arg 2001) at 30 dpv and day 0 (30 days of age, G-30 group). X axis: MDA titres at day 0

104

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 15 LONGEVITY OF PROTECTION IN CATTLE FOLLOWING VACCINATION WITH EMERGENCY FMD FROM THE UK STRATEGIC RESERVE – PRELIMINARY RESULTS

S. Cox 1

, S. Parida1, P. Hamblin1, B. Bankowski1, B Veronica Carr2, D. Paton1 and P. Barnett1

Pirbright Laboratory, Institute for Animal Health, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, UK 2 Compton Laboratory, Institute for Animal Health, Compton, Newbury, Berkshire, RG20 7NN, UK

ABSTRACT Objective To determine the longevity of protective immunity following a single emergency vaccination of cattle and establish whether the immune response could be further optimised by increasing the antigen payload. Materials and methods An experiment was carried out in which cattle were vaccinated with either A22 Iraq vaccine containing1x antigen payload (field dose) or 5x antigen payload. A further 2 cattle remained unvaccinated. Six months post immunisation all cattle received a homologous virus challenge by needle using the intradermolingual route. Blood, saliva and nasal swab fluid samples were taken at regular intervals post vaccination and challenge. Post challenge, clinical observations and rectal temperatures were recorded daily and oropharyngeal samples were also taken. Various serological and cell proliferation assays and assays for virus are presently ongoing. Results All vaccinated animals seroconverted but there was considerable variability with regard to development of peak responses. No significant difference in antibody titre between the groups was noted. Following challenge, all the vaccinated cattle, regardless of antigen payload were protected from clinical disease. However, some cattle in both groups became sub-clinically infected. Conclusions Immunisation with a single shot of vaccine containing high antigen payload will protect cattle from clinical disease at 6 months post vaccination and a boost may be unnecessary. However, some animals may become sub-clinically infected but this is likely to be dependent on the severity of challenge. Increasing antigen payload 5-fold has not, as yet, shown any increased benefits. Quantitative analyses with regard to virus recovery from each group of cattle are presently ongoing and further similar cattle experiments are planned using other serotypes. 1. INTRODUCTION In recent years there have been several changes in FMD control policy (OIE Animal Health Code, 2006; EU Directive 2003/85/EC) to allow emergency vaccination to be more readily considered, and vaccinate-to-live may be implemented in the future. It is thus essential to know what the duration of protective immunity would be following a single vaccination and at what time point there may be a requirement to boost vaccinated animals. In countries where FMD is endemic, it is also desirable to use vaccines that provide as long a period of protection as possible, to avoid frequent re-vaccination. To date, only limited studies have been undertaken in ruminants and pigs indicating that the systemic antibody levels are maintained for up to 6 months or more ( Cox et al., 2003; Selman et al., 2006) and that protection against homologous challenge is possible 7 months following vaccination of pigs (Cox et al., 2003). For cattle, there have been no published studies of the protective durability of emergency vaccines against virus challenge in this important ruminant model. In this study, the longevity of protective immunity following a single emergency vaccination of cattle with A22 Iraq 24/64 vaccine from the UK strategic reserve was investigated, and additionally, whether the immune response could be further optimised by increasing antigen payload.

105

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2. MATERIALS AND METHODS Twenty-two Holstein/Friesian steers, approximately 6 months of age, were used. Cattle were immunised (at IAH Compton in secure accommodation) with Aftopur vaccine (Merial) based on monovalent strain A22 Iraq 24/64. One group of 10 cattle received a 1x bovine dose (antigen concentration determined from previous PD50 tests) and a second group of 10 cattle received a 5x bovine dose, administered as the same 2ml volume. In addition, 2 animals of similar age were also held as unvaccinated controls. All animals were sampled pre-vaccination and periodically over a 6 month period. Heparinised and clotted blood, saliva and nasal swab fluids were taken on days 0, 3, 5, 7, 14, 28 days post vaccination and monthly thereafter until 168 days post vaccination. All the animals were then transferred to IAH Pirbright where they received a homologous challenge, at 179 days post vaccination, by the intradermolingual route, to mimic that normally performed in potency tests as described in the European Pharmacopeia. Cattle were inoculated with 105 TCID50 in 0.2ml. Clinical observations and rectal temperatures were recorded daily up to 10 days post challenge and samples of heparinised blood, clotted blood, saliva, nasal swab fluids and oropharyngeal fluid were collected on 2, 4, 6, 8 and 10 days post challenge before termination of the experiment. Systemic neutralising titres to FMDV were measured by micro-neutralisation assay (Golding et al., 1976). Endpoint titres were calculated as the reciprocal of the last serum dilution to neutralise 100 TCID50 of homologous FMDV in 50% of the wells. FMDV antibody isotype (total specific IgG) analyses were performed using an indirect antibody sandwich ELISA (Salt et al., 1996). Sera were also examined for the presence of antibodies to the FMDV non-structural proteins 3ABC (Ceditest FMDV-NS (Cedi Diagnostics)). Peripheral blood lymphocytes (PBMCs) were recovered form heparinised blood and proliferation assays were carried out with various stimuli including FMDV vaccine antigen, essentially as described in Collen and Doel (1990). Heparinised blood and oropharyngeal fluid samples were examined for the presence of virus by cell culture inoculation (Snowdon, 1966) with confirmatory ELISA (Ferris et al., 1988). 3. RESULTS Neutralising antibody responses for each animal are shown in Figure1 and all vaccinated cattle had seroconverted by 5 days post vaccination. There was however, considerable variability for individual animals with regard to development of peak responses, and these ranged between 28 and 140 days post vaccination. The ELISA for total specific IgG gave similar results, although peak responses were generally demonstrated earlier using this assay (data not shown). By the time of challenge, there had been a decline from peak titres in all animals. Figure 2 provides a comparison of the mean responses for each treatment dose group. ANOVA analysis of the data using Minitab 15 showed no significant difference in antibody titre between treatment groups at any time point. A summary of the T cell proliferation responses is provided in Table 1. Some cattle in both groups demonstrated either no measurable proliferation, a weak response or were strong reactors. A similar profile of responses was seen for each treatment group. Following challenge, all the vaccinated cattle, regardless of antigen payload were protected from clinical disease and no viraemia was detected (Table 2). The unvaccinated control cattle, however, developed FMD. Lesions were evident on their feet from 2 and 3 days post inoculation and viraemia was detected at 2 and 4 days post inoculation. Cattle in both treatment groups became subclinically infected as shown by the recovery of virus from oropharyngeal fluid samples, and/or development of a non-structural antibody response, and/or a 4-fold rise in neutralisation antibody titre by 10 days post challenge (Table 2). The degree of virus replication varied between animals and several showed none or very limited infection. There was however, no clear difference between the two vaccinated treatment groups. 4. DISCUSSION It is generally accepted for conventional, prophylactic vaccination that a second, boosting injection of vaccine is required to maintain immunity at protective levels for at least 6 months. Thereafter, further re-vaccinations at regular intervals, depending upon the epidemiological situation, are required to maintain protective immunity. However, high potency emergency vaccination as a

106

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

means of control during an outbreak in an FMD free country is very much dependent on the administration of a single dose. The priority in such situations is to provide rapid protection to as many at risk animals as possible. From the logistical point of view, neither manpower nor valuable doses of vaccine can afford to be wasted on boosts, if these are not necessary. Many previous studies have shown how effective high antigen payload vaccines are at providing rapid protection from clinical disease following a single shot, but to date only a few have been undertaken to investigate the longevity of protection. With the shift in policy towards vaccination-to-live, it would therefore be useful to know the durability of protection following a single shot of vaccine and when or whether it is necessary to boost animals. This first experiment of a series of four, using A22 Iraq/24/64 vaccine has shown that a single immunisation with the field dose, provided protection from a homologous intradermolingual virus challenge at 6 months, as none of the animals developed clinical disease. This uniform protection was achieved, despite individual animals showing variable responses for both structural antibodies and T-cell proliferation post-vaccination. Virus replication, however, did occur in most animals, although this was limited in several. Since the animals were needle challenged, recovery of virus and/or development of a non-structural antibody response/ 4-fold increase in neutralising antibody titre may not be surprising and therefore it might be extrapolated that lesser challenges which could occur in the field, could result in less or no sub-clinical infection. Further experiments using more natural routes of infection would thus be useful. We do not presently know the PD50 of the vaccine used or the precise antigen content per dose (for confidentiality reasons) but an estimate of the vaccines likely efficacy can alternatively be made from assessing the serological response at 21 days post vaccination. The mean responses of both treatment groups were not significantly different from that of two other A serotype vaccines previously tested for the UK reserve which had PD50 values of ≥32 (internal communication). The responses, however, were significantly higher (P<0.05) than those of a third A serotype vaccine from the UK reserve which had a PD50 of 24. Additionally, a similar A22 Iraq 24/64 vaccine held for emergency use from the same manufacturer, was recently shown to have a PD50 value of 32 (Brehm et al., 2008). From the above comparison we can be fairly sure that the vaccine used was already very high potency but the opportunity was also taken to investigate whether increasing the antigen payload even more than usual for emergency vaccines (typically 4-10 µg per dose – Brehm et al., 2008) would further enhance the measured immune responses and provide greater protection against clinical and sub-clinical FMD at 6 months post vaccination. Previous work investigating early responses using increased antigen payload in cattle has shown that a quicker and better antibody response can be elicited and that sub-clinically infected cattle immunised with higher antigen payload (10x), cleared the virus more readily (Cox et al., 2006), suggesting that the antigen payload was not optimised. In this experiment, no increased benefit was recognised from increasing the antigen payload 5x, as all the animals receiving the single strength field dose were also protected from clinical disease and the post challenge recovery of virus and antibody responses were also similar to those of the field dose group. The difference between experiments may be accountable to the difference in serotype/vaccine, as A serotype was used in this experiment and O serotype in the previous work, and suggests that further optimisation of emergency vaccine performance may only be required for some vaccines. Since the aim of this project is to assess a range of different vaccines, it will be interesting to see in future experiments whether the same holds true for other serotypes, or whether the good result achieved here for the field dose is due to an exceptional A22 vaccine. 5. CONCLUSIONS Immunisation with a single shot of A22 vaccine containing a high antigen payload will protect cattle from clinical disease at 6 months post vaccination and a boost may be unnecessary. Some animals may become sub-clinically infected but this is likely to be dependent on the mode/severity of challenge. Increasing antigen payload 5-fold did not show any increased benefit. 6. RECOMMENDATIONS None at this stage 7. ACKNOWLEDGEMENTS

107

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

This work was supported financially by Defra, UK (Project SE2812). The authors wish to thank the staff of the animal units at both IAH Compton and Pirbright for their assistance and care of cattle used in this study. 8. REFERENCES [1] Collen, T. and Doel, T.R. 1990. Heterotypic recognition of foot-and-mouth disease virus by cattle lymphocytes. J. Gen. Microbiol., 71: 309-315. [2] Cox, S.J., Aggarwal, N., Statham, R.J., and Barnet, P.V. 2003. Longevity of antibody and cytokine responses following vaccination with high potency emergency FMD vaccines. Vaccine, 21: 1336-1347. [3] Cox, S.J., Voyce, C., Parida, S., Reid, S.M., Hamblin, P.A., Hutchings, G., Paton, D.J. and Barnett, P.V. 2006. Effect of emergency FMD vaccine antigen payload on protection, subclinical infection and persistence following direct contact challenge of cattle. Vaccine, 24: 31843190. [4] Brehm, K.E., Kumar, N., Thulke, H.-H and Haas, B. 2008. High potency vaccines induce protection against heterologous challenge with foot-and-mouth disease virus. Vaccine, 26: 16811687 [5] Ferris, N.P. and Dawson, M. 1988. Routine allocation of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Vet. Microbial., 8: 249-56. [6] Golding, S.M., Hedger, R.S. and Talbot, P. 1976. Radial immuno-diffusion and serum neutralisation techniques for the assay of antibodies to swine vesicular disease.Res. Vet. Sci., 20: 142-7. [7] Salt, J.S., Mulcahy, G. and Kitching R.P. 1996. Isotype-specific antibody responses to footand-mouth disease virus in sera and secretions of carrier and non-carrier cattle. Epidemiol. Infect., 117: 349-60 [8] Selman, P., Chenard, G. and Dekker, A. 2006. Cedivac-FMD; Duration of immunity in cattle, sheep and pigs. In: Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease; Paphos, Cyprus, [9] Snowdon W.A. 1966. Growth of foot-and-mouth disease virus in monolayer cultures of calf thyroid cells. Nature, 210: 1079-80 Figure 1: Virus neutralising antibody responses of cattle vaccinated with either 1x antigen payload (a) or 5x antigen payload (b). Each bar represents one animal at each time point. (a)

Antibody titre

1x antigen payload FMD 23

1600 1400 1200 1000 800 600 400 200 0

FMD 24 FMD 25 FMD 26 FMD 27 FMD 28 FMD 29 FMD 30 0

Days post vaccination

(b)

108

119

140

168

179

FMD 31 FMD 32

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Antibody titre

5x antigen payload FMD 34

1600 1400 1200 1000 800 600 400 200 0

FMD 35 FMD 36 FMD 37 FMD 38 FMD 39 FMD 40 FMD 41 0

119

140

168

179

FMD 42

Days post vaccination

FMD 43

Figure 2: Mean neutralising antibody responses of cattle vaccinated with either field dose (1x) or vaccine containing 5-fold antigen payload (5x).

900 800 antibody titre

700 600 500

400

300 200 100 0 0

84 119 140 168 179

Days post vaccination

Table 1: Summary of FMD specific T-cell proliferation responses for each animal following vaccination with either field dose (1x) or 5-fold antigen payload (5x) vaccine. Animal Ref

Treatment

Days post vaccination 0 7 14

172

FMD FMD FMD FMD FMD FMD FMD FMD FMD FMD

23 24 25 26 27 28 29 30 31 32

1xa 1x 1x 1x 1x 1x 1x 1x 1x 1x

-b -

++d + + + ++ + -

++ + + + + -

+c ++ ++ ++ ++ + -

++ ++ ++ ++ + -

+ ++ ++ ++ + ++ + -

++ ++ ++ + ++ + -

FMD FMD FMD FMD FMD FMD FMD FMD FMD FMD

34 35 36 37 38 39 40 41 42 43

5x 5x 5x 5x 5x 5x 5x 5x 5x 5x

+ + + + ++ +

+ + + ++ + +

+ + + + + ++ ++

++ ++ ++ + + +

109

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

FMD 33 FMD 44

0 0

No T cell proliferation detected; ; b No T cell proliferation detected; detected; d Strong T cell proliferation detected

Weak T cell proliferation

Table 2: Outcome of intradermolingual challenge of vaccinated and unvaccinated cattle. Oropharyngeal fluid virus isolation Animal Clinical NS (days post challenge) Treatment Viraemia Ref disease Ab 0 2 4 6 8 10 FMD FMD FMD FMD FMD FMD FMD FMD FMD FMD

23 24 25 26 27 28 29 30 31 32

1xa 1x 1x 1x 1x 1x 1x 1x 1x 1x

No No No No No No No No No No

-b -

4.1 <1.35

1.35 <1.35 4.1 -

1.85c 1.6 <1.35 4.1 <1.35 -

<1.35 <1.35 <1.35 <1.35 -

1.6 <1.35 Yes <1.35 Yes Yes Yes Yes 1.35 Yes

FMD FMD FMD FMD FMD FMD FMD FMD FMD FMD

34 35 36 37 38 39 40 41 42 43

5x 5x 5x 5x 5x 5x 5x 5x 5x 5x

No No No No No No No No No No

2.1 1.35 3.85 <1.35 ≥6.1 <1.35

1.85 1.35 1.6 3.1 1.6 3.6 <1.35

<1.35 1.35 <1.35 1.6 <1.35 <1.35

<1.35 1.35 <1.35 <1.35

1.35 <1.35 Nsd <1.35 ≤1.35 Yes <1.35

Yes Yes

<1.35 3.35 5.35 2.6

FMD 33 0 FMD 44 0 a

110

4-fold rise in VNT

Yes Yes Yes Yes Yes Yes

<1.35 <1.35 Yes Yes <1.35 <1.35 <1.35 Yes Yes

Antigen payload; bNo virus detected; c Log titre (TCID50/ml); d No sample

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 16 HIGH POTENCY VACCINES INDUCE PROTECTION AGAINST HETEROLOGOUS CHALLENGE WITH FOOT-AND-MOUTH DISEASE VIRUS K. Brehm1, N. Kumar1, H.-H. Thulke2 and B. Haas1* 1

Friedrich-Loeffler-Institute, Südufer 10, 17493 Greifswald-Insel Riems, Germany UFZ Helmholtz Centre for Environmental Research, Department of Ecological Modelling Permoserstrasse 15, 04318 Leipzig, Germany

ABSTRACT Introduction While generally serological r-values are used to assess the ability of vaccines to protect against foot-and-mouth disease (FMD) field strains, there is a lack of experimental cross protection studies. Therefore, we investigated the capability of high potency type A vaccines to induce protection against heterologous challenge and also the correlation of protection and neutralization titres recorded for post vaccination sera. Materials and methods A series of three homologous and eight heterologous cattle challenge experiments was performed according to the protocol described in the European Pharmacopoeia monograph. Results It was shown that high potency vaccines against FMDV of serotype A can induce protection even against heterologous challenge infection with viruses that give low r-values with the vaccine strains. Three vaccines (A22Iraq24/64, AIran2/97, AIran22/99) with homologous PD50 values of at least 32 showed significant protection even against heterologous challenge with viruses showing low r-values. The r-values were determined on the basis of full dose group mean titres and did not exceed 0.23. In six out of eight heterologous challenge experiments, the high potency vaccines still conferred a protection of at least six PD50. Therefore, in a situation when vaccination is considered, but no closely related vaccine is available, the usage of a high potency vaccine may be justified despite low r-values. The challenge virus specific neutralizing antibody response on the day of challenge (21 days post vaccination) generally correlated with protection. Discussion While the results of this study do not yet provide a sufficient statistical basis to establish a probit curve (titre vs. probability of protection), they already can be used to support a decision on the use of a vaccine. As the results of a heterologous challenge test would only be obtained after more than a month while the decision to vaccinate usually will have to be made very fast, field virus specific neutralizing antibody titres, which can be obtained within days, will provide valuable information. However, it will have to be investigated if these results for serotype A are also valid for other serotypes. In serotype O cross challenge experiments, poor cross protection was found despite good r-values. Furthermore, VNT titres obtained by different groups with different test systems cannot be directly compared. In order to create a better scientific basis for the choice of vaccines, it is suggested to analyse sera produced during this and related projects also in other laboratories and with other methods for comparison. There also is an urgent need to address protection against newly emerging strains, e.g. recent type A strains circulating in the Middle East. In particular, vaccine trials with the O PanAsiaII strain have to be performed because of the recent increase of clinical apparent type O outbreaks in Turkey despite the use of apparently good vaccines and relatively high r-values. This might be interpreted as an indication that this strain has an increased intrinsic ability to overwhelm protection. It should be examined whether this is true and if so, which mechanisms might be responsible.

111

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[1] Brehm, K.E., Kumar, N, Thulke, H.-H, and Haas, B; High potency vaccines induce protection against heterologous challenge with foot-and-mouth disease virus, Vaccine (2008) 26, 1681—1687

112

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 17 PREDICTING ANTIGENIC SITES ON THE FMDV CAPSID FROM CROSS-REACTIVITY DATA F. F. Maree1, R. Reeve2, B. Blignaut1, 3, J. J. Esterhuysen1, E. Fry4, T. de Beer5, E. Rieder6 and D. Haydon1 1

Onderstepoort Veterinary Institute, Transboundary Animal Diseases Programme, Private Bag X05, Onderstepoort, 0110, South Africa. 2 Boyd Orr Centre for Population and Ecosystem Health, Division of Ecology and Evolutionary Biology, University of Glasgow, Glasgow, G12 8QQ, United Kingdom. 3 Department of Microbiology and Plant Pathology, University of Pretoria, Pretoria, 0002, South Africa. 4 Division of Structural Biology, University of Oxford, Oxford, United Kingdom. 5 Department of Bioinformatics, University of Pretoria, Pretoria, 0002, South Africa. 6 Foreign Animal Disease Research Unit, United States Department of Agriculture, Agricultural Research Service, Plum Island Animal Disease Center, Greenport, NY 11944, United States of America.

ABSTRACT Conventional FMD vaccines consist of chemically inactivated virus preparations that are applied parenterally. The selection of FMD vaccine candidates to be used in emergency and control settings is complicated by the genetic variability characteristic of this virus and the selection of mutants escaping the host immune response. Identifying epitopes for vaccine strains is therefore essential to determine whether the vaccine will protect against newly circulating outbreak strains from the same geographic location. Antigenically significant sites have previously been identified for O, A and C serotypes as well as an epitope for a Southern African Territories type 2 (SAT2) virus by sequencing of monoclonal escape mutants. We took a different approach by combining amino acid variation and in vitro cross-protection titres from virus neutralisation tests, together with crystallographic structural data, to derive similar information indirectly. Examining SAT1 and SAT2 strains, we identified a general negative correlation between genetic distance and serological relatedness (r-values), but more significantly we could identify areas on the surface of the capsid where mutations were strong predictors of antigenic distance. These were consistent within, but not between serotypes and were found to match some of the independently-identified antigenic sites in other serotypes, which reinforces our belief that this technique is effective at identifying epitopes. We predict that these sites should be under positive selection, and plan to verify this by examining non-synonymous to synonymous mutation rates, and further propose a reverse genetics approach to verify the predictions. 1. INTRODUCTION Foot-and-mouth disease (FMD) is endemic in most countries in sub-Saharan Africa where six of the seven immunologically distinct serotypes occur. Although these serotypes cause a clinically indistinguishable vesicular disease in cloven-hoofed animals the serotypes display different geographical distributions and epidemiology. The South African Territories (SAT) types 1, 2 and 3 are confined to Africa although incursions into the Middle East by SAT1 (1961-65 and 1970) and SAT2 (1990 and 2000) viruses have been recorded. The SAT3 serotype has a restricted distribution and essentially occurs only in southern Africa (Bastos et al., 2003). Eradication of the disease in Africa is unlikely due to the presence of large numbers of free-living maintenance host, the African buffalo (Syncerus caffer), which provide a potential source of infection for domestic livestock and wildlife (Dawe et al., 1994; Bastos et al., 2000; Vosloo & Thomson, 2004). The sub-clinical infected buffalo pose a constant threat to susceptible livestock (Vosloo & Thomson, 2004) and due to increasing global movement of animals and animal products to the rest of the world. The SAT types display appreciably greater intratypic genomic and antigenic variation than the traditional “Euro-Asian” types (Vosloo et al., 1992; Vosloo et al., 1995; Vosloo et al., 1996; Bastos et al., 2001; Bastos et al., 2003a, b). Even within a serotype, distinct genetic and antigenic variants exist in different geographic regions with implications for the control of the disease by

113

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

vaccination, since it may render available vaccines less effective. Consequently, the ability to predict vaccine efficacy would be valuable tool in an effective control strategy, since control of FMD in Africa is essentially via strategic vaccination and restriction of animal movements. Antigenic variation results from changes to the viral capsid as a consequence of the high mutation rate of the virus. The outer capsid proteins are directly involved in antigenicity seeing as 30-50% of their residues are more or less exposed on the virion surface many of which constitute neutralising epitopes. Although several antigenic sites have been identified for A and O serotypes (Kitson et al., 1990, Crowther et al., 1993a, b), a dearth of knowledge exists about the epitopes of the SAT types. The identification of those residues that comprise the antigenic determinants of the SAT viruses or other serotypes will allow the identification of those changes in outbreak strains that may cause escape from protection afforded by the vaccine. If enough of these epitopes are identified, it may be possible to predict the protection afforded by a vaccine against a specific outbreak. Traditionally the in vitro neutralization test and statistically calculated r-values are used to determine antigenic relationships (Rweyemamu et al., 1978). The use of r-values to estimate cross-protection has been known to rely on having sufficient repeated measures to overcome the inherent variability of the neutralisation titres (Rweyemamu, 1984). It is also known that measuring the titre ratio to a known control is not sufficient to eliminate the inter-experiment variability, highlighting the necessity for time-consuming duplicate tests undertaken on separate and independent occasions to compensate for day-to-day variations (Rweyemamu et al. 1984). Meanwhile molecular epidemiological techniques have offered the possibility of more detailed analyses of FMD epidemiology. We have investigated the genetic variation of the complete capsidcoding region of representative FMDV strains found in sub-Saharan Africa. Statistical analyses based on these capsid-coding regions of SAT1 and SAT2 isolates are combined with serological relatedness and variable regions that could lead to antigenic changes in the SAT viruses are identified. Here, we propose to use a mixed-effects model which allows the many sources of variability, i.e. repeated measures, and inter-serotype, inter-vaccine strain, inter-lab, inter-day and inter-experiment variability, to be modelled statistically and their impact accounted for. 2. MATERIALS AND METHODS 2.1. Virus isolates, RT-PCR, sequencing and analysis The viruses included in this study were either supplied by the World Reference Laboratory (WRL) for FMD at the Institute for Animal Health, Pirbright (United Kingdom) or form part of the virus databank at Transboundary Animal Diseases Programme, Onderstepoort (South Africa). The 20 SAT1, and 21 SAT2 FMDV isolates from 17 countries in Africa, selected for genetic characterization, represented a broad geographical distribution. Sequences for the Leader-P1-2A coding regions of the isolates were obtained via RT-PCR of viral genomic RNA as previously described (Bastos, 1998; van Rensburg et al. 2002). Direct DNA sequencing of amplicons yielded a consensus sequence representing the most probable nucleotide for each position. Sequences of the ca. 2.2kb P1-coding region were compiled and edited using BioEdit 5.0.9 software (Hall, 1999) and the nucleotide sequences subsequently translated. The deduced amino acid sequences were aligned using clustal X (Thompson et al., 1997). 2.2. Animal sera and virus neutralization test The antigenic diversity of the field isolates was determined using cross-neutralization assays in micro-titer plates on IB-RS-2 cells carried out similarly as described in the OIE Manual of Standards (2004) against cattle sera prepared by two consecutive vaccinations (vaccinated at day 0, boosted at day 28 and bled at day 38) or convalescent sera obtained from cattle 21 days post-infection with reference viruses (SAT1: SAR/9/81, KNP/196/91 and NIG/5/81; SAT2: ZIM/7/83, KNP/19/89/2 and ERI/12/89). The end point titre of the serum against homologous and heterologous viruses was calculated as the reciprocal of the last dilution of serum to neutralize 100 TCID50 in 50% of the wells (Rweyemamu et al., 1978). One-way antigenic relationships (r-value) of the field isolates and engineered viruses relative to the reference strains was calculated, which is expressed as the ratio between the heterologous and homologous serum titers. All neutralization titre determinations were repeated at least twice each time using sera from 2 animals. r-Values were interpreted as proposed by Samuel et al. (1990). Briefly, values between 0-0.19 indicated highly significant antigenic variation from the reference strains; values of 0.20-0.39 showed antigenic relatedness where some protection may be provided by a potent vaccine based

114

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

on the reference strain; while r-values of 0.40-1.0 demonstrated that the reference and field strains are antigenically similar enough for the reference strain to provide good protection. The averages of the r-values were calculated (Table 1 and 2) and plotted against total amino acid changes between the field strains and reference viruses. However, in the final model the end point serum titres were used in the analysis, as too much information was lost in the data processing from neutralising titre to r-value (see Figure 2). 2.3. Structural modeling A crystallographic structure of the capsid proteins of a SAT1 virus was obtained from unpublished work by the Division of Structural Biology in Oxford. SAT1 and SAT2 models were also build using the O1BFS coordinates (1ZBE) as template, alignments were done with clustal X, modeling scripts were generated with the Structural module of FunGIMS and the models with Modeller 9v1. Structures were visualized and the surface-exposed residues identified with PyMol v1.1rc2pre (DeLano Scientific LLC), while the precise accessibility was calculated by naccess (Hubbard et al., 1993). 2.4. Statistical analysis The statistical package R (R Development Core Team, 2007) was used to build mixed-effects models (Grafen et al., 2002) to predict the antigenically significant regions of the capsid proteins. 2.4.1. Linear mixed-effects models The term experiment is used to denote any number of virus neutralisation tests which use one antiserum together with the homologous strain and one or more field strains on one occasion. This corresponds to the usage when r-values have been calculated. Since we have no prior belief about the nature of the relationship between structural changes and antigenic distance, model development begins with the simplest linear regression model with neutralising titre as the response variable. However, for this response variable the variance structure of the residuals is heteroscedastic. Therefore, following previous research (Rweyemamu et al., 1984), log (titre) rather than titre is used in the same linear model, and the variance becomes and remains homoscedastic. A generalised or non-linear model is therefore unnecessary. 2.4.2. Variance structure Two approaches were used to test the effect of outside factors on the variance of the model (i) examining the model terms in two parts as they are described (see results), building up terms and using likelihood ratio tests to compare nested models, or (ii) a stepwise regression of all of the terms. In both cases, Bonferroni corrections are required to take account of the multiple tests. 2.5. Measuring genetic distance The P1 sequences for the SAT strains used in neutralisation tests, allowed the number of amino acid changes (mutations, insertions or deletions) between any two strains to be counted. In addition, the structural data was used to identify those residues which are surface-exposed, allowing the number of changes specifically corresponding to surface-exposed residues between any two sequences to be counted. As baseline measure against which to assess antigenically important changes synonymous mutations were used as these correlate with antigenic distance (p<10-15). Candidate antigenic areas were chosen from the surface exposed residues by identifying the longest contiguous chains of residues which are surface-exposed. The candidate antigenic region were tested with their respective datasets to determine a significant positive correlation between increasing changes to the region and increasing antigenic distance (i.e. decreasing titres). If so, the combined SAT1 and SAT2 dataset was used and significant interaction with serotype and with protection strain was next looked at. 3. RESULTS 3.1. Phylogenetic relationships of the P1-coding region The minimum evolution phylogeny based on P1 nucleotide sequences of representative SAT viruses causing outbreaks throughout the African continent over the last 32 years, is indicated in Figure 1. The SAT viruses clustered according to serotype with high bootstrap support and within a serotype the isolates grouped mostly according to their geographic location (southern, western and eastern Africa). This observation is in agreement with the FMD topotype concept as described for the European and SAT serotypes (Samuel & Knowles, 2001a, b; Bastos et al., 2001; Bastos et al., 2003a, b). Exceptions were observed for SAT1 and SAT2 where isolates from East Africa, i.e. SAT2/KEN/8/99, SAT1/KEN/5/98 and SAT1/TAN/37/99, clustered in the southern topotypes,

115

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

possibly due to historical movement of buffalo and livestock between the two regions (Bastos et al., 2001; Sangare et al., 2003). Phylogenetic analysis and epidemiological tracing of FMDV outbreaks have in the past been performed largely on FMDV sequences from the 1D-coding region. Phylogenetic analysis of the complete P1 or 1B and 1C-coding regions of the SAT isolates indicated similar tree topologies compared to 1D phylogeny. Therefore, although 1D exhibit the most variability, evolutionary selection as a result of immune pressure is exerted equally on the outer capsid proteins. Furthermore, no evidence for genetically dissimilar recombinants was observed within the P1-coding region for the data set used. Using a 16% nucleotide difference cut off value, seven distinct lineages could be identified within the SAT1 serotype, while eleven of the 14 SAT2 topotypes could be resolved using the P1 phylogeny. This correlate with phylogeny observed based on 1D sequence only and P1 phylogeny did not provide significantly more resolution for epidemiology purposes. An exception was the Angolan SAT2 isolate, ANG/4/74, which grouped separately from other southern African SAT2 isolates according to 1D phylogeny and was previously assigned to a different topotype (Bastos et al., 2003b; Sangare et al., 2004). P1 phylogeny (Figure 1) grouped this isolate more closely to the other southern Africa isolates with good bootstrap support. 3.2. Antigenic variation among SAT1 and SAT2 viruses from different topotypes One-way antigenic relationships (r-values) were used to antigenically compare SAT1 and SAT2 viruses from various topotypes in Africa to selected reference strains within each serotype. The virus isolates and the average r-values are summarized in Table 1 and 2. Sera from cattle twice vaccinated with the two SAT1 viruses, i.e. SAR/9/81 and KNP/196/91, both belonging to topotype 1 (Figure 1), demonstrated r-values generally higher with isolates belonging to the same topotype compared to isolates in other topotypes. Among the 21 SAT1 isolates analyzed against the vaccine strains, 9.5% (n=2) and 14.2% (n=3), showed r-values ≥ 0.4 with the SAR/9/81 and KNP/196/91 viruses respectively, all belonging to topotype 1 (Table 1). Furthermore, 52% (n=11) and 66% (n=14) of the isolates had r-values in the range of 0.2 - 0.4, indicating that a high potency vaccine may still afford protection in animals. The remaining isolates reacted poorly to antisera directed against the vaccine strains indicating that the vaccines will most probably not protect animals in the field against these strains. Convalescent sera to a West African SAT1 virus, i.e. SAT1/NIG/5/81, showed reasonable cross-protection (r-values of 0.2-0.4) against 66% (n=14) of the SAT1 viruses (Table 1). Only one isolate had an r-value of ≥ 0.4, i.e. SUD/3/76, belonging to the same topotype than the Nigerian isolate. The difference in cross-reaction with the reference sera might be indicative of some shared epitopes of the isolate in question to the reference virus. The SAT2 vaccine strains, ZIM/7/83 and KNP/19/89, belong to different SAT2 topotypes, i.e. I and II, respectively (Figure 1). Two convalescent antisera, that of ERI/12/89 and RWA/2/01 (topotypes VII and VIII), were also included as references (Table 2). The antigenic variation for the SAT2 isolates was more pronounced with r-values below 0.2 even when vaccine and field strains from the same topotype were compared. With the exception of the reaction of the ZIM/7/83 isolate to the KNP/19/89 sera, none of the other SAT2 isolates showed r-values in the range of 0.4-1.0 against any of the four reference strains. Although none of the SAT2 isolates had an r-value of 0.20.4 against ZIM/7/83, at least 16.6% (n=4) of the isolates fall into this range with KNP/19/89 and 20% (n=5) against ERI/12/89 and RWA/2/01, respectively. From the data in Table 2 the convalescent sera (ERI/12/89 and RWA/2/01) seem to be more cross-reactive in general than the sera derived from the immunized animals (ZIM/7/83 and KNP19/89). Alternatively the two East Africa viruses might have a broader antigenic coverage. Nevertheless, Esterhuysen et al., 1988 have shown that convalescent cattle sera are consistently more cross-reactive than sera from immunized cattle, which explain the better coverage from these sera. The average r-values obtained for each field isolate against the reference sera were plotted against the total amino acid changes between the field strain and reference virus. Figure 2 indicates the first-order effect of genetic distance for SAT2 viruses. However, in the final model the end point serum titres were used in the analysis, as too much information was lost in the data processing from neutralising titre to r-value. 3.3. Variance structure There are a series of effects which may be either a source of variation in titre or error in measurement. In particular, we anticipate an experiment effect since this is what the r-value’s homologous titre denominator most directly corrects for (henceforth experiment). Additional sources of variation include the date/conditions (henceforth group); the cow from which the antiserum was drawn (animal); whether the antiserum is from a vaccinated or infected cow (vacc);

116

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

the strain the cow was infected or vaccinated with (protected); and the serotype of that strain (serotype). A repeated measure is a virus neutralisation test conducted within the same experiment, but additionally using the same challenge strain (challenge). These repeated measures are by definition the lowest level of model, representing as they do replicates of identical tests; the variance between them is therefore represented in the model by the residuals. The variance between different sets of replicates is now represented by a term which is the interaction of experiment and challenge, henceforth “single”. Initially log (titre) is modelled as a constant to build the variance structure as needed from the simplest possible model (containing only the necessary repeated measures term) to include factors identified as significant. Since historically experiment has been considered a significant factor, and experiment is the interaction term between group and animal, we test nested models which are combinations of animal, group and experiment. These are treated as random effects/factors. Other factors which may affect the variance structure are serotype, protected, challenge and vacc. The first three of these are treated as random effects, but vaccination status is a fixed effect with two possible alternatives (vaccinated or infected). There are a variety of ways of testing the effects of these terms on the variance. We considered two approaches (i) examining the model terms in two parts as they are described, building up terms and using likelihood ratio tests to compare nested models, or (ii) a stepwise regression of all of the terms. In both cases, Bonferroni corrections are required to take account of the multiple tests. In both cases, two terms – challenge and experiment – are highly significant (p < 10-15). The remaining terms, after correction, make no significant contribution. As a result our variance structure consists of a repeated measure term and 2 random effects – denoted single, challenge and experiment respectively. 3.4. Measuring genetic distance We have sequenced or have available GenBank sequences for all of the strains involved in neutralisation tests, allowing the number of changes (mutations, insertions or deletions) between any two strains to be straightforwardly counted. In addition, we have identified those residues which are on the surface, allowing the number of changes specifically corresponding to surfaceexposed residues between any two sequences to be counted. These changes are expected to correlate with antigenic distance, but the measure is imperfect because many changes will be antigenically neutral even on surface-exposed residues. Therefore, an analysis designed to identify antigenically important regions requires a baseline measure of genetic change that is antigenically neutral. One option is to select changes to residues which are not surface-exposed on the grounds that they cannot be directly involved in antigenic activity. Exploratory analyses indicate that this approach does work; it is possible however for hidden mutations to alter antigenic activity through some conformational change. For this reason, we prefer to use synonymous mutations. Because these correlate with antigenic distance (p<1015 ), they are used as a baseline against which to assess antigenically important changes. Candidate antigenic areas are chosen from the surface exposed residues by identifying the longest contiguous chains of residues which are on the surface for their whole length. These are expected to contain linear epitopes and parts of conformal epitopes, but are unlikely to contain whole conformal epitopes unless the protein folds back on itself almost immediately. A total of 38 such candidate areas were identified, corresponding mostly to the loops and terminals of the 3 surface proteins (VP1, VP2 and VP3). Of the 38, 11 showed no mutations, leaving 27 candidates, of which 27 in SAT1 but only 24 in SAT2 showed mutations. These candidate areas are tested with their respective datasets to determine if there is a significant positive correlation between increasing changes to the areas and increasing antigenic distance (i.e. decreasing titres). If so, while considering the combined dataset, we also investigated if there is a significant interaction with serotype; or if not, if there is an interaction with protection strain. 4. CONCLUSIONS Phylogenetic analysis in the past has been performed largely on FMDV sequences from the 1Dcoding region. Based on the complete P1-coding region, the phylogeny of 53 SAT isolates with

117

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

diverse geographic distribution in Africa revealed the same clusters within a serotype as previously described for 1D (Bastos et al., 2001; Bastos et al., 2003a,b). Phylogeny based only on the internally located 1A genomic region revealed similar structuring than that of the Leader proteasecoding region (Van Rensburg et al., 2002), although different from the outer capsid-coding regions. In the latter case, isolates grouped accoding to serotype indicating the same selective immune pressure for 1B, 1C and 1D capsid proteins. Darwinian selective pressure for the evolution of the non-structural proteins and the internally located 1A, more reasonably, exist to preserve structure and functionality and escape from immune response is less important (Van Rensburg et al., 2002). Antigenic analysis of field isolates in relation to vaccine strains, based on VNT’s, play a significant role in evaluating the suitability of existing vaccine strains (Jangra et al., 2005) even though significant variation has been reported with VNT’s. Antigenic coverage is also an important role player in the selection of new vaccine strains. It is generally accepted that r-values higher than 0.4 demonstrate a good cross-protection of the vaccine against the field isolate. Although SAT1 viruses displayed larger number of variable amino acids in a complete alignment, antigenic variation within SAT1 was less pronounced than for SAT2 viruses within the dataset. Better antigenic relationships were obtained for the SAT1 viruses belonging to the same topotype of the reference virus. The SAT2 reference strains on the other hand, did not have good antigenic relationships with most SAT2 isolates, even within the same topotype. The antigenic variation observed within the SAT types has serious implications on the choice of a vaccine strain. Furthermore we have developed a linear mixed-effects model of virus neutralisation titre for assessing cross-protection that accounts for the significant sources of variation in titre and errors in measurement, and have shown by cross-validation that this is in practice a good predictor. Using this approach, we can now better assess the protection afforded by a vaccine against a heterologous challenge than by standard calculation of an r-value alone. The techniques described above can be used for any FMDV serotype where cross-protection experiments have been carried out, both to identify potential epitopes and to predict protection for new strains. Indeed, this can also be used to assess the likely efficacy of potential vaccine strains against a group of challenge strains when a new vaccine is required. This can be done by exploiting historical datasets; it is a quick, low cost and therefore a valuable tool for better understanding antigenic relationships. This is a powerful tool that has the potential to be applied to a variety of different infectious agents including influenza A for which large historical records exists. 5. RECOMMENDATIONS No specific recommendations. 6. REFERENCES [1] Bastos, A.D.S., Boshoff, C.I., Keet, D.F., Bengis, R.G. and Thomson, G.R. 2000. Natural transmission of foot-and-mouth disease virus between African buffalo (Syncerus caffer) and impala (Aepyceros melampus) in the Kruger National Park, South Africa. Epid. Infect. 124: 591-598. [2] Bastos, A.D.S. 1998. Detection and characterisation of foot-and-mouth disease virus in subSaharan Africa. Onderstepoort J. Vet. Res. 65: 37-47. [3] Bastos, A.D.S., Haydon, D.T., Forsberg, R., Knowles, N.J., Anderson, E.C., Bengis, R.G., Nel, L.H. and Thomson, G.R. 2001. Genetic heterogeneity of SAT-1 type foot-and-mouth disease viruses in southern Africa. Arch. Virol. 146: 1537-1551. [4] Bastos, A.D., Anderson, E.C., Bengis, R.G., Keet, D.F., Winterbach, H.K. and Thomson, G.R. 2003a. Molecular epidemiology of SAT3-type foot-and-mouth disease. Virus Genes. 27: 28390. [5] Bastos, A.D.S., Haydon, D.T., Sangare, O., Boshoff, C.I., Edrich, J.L. and Thomson, G.R. 2003b. The implications of viral diversity within the SAT-2 serotype for control of foot-and-mouth disease in sub-Saharan Africa. J. Gen. Virol. 84: 1595-1606. [6] Crowther, J. R., S. Farias, W. C. Carpenter, and A. R. Samuel. 1993a. Identification of a fifth neutralizable site on type O foot-and-mouth disease virus following characterization of single and quintuple monoclonal antibody escape mutants. J. Gen. Virol. 74 (Pt 8):1547-1553. [7] Crowther, J. R., C. A. Rowe, and R. Butcher. 1993b. Characterization of monoclonal antibodies against a type SAT 2 foot-and- mouth disease virus. Epidemiol. Infect. 111:391-406. [8] Dawe, P. S., F. O. Flanagan, R. L. Madekurozwa, K. J. Sorensen, E. C. Anderson, C. M. Foggin, N. P. Ferris, and N. J. Knowles. 1994. Natural transmission of foot-and-mouth disease virus from African buffalo (Syncerus caffer) to cattle in a wildlife area of Zimbabwe. Vet Rec. 134:230-232.

118

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[9] Esterhuysen, J. J., G. R. Thomson, W. A. Ashford, D. W. Lentz, M. D. Gainaru, A. J. Sayer, C. D. Meredith, v. R. Janse, and A. Pini. 1988. The suitability of a rolled BHK21 monolayer system for the production of vaccines against the SAT types of foot-and-mouth disease virus. I. Adaptation of virus isolates to the system, immunogen yields achieved and assessment of subtype cross reactivity. Onderstepoort J. vet. Res. 55:77-84. [10] Grafen A. 2002. A first formal link between the price equation and an optimization program. J Theor Biol.; 217(1):75-91. [11] Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98. [12] Kitson, J. D. A., McCahon, D., and Belsham, G. J. Sequence analysis of monoclonal antibody resistant mutants of type O foot-and-mouth disease virus: Evidence for the involvement of the three surface exposed capsid proteins in four antigenic sites. Virology 179, 26-34. 1990. [13] Knowles NJ, Samuel AR. (2003). Molecular epidemiology of foot-and-mouth disease virus. Virus Res., 91; 65-80. [14] Rweyemamu, M.M., Booth, J.C., Head, M. and Pay, T.W.F. 1978. Microneutralization tests for serological typing and subtyping of Foot-and-Mouth Disease virus strains. Journal of Hygiene, Cambridge. 81, 107-123. [15] Rweyemamu, M. M., E. J. Ouldridge, M. Head, and R. Ferrari. 1984. The effect of antiserum quality on strain specificity assessment of foot-and-mouth disease virus by the neutralization reaction. J. Biol. Stand. 12:295-303. [16] Rweyemamu, M. M. 1984. Antigenic variation in foot-and-mouth disease: studies based on the virus neutralization reaction. J. Biol. Stand. 12:323-337. [17] Rweyemamu, M. M. and P. J. Hingley. 1984. Foot-and-mouth disease virus strain differentiation: analysis of the serological data. J. Biol. Stand. 12:225-229 [18] Samuel, A.R., Ouldridge, E.J., Arrowsmith, A.E.M, Kitching, R.O. and Knowles, N.J. 1990. Antigenic analysis of serotype O foot-and-mouth disease virus isolates from the Middle East, 1981-1988. Vaccine, 8: 390-196. [19] Samuel AR, Knowles NJ. (2001). Foot-and-mouth disease virus: cause of the recent crisis for the UK livestock industry. Trends Genet.; 17,421-4.. [20] Samuel, A.R. and Knowles, N.J. 2001. Foot-and-mouth disease type O viruses exhibit genetically and geographically distinct evolutionary lineages (topotypes). J. Gen. Virol. 82: 609621. [21] Sangare, O., Bastos A.D.S., Venter, E.H. and Vosloo, W. 2003. Retrospective genetic analysis of SAT-1 type foot-and-mouth disease outbreaks in West Africa (1975-1981). Vet. Microbiology. 93: 279-289. [22] Sangare, O., Bastos A.D.S., Venter, E.H. and Vosloo, W. 2004. A first molecular epidemiological study of SAT-2 type foot-and-mouth disease viruses in West Africa. Epidemiology and Infection, 132: 525-32. [23] Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25(24):4876-82. [24] Van Rensburg, H., D. Haydon, F. Joubert, A. Bastos, L. Heath, and L. Nel. 2002. Genetic heterogeneity in the foot-and-mouth disease virus Leader and 3C proteinases. Gene 289:19-29. [25] Vosloo, W. and Thomson, G.R. 2004. Natural Habitats in which Foot-and-Mouth Disease Viruses are Maintained. In: Foot-and-mouth disease. Domingo, E. and Sobrino, F. eds. Horizon Scientific Press, pp. 383-410. [26] Vosloo, W., Bastos A.D.S., Michel, A. and Thomson, G.R. 2001. Tracing movement of African buffalo in southern Africa. Rev. Sci. Tech. OIE. 20: 630-639. [27] Vosloo, W., Bastos, A.D., Kirkbride, E., Esterhuysen, J.J., Janse van Rensburg, D., Bengis, R.G., Keet, D.F. and Thomson, G.R. 1996. Persistent infection of African buffalo (Syncerus caffer) with SAT type foot-and-mouth disease viruses: rate of fixation of mutations, antigenic change and interspecies transmission. J. Gen. Virol. 77: 1457-1467. [28] Vosloo, W., Bastos, A.D.S., Sangare, O., Hargreaves, S.K. and Thomson, G.R. 2002. Review of the status and control of foot-and-mouth disease in sub-Saharan Africa. Rev. Sci. Tech. OIE. 21: 437-449. [30] Vosloo, W., Kirkbride, E., Bengis, R.G., Keet, D.F. and Thomson, G.R. 1995. Genome variation in the SAT types of foot-and-mouth disease viruses prevalent in buffalo (Syncerus caffer) in the Kruger National Park and other regions of southern Africa, 1986 1993. Epidem. Infect. 114: 203 218. [31] Vosloo, W., Knowles, N.J. and Thomson, G.R. 1992. Genetic relationships between southern African SAT-2 isolates of foot-and-mouth disease virus. Epidem. Infect. 109: 547-558.

119

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 100 99 85 99 1 100 97 91 79 11 100 2

KNP/02/89/2 KNP/19/89/2 KNP/51/93/2 SAR/16/83/2 ZIM/08/94/2 ZIM/GN/10/91 RHO/01/48/2 KEN/08/99/2 ANG/04/74/2 ZIM/07 /83/2 ZIM/01/88/2

ZIM/34/90/2 ZIM/14/90/2 99 93 ZIM/17/91/2 SAT2 9 100 KEN/03/57/2 12 UGA/02/02/2 100 8 ZAI/01/74/2 92 100 RWA/02/01/2 7 ERI/12/89/2 70 100 SAU/06/00/2 5 SEN/05/75/2 56 100 GHA/08/91/2 6 SEN/07/83/2 KNP/196/91/1 100 KNP/148/91/1 ZIM/HV/03/90 65 SAR/09/81/1 1 100 KNP/41/95/1 ZIM/GN/13/90 74 MOZ/03/02/1 100 3 ZIM/25/90/1 ZAM/02/93/1 92 100 KEN/05/98/1 99 93 TAN/37/99/1 2 79 ZIM/06/94/1 100 NAM/307/98/1 4 UGA/03/99/1 SAT1 84 UGA/01/97/1 100 5 7 SUD/03/76/1 100 NIG/05/81/1 NIG/08/76/1 1008 100 NIG/06/76/1 99 NIG/15/75/1 KNP/10/90/3 80 ZIM/05/91/3 100 BEC/16/53/3 SAT3 ZAM/04/96/3 100 UGA/02/97/3 A22 100 A12 100 O1Kauf 100 O1Tai 100

0.05

120

Figure 1: Minimum evolution tree depicting the gene relationships for P1-coding regions respectively of SAT1, 2 and 3 viruses from southern, western and East Africa. Bootstrap support is indicated.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 1: The average r-values and number of variable amino acids in the capsid proteins of SAT1 isolates as measured against reference strains. The topotypes based on 1D phylogeny are also indicated. SAT1 isolates

KNP/196/91 SAR/9/81 ZIM/GN/13/91 ZIM/HV/3/90 KNP/148/91 KNP/41/95 NAM/307/98 ZIM/6/94 MOZ/3/02 ZIM/25/90 TAN/1/99 TAN/37/99 ZAM/2/93 KEN/5/98 UGA/3/99 UGA/1/97 SUD/3/76 NIG/5/81 NIG/15/75 NIG/6/76 NIG/8/76

Topotype

SAR/09/81/1

KNP/196/91/1

NIG/05/81/1

1 1 1 1 1 1 2 2 3 3 3 3 3 3 4 5 7 7 8 8 8

variable amino acids 49 0 44 42 46 54 73 75 63 68 55 63 56 73 91 103 105 105 104 103

variable amino acids 0 49 38 31 17 50 67 77 59 62 55 60 58 80 102 107 109 108 106 106

variable amino acids 109 105 107 100 106 118 109 113 107 109 105 106 101 100 114 17 0 65 63 63

rvalue 0.44 1.00 0.50 0.40 0.37 0.25 0.21 0.23 0.24 0.27 0.13 0.25 0.16 0.23 0.26 0.24 0.28 0.19 0.17 0.10 0.14

rvalue 1.00 0.44 0.55 0.48 0.36 0.36 0.28 0.33 0.37 0.10 0.28 0.36 0.14 0.34 0.25 0.27 0.25 0.28 0.26 0.12 0.36

rvalue 0.21 0.34 0.23 0.20 0.34 0.23 0.22 0.17 0.32 0.25 0.27 0.10 0.23 0.32 0.29 0.60 1.00 0.21 0.17 0.18

Table 2: The average r-values and number of variable amino acids in the capsid proteins of SAT2 isolates as measured against reference strains. The topotypes are indicated. SAT2

Topotype ZIM/07/83

isolates

KNP/19/89

ERI/12/89

RWA/02/01

variable aa r-value variable aa r-value variable aa r-value variable aa r-value

KNP/19/89 1

0.14

1.00

0.25

0.10

KNP/2/98

0.09

0.11

0.12

0.19

KNP/51/93

0.07

0.09

0.12

0.18

SAR/16/83

0.03

0.05

0.10

ZIM/8/94

0.08

0.12

Z/GN/10/91 1

0.11

0.09

ZIM/7/83

1.00

0.41

0.20

0.19

ZIM/14/90

0.05

0.09

0.14

0.10

ZIM/17/91

0.13

0.05

0.14

0.24

ZIM/1/88

0.18

0.13

0.16

0.13

ZIM/34/90

0.14

0.28

RHO/1/48

0.17

0.12

0.21

0.27

KEN/8/99

0.05

0.07

0.04

0.06

GHA/8/91

0.09

0.18

0.21

0.03

0.05

0.12

0.16

SEN/5/75

100

0.02

0.05

0.03

0.06

SEN/7/83

0.10

0.20

106

0.25

0.35

SAU/6/00

LBR/1/74

0.06

0.26

ERI/12/89 7

0.14

0.31

1.00

0.32

RWA/2/01 8

0.13

0.09

0.23

1.00

KEN/03/57

121

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

ANG/4/74

0.09

0.10

0.04

0.05

UGA/02/02

0.02

0.03

0.04

0.08

ZAI/01/74

0.02

0.05

0.12

0.17

Figure 2: SAT2 average r-values plotted against genetic distance (non-synonymous mutations).

122

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 18 VACCINE MATCHING: RELIABILITY OF FOOT-AND-MOUTH DISEASE (FMD) r-VALUES DETERMINATION N. Mattion 1, f,* N. Goris 2, T. Willems 2, B. Robiolo 1, 6, E. Maradei 3, 6, C. Perez 3, 6, A. Perez Smitsaart 5, 6, N. Fondevila 5, 6, E. Palma 6, K. De Clercq 2 and J. La Torre 1, 6

3, 6

, E.

Centro de Virología Animal, Instituto de Ciencia y Tecnología Dr. César Milstein, CONICET, Saladillo 2468, (1440) Buenos Aires, Argentina. 2 Veterinary and Agrochemical Research Centre, Virology Department, Section of Epizootic Diseases, Groeselenberg 99, 1180 Brussels, Belgium. 3 Servicio Nacional de Sanidad Animal (SENASA), Fleming 1653, (1640) Martínez, Argentina. 4 Instituto Nacional de Tecnología Agropecuaria, Centro de Investigación de Ciencias Veterinarias y Agronómicas (INTA-CICVyA), CC77, (1708) Morón, Argentina. 5 Biogénesis Bagó S.A., Ruta Panamericana Km 38.2, (1619) Garín, Argentina. 6 Argentine FMD Interinstitutional Network for Research and Development in Foot-and-mouth Disease (RIIDFA), Rivadavia 1437, (1033) Buenos Aires, Argentina.

ABSTRACT Sera from animals involved in FMD vaccine potency and cross-protection trials performed using the Protection against Podal Generalization (PPG) test for two serotype A strains were analyzed by the virus neutralization test (VN) and liquid-phase ELISA (lpELISA) in three laboratories. Animals had been vaccinated with A24 Cruzeiro strain and challenged with homologous and heterologous (A/Arg/01) strains. Based on the PPG results, both FMDV strains were regarded as antigenically not closely related. Average VN r-values for medium and high serum titer classes from the A24 Cruzeiro vaccinated animals were in line with the heterologous PPG outcome for the three testing laboratories. The corresponding lpELISA r-values were slightly higher and indicated a closer relationship between both strains. Pooling of serum samples significantly reduced the inter-animal and inter-trial variation and led to r-values that mimicked the in vivo PPG cross-protection outcome. 1. INTRODUCTION Conventional, safe, inactivated vaccines have been and are being used for disease prevention, combat and control in numerous FMD outbreaks. Only several antigenic representative, crossreactive (i.e. broad spectrum of reactivity) FMDV strains are kept over liquid nitrogen in antigen reserves. In order to assess the suitability of the strains for a particular situation (e.g. outbreak scenario), a vaccine matching test should be readily performed. The gold standard matching test is the in vivo cross-protection test which is highly variable [4], time-consuming and not very animal friendly. Consequently, by the time the result is known, the outbreak might have extended and caused irremediable damage. Several indirect vaccine matching tests have been performed based on serological data in which an indirect relationship value (r-value) is considered for each reference bovine post vaccinal sera (reviewed by [8]). There has been a lot of controversy surrounding indirect r-values as the protective immunity to FMDV infection is complex and involves not only humoral antibody responses but also factors derived from innate immunity. The latter are not detected by VN in cells or by antigen binding antibodies in ELISA. Moreover, a correlation between r-values and heterologous protection has not always been observed. There are documented cases where cross-protection was found in spite of low r-values [2] and vice versa. Variation between batches of bovine post vaccinal sera have also led to inconsistent results [6, 8]. Furthermore, it is important to point out that the in vivo or serological assays and the reagents used in different experiments have not been harmonized worldwide, or even within a particular geographic region [4, 8]. 2. MATERIALS AND METHODS

123

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The objective of this work was to use the data obtained from ten PPG FMD vaccine potency trials with homologous and heterologous challenge [4] to help clarify the selection of reference sera for the determination of indirect r-values. 2.1. Vaccine trials and virus strains The FMDV serotype A virus strains used in the study [A24 Cruzeiro/Brazil/55 (A24 Cruzeiro) and A/Argentina/2001 (A/Arg/01)] were provided by the Argentine Animal Health Service (SENASA). The vaccine used was a monovalent formulation of purified A24 Cruzeiro antigen in a water-in-oil emulsion, with an antigenic mass of 10 γg as reported previously [4]. Additionally, FMDV strains O1/Campos/Brazil/58 (O1 Campos) and C3/Indaial/Brazil/71 (C3 Indaial) were also used in liquidphase ELISA (lpELISA) to study the variability of r-value assessment for distantly related strains. The animals and PPG vaccine potency tests were previously described [4]. Six replicate homologous and four replicate heterologous trials were conducted in Argentina according to SENASA Act No. 351/2006 [1], except that animals were challenged at 30 days post vaccination (dpv) instead of at 90 dpv. Sixteen animals (bovine) without previous antibodies to structural and non-structural FMDV proteins were vaccinated with the A24 Cruzeiro vaccine. At 30 dpv the vaccinates plus two unvaccinated controls were challenged with 104 suckling mouse infectious dose 50% (SMID50) of FMDV strain A24 Cruzeiro (six homologous tests) or A/Arg/01 (four heterologous tests). The PPG percentage was calculated according to the following formula: %PPG=

number of protected animals x 100 number of vaccinated animals

(1)

2.2. Participating laboratories and serological tests The serum collection was tested at three different laboratories: the Veterinary and Agrochemical Research Centre (VAR, Brussels, Belgium), Centro de Virología Animal (CEVAN, Buenos Aires, Argentina) and SENASA (Martínez, Argentina). The VAR used the FMDV A24 Cruzeiro and A/Arg/01 VN and lpELISA as previously described [4] whereas CEVAN made use of the FMDV A24 Cruzeiro and A/Arg/01 lpELISA [9]. SENASA applied the FMDV A24 Cruzeiro and A/Arg/01 VN. VN and lpELISA log10 homologous titers were further classified in low, medium and high (Table 1). The limits were assigned arbitrarily, but a SENASA VN titer of 1.60 and a CEVAN lpELISA titer of 2.20 are around the values that may be considered as protective antibody levels for bovines [7]. Table 1: Classification of log10 serum titers in the respective A24 Cruzeiro serological assays VN Low

lpELISA Medium

CEVAN

SENASA

t<1.60 1.6≤t≤2.00

High t>2.00

VAR t<1.81 1.81≤t≤2.26 t>2.26 NA = not applicable; t = log10 serum titer

Low

Medium

High

t<2.20

2.20≤t≤2.80 t>2.80

NA t<2.11

2.11≤t≤2.41 t>2.41

2.3. r-values determination The r-values were determined with VN and lpELISA titers according to the following formula: r1 =

reciprocal titer of reference serum against field virus reciprocal titer of reference serum against vaccine virus

(2)

The interpretation of the results was based on Ferris and Donaldson [3] for lpELISA and on Rweyemamu [10] for VN. Following these guidelines, r-values based on lpELISA between 0.4 and 1.0 or VN r-values superior to 0.3 indicate a close relationship between the vaccine strain and the field isolate. lpELISA r-values between 0.2 and 0.39 signify that the vaccine strain might be suitable for use if no closer match can be found provided that a potent vaccine is used and animals are immunized preferably more than once, whereas a lpELISA r-value inferior to 0.2 or a VN rvalue smaller than 0.3 point toward a vaccine strain that is unlikely to protect against challenge with the field isolate [3, 10]. For each trial (T), a r-value (rT) was calculated from the mean homologous and mean heterologous serum titers of the16 vaccinated animals according to the following formula: rT =

124

reciprocal mean serum titer per trial against A/Arg/01 reciprocal mean serum titer against A24 Cruzeiro

(3)

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

A mean rT ( rT ) was subsequently calculated for the group of trials considered (e.g. only the first 6 trials or only the last four trials):

= 1

with n = number of trials under consideration

∑r n

(4)

T ,i

i =1

When the log10 serum titers were classified into groups of low, medium and high, a r-value for each serum titer class (C) was calculated per trial (rT,C) similarly to formula (3) and the mean per serum titer class for all trials under consideration was also determined ( rT ,C ) in line with formula (4). Furthermore, a r-value of a group (G) of trials was alternatively calculated for each log10 serum titer class using the reciprocal mean serum titers of all the animals belonging to the group of PPG trials considered and the log10 serum titer class (rG,C). The standard deviations (SD) and 95% confidence intervals (95%CI) were calculated for each mean r-value. 2.4. Serum pooling and r-value determination In order to evaluate the effect of pooling of serum samples on r-value determination, different serum pools were constituted and tested in the VAR VN. Pools were formulated based on serum samples of all vaccinated animals per PPG trial, on all protected animals per PPG trial, on all unprotected animals per PPG trial, on ten randomly selected animals, on ten randomly selected protected animals and on ten randomly selected unprotected animals. In total 58 pools were analyzed each containing between 2 to 16 pooled sera. For comparative purposes, a set of r-values were determined. First of all, the r-value of the pool (rpool) was calculated according to formula (2) and the mean of rpool was also determined ( rpool ). Subsequently, the average r-value ( ri ) based on the individual (i) r-values (ri) of each serum sample in the pool was calculated as follows:

1 n ∑ ri with n = number of serum samples per pool n i −1

(5)

A third r-value (rt) was estimated based on the reciprocal mean serum titers (t) against both FMDV strains for all serum samples that make up the pool as follows: rt = reciprocal mean serum titer per pool against A/Arg/01

(6)

reciprocal mean serum titer against A24 Cruzeiro The mean

, the mean rt ( rt ), SD and 95%CI were also determined.

3. RESULTS 3.1. PPG trials and serum collection The mean percentage PPG for six homologous A24 Cruzeiro challenge PPG potency trials was 88.5 % [95%CI: 80.7-93.5] and for four heterologous A/Arg/01 challenge PPG cross-protection trials was 26.6% [17.4-38.5]. All animal trials were performed within an 11-month period (Jan-Nov 2006), during which time a slight decrease in vaccine potency was noted after September 2006 [4]. Given the decrease in vaccine potency as of trial 7, sera from the first six trials (carried out between Jan and Sep 2006) were analyzed separately for r-value determination, instead of using the entire serum collection. Moreover, to study the influence of vaccine potency on the r-value outcome, r-values based on the first four (Jan-Feb 2006) and last four trials (Oct-Nov 2006) were also compared. 3.2. r-values for non-pooled serum samples calculated from VN and lpELISA titers Based on the cross-protection PPG results, A24 Cruzeiro and A/Arg/01 cannot be regarded as closely related FMDV strains, since a single vaccination with the former did not induce a sufficient level of cross-protection against the later (%PPG < 75%). Since the in vivo test is the direct comparison test, discordant VN r-values above 0.3, and lpELISA r-values above 0.4 will be shown in italics in each of the respective tables. For the more distant strains O1 Campos and C3 Indaial (belonging to different serotypes), lpELISA r-values above 0.2 are also shown in italics. For O1 Campos and C3 Indaial strains only lpELISA titers are available. The data derived from A24 Cruzeiro SENASA VN log10 titers inferior to 1.60 resulted in some rT,Cvalues for A/Arg/01 above 0.3 as shown by the mean of the six sets of trials analyzed (Table 2). The same was noted for A24 Cruzeiro VAR VN titers inferior to1.81. The data derived from A24

125

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Cruzeiro CEVAN lpELISA titers smaller than 2.20 resulted in a

rT ,C

-value above 0.4 as well,

whereas this phenomenon was not observed for A24 Cruzeiro VAR lpELISA serum titers inferior to 2.11 (Table 3). The upper limits of the

rT ,C

-value 95%CI were also above 0.3 or 0.4 for the low

A24 Cruzeiro VN and CEVAN lpELISA serum titer class, respectively. For SENASA VN, VAR VN, CEVAN lpELISA and VAR lpELISA log10 serum titers of at least 1.60, 1.81, 2.20 and 2.11, respectively, serological values of relatedness were below 0.3 or 0.4, as shown by the medium or high A24 Cruzeiro serum titers columns in Tables 2 and 3, which is in line with the obtained in vivo PPG cross-protection results. Table 2: VN r-values for FMDV strain A/Arg/01using A24 Cruzeiro vaccination (trials 1-6) SENASA VN Low

rT ,C SD

0.317

0.079

VAR VN

Medium

High

Low

Medium

High

0.145

0.117

0.339

0.196

0.091

0.038

0.083

0.352

0.083

0.058

95%CI 0.253-0.3800.115-0.1750.050-0.183 0.017-0.6610.120-0.2720.038-0.144 a

rG,C 0.300 0.145 0.092 r-values superior to 0.3 are depicted in italics

0.454

0.208

0.081

Table 3: lpELISA r-values for FMDV strain A/Arg/01using A24 Cruzeiro vaccination (trials 1-6) CEVAN lpELISA

rT ,C SD

VAR lpELISA

Low

Medium

High

Low

Medium

High

0.436

0.273

0.245

0.190

0.266

0.225

0.124

0.076

0.131

0.056

0.105

0.032

95%CI 0.337-0.5340.212-0.3340.140-0.350 0.139-0.2410.170-0.3620.196-0.254 rG,C a

0.426

0.261

0.215

0.200

0.243

0.217

r-values superior to 0.4 are depicted in italics

A similar trend was observed for the data derived from O1 Campos and C3 Indaial lpELISA titers where

rT ,C

-values above 0.2 were found for the low titer classes (Table 4). For higher titer classes,

-values and upper limits of 95% CI were always below the 0.2 threshold for distantly related

strains. The bottom lines of Tables 2-3 show rG,C-values calculated on the basis of the mean of the A24 Cruzeiro and A/Arg/01 VN or lpELISA serum titers of all animals involved in trials 1 to 6 per serum titer class. The same calculation carried out using the mean of serum titers for all animals involved in the ten trials (data not shown) also led to high r-values when using low titer sera. For the more distant FMDV strains O1 Campos and C3 Indaial all rG,C-values based on lpELISA titers were found to be below 0.2 regardless of the number of trials considered (i.e. trials 1 to 6 or trials 1 to 10) (Table 4). Table 4: CEVAN lpELISA r-values for FMDV strains O1 Campos and C3 Indaial (trials 1-6)

O1 Campos Low

rT ,C SD

0.227 0.087

C3 Indaial

Medium

High

Low

Medium

High

0.087

0.050

0.188

0.075

0.056

0.019

0.027

0.079

0.028

0.043

95%CI 0.157-0.2960.072-0.1020.028-0.071 0.125-0.2500.052-0.0970.022-0.090 rG,C

126

0.171

0.078

0.044

0.143

0.065

0.044

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

r-values superior to 0.2 are depicted in italics

3.3. Influence of vaccine potency on r-value determination The calculation of r-values were thus based on VN and lpELISA serum titers from the first four (i.e. trials 1 to 4, Jan-Feb 2006) and the last four trials (i.e. trials 7 to 10, Oct-Nov 2006) (Tables 5 and 6). For r-values based on VN titers, the results obtained for trials 1 to 4 were not significantly different from the results obtained for trials 1 to 6 as seen by the highly overlapping 95%CI for both participating laboratories. Slightly, although not-significantly, higher r-values based on SENASA VN were obtained when using data from trials 7 to 10 for the low and medium serum titer class; whereas slightly lower r-values were found in all serum titer classes for the VAR VN (Tables 2 and 5). Interestingly, for the last four trials, no SENASA VN serum titers greater than 1.6 for FMDV strain A24 Cruzeiro were obtained (Table 5). Table 5: VN r-values for FMDV strain A/Arg/01 determined for the first four and the last four trials SENASA VN n=4 (trials 1-4) Low

rT ,C 0.329a 0.099

n= 4 (trials 7-10)

Medium

High

Low

Medium

High

0.160

0.151

0.427

0.200

0.038

0.082

0.161

0.086

95%CI 0.230-0,426 0.122-0.1980.071-0.232 0.269-0.5850.115-0.284na VAR VN n=4 (trials 1-4) Low

rT ,C

0.383

0.488

n= 4 (trials 7-10)

Medium

High

Low

Medium

High

0.204

0.104

0.279

0.194

0.045

0.100

0.064

0.028

0.144

0.027

95%CI -0.164-0.9300.092-0.3160.032-0.176 0.248-0.3100.033-0.3550.015-0.075 na = not available;

r-values superior to 0.3 are depicted in italics

The r-values based on lpELISA titers based on the first four trials are also more in line with the in vivo heterologous PPG results (%PPG < 75%), when sera from the low serum titer class are discarded (Table 6). Moreover, r-values based on lpELISA titers seemed to be more influenced by vaccine potency as shown by the high CEVAN lpELISA

rT ,C

-values for all serum titer classes (Table

6). Interestingly, no VAR lpELISA serum titers greater than 1.81 for FMDV strain A24 Cruzeiro were obtained for the last four trials. Table 6: lpELISA r-values for FMDV strain A/Arg/01 determined for the first four and the last four trials CEVAN lpELISA n=4 (trials 1-4) Low

rT ,C SD

0.439 0.140

n= 4 (trials 7-10)

Medium

High

Low

Medium

High

0.240

0.192

0.524

0.457

0.496

0.059

0.065

0.038

0.078

0.093

95%CI 0.301-0.5770.183-0.2980.128-0.255 0.486-0.5620.381-0.5330.405-0.588 VAR lpELISA n=4 (trials 1-4)

rT,C

n= 4 (trials 7-10)

Low

Medium

High

Low

Medium

High

0.198

0.271

0.233

0.178

0.087

127

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

0.067

0.103

0.021

0.057

0.069

95%CI 0.123-0.2730.156-0.3860.209-0.257 0.114-0.2420.010-0.164na a r-values superior to 0.4 are depicted in italics; na = not available 3.3. Influence of pooling on r-value determination Irrespective of the type of samples and the number of samples pooled, the 58 individual VAR VN rpool-values of the respective pools were found to be below the 0.3 threshold (range: 0.006-0.250). Consequently, the

rpool

-values and upper limits of the 95%CI were also below 0.3, which

correlates to the in vivo cross-protection data. As stated above, the ri-values of the individual sera were highly variable. As a result, the mean

-values of the pools were significantly higher as is

apparent from the non-overlapping 95%CI. Moreover, one third of the upper limits of the

-value

95%CI was greater than 0.3 (Table 7). These violations correlated to the pools of sera from animals that were unprotected against live virus challenge. Similarly higher r-values were also obtained based on the mean serum titers per pool (rt and

) in which the upper limit of the

95%CI was superior to 0.3 in at least one of the six cases, which again corresponded to a group of animals that were found to be unprotected in vivo. Table 7: The effect of pooling on r-value determination for FMDV strain A/Arg/01 using VAR VN 01 and using A24 Cruzeiro vaccination (trials 1-10) No. No of rpool Pool identificationof sera/pool [95%CI] pools 0.048±0.034 a All animals per PPG 10 16 trial [0.0240.072] 0.045±0.029

Protected animals 10 per PPG trial

2-16

Unprotected animals per trial

2-14

Randomly animals

[0.0250.066] 0.055±0.042

PPG 7

selected

[0.0190.091] 0.105±0.067

Randomly selected 10 protected animals

Randomly selected unprotected 6 animals

[0.0660.144] 0.057±0.031 [0.0350.079] 0.107±0.045 [0.0660.148]

mean

[95%CI]

[95%CI] 0.186±0.12 0.207±0.098 5 [0.138[0.0970.277] 0.275]b 0.174±0.12 0.190±0.122 2 [0.104[0.0880.277] 0.260] 0.210±0.12 0.229±0.142 2 [0.109[0.1070.349] 0.313] 0.142±0.06 0.202±0.092 8 [0.148[0.1030.255] 0.181] 0.103±0.04 0.170±0.058 2 [0.129[0.0730.211] 0.133] 0.213±0.06 0.280±0.076 9 [0.210[0.1500.350] 0.276]

Results b

are expressed as mean ± standard deviation r-values superior to 0.3 are depicted in italics

4. DISCUSSION Based on the previously reported in vivo PPG results, FMDV strains A24 Cruzeiro and A/Arg/01 are regarded as antigenically not closely related. Therefore, we hypothesized that corresponding rvalues calculated from VN serum titers must be below 0.3, and inferior to 0.4 or 0.2 when using lpELISA serum titers. The r-values calculated using sera from all A24 Cruzeiro vaccinated animals, irrespective of their in vivo protection status, and from all ten PPG trials performed within 11 months showed a high inter-animal and inter-trial variation (data not shown) with r-values varying from indicating that both strains were antigenically sufficiently similar to indicating that the vaccine

128

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

strain is unlikely to protect against A/Arg/01 challenge. However, when the animals were classified into groups of low, medium or high responders based on the A24 Cruzeiro serum titer, r-values based on medium and high titer were more closely correlated with the in vivo cross-protection studies. Moreover, r-value estimations become less precise when using the data from the last four PPG trials (i.e. trials 7 to 10, Tables 5 and 6). This observation was more evident for lpELISA titers than for VN and might be due to a slight decrease in vaccine potency [4]. The average VN rG,C-values for medium and high serum titer classes from the A24 Cruzeiro vaccinated animals are in line with the heterologous PPG outcome for both testing laboratories, suggesting that the vaccine strain A24 Cruzeiro is unlikely to protect against the field isolate A/Arg/01. The corresponding lpELISA r-values are slightly higher and indicate a closer relationship between both strains. This seems to indicate that the VN test is the preferred test for vaccine matching purposes. In the case of the more distant FMDV strains O1 Campos and C3 Indaial, lpELISA r-values also clearly indicate a complete lack of cross-protection regardless of the trials considered, except for the O1 Campos low titer class (Table 4). Unfortunately, VN titers were unavailable for these strains. Pooling of serum samples significantly reduced the inter-animal and inter-trial variation, irrespective of the number of serum samples in the pool (ranging from 2 to 16) and the type of serum samples pooled (ranging from sera from unprotected animals to randomly selected animals). The upper limit of 95%CI based on alternative r-value calculations for these pools were found to be above 0.3 when sera from unprotected animals were used. This is not surprising, as generally unprotected animals display lower serum titers [5] and this study demonstrates that low titer sera are less suited for r-value determination. 5. CONCLUSIONS According to our results a suitable reference serum for vaccine matching experiments might be a pool of at least five sera or medium to high VN or lpELISA titer sera, from cattle vaccinated with a high potency FMD vaccine. The preferred test for r-value determination would seem to be the VN. 6. ACKNOWLEDGEMENTS The study was funded by the Federal Public Service Health, Food Chain Safety and Environment (grant RT-05/06-ALTANDI-2) and the Argentine Beef Promotion Institute (IPCVA). The authors wish to thank Ina Much, Alejandro Ham and Mariela Guinzburg for their valuable technical assistance. 7. REFERENCES [1] Animal Health Service (SENASA). Act Nº 351/2006 – In: Boletín Oficial Nº 30.940, Argentina, July 5th, 2006 (available at http://infoleg.mecon.gov.ar/infolegInternet/anexos/115000-119999/117636/norma.htm). [2] Brehm KE, Kumar N, Thulke HH, Haas B. 2008. High potency vaccines induce protection against heterologous challenge with foot-and-mouth disease virus. Vaccine 26(13):1681-7 [3] Ferris NP, Donaldson AI. 1992. The World Reference Laboratory for Foot and Mouth Disease: a review of thirty-three years of activity (1958-1991). Rev Sci Tech 11(3):657-84. [4] Goris N, Maradei E, D’Aloia R., Fondevila N, Mattion N, Perez A, et al. 2008. Foot-andmouth disease vaccine potency testing in cattle using homologous and heterologous challenge strains: Precision of the "Protection against Podal Generalisation" test. Vaccine 26:3432-37. [5] Goris, N., Willems, T., Diev V., Merkelbach-Peters, P., Vanbinst, T.,Van der Stede, Y., Kraft H-P., Zakharov, V., Borisov, V., Nauwynck H., Haas,B., De Clercq, K. 2008. Indirect foot-and-mouth disease potency testing based on a serological alternative. Vaccine 26: 3870-79. [6] Kitching, RP, Rendle, R., Ferris, N. P. 1988. Rapid correlation between field isolates and vaccine strains of foot-and-mouth disease virus. Vaccine 6, 403-408. [7] Maradei E., La Torre J.,Robiolo B., Esteves J., Seki C., Pedemonte A., Iglesias M., D’Aloia R., and N Mattion. Updating of the correlation between lpELISA titres and virus challenge for the assessment of the potency of polyvalent aphtovirus vaccines in Argentina. Vaccine 2008, doi:10.1016/j.vaccine.2008.09.033. [8] Paton DJ, Valarcher JF, Bergmann I, Matlho OG, Zakharov VM, Palma EL, Thomson GR. 2005. Selection of foot-and-mouth disease vaccine strains--a review. Rev Sci Tech 24(3): 981-93.

129

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[9] Robiolo B, Grigera PR, Periolo OH, Seki C, Bianchi T, Maradei E, et al. 1995. Assessment of foot-and-mouth disease vaccine potency by liquid-phase blocking ELISA: a proposal for an alternative to the challenge procedure in Argentina. Vaccine 13: 1346-52. [10] Rweyemamu MM. 1984. Antigenic variation in foot-and-mouth disease: studies based on the virus neutralization reaction. J Biol Stand 12(3):323-37.

130

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 19 TOWARDS VACCINE SELECTION GUIDELINES FOR EACH REGIONAL VIRUS POOL OF FOOT-AND-MOUTH DISEASE

D. J. Paton Secretariat for the OIE/FAO FMD Reference Laboratories Network, Institute for Animal Health, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, UK.

1. INTRODUCTION This presentation reports on the outcome of discussions amongst partners of the Network of OIE/FAO FMD Reference Laboratories, at a meeting in Lanzhou China, 15-19 September 2008, to see if vaccine strains tailored to cover the needs of particular regions could be identified and provide targeted, regionalised vaccine recommendations to complement those of the World Reference Laboratory (WRLFMD) which are not presented in a region-specific manner. 2. MATERIALS AND METHODS Areas affected by endemic FMD can be subdivided into seven ecosystems or watersheds associated with particular FMDV serotypes ad topotypes. For each of these virus pools, working groups were asked to assess the position of the watersheds, list the vaccine seed viruses appropriate for each pool and its vaccine priority in 2008 and to consider what additional work is needed to improve these priorities and for better FMD control. 3. RESULTS In some parts of the world, this was relatively straightforward due to the existence of regional control programmes that have already identified vaccine strain requirements. At the other extreme, where there is low demand for and availability of FMDV vaccines, there is consequently little incentive to undertake the research and development needed to provide tailored vaccine strains. 3. CONCLUSIONS AND RECOMMENDATIONS The ecosystem-based, watershed concept is useful, but watershed boundaries sometimes overlap or are uncertain emphasising the need for continuing and in places improved surveillance of circulating viruses. Border areas between pools and neighbouring areas to blind spots might be targeted. A range of measures might facilitate reporting and sample submission to reference laboratories, including: the provision of incentives in the form of vaccine or training in return for samples; submission and analysis of non-infectious materials; fostering regional projects to study the prevalence and genotypes of circulating viruses; establishing new regional laboratories, for example in West Africa. There is variable harmonisation of vaccine strain use at national and regional level and a lack of coherent information on availability and use of different vaccine strains. Local decision makers sometimes have difficulty interpreting non-regionalised vaccine recommendations and regional advice would be useful. However, conflicts of interest that may affect the impartiality of advice given must be avoided. Regional advice on vaccine selection should be provided in future Reference Laboratory Network reports. Vaccine matching requirements differ for emergency use and prophylaxis; the former may require a more exact match, whereas for the latter, generic broadly reactive strains may be more appropriate. Vaccines held in reserves of FMD-free countries often differ from those used in endemic countries. Some areas have no tailored vaccine supply and few measures to control suitability, often related to low demand and public identification of need versus private supply. In such cases, more systematic antigenic matching studies are needed to provide confidence in available vaccine viruses and to develop new ones, but it is not always clear when this is a research

131

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

or a commercial activity. Therefore, there may be benefit in seeking to clarify roles with respect to vaccine development and selection between reference laboratories and vaccine producers. Finally, the affordability and quality control of vaccines are separate but very important issues. 4. ACKNOWLEDGEMENTS The input of all participants at the meeting of the Network of OIE/FAO FMD Reference Laboratories, Lanzhou China, 15-19 September 2008 is gratefully acknowledged.

132

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 20 EXPRESSION OF FMDV CAPSID PROTEINS IN SILKWORM-BACULOVIRUS EXPRESSION SYSTEM AND ITS UTILIZATION AS AN EMPTY CAPSID VACCINE Z. Li1, Y. Yi2, X. Yin1, Zhidong. Zhang3, Y. Li3, Zhifang Zhang2* and J. Liu1 1

State Key Laboratory of Veterinary Etiological Biology, Key Laboratory of Animal Virology of Ministry of Agriculture, Lanzhou Veterinary Research Institute, Chinese Academy of Agriculture Sciences, Lanzhou, Gansu, China 2 Biotechnology Research Institute, Chinese Academy of Agricultural Sciences, Beijing, China 3 Institute for Animal Health, Pirbright, Woking, Surrey, GU24 0NF, UK

ABSTRACT Introduction Foot-and-mouth disease (FMD) is a highly contagious disease of livestock that causes severe economic loss in susceptible cloven-hoofed animals. Although the traditional inactivated vaccine has been proved effective, it may lead to a new outbreak of FMD because of either incomplete inactivation of FMDV or the escape of live virus from vaccine production workshop. Thus, it is urgent to develop a novel FMDV vaccine that is safer, more effective and more economical than traditional vaccines. Methodology and principal findings A recombinant silkworm baculovirus Bm-P12A3C which contained the intact P1-2A and 3C protease coding regions of FMDV Asia 1/HNK/CHA/05 was developed. Indirect immunofluorescence test and sandwich-ELISA were used to verify that Bm-P12A3C could express the target cassette. Expression products from silkworm were diluted to 30 folds and used as antigen to immunize cattle. Specific antibody was induced in all vaccinated animals. After challenge with virulent homologous virus, four of the five animals were completely protected, and clinical symptoms were alleviated and delayed in the remaining one. Furthermore, a PD50 (50% bovine protective dose) test was performed to assess the bovine potency of the subunit vaccine. The result showed the subunit vaccine could achieve 6.34 PD50 per dose. Conclusion The results suggest that this strategy might be used to develop the new subunit FMDV vaccine. Keywords: Foot-and-mouth disease virus; Silkworm-baculovirus expression system; Subunit vaccine 1. INTRODUCTION Foot-and-mouth disease (FMD) is an economically important disease of domestic and wild clovenhoof animals including cattle, swine, goat, sheep and buffalo. It can result in great reduction of productivity in adult animals and death in young animals. At present, vaccination is a major means of FMD control in most endemic areas. Although the inactivated vaccine has been shown to be effective, it may lead to new outbreaks of FMD because of either the incomplete inactivation of FMDV in large-scale production or the escape of the live virus from vaccine production workshops [1] . Therefore several expression systems such as E.coli [2], transgenic plant [3], yeast [4], adenovirus vector [5-9], vaccinia virus vector [10], and DNA vaccine [11], have been used for expression of FMDV antigen to prepare subunit vaccines. But such methods have problems such as poor immunogenic capability or low efficiency. Adenovirus based vaccine known for its best protective effects can protect 5 of 5 vaccinated cattle, but this vaccine is still unacceptable because of safety problem and preservation difficulty. The baculovirus expression system, a valuable expression system to produce virus-like particles, has successfully produced many kinds of empty viral capsid [12, 13], such as rabbit hemorrhagic disease virus, Norwalk-like viruses, SARS and so on [14, 15, 16]. Compared to the baculovirus expression system (AcNPV-Sf cell), silkworm- baculovirus expression system has distinct advantages [17, 18]. First, expression levels in silkworm are 50-1000 times higher

133

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

than that in insect cell lines. Second, silkworm don not have any pathogens that can cross infect with vertebrates and animal serum is not needed to produce foreign proteins in this expression system, so that the expressed antigens are safer to vertebrates. In view of all these advantages, the silkworm-baculovirus expression system was employed for expression of intact P1-2A 3C coding regions of FMDV Asia I/HNK/CHA/05. All five cattle that were vaccinated with diluted expression antigen were induced specific antibody, four of which were considered completely protected. Furthermore, the PD50 (50% bovine protective dose) value of the subunit vaccine was 6.34 in bovine potency test.

、

2. MATERIALS AND METHODS 2.1 Viruses and cell lines FMDV Asia I/HNK/CHA/05 strain (GenBank accession number EF149010) was propagated in BHK21 cell line, and preserved in Lanzhou Veterinary Research Institute of Chinese Academy of Agriculture Sciences. The parental virus BmBacPAK-6 (Chinese patent: 1242428), Bm-N cell line and silkworm variety JY1 used for the experiment were maintained in Biotechnology Research Institute of Chinese Academy of Agriculture Sciences. The BmBacPAK-6 and recombinant virus were maintained in Bm-N cells at 27◦C in TC-100 insect medium (Sigma) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen). 2.2 Construction and screening of recombinant baculovirus The intact P1-2A and 3C protease coding regions were inserted into baculoviral transfer vector pVL1393 and named pVL-P12A3C. The baculoviral transfer plasmid pVL-P12A3C was co-transfected with linearized Bm- BacPAK6 DNA into Bm-N cells by liposome- mediated method [19]. The cotransfection supernatant was subject to plaque assays to screen the individual viral plaques. 2.3 Expression of FMDV polyprotein in Bm-N cells The expression of FMDV polyprotein in Bm-N cells infected with Bm-P12A3C was analyzed by immunofluorescence test (IFAT) and sandwich-ELISA. The Bm-P12A3C was multiplied in Bm-N cells. Bm-N cells (2.0×105) were cultured on cover slips and inoculated at a MOI of 10 pfu with Bm-P12A3C. After 48 hours post infection (hpi), IFAT was conducted to analyze the expression of FMDV proteins. When Bm-N cells infected with Bm-P12A3C were partial floating, they were detached and collected (about 72hpi), the cells pellet was freezed and thawed at -70/37◦C in PBS for three times and centrifuged at 10,000g for 5 min at 4°C. The supernatant was tested using the sandwich-ELISA method. 2.4 Expression of polyprotein in silkworm Early fifth-instar silkworms were infected with the recombinant virus at about 105 pfu per larva. The dying silkworm’s haemolymph was collected on ice and stored at -20°C for sandwich-ELISA. In order to determine the time course of expressed antigen in silkworm and the optimum acquisition time for large scale’s production, the infected silkworm’s haemolymph was collected every 12h starting at 60hpi for determining the expression course of target antigen.

2.5 Detection of specific antibody by LPBE and serum neutralization test (SNT) Silkworm haemolymph was lysed ultrasonically and cell debris was removed by centrifugation. The diluted supernatant was used to produce vaccine. Liquid-phase blocking ELISA LPBE was performed to determine the antibody titer for screening of candidate cattle according to the standard method of World Organization for Animal Health, Office International desEpizooties (OIE) before vaccination. Seven cattle (6-8 months old) were immunized by intramuscular inoculation at the site in the neck. Five cattle were vaccinated with 3ml/animal of vaccine with Bm-P12A3C’s, while two control cattle were vaccinated with the same dose of vaccine with BmBacPAK-6’s. Cattle serum were collected at 7, 14, 21and 28 days postvaccination (dpv). Antibody against FMDV was detected by LPBE method and SNT.

（

）

2.6 Challenge with virulent homologous FMDV According to the descriptions by standard protocol of OIE, all animals were challenged by intradermal inoculation at two sites in the tongue with 10,000 bovine infectious doses (BID50) of Asia I /HNK/CHA/05 at 28 dpv. The body temperature of the animals was monitored daily. The restrained animals were carefully examined in the mouth, and feet every day for the first 10 days after challenge.

134

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.7 PD50 test We followed the bovine potency test protocol described by the OIE to test this subunit vaccine potency. Three groups of five cattle per group and a control group of two non-vaccinated animals were vaccinated. All animals were challenged 3 weeks after vaccination with 10,000 BID50 of Asia I /HNK/CHA/05. The animals were observed daily for 10 days after challenge for clinical signs of FMD. From each animal protected in each group, the PD50 (50% bovine protective dose) content of the vaccine was calculated based on the Reed-Muench method. 3 RESULTS 3.1 Construction of the recombinant virus Bm-P12A3C Intact P1-2A and 3C protease coding regions from FMDV Asia I/HNK/CHA/05 strain were inserted into the transfer vector pVL1393 to generate plasmid pVL-P12A3C. Bm-N cell line was cotransfected with pVL- P12A3C and linearized BmBacPAK-6 DNA. The supernatant was collected 4 days post transfection as the viral stock for screening of recombinant virus. Twenty four isolated viral plaques from the plaque assays were cultivated in a 24-well plate and were inoculated into silkworms. The plaque of recombinant virus expressed at maximal activity was selected to purify. The pure recombinant virus from the last round was used as stock virus and confirmed to contain the full expression cassette. The recombinant virus Bm-P12A3C was used to express FMDV protein in cells or silkworm. 3.2 Expression of polyprotein in Bm-N cell The expression of polyprotein in Bm-N cells was analyzed by IFAT and sandwich-ELISA. IFAT pictures demonstrated that Bm-N cells infected with Bm-P12A3C produced specific fluorescence, while only very weak background fluorescence appeared in the control cells (Fig.1). This indicated that polyprotein was indeed expressed in Bm-N cell. The sandwich-ELISA results indicated that the FMDV antigen in Bm-P12A3C infected cells was expressed at levels about equivalent to the positive control, but was not detected in BmBacPAK-6 infected cell lysate. 3.3 Expression of polyprotein in silkworm Sandwich- ELISA was conducted to evaluate the expressed antigen in silkworm’s haemolymph. The results indicated that OD value of the harvested haemolymph from silkworm infected by BmP12A3C decreased as the dilution rate increased, which was in good agreement with variation of positive control of FMDV antigen. The expression yield was about 100 fold more than the positive control (BHK-21 cell vaccine which had a PD50 value of 3.6), but was not detectable in the negative control (BmBacPAK-6 infected silkworm’s haemolymph) (Fig. 2). In order to determine the time course of expressed antigen in silkworm and the optimum acquisition time for large scale’s production, haemolymph from 10 silkworms was harvested every 12h beginning at 60hpi. Subsequently, the haemolymph was diluted to 1000 folds for detection of expression products (Fig. 3). There was a little at 60hpi, and the accumulation of recombinant products were dramatically increased from 84 hpi and kept at the high levels during the late phase of infection. So, expressed antigen could be harvested at 108-120 hpi (at the condition about the mean rearing temperature of 25°C). 3.4 The anti-FMDV antibody in cattle LPBE-antibody titer was determined at 7, 14, 21 and 28dpv. It was found that all five cattle vaccinated with Bm-P12A3C antigen developed a detectable FMDV-antibody response at 7 dpv, and dramatically reached to high level at 14dpv. By 21and 28dpv , the antibody level was maintained at the same level or higher, and reached to a titer of 360 in cattle No33 and No50. In contrast, antibody level in the two control cattle was not boosted (Table 1). Furthermore, sera were analyzed for neutralizing antibodies against FMDV. The result was in agreement with the LPBE-antibody titer (Table 2). 3.5 Challenge with FMDV Asia I /HNK/CHA/05 All vaccinated cattle were challenged with 10,000 BID50 of Asia I /HNK/CHA/05 at 28dpv. Body temperature, mouth and feet were observed consecutively for ten days to evaluate the incidence of disease (Table 3). Four of the five cattle were considered completely protected. Only one vaccinated cattle, No45, developed lesions. The lesions were detectable by 6dpc and the clinical signs were less severe compared to control group. By contrast, vesicles developed in the control animals by 2dpc at the sites of all feet and mouth. This indicated that antigen produced in silkworm could be effectively protective.

135

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

3.6 The result of PD50 test The PD50 test was performed to assess the subunit vaccine potency by following the bovine potency test protocol described by the OIE to test the traditional inactivated FMD vaccines. In this research, the result showed the vaccine potency of the batch immunized with the expressed antigens reached 6.34 PD50 per dose (Table 4). 4. DISCUSSION The FMDV serotype Asia which was first isolated in Pakistan is epidemic within Southeast Asia and Indian peninsula, disseminating among Near East, Middle East and Far East [20]. In March 2005, FMDV serotype Asia was found in HongKong (Asia I/HNK/ CHA/05 strain). Subsequently, this type of the virus was reported from mainland of China in April 2005[21]. The P1 sequence of Asia 1 /HNK/ CHA/05 isolate was aligned and compared with 9 reference sequences. The result confirmed that Asia 1/HNK/CHA/05 has a high identity with nine Asia I reference sequences from 85.9 to 92.6% [22]. Expression products of baculovirus expressing system are generally considered to be well immunogenic and possess the ability to assemble empty viral capsid. When the same FMDV expression cassette were expressed in E.coli and baculovirus expression system, the expression products from baculovirus excels that from E.coli in terms of the immunogenic[23] and protective effects[24]. Empty capsid comes into being only when capsid precursor P1-2A, protease L and 3C coding region from FMDV O1K serotype were all expressed in baculovirus expressing system AcMNPV-Sf cell . Truncated protease L can not be self-cleaved from VP0. But, the expressed protease L is harmful to host cell growth, reducing the expression efficiency [25]. In adenovirus expression system, P12A3C expression cassette, including full structure of P1-2A 3C and portion of 2B and 3B, can be expressed and assembled into empty capsid [5]. Myristoylation of the animo terminus of P1-2A is of great importance to the assembly of viral particles [26]. It has been reported that the expression products of AcMNPV-Sf cell can be myristoylated well [27]. Based on the above studies, and using the design previously published by Mayr et al [5], the P12A3C expression cassette of FMDV serotype Asia I was constructed. After two sorting rounds of recombinant virus and measurements of expression efficiency for more than 20 viral clones, the over-expressed recombinant virus Bm-P12A3C was obtained. It can express with very high efficiency in the hyperexpression variety of silkworm (JY1). The specific antigen produced per milliliter in silkworm haemolymph at least 100 folds more than the BHK-21 cell vaccine which had a PD50 value of 3.6. Because cattle are the most important economic and susceptible cloven-hoof animal, we designed an experiment to verify whether this produced antigen can be used for preparing a cattle FMD vaccine. We followed the bovine potency test protocol described by the OIE to test this subunit vaccine potency. We used 1/30 diluted dosage to vaccinate five cattle and two controls were vaccinated with vaccine prepared from BmBacPAK-6’s. By two weeks post vaccination, the antibody level of the five vaccinated cattle reached a high titer. The antibody level has some ascension but maintained thereafter two weeks, while the control group maintained lower than titer 8. After virulent homologous virus challenge, four of the five were considered protected, and one delayed the disease and ease the clinical symptom, but two unvaccinated cattle developed lesions on all the feet and in the inside of mouth on the second day. Cell-mediated -immune response was probably involved in the protection: that would explain why animal 45 has the same neutralizing antibody titers as 122 but is not protected. This demonstrated that the expression products from silkwormbaculovirus expression system were immunogenic as well. Based on above result, we did the PD50 test to assess the bovine potency of the subunit vaccine. When employed for routine prophylactic use, the vaccine should contain at least 3 PD50 per dose for cattle by OIE recommended. The result showed the subunit vaccine potency could get 6.34 PD50 a dose for cattle. This leads to a conclusion that it is feasible to use the silkworm-baculovirus expression system for FMD vaccine production.

（

）

、

5. ACKNOWLEDGMENTS We are thankful to Ms. Shu Yun Qi for helping with the LPBE-ELISA and the laboratory of diseasesecure isolation facilities of Lanzhou Veterinary Research Institute for their assistance with the vaccination and virus challenge of cattle. 6. REFERENCES [1] Doel, T.R. 2003. FMD vaccines. Virus Res 91:81– 99

136

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[2] Kleid, D.G., Yansura, D., Small, B., et al. 1981. Cloned viral protein vaccine for foot-andmouth disease: responses in cattle and swine. Science 214(4525):1125-9 [3] Dus Santos, M.J., Carrillo, C., Ardila, F., et al. 2005, Development of transgenic alfalfa plants containing the foot-and-mouth disease virus structural polyprotein gene P1 and its utilization as an experimental immunogen. Vaccine 23(15):1838-1843. [4] Balamurugan, V., Renji, R., Venkatesh, G., et al. 2005. Protective immune response against foot-and-mouth disease virus challenge in guinea pigs vaccinated with recombinant P1 polyprotein expressed in Pichia pastoris. Arch Virol 150:967–79. [5] Mayr, G.A., Jarasvech, C., Grubman, M.J. 1999. Development of replication-defective adenovirus serotype 5 containing the capsid and 3C protease coding regions of foot-and-mouth disease virus as a vaccine candidate. Virology 263(2):496-506. [6] Mayr, G.A., O'Donnell, V., Chinsangaram, J., et al. 2001. Immune responses and protection against foot-and-mouth disease virus (FMDV) challenge in swine vaccinated with adenovirus-FMDV constructs. Vaccine 19(15-16):2152-62. [7] Moraes, M.P., Chinsangaram, J., Brum, M.C.S, et al. 2003. Immediate protection of swine from foot-and-mouth disease: a combination of adenoviruses expressing interferon alpha and a foot-and-mouth disease virus subunit vaccine. Vaccine 22(2):268-79. [8] Moraes, M.P., Mayr, G.A., Mason, P.W., et al. 2002. Early protection against homologous challenge after a single dose of replication-defective human adenovirus type 5 expressing capsid proteins of foot-and-mouth disease virus (FMDV) strain A24. Vaccine 20(11-12):1631-9. [9] Pacheco, J.M., Brum, M.C., Moraes, M.P., et al. 2005. Rapid protection of cattle from direct challenge with foot-and-mouth disease virus (FMDV) by a single inoculation with an adenovirusvectored FMDV subunit vaccine. Virology 337(2):205-9. [10] Abrams, C.C., King, A.M. &Belsham, G.J. 1995. Assembly of foot-and-mouth disease virus empty capsids synthesized by a vaccinia virus expression system. J Gen Virol 76:3089–98. [11] Guo, H., Liu, Z., Sun, S., et al. 2005. Immune response in guinea pigs vaccinated with DNA vaccine of foot-and-mouth disease virus O/China99. Vaccine 23 (25):3236-42. [12] Noad, R., Roy, P. 2003. Virus-like particles as immunogens. Trends Microbiol 11(9):438-44. [13] Maranga, L., Cruz, P.E., Aunins, J.G., et al. 2002.Production of core and virus-like particles with baculovirus infected insect cells. Adv Biochem Eng Biotechnol 74:183–206. [14] Laurent, S., Vautherot, J.F., Madelaine, M.F., et al. 1994. Recombinant rabbit hemorrhagic disease virus capsid protein expressed in baculovirus self-assembles into viruslike particles and induces protection. J Virol 68(10):6794–98. [15] Mortola, E. &Roy, P. 2004. Efficient assembly and release of SARS coronavirus-like particles by a heterologous expression system. FEBS letters 576(1-2):174–8. [16] Belliot, G., Noel, J.S., Li, J.F., Seto, Y., et al. 2001. Characterization of capsid genes, expressed in the baculovirus system, of three new genetically distinct strains of Norwalk-like viruses. J Clin Microbiol 39(12):4288–95. [17] Choudary, P.V., Kamita, S.G. &Maeda, S. Baculovirus expression protocols. “In: Expression of foreign genes in Bombyx mori larvae using baculovirus vectors”. Ed. Richardson CD. Humana press, Totowa, 1995: 243-264.’’ [18] Wu, X.F. &Zhang, Z.F. Gene expression technology. “In: Insect expression system”. Ed. Li YY. China Scientech Press, Beijing, 2001: 135-146. [19] Lin, X., Zhang, W., Chen, Y., et al. 2006. Overexpression of celB Gene Coding for βGlucosidase from Pyrococcus furiosus Using Baculovirus Expression Vector System in Silkworm, Bombyx mori. Z Naturforsch [C] 61(7-8):595-600. [20] Brown, F. (2003). The history of research in foot-and-mouth disease. Virus Res 91(1): 3–7 [21] Guo, H., Liu, X., Liu, Z., et al. 2006. Recent Outbreaks of Foot-and-Mouth Disease Type Asia 1 in China. J Vet Med B 53:29–33. [22] Li, D., Shang, Y.J., Liu, Z.X., et al. 2007. Molecular relationships between type Asia 1 new strain from China and type O Panasia strains of foot-and-mouth-disease virus. Virus Genes 35: 273–279 [23] Lewis, S.A., Morgan, D.O. &Grubman, M.J. 1991. Expression, processing, and assembly of foot-and-mouth disease virus capsid structures in heterologous systems: induction of a neutralizing antibody response in guinea pigs. J Virol 65(12):6572-80. [24] Grubman, M.J., Lewis, S.A. &Morgan, D.O. 1993. Protection of swine against foot-andmouth disease with viral capsid proteins expressed in heterologous systems. Vaccine 11(8):825-9. [25] Roosien, J., Belsham, G.J., Ryan, M.D., et al. 1990. Synthesis of foot-and-mouth disease virus capsid proteins in insect cells using baculovirus expression vectors. J Gen Virol 71( 8):170311. [26] Krausslich, H.G., Holscher, C., Reuer, Q., et al. 1990. Myristoylation of the poliovirus polyprotein is required for proteolytic processing of the capsid and viral infectivity. J Virol 64:2433– 6.

137

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[27] Belsham, G.J., Abrams, C.C., King, A.M.Q., et al. 1991. Myristoylation of foot-and-mouth disease virus capsid protein precursors is independent of other viral proteins and occurs in both mammalian and insect cells. J Gen Virol 72:747-5 Table 1: FMDV-specific antibody response after vaccination with Bm-P12A3C's Vaccinea Animal number 33 124 122 45 50 2 11

Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C BmBacPAK-6 BmBacPAK-6

LPBEantibodyb -7 <8 <8 <8 <8 <8 <8 <8

dpvc 7 32 20 20 20 45 <8 <8

14 180 90 45 90 360 <8 <8

21 360 90 45 90 360 <8 <8

28 360 90 45 90 360 <8 <8

a Bovines were vaccinated with vaccine prepared from 30 folds diluted expressed antigens (BmP12A3C) or the control (BmBacPAK-6) and challenged 28 days later. b FMDV-specific antibody titer reported as the serum dilution by LPBE method. c Days postchallenge Table 2: Results of neutralizing antibody response against FMDV after inoculation Vaccinea Animal # number 33 124 122 45 50 2 11

Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C BmBacPAK-6 BmBacPAK-6

neutralizing antibodyb -7 <8 <8 <8 <8 <8 <8 <8

dpvc 7 16 8 8 8 8 <8 <8

14 64 64 32 32 64 <8 <8

21 90 64 32 32 90 <8 <8

28 90 64 32 32 90 <8 <8

a Bovines were vaccinated with vaccine prepared from 30 folds diluted expressed antigens (BmP12A3C) or the control (BmBacPAK-6) . b FMDV-specific antibody titer reported as the serum dilution by neutralization tests. c Days postchallenge. Table 3: Protection and clinical signs in cattle after challenge with FMDV Asia I /HNK/CHA/05

Animal number # 33 124 122 45 50 2 11

vaccine

Days of onset of pyrexiaa

Duration of Pyrexia(days)

Lesion scoresb

Protectionc

Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C Bm-P12A3C BmBacPAK-6 BmBacPAK-6

Day 6 Day 2 Day 2

2 3 3

2 4+mouth 4+mouth

+ + + + -

a Pyrexia defined as body temperature 40°C. b The lesion score is the number of feet on which the cattle exhibited. c Protection was determined that cattle did not take on the clinical signs of FMD during observation period (10 days post-challenge). Table 4: The result of PD50 test Immunize dose 1

138

Rate of protection (%) 5/5 (100)

PD50

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

1/3 1/9

4/5 (80) 2/5 (40)

6.34

All of cattle were challenged with 10000 BID50 of Asia I /HNK/CHA/05 after vaccination by inoculating the equivalent of a total of 10,000 BID50 of Asia I /HNK/CHA/05 intradermally into two sites on the upper surface of the tongue. All cattle were observed for 10 days. Vaccinated animals are protected if they do not develop lesions on the feet and areas other than the injection sites on the tongue. Rate of protection (%) = number of cattle no lesions /total number of cattle. The PD50 value was calculated by the Reed-Muench method. Figure 1: Expression of FMDV polypeptides in Bm-N cells was analysed by IFAT

Bm-N cells infected with Bm-P12A3C. (B) Bm-N cells infected with BmBacPAK-6 Figure 2: Expression of FMDV polypeptides in silkworm larvae was estimated by the sandwichELISA. The haemolymph was diluted with two-fold series.

Figure 3: The time courses of FMDV polyprotein expressed in silkworm The larval haemolymph was collected on ice every 12 h from 60 hpi. The FMDV antigen in haemolymph was analyzed by the sandwich-ELISA. The hemolymph was diluted with 1,000 folds dilution.

139

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

140

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 21 POTENTIAL OF MODIFIED VACCINIA ANKARA (MVA) AS A VACCINE DELIVERY VECTOR FOR FOOT-AND-MOUTH DISEASE VIRUS (FMDV).

J. Castillo-Olivares*, D. Paton, B. Charleston and S. Parida Institute for Animal Health, Pirbright Laboratories, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, England

ABSTRACT Although FMDV inactivated vaccines are currently used with success there are a number of modifications that could improve their efficacy. These improvements aim at increasing the duration of immunity, stimulating strong cellular immune responses and / or providing differential diagnostic capability, and can be achieved by the use of live viral vectors, such as MVA, expressing selected FMDV antigens. Our aim is to elucidate whether MVA viruses expressing FMDV proteins could be used, either on their own or in combination with conventional vaccination, to improve the immunity against the disease. The P1 and the 3C/D regions of the A22-Iraq64 FMDV strain were cloned in the standard vaccinia transfer vector pSC11 downstream of the vaccinia early / late P7.5 promoter to generate MVA-P1 and MVA-3CD viruses. Expression of the FMDV expression cassettes was characterised by RT-PCR, immunofluorescence and Western blotting prior to vaccination of cattle with the recombinant viruses. The recombinant MVA-P1 were generated and grown to high titres in chicken fibroblasts (CF). Attempts to recover a recombinant MVA-3CD were unsuccessful up to date. Transcription of the FMDV genes from recombinant MVA-P1 infected cells was confirmed by RT-PCR amplification of P1 sequences from extracted RNA. Levels of expression vary considerably between cell lines, as revealed by immunofluorescence. Thus, P1 expression was detected in CF, LB-9 (bovine skin dermis cells) and P815 (mouse mastocytoma cell line) up to multiplicities of infection of 0.01, 1 and 10 respectively. Expression of P1 could not be detected in BK (bovine kidney cells). A protein band of approximately 80 KDa, corresponding to P1, could be detected with FMDV A22-specific rabbit polyclonal antiserum in immunoblots of MVA-P1 infected CF lysates. Furthermore, when MVA-P1 infected cells were co-transfected with a pSC11 plasmid encoding 3CD downstream of the vaccinia P7.5 promoter evidence of P1 processing was obtained. We conclude that recombinant MVA-P1 can be easily generated and grown to high titres and that it expresses the FMDV P1antigen in bovine cells in vitro showing therefore its potential to be used in FMD vaccination. Further work is currently underway to test its efficacy as a vaccine in vivo. 1. INTRODUCTION Foot-and-mouth disease is a highly infectious disease of cloven-hoofed animals caused by (FMDV) which belongs to the Aphtovirus genus of the family Picornaviridae. The disease causes vesicular lesions in the tongue and oral mucosa, feet, snout and teats resulting in high morbidity and low mortality. However, in young animals the infection can be fatal due to myocarditis. The economic impact of FMD is huge due to export loses and sharp decrease in productivity of the affected farms (Grubman and Baxt, 2004). Control of FMD is achieved by slaughtering infected and in-contact animals and restricting movement of infected products or by means of emergency vaccination in free countries or mass vaccination in endemic countries. However, vaccination is not always implemented, especially in disease-free countries, due the costs associated with the ban on exports from these countries until the disease-free status is re-gained. This happens only after 6 months from the last case of FMDV if vaccination is employed as opposed to only 3 months if slaughter without vaccination is implemented (Kitchin et al., 2007).

141

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

All currently available FMDV vaccines are based on cell culture derived preparations of whole virus, chemically inactivated and blended with oil or aluminium-saponin adjuvant to potentiate the immune response to vaccination. Killed vaccines are unstable at ambient temperature, and after formulation must be kept cool until administered. Since they are produced from live FMDV, strict biocontainment is necessary to exclude the risk of spreading virus from manufacturing plants (Grubman and Baxt, 2004; Kitchin et al., 2007; Doel, 2003). An animal immunised against one strain of FMDV may still be susceptible to another. Therefore, a number of vaccine strains for each serotype, particularly A and O, are required to cover the antigenic diversity and it is essential to monitor for the appearance of new strains internationally (one of the prime functions of the World Reference Laboratory). Depending on the dose and on the severity of challenge, FMDV vaccines may protect against disease within 4 to 5 days of vaccination and such protection may endure for approximately 6 months (Barnett et al., 2002). While existing vaccines have been associated with notable success, there are a number of areas where improvements would dramatically enhance the prospects for control of a FMD outbreak in a previously free country. 1) 2) 3) 4) 5) 6) 7)

More rapid onset of protection. Wider spectrum of protection against different strains. More potent immunity preventing virus replication and development of viral carriers. Better discrimination of vaccinated animals that go on to become infected. More thermostable. Easier and safer to make. Easier to administer.

Currently we are attempting to improve the efficacy of killed vaccines by co-administering them with recombinant modified Vaccinia Ankara (MVA) viruses expressing FMDV antigens, in particular P1 (precursor of capsid proteins) and 3C/D (encoding the viral protease and RNA polymerase respectively). This may provide more T-cell mediated antibody responses as current vaccine is chemically inactivated and devoid of non-structural component. Recombinant MVA has been utilised successfully as a viral vaccine vector for many diseases being especially effective at inducing cellular immune responses when administered in vaccination regimes in combination with other antigen delivery systems (i.e. DNA, sub-unit vaccines) (Gilbert et al., 2006). 2. METHODS 2.1. Cells, viruses and plasmids Primary chicken embryo fibroblasts (CEF) from gnotobiotic chickens were obtained from the Microbiological Services, Institute for Animal Health (IAH), Compton. The avian cell line DF-1, primary bovine thyroid cells (BTY), bovine dermal cells (LB-9) and bovine kidney (BK) were obtained from the Central Services Unit, (IAH), Pirbright. The P815-BLA cell line, a mouse mastocytoma cell line stably expressing bovine MHC-I, was obtained from Dr Shirley Ellis, IAH, Compton. All cell lines were grown in Dulbecco Minimal Essential Medium supplemented with Hepes, penicillin-streptomycin, L-glutamine and foetal calf serum and propagated using standard cell culture techniques. The standard vaccinia shuttle vector used was pSC11 (Chakhrabarti et al., 1985). Modified Vaccinia Ankara (MVA) was obtained from Dr Barbara Blacklaws (University of Cambridge). 2.2. Cloning and generation of recombinant MVA The cDNA of the complete P1 and 3CD regions of FMDV were derived from total RNA (extracted with Trizol Reagent) from BTY cells infected with the FMDV A22 Iraq 64 strain. A one-step RT-PCR reaction (SuperScript™ III One-Step RT-PCR System with Platinum® Taq High Fidelity, Invitrogen) using FMDV specific primers bearing SmaI restriction sites and start and stop codons was used. After digesting the P1 and 3CD DNA amplicons with SmaI, the DNA fragments were inserted in the SmaI site of pSC11, downstream of the P7.5 vaccinia early / late promoter. Purified plasmids were subsequently used to transfect MVA infected CEF and / or DF-1 cells. Briefly, 24h-old cell monolayers were washed with Hank’s balanced salt solution (HBSS) before infection with MVA using a multiplicity of infection of 1. Two hours later the cells were washed again and transfected using the recombinant pSC11-P1 or pSC11-3C/D plasmids using lipofectamine 2000 (Invitrogen) according to the manufacturer recommendations. The cells were scrapped when CPE reached 80% of the cell sheet and the cells were harvested, centrifuged at low speed and resuspended in HBSS. The harvested cells were stored at -80C until used in plaque assays.

142

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.3. Plaque assays Nearly confluent cell monolayers were inoculated with serial dilutions of the transfection harvest and after 2 hours at 37°C and removing the inoculum the cells were overlayed with 10% DMEM containing 1% agarose. The cells were incubated at 37°C and a second agarose overlay containing 400 ug / ml of X-gal was added when cells began to show CPE (usually 48-72 hours). Blue plaques produced by recombinant MVA were picked and subjected to various rounds of plaque purification until a pure stock was obtained. This was grown to high titres in DF-1 cells and used to characterise the expression of the inserted FMDV sequences. 2.4. Immunofluorescence Cells grown in glass cover slips were fixed with PBS containing 4% formaldehyde and 0.4 % TritonX-100, washed with PBS and blocked with PBS containing 1% bovine serum albumin. The primary antibody used was a rabbit polyclonal antiserum raised against inactivated FMDV A22 (a gift from Nigel Ferris, IAH, Pirbright) diluted in blocking buffer. After washing with PBS the cells were incubated with FITC-conjugated goat antibody to rabbit IgG (Zymed) diluted 1/200 in blocking buffer. After a final wash with PBS and then water the cells were mounted in Vecta-shield mounting medium with DAPI and observed on a UV light microscope. 2.4. Western blot Cells were lysed in lysis buffer [(20mM Tris-HCL, 150mM NaCl, 1% sodium deoxycholate, 1% Tergitol, 0.1%, SDS, 2mM EDTA, supplemented with protease inhibitors (all supplied by SigmaAldrich)] and diluted 1:1 in 2x protein loading buffer (National Diagnostics) before heating at 95°C for 5 minutes. Samples were run on 12% SDS-polyacrylamide gels and transferred to a nitrocellulose membrane using the iBlot Dry blotting device and iBlot membranes (Invitrogen). The FMDV A22 specific rabbit antiserum was used as the primary antibody. The Western Breeze Chromogenic kit-Anti Rabbit was used for the rest of the procedure according to the manufacturer recommendations. 3. RESULTS A recombinant MVA virus encoding the P1 segment of FMDV A22 strain was generated and grown to high titres following the procedures indicated in the materials and methods section. The expression of the P1 protein in various cell lines was evaluated and characterised. 3.1. Detection of P1 sequences by RT-PCR from MVA-P1 infected CF cells In order to determine whether the P1 expression cassette was functional we performed an experiment aimed at detecting P1 sequences by RT-PCR from extracted RNA samples of MVA-P1 infected cells using P1 specific primers (forward primer: 5’-tacatggtggcgtacgtt-3’; reverse primer: 5’-ccgtagttttcaacagtggt-3’). To maximise the chances of detection of P1 RNA we infected CF cells with MVA-P1 at an m.o.i. of 1 and extracted the RNA after 24 hours of infection (Fig. 1). Sensitive X-gal staining of fixed MVA-P1 infected CF cells at 24 hours post-infection revealed that the reporter Lac-Z gene was expressed in 100% of the cells indicating that P1 is also probably expressed abundantly. Indeed, P1 amplicons were easily detected from the RNA extracts of MVAP1 infected CF cells. The detection of P1 amplicons from the DNA-se treated samples confirmed the P1 expression was not due to contaminating MVA-P1 viral DNA. 3.2. Detection of P1 expression by indirect immunofluorescence Immunofluorescence experiments on fixed MVA-P1 infected cells were performed to determine the levels of protein expression in different cell lines, including cells from bovine origin. Detection of P1 was detected with very high sensitivity in CF (at 24 and at 48 hours) up to multiplicities of infection of 0.001 (Fig 2). Expectedly, expression levels were reduced in bovine dermal cells (LB-9) and bovine dendritic cells (up to an m.o.i. of 10) in comparison to those achieved in CF. Expression of P1 in bovine kidney cells could not be detected. We also performed experiments in P815 cells that have been stably transfected with bovine MHC-I. This cell line will be a very useful tool to examine the cytotoxic cell responses of FMD infection in cattle. Expression of P1 was achieved in this cell line up to m.o.i. of 0.1 (Fig. 3). 3.3. Western blot analysis of MVA-P1 infected cells The analysis of P1 expression in MVA-P1 infected CF and BK cells (Fig.4) are consistent with the results obtained by immunofluorescence. Bands of approximately 76 kDa were obtained in MVA-P1 infected CF cell lysates prepared at 24 and 48 hours post-infection. This was not the case for the BK cells. MVA-3CD was not available at the time these experiments were performed but a vaccinia

143

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

transfer vector encoding the 3CD segment of FMDV A22 (pSC11-3CD) was made and used to transfect MVA-P1 infected CF cells to check whether the P1 protein could be cleaved by the expressed 3C. Western blot experiments using lysates of MVA-P1 infected and pSC11-3CD transfected cells revealed, apart from P1, extra bands of approximately 31 kDa and 20 kDa, corresponding to the sizes of VP0 and VP1 and VP3. These extra bands could not be observed in any other cell lysate samples (Fig 5). 4. DISCUSSION Recombinant MVA viruses expressing antigens have been utilised profusely as a vaccine strategy to induce immunity against viral diseases. Specifically they have been particularly effective at inducing cell mediated immune responses in combination with other vaccine antigen delivery methods or formulations when administered as a boosting agent. It is our goal to administer MVA recombinant viruses expressing FMDV proteins to cattle on their own or in combination with inactivated FMDV vaccines to enhance the cellular component of the immune response and therefore increase the strength and duration of immunity to FMDV. The primary objective of our research is to provide additional T cell help stimulus to antibody formation during the induction phase of the immune response as well as inducing cytotoxic T cell responses. For this, we use the capsid precursor protein (P1) (capsid antigens are present in the conventional vaccine) and the 3CD proteins (known to hold epitopes for CTL) of FMDV. Therefore, the processing of the P1 into individual capsid protein components of the virion is not a requirement of our strategy since the B cell epitopes will be provided by the conventional FMDV vaccine. We have generated a recombinant MVA virus expressing P1 using standard molecular biology techniques. The virus was grown to high titres in the DF-1 chicken fibroblast cell line. The expression of the foreign gene is driven from the constitutive vaccinia promoter P7.5 and the protein was detected by immunofluorescence in various cell lines and also by Western blot. The latter experiments confirmed the identity of the protein by its size and further processing when the protease 3C was co-expressed with P1 in MVA-P1 infected cells transfected with pSC11-3CD. This processing was not strictly necessary to test our hypothesis that MVA-P1 and or MVA-3CD can improve FMDV vaccine efficacy but it could represent a strategy to produce FMDV capsids in vitro. Further studies are currently being performed to characterise the expression and processing of P1 into capsid components using the techniques described in this paper. We have shown that the expression of P1 was achieved in avian as well as bovine cell lines. However, the levels of expression were reduced in the latter, reflecting the known replication characteristics of the avian adapted MVA virus. Despite these differences, we do not anticipate a failure of MVA-P1 vaccinated cattle to mount a measurable immune response to FMDV since a recombinant MVA carrying African horse sickness (AHS) virus VP2 was capable of stimulating virus neutralising antibodies to AHS virus in ponies after vaccination despite expression levels of VP2 were reduced in equine dermal cells in comparison with avian fibroblasts (Castillo-Olivares J., unpublished observations). 5. CONCLUSIONS

We have generated an MVA-P1 expressing the P1 antigen and that has the potential to induce FMDV specific immunity in cattle. Vaccination experiments using MVA-P1 in combination with conventional vaccine are currently being performed.

REFERENCES [1] Gilbert, S. C., V. S. Moorthy, et al. (2006). "Synergistic DNA-MVA prime-boost vaccination regimes for malaria and tuberculosis." Vaccine 24(21): 4554-61. [1] Grubman, M. J. and B. Baxt (2004). "Foot-and-mouth disease." Clin Microbiol Rev 17(2): 465-93. [1] Doel, T. R. (2003). "FMD vaccines." Virus Res 91(1): 81-99. [1] Kitching, P., J. Hammond, et al. (2007). "Global FMD control--is it an option?" Vaccine 25(30): 5660-4. [1] Barnett, P. V. and H. Carabin (2002). "A review of emergency foot-and-mouth disease (FMD) vaccines." Vaccine 20(11-12): 1505-14.

144

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig 1a)

Fig 1b)

145

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig 2a)

0.01

0.1

0.001

Neg

Neg Fig 2b)

Figure 2: Detection of P1 in MVA-P1 infected chicken embryo fibroblasts (a) and bovine skin fibroblasts (LB-9) (b) using various multiplicities of infection (indicated in each photo of each panel). Cells were fixed with 4% formaldehyde, 0.4% TX100 at 48 hours postinfection and probed with rabbit anti-FMDV A22 polyclonal antiserum. After incubation with the appropriate FITC-conjugate cells were mounted with DAPI mounting medium and observed under fluoescence microscope.

146

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig. 4 24 h p.i. BK cells

MWM

MVAP1 MVA MVA -

48 h p.i. CEF

MVAP1

BK cells MVA

MVAP1

MVA

CEF MVAP1

76 52 38 31 24

17 12

Fig.4: Western blot of MVA-P1 (lanes 1, 3, 5, 7) and MVA (2, 4, 6, 8) infected cell lysates prepared at 24 (1, 2, 3, 4) and 48 (5, 6, 7, 8) hours post-infection. Proteins were probed with A22 FMDV specific rabbit polyclonal antiserum. The secondary antibody was an Alkaline Phosphatase conjugated anti-rabbit IgG. Results were revealed using a chromogenic substrate. The ‘Rainbow’ molecular weight marker was used

MVAP1 pSC113CD No plasmid

MVA pSC113CD

No plasmid

No virus pSC113CD

No plasmid

52 38 31

VP0

24 VP1 VP3

Fig.5. Western blot of MVA-P1, MVA and mock infected cell lysates prepared at 14 hours post-infection. Half of the cell samples were transfected with pSC-3CD two hours postinfection. Proteins were probed with A22 FMDV specific rabbit polyclonal antiserum. The secondary antibody was an Alkaline Phosphatase conjugated anti-rabbit IgG. Results were revealed using a chromogenic substrate. The ‘Rainbow’ molecular weight marker was used.

147

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

148

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 22

MARKER POTENTIAL OF A VACCINE PREPARED FROM A FOOT-AND-MOUTH DISEASE VIRUS WITH A PARTIAL VP1 G-H LOOP DELETION V.L. Fowler1, N.J. Knowles1, D.J. Paton1, E. Rieder2 and P.V. Barnett1*. 1

Institute for Animal Health, Pirbright Laboratory, Ash Road, Surrey, GU24 0NF. Foreign Animal Disease Research Unit, USDA, ARS, Plum Island Animal Disease Center PO Box 848, Greenport, New York, 11944-0840, USA. *Corresponding author, Tel +441483 231153, Fax +441483 232448 or e mail paul.barnett@bbsrc.ac.uk (P V Barnett).

ABSTRACT Introduction Full protection against Foot-and-Mouth Disease (FMD) can be achieved following vaccination with chimeric vaccines, in which the VP1 G-H loop had been substituted with that from another serotype. This suggests potential for other marker vaccines without some of the VP1 G-H loop region since the immunodominance of this epitope provides basis for distinguishing animals that subsequently become sub-clinically infected with viruses with the VP1 G-H loop region. Serological profiles generated from a partial G-H loop deleted virus were evaluated on probability to protect cattle and as a diagnostic tool. Materials and methods Inactivated, vaccines containing either an A serotype strain (A+) or its equivalent with a partially deleted VP1 G-H loop region (A-), were used to immunise cattle, and serum was collected at regular intervals. Antibody titres, using a range of commonly used assays were used to estimate the kinetics of the antibody response, the likelihood of protection and level of cross reactivity against a range of field isolates. Assays were developed to decipher such responses of A+ and Avaccinated animals as a marker strategy. Results The kinetics of the antibody responses were similar for both vaccines and individual neutralising antibody titres against the A+ virus were considered to be above the threshold required for protection in this species. Cross-reactivity studies were not conclusive in proving that the Avaccine was any less cross reactive than the A+, but the potential of the A- vaccine as a marker was. Discussion Antibody responses from an FMDV vaccine with a partially deleted VP1 G-H loop region provided incites to its potential to protect cattle from disease, and as a diagnostic marker. Further ‘proof of principle’ experiments are planned. 1. INTRODUCTION Foot-and-mouth disease virus (FMDV) shares many structural features with other picornaviruses but unlike rhino- and enteroviruses, has a relatively ‘smooth’ surface with a cell binding motif consisting of a prominent structure, the so called G-H loop, originating from capsid protein VP1. Early work established that the immunogenicity of FMDV was largely associated with the capsid protein VP1 and that the hypervariable G-H loop was immunodominant and an important region for the generation of virus neutralising and protective antibodies. However, the VP1 G-H loop region and its role toward protective immunity in target species is still unclear and there is increasing evidence that other sites on the capsid are also important for protection [1, 2, 3, 4, 5, 6, 7 and 8]. Another study which experimentally substituted the VP1 G-H loop with 10 glycine residues showed that the removal of this region could dramatically enhance the immune response to less “dominant” regions leading to broader cross reactivity within and between serotypes [9], and we have shown that an FMDV chimera, where the VP1 G-H loop on the A12119 is replaced with that of another

149

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

serotype, induces responses that protect cattle against the A serotype virus (10). This leads to the question of the potential of negative FMDV marker vaccine characterised by an absence of some, or most, of the VP1 G-H loop. We examined whether a vaccine lacking a significant portion of this loop could potentially protect cattle against its homologous counterpart with the loop and whether a novel supportive diagnostic test approach has the potential to differentiate vaccinated animals that subsequently become infected. 2. MATERIALS AND METHODS 2.1 Viruses and antigen production Two FMD viruses, namely A-, with a 13 amino acid deletion within the VP1 G-H loop and A+, with the native VP1 G-H loop, were isolated and harvested following three plaque purifications from a single origin. Both viruses were grown on roller bottles, harvested, BEI inactivated and sucrose density gradient purified for assay purpose and vaccine formulation. A selection of A serotype field isolates were also obtained for cross-reactivity studies. 2.2 Sequencing of viruses The capsid-coding regions of both A+ and A- were sequenced following RNA extraction, cDNA synthesis and PCR and additionally sequenced through the VP1 G-H loop for confirmatory purpose following serial passage. 2.3 Cattle immunisations Two separate groups of 5 Holstein Friesian cross-bred cattle, 6-7 months of age, were intramuscularly immunised with water-in-oil-in-water (WOW) vaccines prepared from either A- and A+ inactivated virus (15 µg per bovine dose) using Montanide ISA 206 oil adjuvant. Clotted and heparinised bloods were regularly collected. 2.4 Serology Individual sera were used to screen against A+ and A- viruses, and A+ or A- pooled sera were also used for strain relationship studies. For anti-FMDV neutralising antibodies [11], titres were calculated as the log10 of the reciprocal antibody dilution required for 50% neutralisation of 100 TCID50 of virus. A liquid phase blocking ELISA (LPBE) was carried out as previously described [12]. 2.5 Novel marker differentiation assay In order to also examine the diagnostic potential of the A- virus as a marker vaccine, an indirect integrin αvβ6 [13] capture ELISA was developed. Essentially, the αvβ6 integrin provided the ability to capture any FMD virus and hence improve the versatility of such a test. In the absence of being able to experimentally challenge the A- vaccinated cattle, A+ serum was considered as an ‘infected’ animal serum, since it contained antibodies to the antigenic site not found in A- virus, whereas Aserum was considered the ‘vaccinated’ animal serum. Based on pre-absorption of appropriately diluted ‘test’ polyclonal serum with vaccine antigen (in this case A-) and subsequent incubation, using an indirect test format to identify this G-H loop response, animals which were theoretically infected (A+ serum) should give a strong OD, whereas those animals which were vaccinated with the A- strain should not. 2.5 Statistical analysis A two sample t-test was used to determine the statistical significance between the homologous VNT titres and the LPBE titres. Statistical analysis was performed using Minitab version 14.15. 3 RESULTS 3.1 Sequence and conformational analysis of VP1 G-H loop The capsid sequences still require completion; A- capsid sequence consisted of 2161 resolved nucleotides and 8 unresolved nucleotides and A+ capsid sequence consisted of 2208 resolved nucleotides and 338 unresolved nucleotides. However, all known antigenic sites for this serotype had sequence available for comparative purpose. Evaluation of the capsid coding region of A+ and A- confirmed the partial absence of the VP1 G-H loop in A- and revealed other changes in VP1, specifically, 4 substitutions located at residues 138, 141, 155 and 203 and 13 deletions located at residues 142-154 of A-. All deletions and substitutions involving site 1 (residues 144, 148 and 154) and site 5 (residue 149) are critical residues. Changes at residues 17, 37, 53 and 203 of VP1 have not been shown to be antigenically significant [12], [5], [13] and [14]. Structural analysis of VP1

150

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

from A+ and A-, using ESyPred3D [14], based on the co-ordinates of A10/Arg/61 (database accession no. 1ZBE) [15], predicted no structural differences between A+ and A- at sites outside the VP1 G-H loop. The G-H loop deletion appears to have simply led to the joining of the two ends into a shortened loop. 3.2 Immunogenicity and predicted protection of the A+ and A- vaccine strains The individual sera assessed by virus neutralisation, showed that for the majority of the animals, levels were above a threshold normally considered to be protective in this species at day 21 [16]. The only exception being the A- animal RZ65 which had a titre of 1.1 against the A+ virus (Table 1). Table 1: Day 21 virus neutralisation titres expressed as the log10 reciprocal antibody dilution required for 50% neutralisation of 100 tissue culture infectious units. a BEI-inactivated virus, 15µg/2ml dose, b VNT titres, Means were calculated from unrounded data. Titre values presented in this table are mean values from two repeat tests. Vaccine Typea

Animal Number RZ59 RZ60 RZ61 RZ62 RZ63 Mean CI 95% RZ64 RZ65 RZ67 RZ68 RZ81 Mean CI 95%

A+ Sera

A– Sera

Titres against homologous heterologous vaccine virusesb A + Virus A - Virus 2.2 2.7 2.9 3.1 2 2.5 2.1 2.6 2.7 3 2.4 2.8 2.05-2.75 2.57-3.03 1.8 2.1 1.1 1.8 2 2.6 1.9 2.5 2 3 1.8 2.4 1.45-2.13 2.0-2.81

and

These titres also indicated that the response generated against the homologous virus was not significantly different (p=0.94) between A+ and A- vaccinated cattle. Based on individual (r) value relationships, protection from challenge with A+ would be achieved in three out of the five animals vaccinated with A- (data not shown). Pooled sera were used to estimate the relative immunogenicity of both vaccine strains. The vaccines prepared from A- or A+ produced similar and measurable levels of anti-FMDV neutralising antibody which were detectable as early as 7 days post vaccination and identical by day 21 (Figure 1). 3

2.5

Log10 Virus Neutralising Titre

1.5

A+

A-

0.5

0 Day 0

Day 7

Day 14

Day 21

Days Post Vaccination

Figure 1: Homologous virus neutralising antibody responses of pooled serum from cattle either vaccinated with A - (▲) or A + (■). VNT: virus neutralising titre. Error bars represent 1 standard error (SE) above and below the mean. LPBE was also used to calculated total antibody titres, in this case homologous titres were significantly different (p=0.018), with the A+ sera titre being 3 fold greater than the A- sera titre (Table 2). However, relationship (r) values indicated that protection from challenge with A+ would be achieved in five out of the five animals vaccinated with A- (data not shown).

151

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 2: Individual day 21 cattle serum titres expressed as the log10 reciprocal antibody dilution derived from the LPBE. Titre values presented in this table are mean values from two repeat tests. Vaccine Type

A+ Sera

A– Sera

Animal Number RZ59 RZ60 RZ61 RZ62 RZ63 Mean CI 95% RZ64 RZ65 RZ67 RZ68 RZ81 Mean CI 95%

Titres against homologous heterologous vaccine viruses A + Virus A - Virus 2.2 1.9 2.8 2.2 2.2 1.9 1.9 1.9 2.5 1.6 2.2 1.9 1.9-2.5 1.7-2.1 1.0 1.3 1.9 1.9 1.6 1.6 1.6 1.9 1.9 1.9 1.6 1.7 1.3-1.9 1.5-1.9

and

3.3 Cross-reactivity of the A+ and A- vaccine strains A+ serum generally had a greater relationship with field isolates than A- serum by virus neutralisation (Table 3), indicating that it could be effective against at least five of the isolates examined compared to one for the A- vaccine. However, using LPBE (Table 3), twelve out of seventeen isolates examined, had r values either identical between A+ and A- serum (A+, A/IRN/2/87, A/TUR/4/2002, A/IRN/41/2003, A/IRN/4/2005, A/PAK/9/2003, A/TUR/5/2003, A/MAY/2/2002 and A/IRN/32/2004) or improved from those calculated for A+ (A-, A/IRN/6/2002, and A/IRAN/5/2005). Nevertheless, better coverage was indicated by the A+ serum LPBE against two isolates (A/IRN/31/2001 and A/IRN/7/2003). Table 3: Relationship (r) values predicted from A+ and A- serum titres generated by VNT and LPBE when tested against 15 field isolates. Values in bold indicate r values which would be considered protective, for VNT > 0.3, for LPBE > 0.4 (Paton et al., 2005). Values shaded in grey indicate where the either the A- sera has a greater relationship with field isolates that the A+ sera or is equal to. A + Sera A - Sera Virus VNT LPBE VNT LPBE A1 0.5 1 1 A+ 1 1 0.23 1 A/MAY/2/2002 0.2 0.13 0.1 0.13 A/TUR/5/2003 0.32 0.13 0.2 0.13 A/TAI/10/2003 0.09 0.3 0.06 0.02 A/LAO/36/2003 0.09 0.3 0.06 0.13 A/IRN/31/2001 0.52 0.5 0.3 0.3 A/PAK/9/2003 0.05 0.13 <0.009 0.13 A/IRN/4/2005 0.32 1 0.07 1 A/IRN/5/2005 0.03 0.5 0.01 1 A/IRN/41/2003 0.04 0.3 0.03 0.3 A/IRN/7/2003 0.01 1 0.02 0.125 A/TUR/4/2002 0.2 0.13 0.03 0.13 A/IRN/6/2002 0.2 0.03 0.17 0.13 A/IRN/2/87 1 1 1 1 A/IRN/32/2004 0.02 0.3 0.02 0.3 A/PAK/11/2003 0.13 0.3 0.13 0.13 3.4 Differentiation assays

152

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

With reference to table 4 and with a cut off defined as values greater than two times the day 0 serum values, it was clear that by using the integrin based assay, all animals which were vaccinated with A- would be considered negative whereas all animals vaccinated with A+ would be considered positive at a serum dilution of 1/200 (Table 4). Table 4: OD values of day 21 serum (1/200) from A+ and A- vaccinated cattle using the integrin capture ELISA. Shading indicates samples considered as positive. Standard deviation in italics A+ vaccinated cattle

A- vaccinated cattle

Day 0 sera X2

0.23 16

0.739

0.826

0.616

0.459

0.05 2

0.104 652

0.050 912

0.012 728

0.079 196

63 0.7 19 5 0.0 60 10 4

0.138

0.168

0.111 5

0.116

0.229

0.004243

0.022 627

0.010 607

0.001 414

0.035 355

4. DISCUSSION Sequence analysis so far confirms that the only significant differences between A+ and A- viruses were located in VP1 and predominantly restricted to the VP1 G-H loop. Modelling the predicted structure of A+ and A- viruses demonstrated that this loop deletion does not influence the surrounding structures and that the two ends can simply join together. The kinetics of the antibody responses following immunisation of either A+ or A- were very similar, indicating that the deleted region had not impaired the immunogenicity. The ability of A+ and Apolyclonal sera to bind to, or neutralise, A- virus indicated that the results very much depended on the assay used, homologous titres being significantly different (p=0.018) in the LPBE test but not in the virus neutralisation test (p=0.94). In the absence of being able to conduct an in-vivo virus challenge experiment, serum antibody titres and subsequent r value calculations from this vaccination trial were used to predict protection and cross-reactivity of the A+ and A- viruses. There are many publications relating to predicting protection against FMDV by in vitro analysis [16], and good correlation has been observed between virus neutralisation titres in cattle and protection. It was demonstrated that you can predict the probability of protection in the absence of challenge from serum neutralising titres [16] and that r values derived from serum titres can also be used to predict the likelihood of protection against other strains [17]. For the purpose of this paper, however, we interpreted our findings based on both. There are several in vitro vaccine matching assays but for this particular analyses both VNT and LPBE [17] [18] were selected. The antigenic relationship between vaccine(s) and the field isolate(s), termed r value, can be estimated from their comparative reactivity with vaccine antisera. In the case of the VNT, r values greater than 0.3 are indicative of a reasonable antigenic match likely to give protection, whereas r values less than 0.3, indicate that the field isolate is sufficiently different to the vaccine strain and that the vaccine may not protect. When analysing the r values derived from LPBE titres, it is considered that values greater than 0.4 are indicative of a good relationship to the vaccine virus. Sera from animals vaccinated with either A+ or A- inactivated FMDV antigen were compared to A+ and A- viruses and both could bind to and neutralise their homologous viruses to comparable titres. However, the A+ neutralising response appeared to be better against the A- virus in contrast to the A- response which only neutralised the A+ virus at a lower dilution. Based on these observations, one might predict that the A- vaccine would not be as effective at protecting cattle from challenge. However, serum titres do not support this. The mean serum titre from the A- group of cattle was 1.8 log10 which indicates a good probability of protection (16) and additionally the r values calculated from virus neutralising antibody titres, indicate that, three out of the five animals have antibody titres which would be considered likely to achieve protection. However, this assessment should be considered cautiously since by its nature, the VNT is preferentially selecting certain specificities of antibody with the principal mechanism being one which blocks the virus from

153

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

attaching to its cellular receptor. Therefore a serum which lacks antibodies against the VP1 G-H loop is likely to be less efficient at prevent viral attachment and thus compromising its ability to neutralise wild type viruses such as A+ in vitro. The fact that A- serum can measurably bind and neutralise A+ virus is very encouraging, and demonstrates that other mechanisms of neutralisation are working effectively, such as viral aggregation and/or conformational alteration of the capsid structure leading to the release of viral RNA [19], and that the A- antibody response is still binding to other important antigenic sites on A+ virus. The r values calculated from the LPBE serum titres indicate that all cattle vaccinated with A- vaccine virus might protect against an A+ challenge. Combined, the serological data provides a strong argument that a vaccine lacking the VP1 G-H loop would protect cattle from challenge of a similar strain that included the G-H loop, though how efficiently remains to be evaluated. Serum responses from the A+ and A- vaccinated cattle, were compared against a random selection of A serotype field isolates to determine whether the loss of the VP1 G-H loop compromised or improved the antigenic relationships observed [9]. Again, due to the nature of the test, r values predicted from virus neutralising titres were low. However, when considering the r values predicted from LPBE titres the results, apart from two cases, were more promising, with the r values from the A- serum being identical or sometimes better to those calculated for A+ serum. Interestingly, and in support of previous work [9], it was found that the binding relationship had been improved above that of A+ serum against A/IRN/6/2002 and A/IRN/5/2005. It is possible that this ‘improvement’ in reactivity has occurred due to sequence homology in the backbone of the virus which was shared between A- and the other viruses, and that because A- serum lacked any VP1 GH loop antibodies, other more specific antibodies to the other parts of the capsid could bind more efficiently. If homology to the residual capsid structure without the VP1 G-H loop is greater than VP1 G-H loop itself an enhanced non-VP1 G-H loop antibody response, might lead to an improved vaccine and improved efficacy to heterologous viruses. This needs to be investigated further by obtaining complete capsid sequences for the isolates tested in this paper and relating this data to serological responses. Though it was not possible to experimentally challenge either the A- or the A+ vaccinated cattle, it was possible to theoretically demonstrate that a vaccine generated from the A- virus could be used to identify ‘infected’ from ‘vaccinated’ animals using an αvβ6 integrin based assay format. We demonstrated that in all cases, the individual steers could be categorised into the correct classification of ‘infected’ or ‘vaccinated’. This is particularly encouraging given that current DIVA tests can only discriminate animals at the herd level (20). Therefore, a vaccine construct of this type could have clear advantages over other marker vaccine prototypes, not only because it produces a serum response which can be easily discriminated from an infected animal response, but because it would include nearly 100% of its genome, thereby maximising its chances of conferring protection and reducing the concern of ineffectiveness through natural variability and key mutational changes that will occur. In addition, the integrin capture assay as a diagnostic approach offers additional advantages in that it utilises both the vaccine virus and any field isolates, minimising the need to generate pre-prepared reagents and making it commercially viable and easy to implement. Both vaccine efficacy and diagnostic approach need to be investigated further, as these could ultimately provide us with more efficient tools for FMD control. 5. AUTHOR’S CONCLUSIONS A short partial deletion involving the VP1 G-H loop region of an A serotype vaccine strain (A-) did not impair the vaccines antigenicity and may not reduce its cross-reactivity. The humoral responses from this partially deleted vaccine (A-) were generally at levels considered to be protective. Alongside a versatile ELISA format, that incorporates the αvβ6 integrin as a capture system, a vaccine encompassing a partial deletion of the VP1 G-H loop region may offer considerable potential for differentiating infected from vaccinated animals. 6. AUTHOR’S RECOMMENDATIONS A ‘proof-of principle’ protection study should be carried out in cattle with this VP1 G-H loop deleted vaccine virus. 7. ACKNOWLEDGEMENTS

154

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Veronica Fowler was in receipt of a BBSRC PhD studentship and this work was also from the European Commission (FMD_Improcon project of the EU 6th Framework Programme, SSPE-CT2003-503603).The authors would like to thank IAH staff for their assistance with the handling and care of the cattle used in this study. We would also like to thank Jean-François Valarcher who was involved in the plaque purification of the A+ and A- viruses. Dr David Paton and Dr Paul Barnett are Jenner Investigators. The marker approach is under a UK Patent Application GB0810710.4 "Vaccine" 8. REFERENCES [6] Aggarwal, N. and Barnett, P. V. (2002) Antigenic sites of foot-and-mouth disease virus (FMDV): an analysis of the specificities of anti-FMDV antibodies after vaccination of susceptible host species. Journal of General Virology. 83. 775-782. [7] Barnett, P. V., Samuel, A. R., Pullen, L., Ansell, D., Butcher, R. N. and Parkhouse, R. M. (1998) Monoclonal antibodies, against O1 serotype foot-and-mouth disease virus, from a natural bovine host, recognise similar antigenic features to those defined by the mouse. Journal of General Virology. 79. 1687-1697. [16] Barnett, P. V., Statham, R. J., Vosloo, W. and Haydon, D. T. (2003) Foot-and-mouth disease vaccine potency testing: determination and statistical validation of a model using a serological approach. Vaccine. 21. 3240-3248. [3] Bergmann, I. E., Tiraboschi, B., Mazzuca, G., Fernandez, E., Michailoff, C. A., Scodeller, E. A. and La Torre, J. L. (1988) Serological and biochemical analysis of foot-and-mouth disease virus (serotype C3) isolated in Argentina between 1981 and 1986. Vaccine. 6. 245-252. [10] Fowler, V.L., Paton, D.J., Reider, E., Barnett, P.V. (2008) Chimeric foot-and-mouth disease viruses:Evaluation of their efficacy as potential marker vaccines in cattle. Vaccine 26, 1982-1989 [9] Frimann, T. H., Barfoed, A. M., Aasted, B., and Kamstrup, S. (2007) Vaccination of mice with plasmids expressing processed capsid protein of foot-and-mouth disease virus—Importance of dominant and subdominant epitopes for antigenicity and protection. Vaccine. 25. 6191-6200. [15] Fry, E. E., Newman, J. W., Curry, S., Najjam, S., Jackson, T., Blakemore, W., Lea, S. M., Miller, L., Burman, A., King, A. M. and Stuart, D. I. (2005). Structure of Foot-and-mouth disease virus serotype A10 61 alone and complexed with oligosaccharide receptor: receptor conservation in the face of antigenic variation. Journal of General Virology. 86. 1909-20. [12] Hamblin, C., Barnett, I. T. and Crowther, J. R. (1986) A new enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease virus. II. Application. Journal of Immunological Methods. 1. 123-129. [14] Lambert, C., Leonard, N., De Bolle, X. and Depiereux, E. (2002) ESyPred3D: Prediction of proteins 3D structures. Bioinformatics. 18. 1250-1256. [4] Mateu, M. G., Camarero, J. A., Giralt, E., Andreu, D. and Domingo, E. (1995) Direct evaluation of the immunodominance of a major antigenic site of foot-and-mouth disease in a natural host. Virology. 206. 298-306. [8] McCullough, K. C., Crowther, J. R., Butcher, R. N., Carpenter, W. C., Brocchi, E., Capucci, L. and De Simone, F. (1986) Immune protection against foot-and-mouth disease virus studied using virus-neutralising and non-neutralising concentrations of monoclonal antibodies. Immunology. 58. 421-428. [19] McCullough, K. C., Smale, C. J., Carpenter, W. C., Crowther, J. R., Brocchi, E., and Simone, F. (1987) Conformational alteration in foot-and-mouth disease virus virion capsid structure after complexing with monospecific antibody. Immunology. 60. 75-82. [18] OIE Manual of Diagnostic Tests and Vaccines for Terrestrial Animals 2004. Updated 2006. [17] Paton, D. J., Valarcher, J. F., Bergmann, I., Matlho, O. G., Zakharov, V. M., Palma, E. L., Thomson, G. R. (2005) Selection of foot-and-mouth disease vaccine strains-a review. Revue scientifique et technique (International Office Epizootics). 24. 981-993. [20] Paton, D.J., de Clercq, K., Greiner, M., Dekker, A., Brocchi, E., Bergmann, I., Sammin, D. J., Gubbins, S. and Parida, S. (2006) Application of non-structural protein antibody tests in substantiating freedom from foot-and-mouth disease virus infection after emergency vaccination of cattle Vaccine 24 6503-6512. [2] Rieder, E., Baxt, B., Lubroth, J. and Mason, P. (1994) Vaccines prepared from Chimeras of Foot-and-Mouth Disease virus (FMDV) Induce Neutralising Antibodies and Protective Immunity to Multiple Serotypes of FMDV. Journal of Virology. 68. 7092-7098. [11] Rweyemamu, M. (1984) Antigenic variation in foot-and-mouth disease: studies based on the virus neutralisation reaction. Journal of Biological Standardisation. 12. 323-337. [1] Taboga, O., Tami, C., Carrillo, E., Nunez, J. I., Rodriguez, A., Saiz, J. C., Blanco, E., Valero, M. L., Roig, X., Camarero, J. A., Andreu, D., Mateu, M. G., Giralt, E., Domingo, E.,

155

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Sobrino, F. and Palma, E. L. (1997) A large-scale evaluation of peptide vaccines against footand-mouth disease: lack of solid protection in cattle and isolation of escape mutants. Journal of Virology. 71. 2606-2614. [5] Thomas, A. A. M., Woortmeijer, R. J. Puijk, W., Barteling, S. J. and Meloen, R. H. (1988) Antigenic sites of foot-and-mouth disease virus type A10. Journal of Virology. 62, 2782-2789. [13] Weinacker, A., Chen, A., Agrez, M., Cone, R.I., Nishimura, S., Wayner, E., Pytela, R., Sheppard, D. Role of the integrin alpha v beta 6 in cell attachment to fibronectin. Heterologous expression of intact and secreted forms of the receptor. J Biol Chem. 1994 Mar 4;269(9):6940-8.

156

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 23

FRENKEL LECTURE: FMD VACCINE DEVELOPMENT - PAST AND FUTURE L. Robinson1, M. Windsor, J. Hope1, G. MacPherson2, B. Charleston1* 2

1 Institute for Animal Health, Pirbright Laboratory, Guildford GU24 0NF. Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE.

INTRODUCTION Foot-and-mouth Disease Virus (FMDV) causes a highly contagious acute vesicular disease affecting a number of economically important animal species. Little is known about the interaction of the virus with cattle dendritic cells (DC). Development of a comprehensive protective T and B cell response requires antigen capture, migration, maturation and antigen presentation by DCs. The capacity to stimulate CD8 and CD4 T cells relies upon the presentation of antigen through MHC class I and II molecules respectively. There is stimulation of specific CD4 and CD8 responses to live and killed FMDV antigen, suggesting presentation occurs through the class I and II pathways. However, the response of CD4 and CD8 T cells isolated from infected cattle are consistently low compared to the response to control antigens, despite the absence of generalised immunosuppression in the FMDV infected cattle. The specific CD4 response to vaccination is variable. MATERIAL AND METHODS Bovine dendritic cells generated from CD14+ monocytes were (MoDC) were produced by published methods. Experiments were performed to determine integrin expression on these cells and whether serotype O FMDV could productively infect these cells. In addition, the effect of adding immune complexed virus to the DC was studied. Furthermore, the capacity of DC, pulsed with virus or inactivated antigen, to stimulate specific CD4+ T cell proliferation was determined. RESULTS MoDC do not express the integrin used by FMDV for entry and non-structural proteins are detected in 5% of the cells by flow cytometry. In contrast, 70% of the MoDC are infected after the addition of immune complexed virus, causing cell death in 6-8 hours, the production of type-I interferons and decreasing the ability to stimulate specific CD4+ T-cell proliferation. Using immune complexed inactivated virus to target uptake by MoDC, resulted in enhanced CD4+ T-cell proliferation. DISCUSSION These in vitro observations improve our understanding of the development of the immune response to FMDV infection in vivo and suggest alternative strategies to improve vaccine efficacy.

157

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

158

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 24

KEYNOTE: VACCINE QUALITY TESTS: THE VALUE OF ALTERNATIVE METHODS?

N. Goris 1

*, 1

and K. De Clercq

Epizootic Diseases Section, Virology Department, Epizootic Diseases Section, Veterinary and Agrochemical Research Centre, Groeselenberg 99, 1180 Brussels, Belgium

ABSTRACT In vivo vaccine potency and cross-protection tests for foot-and-mouth disease are highly variable. This is equally true for the European Pharmacopoeia 50% protective dose test and the SouthAmerican “Protection against Podal Generalisation” test. In light of the 3Rs (Reduction, Refinement, Replacement), alternative potency and matching tests have been described based on serology for instance. Promising prospects include correlating innate immunity to protection. From serological potency data it is evident that each laboratory needs to set up its own pass-level for protection. This may not be feasible and the use of standardised/harmonised assays and reference standards should thus be explored. In vitro measures of relatedness between vaccine and field strains should also be further harmonised. Guidelines for the selection of sera for vaccine matching purposes are urgently needed as well as standardised assay protocols. Other in vitro tests for vaccine quality (vaccine stability, measurement of 146S content etc) need further research. 1. INTRODUCTION Vaccination is the cornerstone in foot-and-mouth disease (FMD) control. In Europe, vaccine production has to comply with the requirements of the European Pharmacopoeia (Ph.Eur.) Monograph 01/2008:0063 [7]. In Argentina (South-America), FMD vaccine manufacturers must adhere to Act No. 351/2006 of the Argentine Animal Health Service (SENASA) [24]. In order to fully assess the value of alternative methods for FMD vaccine quality control, it is imperative to first understand and appreciate these Gold Standards in terms of their characteristics, feasibility, costs and acceptance. The Ph.Eur. Monograph 01/2008:0063 prescribes a series of vaccine quality tests, some of which are in vitro assays while others are performed in the target species. The quality parameters include in vitro assays to (i) assess residual live virus by using highly sensitive cell lines able to detect 1 µg of 146S of FMD virus (FMDV) at a titre of at least 106 cell culture infective dose 50%, (ii) quantify the 146S antigen content of the vaccine using the sucrose-gradient method, and (iii) determine vaccine batch potency by a validated serological alternative. General safety and immunogenicity (i.e. the vaccine potency 50% protective dose (PD50) test) on the other hand are prescribed in vivo assays involving 10 and 17 cattle, respectively [7]. The latter requires expensive bio-safety level 3 (BSL-3) animal facilities which also limit the number of locations where the test could be performed. In light of the 3Rs (Reduction, Refinement, Replacement) on the use of animals for regulatory purposes [12], alternative potency tests have been described based on serology or antigen payload. In South-America, the “Protection against Podal Generalisation” (PPG) test, rather than the PD50 test, is used for assessing FMD vaccine immunogenicity. The PPG potency test is based on vaccinating 16 cattle with the full vaccine dose and subsequently challenging them with live FMDV. Again, BSL-3 facilities are needed to perform the challenge. The present paper will focus on the in vivo vaccine immunogenicity tests at hand and then suggest potential alternatives to in vivo potency and cross-rotection. Ways will be explored to move forward in getting these alternatives accepted by regulatory bodies. 2. VARIABILITY OF IN VIVO POTENCY TESTING

159

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

For routine vaccination programmes FMD vaccines must have a demonstrated potency of at least 3 PD50, in which 1 PD50 equals the vaccine dose that protects 50% of the vaccinated animals against subsequent live viral challenge at 21 days post vaccination. The PD50 value is determined by vaccinating three groups of at least 5 cattle with reduced volume doses of the vaccine. Generally, the full vaccine dose (2 ml), a quarter (0.5 ml) and a sixteenth vaccine dose are administered (0.125 ml) although other volumes may also be used. Eight days post challenge with 104 BID50 (50% bovine infective dose), animals not exhibiting FMDV-induced lesions on the feet are considered to be protected, whereas animals having one or more foot lesions are scored as unprotected [7]. From the number of protected animals for each volume dose group the PD50 content is estimated, frequently by using the method of Kärber [13]. In case of a FMD emergency in Europe, vaccination must be carried out with a vaccine of no less than 6 PD50 potency (Council Directive 2003/85/EC). Back in 2003, the mathematical theoretical 90% confidence intervals (CI) were calculated for a 10 PD50 vaccine. They ranged from 4.5 to 22 PD50 [4]. Recently, studies were conducted to investigate what would be the additional effect of the biological variation of animals to vaccination on these already wide CI. In other words, is it possible to distinguish between a potency of 3, 6 or even 10 PD50 based on the outcome of a single potency test? To answer this question, Goris et al. [9] performed ten replicate Ph.Eur. PD50 tests with a commercial, double-oil emulsion, inactivated FMD vaccine against O1 Manisa having an average potency of 9.99 PD50 [95% CI: 7.45-13.27]. The observed in vivo PD50 values ranged from 4.59 to 24.25 PD50 (Table 1). Table 1: Number of protected animals per vaccine dose group and obtained PD50 value for ten replicate FMDV O1 Manisa vaccine potency tests Trial

Number of protected animals per dose group 2 ml 0.5 ml 0.125 ml

PD50 value

95% CI

1 2

5 3

3 4

3 2

10.56 6.06

4.27-21.82 2.75-15.45

3 4

5 4

2 3

1 3

4.59 8.00

2.52-12.20 3.31-18.72

5 6

5 5

3 4

3 5

10.56 24.25

4.24-21.87 8.07-35.18

7 8

5 4

4 5

1 0

8.00 6.06

3.69-17.62 3.11-13.71

9 10

5 5

2 4

13.93 24.25

5.49-24.89 7.99-35.12

Overall

9.99

7.45-13.27

In order to check the serotype-independence of these findings, five replicate Ph.Eur. PD50 tests were performed in 2008 with a double-oil emulsion, inactivated FMD vaccine against A Iran 1997 having an average potency of 20.49 PD50 [Goris et al., unpublished data]. The observed in vivo PD50 values ranged from 10.56 to at least 32.00 PD50 (Table 2). Two additional parameters were subsequently introduced to quantify the precision of the FMD vaccine potency test. Vaccine accordance (VACC) is defined as the intra-potency test repeatability or the percentage chance of finding the same result in terms of the animals’ protection status for two similarly vaccinated animals within the same potency tests (e.g. two animals having received the full vaccine dose are both scored as protected or are both scored as unprotected). Secondly, vaccine concordance (VCON) or inter-potency test reproducibility was estimated. This is the percentage chance of finding the same result for two similarly vaccinated animals in different vaccine potency tests performed under standardised conditions (e.g. two animals vaccinated with a quarter dose both have lesions on their feet, but one belongs to the first potency test and the second animals was vaccinated in another potency test). Table 3 summarises the obtained values for the O1 Manisa and A Iran 1997 replicate PD50 trials. The precision of the Ph.Eur. FMD vaccine potency test is lowest for the 0.125 ml group. Moreover, the results suggest a serotypeindependent variability of the potency test as the overall values for VACC and VCON are highly similar.

160

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 2: Number of protected animals per vaccine dose group and obtained PD50 value for five replicate FMDV A Iran 1997 vaccine potency tests PD50 value

1 2 3 4 5

Number of protected animals per dose group 2 ml 0.5 ml 0.125 ml 5 4 4 5 5 1 4 5 5 5 5 5 5 4 4

Overall

20.49

Trial

18.38 10.56 24.25 32.00 24.25

Table 3: Vaccine accordance and vaccine concordance of the FMD PD50 vaccine potency test and associated 95% confidence interval Parameter Vaccine accordance

Vaccine concordance

Vaccine group 2 ml 0.5 ml 0.125 ml Overall

dose

O1 Manisa

A Iran 1997 trials

73.7 65.7 63.6 67.6

71.9 71.2 58.8 67.3

[65.2-81.8] [58.9-73.1] [56.8-70.6] [63.2-72.1]

[57.7-85.1] [56.4-84.5] [50.0-72.3] [59.1-75.1]

2 ml

68.4 [58.7-78.7]

72.1 [51.9-89.8]

0.5 ml 0.125 ml Overall

57.3 [51.5-65.6] 50.6 [50.0-52.2] 58.8 [54.8-63.1]

71.4 [51.1-89.7] 54.0 [50.0-72.9] 65.8 [54.6-76.7]

In conclusion, it is impossible to distinguish between a potency of 3, 6 or 10 PD50 based on the outcome of a single Ph.Eur. potency trial. Increasing the number of animals per vaccine dose group to 25 would make such distinction possible, but would also render vaccine potency testing logistically, technically and financially unfeasible. Possible alternatives might involve (i) vaccinating all cattle with the full vaccine dose (i.e. performing a PPG test) or devise alternative methods not requiring needle challenge for instance based on (ii) serology and (iii) 146S antigen content. The latter two will be discussed in section 3. As explained above, the PPG test involves 16 cattle all vaccinated with the full vaccine dose and subsequently challenged with 104 BID50. Seven days post challenge all cattle are read clinically. In the absence of FMDV-induced secondary lesions on the feet, the animals are protected. As an acceptance criterion, at least 12 out of the 16 vaccinated animals have to be protected (i.e. 75% PPG) [24]. In literature, it has been suggested that the PPG test is reproducible [25]. However, the level of reproducibility has not been quantified. Recently, six replicate PPG FMD vaccine potency tests were performed with a water-in-oil emulsion, inactivated vaccine against FMDV A24 Cruzeiro having an average potency of 88.5% PPG [95% CI: 80.7-93.5]. The obtained PPG potency values ranged from 75.0 to 100.0% (Table 4) [10]. VACC and VCON were estimated to be 75.9% [95% CI: 64.9-86.2] and 73.7% [95% CI: 62.1-84.3], respectively, values which correspond to those obtained for the full vaccine dose group of the PD50 test. Table 4: Summary of six replicate FMDV A24 Cruzeiro vaccine potency tests Trial

Vaccinated animals

PPG%

95% CI

Protected

Total

100.0

80.6-100.0

93.8

71.8-98.6

161

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

93.8

71.1-98.5

81.3

56.2-93.0

87.5

64.2-96.3

75.0

50.2-89.4

Overall

88.5

80.7-93.5

To summarise, the Gold Standard vaccine potency tests have proven valuable in the past, but their statistical limitations (e.g. the width of the 95% CI and low level of precision) should be better recognised. It, therefore, seems unfair to criticise alternative potency tests for not perfectly correlating to an imperfect test [4]. In light of the 3R principle, in vitro alternatives merit more attention. 3. IN VITRO ALTERNATIVES TO IN VIVO POTENCY TESTING Over the course of the last 20 years, FMD researchers have reported on the correlation between FMD vaccine-induced neutralising antibodies and protection against challenge and on the use of 146S data to predict protection [e.g. 19-20, 22]. It is beyond the scope of this paper, however, to give an extension review of all methods proposed, most of which are based on probit analysis. Two studies have used logistic regression curves to statistically evaluate the correlation between post vaccination serum titres and protection against live viral challenge [2, 11]. This section will focus on an example of the logistic regression model developed by Goris et al. [11] for FMDV serotype O based on the serological data obtained from the above described ten replicate PD50 tests [9]. Figure 1 depicts the logistic regression model generated for the solid-phase competition ELISA (SPCE). On the x-axis the log10 serum titres of sera sampled at 21 days post vaccination are shown, whereas the corresponding protection status of the animal is given on the y-axis. An animal not protected is represented by zero probability of protection and a protected animal has a probability of protection of one. For certain log10 SPCE serum titres, animals were found to be protected and unprotected against in vivo live viral challenge at 21 days post vaccination. The regression model, thus, estimates the best fitted line between these two possibilities. The fit of the model was assessed using the Aikake’s Information Criterion. The association between the log10 serum titre and the probability of protection was determined by the Somers’ D rank correlation factor as decribed by Goris et al. [11].

Figure 1: Logistic model for the solid-phase competition ELISA (full line). The dotted line represents the 95% confidence interval around the predicted probability of protection. Subsequently, a Receiver Operating Characteristics (ROC) analysis was carried out to estimate the log10 serum titre for which the probability of predicting in vivo protected animals as protected based on their serological response was maximised without resulting in too many false predictions of protection (i.e. antibody pass-level for protection). The ROC analysis thus allows quantifying the sensitivity (i.e. predicting in vivo protected animals as protected using serology), specificity (i.e. predicting in vivo unprotected animals as unprotected using serology) and accuracy (ACC) of the

162

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

indirect potency test. An illustration of the ROC analysis for the SPCE logistic regression model is given by Figure 2.

Figure 2: The Receiver Operating Characteristics curve for the indirect FMD vaccine potency test based on the solid-phase competition ELISA logistic regression model. The point closest to the upper left corner corresponds to the ideal situation (100% sensitivity and 100% specificity). This trade-off point results in the highest level of ACC (Table 5) and is considered to be the antibody pass-level for protection. In other words, animals with a log10 serum titre of at least 1.45 in the SPCE are considered to be protected using the developed logistic regression model. All other animals are scored as unprotected. This antibody pass-level corresponds to about 79.5% probability of protection. Using this antibody pass-level for protection, the indirect potency assessment model based on logistic regression followed by ROC analysis exhibits a sensitivity of 82.4% and a specificity of 78.0%. The ACC was estimated to 80.2% (Table 5). Table 5: The Receiver Operating Characteristics analysis for the solid-phase competition ELISA Probability protection 0.406 0.547 0.687 0.795 0.876 0.928 0.958 0.976 0.986 0.992

Log10 titer 1.00 1.15 1.30 1.45 1.60 1.76 1.90 2.05 2.20 2.35

serum

1Specificity 0.634 0.439 0.317 0.220 0.098 0.024 0.024 0.000 0.000 0.000

Sensitivit y 0.991 0.981 0.880 0.824 0.676 0.500 0.315 0.194 0.065 0.028

ACC (%) 67.8 77.1 78.1 80.2 78.9 73.8 64.5 59.7 53.2 51.4

Subsequently, the validity of the indirect potency test was assessed by using serological SPCE data from 15 animals vaccinated according to Ph.Eur. standards with a FMD vaccine batch against FMDV O1 Manisa which was not related to the batch used to generate the model. The SPCE was performed ten times independently to estimate the precision (VACC and VCON) of the indirect potency test based on serology. In Table 6, protected animals (i.e. log10 serum titre ≥ 1.45) are represented by “1”, whereas unprotected animals are represented by “0” (i.e. log10 serum titre < 1.45). The indirect potency ranged from 3.48 to 8.00 PD50, with an average indirect potency of 4.86 PD50 [95% CI: 2.66-12.00]. Based on in vivo estimations, the vaccine batch had a potency of 10.56 PD50 [95% CI: 4.27-21.82]. From the highly overlapping 95% CIs, it cannot be concluded that the indirect potency is significantly different from the in vivo potency obtained. VACC and VCON were estimated to be 65.8 and 60.7%, respectively.

163

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 6: External validation of the indirect vaccine potency assessment based on the solid-phase competition ELISA logistic regression model full vaccine dose 1 2 3 4

quarter vaccine dose 6 7 8 9 10

sixteenth vaccine dose 11 12 13 14 15

in vivo

SPCE SPCE SPCE SPCE SPCE

1 2 3 4 5

1 1 1 1 0

1 1 1 1 1

0 0 0 0 0

1 1 1 1 1

0 0 0 0 0

1 1 1 1 1

0 0 0 0 0

0 1 0 1 0

0 0 0 0 0

0 1 0 0 0

1 1 1 1 1

0 0 0 0 0

SPCE 6

SPCE 7

SPCE 8 SPCE 9

1 1

0 0

1 1

0 0

1 1

0 0

1 1

0 0

SPCE 10

In conclusion, a valid, alternative indirect PD50 test was developed. The proposed method is quantitative, sensitive, specific, accurate and at least as precise as the prescribed in vivo test. Alternative PPG potency tests based on liquid-phase blocking ELISA expected percentage protection (EPP) have been previously described by Robiolo et al. [21] and will not be discussed in this paper.

4. VARIABILITY OF IN VIVO CROSS-PROTECTION TESTING FMDV is a highly variable RNA virus, and in general, there is little or no cross-protection between serotypes and even among different strains of the same serotype [Paton et al., 2005]. Therefore, it is impossible for individual countries or consortia of countries to store every possible strain. Consequently, only several antigenic representative, cross-reactive (i.e. broad spectrum of reactivity) FMDV strains are kept over liquid nitrogen in antigen reserves. In order to assess the suitability of the strains for a particular situation (e.g. outbreak scenario), a vaccine matching test should be readily performed. The direct comparative matching test is the in vivo cross-protection test which is closely related to the FMD in-field situation. Recently, a series of in vivo crossprotection assays have been performed in vivo according to Ph.Eur. standards with the exception that animals were challenged with intra-typic heterologous FMDV strains (e.g. vaccination against FMDV A Iran 99 and challenge with FMDV A Iran 96) [3]. In light of the observed variability for homologous vaccine testing (see sections 2), a series of replicate in vivo cross-protection studies were set up for both major FMD vaccine potency tests (PD50 and PPG). Goris et al. [unpublished data] performed five replicate Ph.Eur. PD50 test with a double-oil emulsion inactivated FMD vaccine against A Iran 1997. At 21 days post vaccination, all cattle were challenged with 104 BID50 of FMDV A22 Iraq 24/64. The average cross-protection PD50 value was estimated to be 1.03 PD50, but ranged from 0.50 to 3.48 PD50 (Table 7). Table 7: Number of protected animals per vaccine dose group and obtained PD50 value for five replicate FMDV A Iran 1997 vaccine cross-protection tests using intra-typic heterologous challenge conditions with FMDV A22 Iraq 24/64 Trial

164

PD50 value

1 2 3 4 5

Number of protected animals per dose group 2 ml 0.5 ml 0.125 ml 3 2 2 2 2 0 1 0 0 1 0 0 0 0 0

Overall

1.03

3.48 1.52 0.66 0.66 0.50

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Similar studies were performed for the PPG test. In total, four replicate PPG FMD vaccine crossprotection tests were performed in vivo with a water-in-oil emulsion, inactivated vaccine against A24 Cruzeiro. At 30 days post vaccination, all cattle were challenged with FMDV A/Arg/01. The average cross-protection PPG% was estimated to be 26.6% [95% CI: 17.4-38.5], but ranged from 12.5 to 56.3% PPG (Table 8) [10]. Table 8: Summary of four replicate FMDV A24 Cruzeiro vaccine cross-protection tests using intratypic heterologous challenge conditions with FMDV A/Arg/01 Trial

Vaccinated animals

PPG%

95% CI

16 16 16 16

56.3 25.0 12.5 12.5

33.2-76.6 10.5-50.0 3.8-36.4 3.9-36.6

26.6

17.4-38.5

Protected

Total

1 2 3 4

9 4 2 2

Overall

In conclusion, although valuable, in vivo cross-protection studies are highly variable and their statistical limitations (e.g. the width of the 95% CI) should be better recognised. In light of the 3R principle, it, therefore, seems unfair to criticise alternative in vitro vaccine matching (see section 5) tests for also exhibiting a certain degree of variability. 5. IN VITRO ALTERNATIVES FOR VACCINE MATCHING For vaccine matching purposes, FMDV strain selection is based on indirect serological methods (rvalues) [8, 23], on sequence data [17] or alternatively on the calculation of the relatedness between the field isolate and available vaccine strains using the EPP method [26]. Most of these methods have been extensively reviewed in recent literature [e.g. 17] or are OIE-prescribed tests [26]. Moreover, a paper proposing some guidelines for determining FMD vaccine strain matching by serological r-values has recently been accepted for publication [14]. The reader is thus referred to these papers for more detailed information. 6. FUTURE RESEARCH ON VACCINE QUALITY TESTING Even though scientific evidence to replace in vivo FMD vaccine potency testing by a serological approach seems overwhelming, Ahl and colleagues [1] have urged caution when interpreting data obtained by pooling serological results obtained over the course of time with different assay methods and/or cell lines. Others have criticised any and all serological alternative [5, 15]. They argued that virus neutralisation tests and ELISAs merely measure the capacity of antibodies to neutralise the virus in cell cultures and/or to interact with it. Therefore, these tests constitute no measure of the protective immune response against FMDV, in which an important role is attributed to innate immunity as well. As such, the use of promising tests such as for instance the measurement of the interferon-γ production in vitro from whole blood [16] should be further explored. A collaborative study on the quantification of the 146S antigen content in FMD vaccines in 11 different FMD laboratories using a standardised protocol based on the sucrose gradient method indicated low levels of within laboratory variation whereas between laboratory variation was greater [6]. Nevertheless, the variability is generally considered too high. Alternative tests for measuring the 146S antigen content based on ELISA methods in the final product are urgently needed. The same applies to reliable assays to measure vaccine stability and VP1 integrity, for instance by developing ELISAs that quantify the amount of 12S and 146S particles. Moreover, to allow application of a vaccinate-to-live policy and the subsequent use of non-structural protein (NSP) tests to substantiate freedom from FMD [18], vaccine purity and the removal of NSP are essential. ELISAs able to measure the reduction of NSP during vaccine purification, formulation and in the final product could be handy tools for producer-independent assessment of vaccine quality. All proposed new assays should, however, be reliable and biologically relevant (fit-for-purpose). 7. SUMMARY CONCLUSIONS AND RECOMMENDATIONS

165

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Given the variability of in vivo FMD vaccine potency and cross-protection testing (PD50 and PPG) and in light of the 3Rs, more attention should be paid to the acceptance and further development of in vitro alternatives. In light of the controversy surrounding in vitro FMD vaccine potency testing based on serology and the time and effort it takes to establish a statistically sound correlation between in vivo protection afforded by a certain vaccine and the serum titres measured in a particular test, it might be advisable to only conduct alternative potency assessment in accredited laboratories having demonstrated statistically sound correlation curves. Moreover, the use of standardised reagents and test procedures should be explored to allow a more harmonised approach globally. Furthermore, reference standards for in vitro vaccine potency and vaccine matching purposes should be developed. Following a subsequent calibration phase, a proficiency testing scheme (PTS) for both vaccine potency and matching should be organised. Nonetheless, it must be recognised that different laboratories around the world have different expertise (e.g. African countries have more experience with FMDV SAT serotypes than European of American laboratories). The development of the above mentioned reagents, protocols and PTS should thus be done in a joint effort. As such, the responsibility of success is shared among the interested laboratories. Participation of vaccine producers in supplying samples, data and reagents is highly encouraged. 8. ACKNOWLEDGEMENTS The review was made possible by several studies jointly funded by the FMD_ImproCon project (www.fmdimprocon.org) under the Sixth Framework Programme (grant SSPE-CT-2003-503603), the Belgian Federal Public Service Health, Food chain safety and Environment (grant RT-05/06ALTANDI-2) and the Argentine Beef Promotion Institute (IPCVA). Special thanks go to Bayer HealthCare for the O1 Manisa vaccine, Intervet International for the A Iran 1997 vaccine, and Biogenesis Bago for the A24 Cruzeiro vaccine used in the references studies. Sincere thanks goes to Tom Willems for his statistical support. 9. REFERENCES [1] Ahl, R., Haas, B., Lorenz, R.J., Wittman, G. 1990. Alternative potency test of FMD vaccines and results of comparative antibody assays in different cell systems and ELISA. Report of the Session of Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease. FAO, Rome, Rio Lindholm, Denmark, June 25-29. pp 51– 54. [2] Barnett, P.V., Statham, R.J., Vosloo, W., Haydon, D.T. 2003 Foot-and-mouth disease vaccine potency testing: determination and statistical validation of a model using a serological approach. Vaccine 21(23): 3240–3248. [3] Brehm, K.E., Kumar, N., Thulke, H.H., Haas, B. 2008 High potency vaccines induce protection against heterologous challenge with foot-and-mouth disease virus. Vaccine 26(13):1681-1687. [4] Doel, T. 2003. Potency testing of foot-and-mouth disease vaccines: an industrial perspective. pp 41–47. [5] Doel, T.R. 2005 Natural and vaccine induced immunity to FMD. Curr Top Microbiol Immunol 288: 103-131. [6] Doel, T.R., Mowat, G.N. 1985. An international collaborative study on foot-and-mouth disease virus assay methods. 2. Quantification of 146S particles. J Biol Stand 13(4):335-344. [7] European Pharmacopoeia version 6.0. 2008. Monograph 01/2008:0063. Foot-and-mouth disease (ruminants) vaccine (inactivated). [8] Ferris, N.P., Donaldson, A.I. 1992. The World Reference Laboratory for Foot and Mouth Disease: a review of thirty-three years of activity (1958-1991). Rev Sci Tech 11(3):657-684. [9] Goris, N., Merkelbach-Peters, P., Diev, V.I., Verloo, D., Zakharov, V.M., Kraft, H.P., De Clercq, K. 2007. European Pharmacopoeia foot-and-mouth disease vaccine potency testing in cattle: between test variability and its consequences. Vaccine 25(17): 3373-3379. [10] Goris, N., Maradei, E., D'Aloia, R., Fondevila, N., Mattion, N., Perez, A., Smitsaart, E., Nauwynck, H.J., La Torre, J., Palma, E., De Clercq, K. 2008. Foot-and-mouth disease vaccine potency testing in cattle using homologous and heterologous challenge strains: precision of the "Protection against Podal Generalisation" test. Vaccine 26(27-28): 3432-347. [11] Goris, N., Willems, T., Diev, V.I., Merkelbach-Peters, P., Vanbinst, T., Van der Stede, Y., Kraft, H.P., Zakharov, V.M., Borisov, V.V., Nauwynck, H.J., Haas, B., De Clercq, K. 2008. Indirect foot-and-mouth disease vaccine potency testing based on a serological alternative. Vaccine 26(31): 3870-3879.

166

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[12] Hendriksen, C.F. 2002. Refinement, reduction and replacement of animal use regulatory testing: current best scientific practices for the evaluation of safety and potency of biologicals. ILAR J 43(Suppl):S43-S48. [13] Kärber, G. 1931. Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche. Arch Exp Pathol Pharmakol 162: 480–483. [14] Mattion, N. Goris, N., Willems, T., Robiolo, B., Maradei, E., Perez Beascoechea, C., Perez, A., Smitsaart, E., Fondevila, N., Palma, E., De Clercq, K., La Torre, J. In press. Some guidelines for determining foot-and-mouth disease vaccine strain matching by serology. Vaccine in press. [15] McCullough, K.C., De Simone, F., Brocchi, E., Capucci, L., Crowther, J.R., Kihm, U. 1992. Protective immune response against foot-and-mouth disease. J Virol 66(4): 1835-1840. [16] Parida, S., Oh, Y., Reid, S.M., Cox, S.J., Statham, R.J., Mahapatra, M., Anderson, J., Barnett, P.V., Charleston, B., Paton, D.J. 2006. Interferon-gamma production in vitro from whole blood of foot-and-mouth disease virus (FMDV) vaccinated and infected cattle after incubation with inactivated FMDV. Vaccine 24(7): 964-969. [17] Paton, D.J., Valarcher, J.F., Bergmann, I., Matlho, O.G., Zakharov, V.M., Palma, E.L., Thomson, G.R. 2005. Selection of foot-and-mouth disease vaccine strains--a review. Rev Sci Tech 24(3):981-993. [18] Paton, D.J., De Clercq, K., Greiner, M., Dekker, A., Brocchi, E., Bergmann, I., Sammin, D.J., Gubbins, S., Parida, S. 2006. Application of non-structural protein antibody tests in substantiating freedom from foot-and-mouth disease virus infection after emergency vaccination of cattle. Vaccine 24(42-43): 6503-6512. [19] Pay, T.W., Hingley, P.J. 1992. A potency test method for foot-and-mouth disease vaccine based on the serum neutralizing antibody response produced in cattle. Vaccine 10(10): 707–713. [20] Pay, T.W., Hingley, P.J. 1992. Foot-and-mouth disease vaccine potency tests in cattle: the interrelationship of antigen dose, serum neutralizing antibody response and protection from challenge. Vaccine 10(10): 699–706. [21] Robiolo, B., Grigera, P.R., Periolo, O.H., Seki, C., Bianchi, T., Maradei, E., La Torre, J.L. 1995. Assessment of foot-and-mouth disease vaccine potency by liquid-phase blocking ELISA: a proposal for an alternative to the challenge procedure in Argentina. Vaccine 13(14):1346–1352. [22] Rweyemamu, M.M., Black, L., Boge, A., Thorne, A.C., Terry, G.M. 1984. The relationship between the 140S antigen dose in aqueous foot-and-mouth disease vaccines and the serum antibody response of cattle. J Biol Stand 12(1): 111–120. [23] Rweyemamu, M.M. 1984. Antigenic variation in foot-and-mouth disease: studies based on the virus neutralization reaction. J Biol Stand 12(3):323-337. [24] SENASA - Argentine Animal Health Service. Act no. 351/2006—boletin official no. 30.9405/7/06. Available at http://infoleg.mecon.gov.ar/infolegInternet/anexos/115000-119999/117636/norma.htm; 2006. [25] Vianna Filho, Y.L., Astudillo, V., Gomes, I., Fernández, G., Rozas, C.E.E., Ravison, J.A., Alonso, A. 1993. Potency control of foot-and-mouth disease vaccine in cattle. Comparison of the 50% protective dose and the protection against generalization. Vaccine 11(14): 1424–1428. [16] World Organisation for Animal Health. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Chapter 2.1.5, Paris, 2008.

167

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

168

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 25

A STANDARDIZED DIRECT CONTACT CHALLENGE METHOD FOR FMDV IN SWINE J. M. Pacheco1, M. Tucker1, E. Hartwig1, L. Rodriguez1* 1

Foreign Animal Disease Research Unit, Agriculture Research Service, U.S. Department of Agriculture, Plum island Animal Disease Center, Greenport, New York, USA.

INTRODUCTION Pigs are excellent viral amplifiers of Foot-and-mouth Disease Virus (FMDV) and play an important role in the airborne transmission and spread of FMDV to other species and among different premises. Thus protection of pigs either by vaccination or biotherapeutic treatment could be essential to control FMD outbreaks. Currently the challenge method for pigs consists of direct virus inoculation in the heel bulb. A swine challenge model that better reflects natural infection of FMDV is necessary in order to accurately assess the effectiveness of vaccine and biotherapeutic candidates. There is ample evidence in the literature that pigs are highly resistant to natural aerosol infection of FMDV while direct contact transmission is more common. In order to develop a successful swine challenge model, first we must define the parameters determining infection in pigs during direct exposure. MATERIALS AND METHODS In this study we describe a methodology for direct contact transmission in pigs for two FMDV strains (serotypes A and O). For each strain experimental groups containing 4 pigs were exposed to directly inoculated pigs that had received 100 pig heel infectious doses 50 (PHID50) each by heel-bulb inoculation. In each case we controlled the ratio of inoculated to naïve pigs (1:2) and time of exposure (4 h and 18 h). After contact exposure pigs were housed two per room but separated from direct contact. The criteria for deciding when the donor pigs were infectious were based on viral RNA detection in serum, saliva and nasal swabs. We utilized a standardized FMDV real-time RT-PCR to monitor viral infection in clinical samples as well as in air samples collected from animal rooms housing donors and recipients throughout the studies utilizing dry filters. Clinical observations were done and samples obtained at 24h intervals for 10 days. RESULTS AND DISCUSSION For both serotypes directly inoculated animals had viral RNA in saliva a nasal swabs at 48hpi and were considered infectious and moved to the contact room at that time. For serotype A (subtype A24) both 4h and 18h exposure resulted in infection of all contact animals. In contrast for serotype O (O1- Manisa) 4h exposure time was insufficient and none of 4 contact pigs became infected but 18h exposure resulted in infection of all animals. Viral RNA was readily detectable in air samples and was consistent with the transmission results. CONCLUSIONS We have established a consistent direct contact exposure methodology for FMDV in pigs for two FMDV strains. Each strain demonstrated different transmission characteristics and required different exposure times for successful contact transmission. The data emphasize the need for a thorough evaluation of each viral strain in a well defined contact challenge methodology. We are currently testing this methodology in vaccine efficacy trials.

169

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

170

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 26 CAN IN VITRO METHODS RELIABLY REPLACE EXISTING VACCINE CHALLENGE TESTS? R. Reeve1*, P. Barnett2 and D. Haydon1 1

Boyd Orr Centre for Population and Ecosystem Health, Division of Ecology and Evolutionary Biology, Graham Kerr Building, University of Glasgow, Glasgow, G12 8QQ, United Kingdom 2 FMD Vaccine Group, Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Surrey, GU24 0NF, United Kingdom *Corresponding author: r.reeve@bio.gla.ac.uk

ABSTRACT The OIE Terrestrial Manual and the European Pharmacopoeia (EP) still describe live challenge experiments for FMDV vaccine potency tests. However, the EP allows for other validated tests and specifically in vitro tests if a “satisfactory pass level” has been determined. Much research is now focused on validating the replacement of live challenge with in vitro tests. Three issues have to be resolved first: the definition of a satisfactory vaccine; a record of the effectiveness of existing live challenge tests; and a validation of new tests to show that they can safely replace the old ones. This paper looks at all three issues. It describes a metric, based on vaccine potency requirements, against which tests can be measured; we then characterise existing tests, and recommend that these be the minimum standard for a replacement in vitro test; finally, we propose using this same metric to validate such a test. 2. INTRODUCTION The vaccine requirements for ten of the fifteen former List A diseases covered by the OIE Terrestrial Manual[18] include some form of live challenge experiment for either efficacy or potency tests; the former List B and the European Pharmacopoeia (EP)[8] cover significantly larger numbers of animal vaccines for which live challenge experiments are still required. The most well-known of these is the dose-response experiment, which together with the statistic now known as Spearman-Kärber[23] is still widely used for calculating the PD50 of vaccines. This method is cited by Doel[11] as the most commonly used test for foot-and-mouth disease (FMD) vaccine potency. The dose-response experiment mandated by the OIE consists of three groups of at least five cattle each, which are injected with different volumes of the vaccine (a full dose for one group, 1/4 for another, and 1/16 for the last is suggested by the EP[9]), and are subsequently challenged with 10,000 ID50 (50% infectious dose) of the homologous live virus. From records of subsequent generalisation of the disease to secondary sites in these animals, an estimate is made of the PD50 for the vaccine using Spearman-Kärber, and this is then compared to the desired threshold – usually 3 (1/3 of a full dose) or 6 (1/6 of a full dose). If the PD50 is at or above this threshold (i.e. at most this amount is required to protect them 50% of the time) then the vaccine is deemed to have passed the test. The EP recommends the same dose-response experiment and threshold, but a different statistic to estimate the PD50 for the vaccine – it describes Maximum Likelihood (ML) techniques[10], specifically probit analysis[4, 12] and logistic regression[3, 16]. These differ in making marginally different assumptions about the relationship between dose and response – that they are connected by the probit and logit functions respectively. The OIE also allows the Protection against Podal Generalisation (PPG) [26] test to be carried out for FMD. Under this test, a group of animals is vaccinated with a full dose of the vaccine and is subsequently challenged by the live virus – to pass the test the proportion protected must be above a threshold. Currently, although the diagnostic tests are well quantified in terms of their specificity and sensitivity for detection of infection, the potency tests described above have not been subject to the same analysis. A descriptive overview of different tests was given by Tanner[25], and a preliminary attempt was made by Sutmöller[24] to measure their theoretical repeatability. This was followed up by a large-scale empirical comparison of repeatability by Vianna Filho [26], and

171

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

then a smaller but more detailed analysis of dose-response experiments by Goris[13-15]. Until now, however, it has proved computationally intractable to quantify the sensitivity, specificity and precision (the probability of a positive test result being a true positive) of these tests. Comparing the true quality of the different tests requires us either to have a gold standard against which we can benchmark them; to be able to generate vaccines of known quality; or to agree on a model for the protection afforded to an individual by a vaccine and then to simulate the process of generating and testing vaccines. Insofar as a gold standard exists, it involves increasing the group sizes and/or the number of groups in the existing tests. Vianna Filho [26] and Goris [13, 14] have indirectly done this for a few specific vaccine batches while examining between-test variability, but in general it requires unjustifiably many animal challenge experiments. Nor can we generate vaccines of known potency – indeed; we are required to test not just individual vaccines, but individual batches of vaccine to confirm their potency levels. The remaining option is to agree on a model and simulate the process. This is the approach taken here. The advantage of this approach is that it does not involve the live challenge of further animals. It does, however, rely on explicit assumptions about the process being modelled. As we shall see, these assumptions are mostly the same as those that already underlie the statistical tests themselves. It therefore provides no information about the validity of the assumptions. However, it does expose them to ready inspection. To make a comparison between the tests we require the simulation to generate a measure of confidence in the pass/fail responses returned by each test. We can then use these confidence levels to compare the tests directly. If we accept the assumptions underlying the simulation we are using, this allows us straightforwardly to select the test in whose results we have the highest confidence. One step remains missing from this approach: this is the definition of a good vaccine. The Spearman-Kärber test, probit analysis and logistic regression all answer the following question: “What is the value of the PD50 for this vaccine?” This answer is then used to address a second question: “Is the PD50 at least 3 [or 6]?” The PPG test addresses a perhaps simpler question: “Will the vaccine protect 75% [or 90%, etc.] of animals?” However, accurately quantifying our confidence in the answer, as well as calculating the sensitivity and specificity of the tests, requires assumptions about the distribution of possible vaccines, which we will discuss in more detail later. Which, if either, of these questions is appropriate for assessing a good vaccine? Pay argues that we should be focussing on “estimated percentage protection” rather than PD50[19], and Vianna Filho agrees [26]. However, there is a long history of using PD50 in Europe, and so it is unlikely that an answer will be universally agreed to this question in the immediate future. If animals are always to be inoculated with a full dose of the vaccine, the protection afforded by that dose would seem to be a useful parameter to estimate. Alternatively, if there is a desire to assess how concentrated a vaccine is, or a belief that the dose-response slope is known, so the PD50 is better for calculating the full dose protection, then the PD50 should be used. In fact it transpires that either experiment can be used to answer either question, through the use of different statistics. A more serious failing is that it is ambiguous in current usage whether the definition of a good vaccine is that its PD50 is above 3 (or the protection it affords for a full dose is above 75%), or whether these are just the thresholds for batch acceptance. This has important implications for the sensitivity and specificity of the potency test: for instance, if we set a threshold for success of 75% in our potency test but define a good vaccine as one that protects 50% of animals, then the specificity will be high (few poor vaccines will pass) but the sensitivity will be relatively low (because a number of good vaccines will fail). Similarly, if a good vaccine has a PD50 of at least 3 the threshold could be set higher, say to 4, in order to increase the specificity at the expense of sensitivity. We remain agnostic on what the definition of a good vaccine should be, we merely note that a decision has to be taken. 3. METHODS Our approach is as follows. We calculate the likelihood functions for the experiments that are currently used (dose-response and protection against podal generalisation). We analyse existing statistics and estimation techniques used with these experiments (such as probit analysis, logistic regression, Spearman-Kärber, etc.) and propose another statistic for comparison. We create a

172

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Bayesian Markov Chain Monte Carlo[1] (MCMC) framework to allow use of domain knowledge and the data from other experiments to inform the analysis of the performance of new vaccines. For a specific definition of a good vaccine, we choose possible thresholds for accepting or rejecting results. Using the results of the MCMC simulation, we calculate which results achieve the threshold and should pass the different proposed potency tests (each test is a combination of a challenge experiment, a statistic for estimating efficacy, and a pass/fail threshold). We then characterise the potency tests in terms of their sensitivity, specificity and precision in different scenarios based on the results of the simulations. This provides an objective criterion through which to determine the best test (combining experiment, statistic and threshold) for the chosen definition of a good vaccine given our prior knowledge. Finally we propose using the characteristics of this best test as the minimum requirement for a replacement in vitro test, and suggest how to extend the above technique to characterise such a test. 3.1 Experiments The probabilities of every result given specific vaccine efficacies are governed by simple equations. These can be used to generate the likelihood functions for the efficacies given a result, which can in turn be inspected to determine those important parameters (like PD50, probability of protection of a full dose) that best characterise the behaviour of the vaccine. These in turn can be used to determine the summary statistics that allow us to best use the results of vaccine efficacy trials to estimate the parameters. In order to write down the equations, the following assumptions about how protection works for vaccines are necessary: The probability of protection is related to the logarithm of the dose through the probit or logit function; Individual animals have the same probability of protection for a given dose. These assumptions are precisely those made in probit analysis and logistic regression to obtain maximum likelihood estimates for the PD50. Note that under the first assumption, the use of these functions requires that there is no natural immunity to the disease and the vaccine offers perfect protection in sufficiently high dose. For PPG and the dose-response experiment using the logit function, this work produces natural parameters for the system and summary statistics that are useful for analysing experimental results. For probit, however, the equations do not simplify. As a consequence, we can do very little analytically if we assume that the probit rather than the logit function connects the probability of protection to the logarithm of the dose. Moreover, these two functions are extremely similar, and for this reason we prefer the logit function. However, to confirm consistency of results, it is possible to duplicate the analyses using both functions. 3.2 Statistics The next area to study is the statistics and estimation techniques used to derive estimates for the vaccine efficacy. Better understanding the estimates and the uncertainty associated with them will provide a clearer understanding of their relative merits. The only statistic that has been used in the past for the PPG experiment is the ML estimate for the proportion protected by a full dose, which is just the fraction protected. This is easy to calculate and is provably optimal for this kind of experiment. There are also well-established techniques for calculating the uncertainty in these estimates. For dose-response experiments, however, the situation is less clear, and statistics that have been used in the past include: Reed-Muench[22]; Spearman-Kärber [17, 23]; ML estimates via probit analysis [12]; and ML estimates via logistic regression[16]. The Reed-Muench and Spearman-Kärber statistics, and the estimates for their standard errors, are very straightforward to calculate from the experiment data[11]. These are not unbiased estimators however, and the Spearman-Kärber variance estimate is only an approximate lower bound for the true variance, and as such can be improved upon empirically. The Reed-Muench variance has been estimated from previous experimental observations[21], and so is also far from perfect. The use of MCMC simulation to generate empirically the distributions of parameter values for existing vaccines allows us to directly calculate confidence intervals for tests that use all of the above statistics rather than relying on these estimates.

173

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Clearly protection is expected to increase as the dose increases, so meaningful ML estimates should be associated with positive slopes. In practice, however, Vianna Filho[26] cite 22% of trials in a large study not having a dose-response relationship. Negative slopes can arise, and the ML estimate is then that the vaccine causes infection, and failure to account for this can result in accepting a PD50 that is in fact a nonsensical ID50. Similarly, if there is no systematic change in protection as dose changes, then we will normally find that iterative ML techniques will not converge. In both of these cases (except the special case of every animal being protected) our domain knowledge immediately tells us that these ML conclusions are false and that our ML estimates (if any) for the PD50 are unusable. However, there is no general method for adding domain knowledge to ML techniques. The standard approach to this is to discard and repeat the experiment or to use a different statistic. However, it is by no means clear that the experiment needs to be discarded, as the results are not inconsistent with a valid test, and so discarding them would at the very least be wasteful and inconsistent with the three Rs. A Bayesian approach offers a way to use these results – it allows the addition of prior domain knowledge to the system and the evaluation of new experimental data in that context. Using this framework, the difficulties outlined above can be avoided by posing a question of the form: “Given that we know that this is a working vaccine and hence protective in sufficiently high quantities, and that there is no natural immunity to the disease, what is the most likely value [or the distribution of possible values] for the PD50 given the results of this experiment?” 3.3 Bayesian inference All of the statistics examined so far provide estimates for either the PD50 or the probability of protection of a full dose; however, the accuracy of these estimates is uncertain; confidence intervals rely on approximations which we cannot be confident hold with such small sample sizes (15 or 16 animals). Prior knowledge can instead be incorporated to calculate empirical values. The standard approach for incorporating prior knowledge into experimental analysis is Bayesian inference[7]. Provided our prior information and our assumptions are appropriate, this will maximise the information that can be extracted from our new result. In addition to those mentioned previously, we make the following further assumptions: The vaccine provides an effective protection against the disease: i.e. the slope is strictly positive so that a higher dose provides a higher probability of protection. There is some level at which vaccines are likely to have similar properties, whether it be different batches of the same vaccine, different vaccines manufactured from the same strain or the same serotype, all FMDV vaccines, or even vaccines of a particular type (e.g. using a specified adjuvant). The first assumption is uncontroversial. With regard to the second assumption, this is critical to our ability to generalise about foot-and-mouth disease vaccines, and particularly to our ability to move to in vitro tests; however, it is widely supported by data at the batch level [2]. 3.4 Thresholds With the two types of experiments we have (PPG, and dose-response), we can define criteria for accepting a vaccine – a threshold and a confidence level: for protection against generalisation, the probability is at least x% that the protection of a full dose is above y%for dose-response, the probability is at least x% that the PD50 is at or above y Previously, for a given experiment we could calculate the associated statistics for every possible result (Reed-Muench, Spearman-Kärber and Maximum Likelihood estimates from probit analysis and logistic regression) and compare that to the threshold, y, for a pass-fail answer; however, we would not know the confidence, x, that we could have in that answer. Although there are estimates for confidence intervals, they are generally without reliable information on their own accuracy, especially at low sample sizes. Now, however, we can empirically generate the posterior distributions of the parameters from our existing data using MCMC techniques, and we can use these to decide which results to accept. In fact, the output of the MCMC simulation is a sample from the posterior distributions, so we directly measure the probability that a vaccine exceeds the threshold (y) by counting the proportion of samples where the efficacy is above it; this can be directly compared to the required probability x to determine whether or not a vaccine generating this result should pass the test. This is in fact also our new Bayesian statistic, which we will compare with the existing ones. To characterise the individual tests, we simulate new vaccines by drawing efficacies at random from the distributions for unknown vaccines (vaccines with no current challenge results) that we also also generate in the same MCMC simulation – both good (better than our definition of a good vaccine) and bad (worse) – and calculating the probabilities of individual test results for them. We

174

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

then calculate the pass/fail results for each result, and compare that to the “truth” of whether the vaccine is good or bad. From this we can directly calculate the sensitivity, specificity and precision of each test. 3.5 Test Selection We now have a method for characterising any test. Now for any given experiment and thresholds we can compare all of the different statistics, but we can vary different aspects. In particular, we can see how the characteristics of the test change when we: Keep the threshold the same, and change our definition of a good vaccine; Keep the definition and change the threshold or confidence level; Keep the definition and vary the threshold and confidence level systematically to maintain a fixed precision for the test. The first is clearly distinct from the others, which attempt to judge the best way of achieving a given standard. However, it is important to appreciate that it often not clear exactly what constitutes a good vaccine. In each case we can plot graphs of how sensitivity and specificity trade off – so called receiver-operator characteristic (ROC) curves [27] – and we can then directly compare the different tests. We might, for instance, say that we cannot accept a false positive rate of over 10%, so then we can use the true positive rate or the precision as a metric to choose the best test amongst those that satisfy this requirement. Whatever requirements we finally decide on, we can create a metric that we can use to directly compare all possible tests. 3.6 Immunological assays We can then use the same framework to examine in vitro assays, such as Virus Neutralisation Tests (VNT), ELISA, and so on. In the case of such a test where we already know that it holds a specific relationship with protection (e.g. VNT titres for FMDV are linked by the logit function to probability of protection [2]), we can incorporate the model and the associated data into the Bayesian framework to learn the distribution of possible relationships between dose and titre. Once this is done, then for a new experiment we can run the simulation with just that data, and we can calculate a distribution for the vaccine potency using the information we have already extracted about the relationship between protection and the assay. This is then used to pass or fail the vaccine, using a threshold set so that the characteristics of the in vitro test are no worse than the best challenge test. By this means, we can assure that there is no diminution in our confidence in the efficacy of vaccines as a result of the switch. The main difference from what we have presented for challenge experiments is the added complexity of modelling the connection between the assays and challenge results. Note that data used to determine this relationship may need to be partitioned appropriately (in FMD, for instance, the relationship is serotype-specific [2, 20]). 4. DISCUSSION The European Pharmacopoeia states that “alternative methods of analysis may be used for control purposes provided that the methods used enable an unequivocal decision to be made as to whether compliance with the standards of the monographs would be achieved if the official methods were used”[8]. The Committee for Veterinary Medicinal Products in the European Medicines Agency has produced guidance on the interpretation of this: for batch potency tests it states that “it is important ... that the accuracy and precision is determined if an acceptable range of observed values for the results ... is to be defined”[5]; however, for designers of new tests it merely states “they should validate their potency test setting the pass level with reference to a representative batch shown efficacious”[6]. The first of these is too strict as the official method will in general offer no certainty, and the last is too lax offering no method of appraising whether any standard has been reached. We choose instead the middle ground. We describe a process for determining the best vaccine potency test in a given situation. We construct a careful specification of what we want to achieve and what assumptions we are willing to make. We then use that to produce a metric against which different tests can be measured. In particular we calculate the specificity, sensitivity and precision of any given test. At the end, we have both a decision as to which test to use, and an assessment of its quality. The process is nontrivial, but appears to be necessary if we wish to know how good current tests are. As well as minimising the risk of accepting substandard vaccines, knowing the characteristics of the chosen test allows us to specify them as the standard for new tests as they are developed (for instance, in vitro tests).

175

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The second issue we have addressed is our definition of a good vaccine. We have separated our threshold for accepting a vaccine for our definition of good one. We have shown that they can be easily confused and that this may have led to confusion in assessing the quality of potency tests in the past. We propose taking existing challenge data, calculating the best available test for a given definition of a good vaccine, characterising this test, and then we then propose using this as the standard for replacement in vitro tests. 5. ACKNOWLEDGEMENTS RR is grateful to BBSRC (grant BBE0103261) for financial support. 6. REFERENCES [1] C. Andrieu, N. de Freitas, A. Doucet, and M. I. Jordan, "An Introduction to MCMC for Machine Learning," Machine Learning, vol. 50, pp. 5-43, Jan 1 2003. [2] P. V. Barnett, R. J. Statham, W. Vosloo, and D. T. Haydon, "Foot-and-mouth disease vaccine potency testing: determination and statistical validation of a model using a serological approach," Vaccine, vol. 21, pp. 3240-8, Jul 4 2003. [3] J. Berkson, "Application of the logistic function to bio-assay," Journal of the American Statistical Association, vol. 39, pp. 357-365, Sep 1944. [4] C. I. Bliss, "The method of probits - A correction," Science, vol. 79, pp. 409-410, MAY 4 1934. [5] Committee for Veterinary Medicinal Products, "Position paper on batch potency testing of immunological veterinary medicinal products," European Medicines Agency, 1998. [6] Committee for Veterinary Medicinal Products, "Position paper on compliance of veterinary products with veterinary vaccine monographs of the European Pharmacopoeia," European Medicines Agency, 1999. [7] P. Congdon, Bayesian statistical modelling. Chichester; New York: Wiley, 2001. [8] Council of Europe, "European Pharmacopoeia," 6th ed Strasbourg: Council of Europe, 2008. [9] Council of Europe, "Foot-and-mouth disease (ruminants) vaccine (inactivated)," in European Pharmacopoeia, 6th ed Strasbourg: Council of Europe, 2008, pp. pp.918-920. [10] Council of Europe, "Statistical analysis of results of biological assays and tests," in European Pharmacopoeia, 6th ed Strasbourg: Council of Europe, 2008, pp. pp.571-600. [11] T. R. Doel, "Potency assessment of inactivated viral vaccines," in Vaccine manual : the production and quality control of veterinary vaccines for use in developing countries, N. Mowat and M. M. Rweyemamu, Eds. Rome: Food and Agriculture Organization of the United Nations, 1997, pp. 395-409. [12] D. J. Finney, Probit analysis, 3rd ed. London: Cambridge University Press, 1971. [13] N. Goris, P. Merkelbach-Peters, V. I. Diev, D. Verloo, V. M. Zakharov, H. P. Kraft, and K. De Clercq, "European Pharmacopoeia foot-and-mouth disease vaccine potency testing in cattle: Between test variability and its consequences," Vaccine, vol. 25, pp. 3373-9, Jan 13 2007. [14] N. Goris, E. Maradei, R. D'Aloia, N. Fondevila, N. Mattion, A. Perez, E. Smitsaart, H. J. Nauwynck, J. La Torre, E. L. Palma, and K. De Clercq, "Foot-and-mouth disease vaccine potency testing in cattle using homologous and heterologous challenge strains: Precision of the "Protection against Podal Generalisation" test," Vaccine, vol. 26, pp. 3432-7, Jun 25 2008. [15] N. Goris, T. Willems, V. I. Diev, P. Merkelbach-Peters, T. Vanbinst, Y. Van der Stede, H. P. Kraft, V. M. Zakharov, V. V. Borisov, H. J. Nauwynck, B. Haas, and K. De Clercq, "Indirect foot-and-mouth disease vaccine potency testing based on a serological alternative," Vaccine, vol. 26, pp. 3870-9, Jul 23 2008. [16] D. W. Hosmer and S. Lemeshow, Applied logistic regression, 2nd ed. New York; Chichester: John Wiley & Sons, 2000. [17] G. Kärber, "Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche," NaunynSchmiedebergs Archiv Fur Experimentelle Pathologie Und Pharmakologie, vol. 162, pp. 480-483, 1931. [18] OIE, "List A Diseases," in Manual of Diagnostic Tests and Vaccines for Terrestrial Animals 2004, 5th ed Paris: Office International des Epizooties, 2004. [19] T. W. Pay and P. J. Hingley, "A potency test method for foot-and-mouth disease vaccine based on the serum neutralizing antibody response produced in cattle," Vaccine, vol. 10, pp. 70713, 1992. [20] T. W. Pay and P. J. Hingley, "Foot-and-mouth disease vaccine potency tests in cattle: the interrelationship of antigen dose, serum neutralizing antibody response and protection from challenge," Vaccine, vol. 10, pp. 699-706, 1992. [21] M. Pizzi, "Sampling variation of the fifty percent end-point, determined by the Reed-Muench (Behrens) method," Hum Biol, vol. 22, pp. 151-90, Sep 1950.

176

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[22] L. J. Reed and H. Muench, "A simple method of estimating fifty per cent endpoints," American Journal of Hygiene, vol. 27, pp. 493-497, May 1938. [23] C. Spearman, "The method of 'right and wrong cases' ('constant stimuli') without Gauss's formulae," Brit J Psych, vol. 2, pp. 227-242, Jan 1908. [24] P. Sutmöller, "Computer simulation of foot-and-mouth disease vaccine potency tests," Prev Vet Med, vol. 4, pp. 329-339, Dec 1986. [25] J. E. Tanner and A. P. Morgan, "Design and analysis of veterinary vaccine efficacy trials," Veterinary microbiology, vol. 37, pp. 221-30, Nov 1993. [26] Y. L. Vianna Filho, V. Astudillo, I. Gomes, G. Fernandez, C. E. Rozas, J. A. Ravison, and A. Alonso, "Potency control of foot-and-mouth disease vaccine in cattle. Comparison of the 50% protective dose and the protection against generalization," Vaccine, vol. 11, pp. 1424-8, Nov 1993. [27] M. H. Zweig and G. Campbell, "Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine," Clin Chem, vol. 39, pp. 561-77, Apr 1993.

177

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

178

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 27

CORRELATING VACCINE INDUCED PROTECTION WITH HUMORAL AND CELLULAR IMMUNE RESPONSES TO FMDV: POTENTIAL IN VITRO ASSAYS FOR REPLACEMENT OF POTENCY TEST Y. Oh1, P. Hamblin1, B. Statham1, B. Charleston1, D. J. Paton1, J.H. Park2, Y.S. Joo2, S. Parida1* 1

Pirbright Laboratory, Institute for Animal Health, Ash Road, Surrey, GU24 0NF, United Kingdom. 2 FMD Research Laboratory, FADRD, NVRQS, Republic of Korea.

INTRODUCTION It is well known that neutralising antibody titres are important in protecting against Foot-andmouth Disease (FMD) infection. However, it has often been shown the humoral antibody titre is not always fully predictive of vaccine-induced protection against FMD. Therefore, this study has looked for a correlation between cell mediated immune responses, humoral immune responses and postvaccination protection against FMDV infection. MATERIALS AND METHODS Samples were collected from 5 vaccine challenge experiments conducted at Pirbright, United Kingdom. Blood samples from FMDV vaccinated, non-vaccinated and vaccinated-and-challenged cattle were re-stimulated overnight with inactivated FMDV antigen and the level of induced IFN-γ was measured. Humoral antibody levels were measured by virus neutralisation test. RESULTS A positive correlation was found between humoral antibody response, IFN-γ response and protection against the clinical disease. CONCLUSION T cell stimulation assays such as the whole blood IFN-γ assay along with VNT are potential candidates for vaccine evaluation and could reduce the need for in vivo challenge in the future.

179

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

180

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 28 DEVELOPMENT OF A FOOT-AND-MOUTH DISEASE VACCINE POTENCY TEST WITHOUT CONDUCTING ANIMAL CHALLENGE EXPERIMENT M. Alkan1, S. Gurcan2, M. F. Sarac1, Y. Gultekin1, A. Arslan1, E. Uzunlu1, S. Akyuz1 and G. Aynagoz1* 2

1 FMD Institute, P.O. Box 714, 06044 Ulus, Ankara, Turkey Ankara University Faculty of Veterinary Medicine, Department of Biostatistics, Ankara, Turkey

ABSTRACT Introduction Vaccine manufacturers evaluate the efficiency of their vaccine according to the method which is defined in the European Pharmacopoeia. Regarding foot-and-mouth disease (FMD) vaccine there are some difficulties to find animals for potency tests in the countries like Turkey where FMD is endemic and vaccine campaigns are carried out. In addition, potency tests must be carried out in containments having high biosecurity levels. There are many publications indicating a correlation between protection from virus challenge and neutralizing antibody response. However, up to now, none of the suggested method has been found valid. Materials and methods An analysis was made of data from potency tests on four batches O1 Manisa and two batches of Asia-1 Tur 73 monovalent oil adjuvanted foot-and-mouth disease vaccine. Regression were calculated for the relation of protection from virus challenge versus antigenic load (Log 146S) and neutralizing antibody response (Log SN50), versus only Log 146S, versus only Log SN50. Results For the relation of protection from virus challenge versus Log 146S and Log SN50, R square was determined as 0,809 for O1 Manisa vaccines, 0,866 for Asia-1 Tur 73 vaccines. In addition, the amount of required antigen for % 50 protection in O1 Manisa and Asia-1 Tur 73 vaccines was found 1,15 µg and 0,75 µg respectively. Conclusion We have shown that Log SN50 and Log 146S can give good correlation with protection from virus challenge and can be used as an indirect indicator of potency. 1. INTRODUCTION Foot-and-mouth disease (FMD) is highly contagious and economically important viral disease of cloven hoofed animals due to international trade restrictions and loss of productivity (OIE, 2004). European manufacturers have to perform a number of tests including potency to evaluate vaccine quality. Potency of FMD vaccines in Europe have been assessed by the method described in European Pharmacopoeia (EP) Monograph 04/2005:0063 (Ph. Eur., 2006). Non-vaccinated and non-infected 17 cattle which were free from FMD antibodies and at least 6 months old are used for potency control of FMD vaccines (Ph. Eur., 2006). It is difficult and costly to find these animals in countries like Turkey where the disease is endemic and routine mass vaccination is operational. Although cattle challenge method is accepted as a gold standard for potency evaluation of FMD vaccines, it has some disadvantages such as high cost, difficulties of finding suitable animals, requirements for high security laboratories and people’s growing awareness on animal welfare. Therefore many researchers are trying to develop an alternative method. However no method has been accepted complete until now.

181

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The well established test for assessing antibodies to FMD vaccines is virus neutralization test (VNT). A good correlation was reported between VNT and protection (Ahl et al., 1990, Pay and Hingley, 1992b). Another alternative test to estimate a potency of vaccine could be antigen payload of the vaccine measured by 146S density gradient procedure. The intact 146S particle is essential for efficacy of a vaccine and is one of the most important parameter for calculating antigen payload within a vaccine formulation. The test’s precision and sensitivity is high and very straightforward for operators. Finally, some elements of the procedure can easily be standardised as demonstrated by the author and colleagues with an international joint project involving all of the main FMD laboratories at that time. Vaccine manufacturers routinely use it for vaccine formulation. Many manufacturers have been already set a correlation between 146S concentration and protection (Doel, 2003). In this study it was investigated whether there was any statistical relationship between protection from virus challenge, antigenic load and neutralizing antibody response, and whether it was significant enough to omit animal challenge experiment. 2. MATERIALS AND METHODS Vaccine FMDV antigen was produced in BHK21 cell culture, clarified and inactivated using binary etyleneimine. Inactivated antigen concentrated by PEG 6000. Then the antigen formulated with Montanide ISA 206 (Seppic/France) 1:1 to obtain ready to use DOE vaccine. Sterility and safety tests were performed to ensure suitability for using in the field. In this study, four series of O1 Manisa and two series of Asia-1 Tur 73 monovalent oil adjuvanted FMD vaccine with different concentrations of 146S were used (Figure 1). Cattle Holstein-Friesian, 8-12 months old, free from FMD antibodies, non-vaccinated cattle were provided from a state farm under strict veterinary control. Potency Test Cattle were divided into 3 groups, each consist of five animals for potency testing of each series. Full dose (2 ml) administered to five animals by intramuscular route. Five animals received 1/4 dose (0, 5 ml) and other five animals 1/16 dose (0,125 ml) of vaccine. 2 control animals remained non-vaccinated. Animals were challenged by two separate intradermolingual inoculation with 104 bovine infective dose 50 of fmdv on day 28th post vaccination. They were monitored for 8 days after challenge with daily records of rectal temperature and clinical examinations for specific lesions. 146S Antigen assay The concentration of 146S of each series was measured by method described by Doel et al. (1981). Virus neutralizing antibody assay The cattle were bled at 28. Days after vaccination for serum neutralising antibody assay. Virus neutralisation test were performed against O1 Manisa and Asia-1 Tur 73 homologue viruses by using BHK 21 cell culture as described in World Organisation for Animal Health (OIE) Manuel of Diagnostic Tests and Vaccines for Terrestrial Animals (OIE, 2004). Statistical analysis Linear regression models were used to predict protection by 146S and SN50. Pearson correlation analysis was performed to relationship between 146S and SN50.

3. RESULTS Neutralising antibody titers and the results of potency tests against four series of O1 Manisa and two series of Asia-1 Tur 73 at the day 28 after vaccination were shown on Figure 1. Relationship between protection and Log 146S

182

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The relationship between protection and Log 146S value was shown at Figure 2 and 3. Both type O1 Manisa and Asia-1 Tur 73 regressions was statistically significant (p<0, 01 for both types). However R square was determined as 0,866 for Asia-1 Tur 73 whereas type O1 Manisa was 0,613. It was concluded that to achieve a 50% protection level was needed 1, 15 µg O1 Manisa and 0,75 µg Asia-1 Tur 73 146S antigen Interrelationship between protection, 146S and Log SN50 Interrelationship between protection and two variables Log 146 S and Log SN50 were analysed. R square was calculated 0,809 for type O1 Manisa and 0,866 for Asia-1 Tur 73 according to analysis. This has shown that there was a strong correlation between protection, Log 146S and Log SN50 (O1 Manisa p<0, 01, Asia-1 Tur 73 p<0, 05). From the datas in Fig. 1, the protection rate can be determined according to the following formula for O1 Manisa Protection = - 0,077 + 0,177 Log146S + 0,487

and

for Asia-1 Tur 73 Protection = 0,521 + 0,554 Log146S + 0,036

i =n

∑ LogSN 50 where

i =1

in which 146S is defined as measured antigen concentration of each batch vaccine by the density gradient procedure and x is defined as arithmetic mean of neutralizing antibody response of animals which vaccinated with each batch vaccine. As a result, formula 1 means 100% protection and 0 means 0% protection. 4. DISCUSSION FMD vaccine potency testing is a highly important, but expensive matter. Due to its high variability and low repeatability it only yields an approximate estimate of the PD50 vaccine content. A number of researchers have tried to develop alternative approaches which use both animal models and in vitro methods (Barnett, 2003a) Furthermore according to Pay and Hingley (1987) a correlation could be found between antigen load in vaccine (146S) and protection. In this study we design a strategy seeking a correlation among SN50, antigen payload and protection in order to omit animal challenge experiment. One promising alternative of potency test is VNT. According to homologous challenge study of Ahl et al.(1990) the BHK21-CT cell used titres was low compared to the IBRS2 cells used, but correlated well with protection. Indeed, only 2 of the 85 were protected had low challenge virus specific neutralizing antibodies in this study. Many factors can effect of VNT titres of a serum: the cell substrate; the maturity of cells, medium and pH, the virus dose; the antigenic relationship of the assay virus to the vaccine virus; whether serum dilutions are encountered before or after with the virus inoculum; etc. Therefore, different laboratories may have different antibody pass-levels or log10 serum titers. (Barnett, 2003b). Pay and Hingley (1992a) announced that there is a big difference existed in the correlation of antibody to protection between laboratories, particularly in the case of the O serotype. Hence every laboratory should set their own correlation if VNT was thought as the alternative method. Moreover 146S density gradient procedure is extremely susceptible, reliable, reproducible and straightforward test. By setting the correlation, Log 146S only, or combine with SN50 value, may assess the potency indirectly. Nevertheless as the VP1 polypeptide composes only ≈20% of total 146S virion and as the neutralizing epitopes most likely compose less than 10% of total VP1 polypeptide, it can be deduced that a few hundred picograms of epitope of some strains in vaccine can be anticipated to protect 50% of cattle (Pay and Hingley, 1987). Therefore, same amount of 146S can not be given same percentage of protection because of stabilty of 146S or portion of VP1 in 146S concentration.

183

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Numerous authors and manufacturers accepted that more antigen is needed for protection against type O than other types (Doel, 2003, Pay and Hingley, 1987) Same results were obtained in this study, for 50% protection, 1,15 µg and 0,75 µg of antigen of O1 Manisa and Asia-1 Tur 73 type were needed respectively (Figure 2 and 3). A few previous publications which have given estimates for the 140S concentrations which will supply 50% protection of cattle were produced that by Pay and Hingley (1987) who reported an estimate 220 ng for type O1 BFS strain vaccine, and that Mowat (1972) gave estimate 6 ng for a type Asia-1 Israel 3/63 strain vaccine. These differences could come from strain variations or different process of manufacturers used in production. Manuel of Diagnostic Tests and Vaccines for Terrestrial Animals points out that the expected percentage of protection in indirect tests should be equal to or greater than 75% when 16 animals are used or 70% when 30 animals are used in the experimental group (OIE, 2004). In this study 3, 55 µg and 2, 03 µg of 146S of O1 Manisa and Asia-1 Tur 73 strain were needed respectively for 75% protection. If Log 146S and Log SN50 are used together to predict the potency, expected percentage of protection can be used as 70% when 30 animals are used. We strongly recommend using 146S assay only or some serological tests which together increase the reliance on estimating potency of specific vaccine batch. They can also be assembled with an interconnected method which is based on the 146S concentration of the final vaccine batch. 5. CONCLUSIONS Log SN50 and Log 146S can give good correlation with protection from virus challenge and can be used as an indirect indicator of potency. To release FMD vaccine batch by an alternative test of European Pharmacopoeia vaccine PD50 test is recommended. Only 146S test or combined with VNT as an alternative of potency is enough to omit animal challenge experiment. 6. RECOMMENDATION Researches on other types are needed. An in-vitro method should be developed and standardized for monitoring and measuring the antigenic integrity of 146S. 7. ACKNOWLEDGEMENTS The study was funded by the FMD_ImproCon project under the Sixth Framework Programme (Grant SSPE-CT-2003-503603) and by the General Directorate of Agricultural Research of Turkish Ministry of Agriculture and Rural Affairs 8. REFERENCES [1] Ahl, R., Haas, B., Lorenz, R.J. & Wittmann, G. 1990. Alternative potency test of FMD vaccines and results of comparative antibody assays in different cell systems and ELISA. In: Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-mouth Disease, FAO, Rome, Lindholm, Denmark, 25-29 June, pp. 51-60. [2] Barnett, P. 2003a. Potency testing; alternatives to the challenge test: serological assays and small animal models. In: Proceedings of International Symposium organised by the European Directorate for the Quality of Medicines. Strasbourg, France, EDQM, March 17-18. pp. 33-40 [3] Barnett, P.V., Statham, R.J., Vosloo, W. & Haydon, D.T. 2003b. Foot-and-mouth disease vaccine potency testing: determination and statistical validation of a model using a serological approach. Vaccine 21:3240-3248. [4] Doel, T.R., Fletton, B. & Staple, R.F. 1981. Further developments in the quantification of small RNA viruses by UV photometry of sucrose density gradients. Dev. Biol. Stand. 50:209-19. [5] Doel, T. 2003. Potency testing of foot-and-mouth disease vaccines: an industrial perspective. In: Proceedings of International Symposium organised by the European Directorate for the Quality of Medicines. Strasbourg, France, EDQM, March 17-18, pp. 41-47. [6] European Pharmacopoeia. 2006. Foot-and-mouth disease (ruminants) vaccine (inactivated) 04/2005:0063, 5th Edition, version 5.5. [7] Mowat, G.N. 1972.Quantities of purified antigen required to immunise swine against foot-andmouth disease. Bull. Off. Int. Epizot. 82:1151

184

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[8] OIE (Office International des Epizooties/World Organisation for Animal Health). 2004. Foot-and-mouth disease. In: OIE Standards Commission, editor. Manual of diagnostic tests and vaccines for terrestrial animals. 5th ed., Paris, France: Office International des Epizooties [chapter 2.1.1]. [9] Pay, T.W.F. & Hingley, P.J. 1987. Correlation of 140S antigen dose with the serum neutralizing antibody response and the level of protection induced in cattle by foot-and-mouth disease vaccines. Vaccine, 5, 60-64. [10] Pay, T.W.F. & Hingley, P.J. 1992a. Foot-and-mouth disease vaccine potency tests in cattle: the interrelationship of antigen dose, serum neutralizing antibody response and protection from challenge. Vaccine, 10:669-706. [11] Pay, T.W.F. & Hingley, P.J. 1992b. A potency test method for foot-and-mouth disease vaccine based on the serum neutralizing antibody response produced in cattle. Vaccine, 10:707713 Figure 1: Potency test data for O1 Manisa (A) and Asia-1 Tur 73 (B) (A)

Potency test data for O1 Manisa 3

Payload

LogSN50 (Protected)

2,5

12 10 8

LogSN50(Unprotected)

1,5

Log SN50

Payload (146Sug)

4 0,5

2 0

0 51 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1

Animal no

LogSN50(Unprotected)

(B)

Potency test data for Asia-1 Tur 73 6

Payload

2,5

1,5

0,5

Log SN50

Payload (146Sug)

LogSN50 (Protected)

30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Animal no

Figure 2: Relation of protection from virus challenge versus Log146S for O1 Manisa

185

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

protection = 0.469+0.510*Log 146S 1,2

1,0

0,8

0,6

protection

0,4

0,2

0,0

-0,2 0

146S (µg )

Figure 3: Relation of protection from virus challenge versus Log146S for Asia-1 Tur 73 protection = 0,572+0,575*Log 146S 1,1 1,0 0,9 0,8

protection

0,7 0,6 0,5 0,4 0,3 0,2 0,1 0

3 146S (µg )

186

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 29 KEYNOTE BIOSECURITY: WHAT’S IN IT FOR ME? N. Honhold1*, P. Ankers1, A. McLeod1 1

AGA, FAO, Rome.

Biosecurity is an essential tool in disease control but its application depends upon the private activities of large numbers of individuals and organisations within a framework of public regulations. They will only adopt biosecurity measures if they believe that there is significant risk to be avoided and that the result will benefit them. Ideally biosecurity measures will be integrated into management practice to improve productivity and profit. Important factors affecting motivation include perceptions of risk, the different needs in endemic and epidemic situations, perceived and real economic benefits, practicality of applying different measures, and the availability of options other than biosecurity. For biosecurity to reach its potential as a tool of real value in FMD control across the wide range of countries that experience problems with the disease, a balance needs to be struck that accommodates the needs of different stakeholders who are all behaving rationally from their own point of view. This paper discusses the roles and motivations of different stakeholders in the value chain with regard to biosecurity, using examples taken from different production systems and disease situations. It also discusses the contrast between government desire for mass action and the interest of individuals. Governments sometimes seek to use regulations to improve biosecurity, but may not fully appreciate the potential damage to the livelihoods of small scale producers. The paper calls for all of these issues to be taken into account in developing and facilitating balanced and sustainable biosecurity initiatives that match the disease situation. Biosecurity measures should, wherever possible, create returns that are greater than their costs. It suggests some possible approaches to improving buy-in, identifies gaps in knowledge of the impact of biosecurity measures, and calls for more field-based, multi-sectoral and participatory research to address these deficiencies.

187

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

188

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 30 FOOT-AND-MOUTH DISEASE CONTROL IN ENDEMIC SETTINGS: COMBINING EPIDEMIOLOGY, RISK ASSESSMENT AND VALUE CHAINS ANALYSIS TO IDENTIFY RISK CONTROL POINTS J.o Pinto1, G. Ferrari1; N. Taylor2, Jn Rushton1, K. Sumption1 and J. Lubroth1 1

FAO Animal Production and Health Division, Food and Agriculture Organization of the United Nations, Rome, Italy 2 Veterinary Epidemiology and Economic Research Unit (VEERU), School of Agriculture, Policy and Development, The University of Reading, United Kingdom.

ABSTRACT Foot-and-mouth disease control in endemic settings poses a number of challenges, particularly the extreme contagiousness of the infection, host range, the diversity of production sectors at risk, and the optimum use of the limited financial and human resources available to veterinary services. Targeting of these resources and regulatory controls to interrupt or reduce transmission at critical points in FMD transmission cycles could achieve significant advantages over programs that do not discriminate between risk groups, as occurs in most national mass vaccination programs which developing countries could hardly sustain. Traditional FMD surveillance is focused on the density and distribution of livestock population without taking into account the level of risk of different ecosystems, production systems and market chains. Risk based disease surveillance using analysis of livestock markets, livestock and animal product circuits must be a guiding element in the analysis of FMD risks and decision-making for control interventions addressing those risks. This paper will provide examples from FAO programs of how these critical control points can be identified, and will outline ideas for expanding the use of the methods in endemic countries as part of national FMD control strategy development. In particular, it will: 1) Illustrate how a risk factor/ production and marketing chain approach can be used to develop consensus on risk and critical control points within a livestock marketing (value) chain, using examples on ongoing FAO activities in Ecuador, Venezuela (GTFS/RLA/172/ITA) and Vietnam; 2) Illustrate how a combination of outbreak investigation, marketing chain analysis, and serological surveys at markets and slaughterhouses has been used to identify critical control-points in FMD control in Turkey; 3) Illustrate how a serological (NSP-antibody) survey approach can be used to identify how FMD exposure is related to production/marketing chains, species and age, using examples from the multi-country FMD surveillance program in five Central Asian countries (GTFS/INT/907/ITA); 4) Outline how the use of the methods could be expanded to assist other endemic countries to identify strategies for reducing FMD risk. 2. CASE STUDY NO1 Foot-and-mouth disease risk assessment in endemic settings of Viet Nam, Ecuador and Venezuela using value chain analysis 2.1. Background Often, the focus of disease control in response to outbreaks or prevention in response to perceived threat is on spatial or geographical, so-called ‘local’ means of disease spread, plus sometimes spread through air currents. This leads to such disease control and prevention measures as quarantine areas (‘protection zones’), local area culling, ring vaccination and buffer-zone vaccination. However, these measures fail to take into account the fact that animal disease is often spread through the actions of people involved in the production and marketing of animals and their products. Thus, infected animals can be rapidly trucked across vaccinated buffer zones, traders carrying virus on their vehicles, can move from farms within ring vaccinated zones to farms outside and pigs can be infected with swine fever as a result of feeding waste feed (swill) with origins outside a country. In this way, disease can be seen to spread by direct animal to animal contact or indirect contact (by carriage of infectious virus on people, vehicles, equipment etc.) over variable

189

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

distances and along routes that are determined by patterns of production and trade, rather than disease spreading between contiguous animal populations as a result of proximity (local or contiguous spread). A better understanding of the dynamics of livestock production and marketing chains in FMD endemic settings could improve the understanding of transference of risks within chains. Value chain analysis in combination with risk assessment could be used as a tool to identify critical risk points within the different livestock chains and therefore risk based control interventions used to tackle disease risks. By definition a value chain ‘describes the full range of activities which are required to bring a product or service from conception, through the different phases of production (involving a combination of physical transformation and the input of various producer services), delivery to final consumer, and final disposal after use’ (Kaplinski et al., NK). Value chains can be described pictorially, orally and in some cases quantitatively. Such structures provide a good starting point and can create good discussions particularly to identify hazards and risk issues. The first result of value chain studies is often one or several diagrammatic value chain ‘maps’ consisting of boxes representing different actors and/or production/marketing sites in the chain and lines or arrows indicating flows of poultry and products between these boxes. 2.2 Methodology This project has been divided in three main phases:

The first phase started with an initial desk based data collection of disease information and characteristics of production systems in each country and ended with one workshop in each of the countries selected. Based on the data gathered, the second phase concentrated on the development of detailed value chains for different production systems and products in each country. The third and last phase is the identification of risk control points and their assessment.

For every workshop participants represented main actors of livestock production systems and marketing chain in livestock production, for example producers, representatives from slaughterhouses and dairy plants, private veterinarians, representatives of national veterinary services, traders and animal health services and representatives of wholesale, retail and trade. Before the workshop a questionnaire with relevant questions was addressed to the participants to in order to give a guideline on the information that needed to be collected during the workshop. The risk assessment approach is based on gaining detailed understanding of the ‘value chains’ associated with production, trade, marketing, and consumption of meat and other products of the relevant species (pigs, bovines, small ruminants). The following information relevant to FMD risk assessment was collected during discussions with stakeholders:

Overview of the FMD epidemiological situation; Overview of local farming systems; Data on production systems for susceptible species such as bovines, pigs and small ruminants; what are current and future trends? A detailed description of the different value chains identified by participants that varies between countries.

Using three of the four basic principles of risk assessment described in the OIE Code (release, exposure and consequences; the fourth being ‘risk estimation’’), the risk factors which are likely to affect the probability and amount of FMD occurring were identified throughout the description of value or market chains derived from the workshops. Risk factors were subdivided into those which have their main effects on FMD introduction, exposure of local livestock or related to FMD spread. A qualitative assessment of the level of risk was presented and each risk factor identified. 2.3. Preliminary results of the study An initial draft value chain was mapped and described by participants for each of the three workshops. Extensive production systems, beef chains, milk production chains, pig, sheep and goat production and marketing chains were discussed and described by the participants in each of the country of study.

190

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Value chains are not uniform and static; for instance in Viet Nam, three different chains were described for pig production and there was a marked seasonality of pork supply identified, in particular from smallholder production systems which contribute to almost 80% of the pork production. In Venezuela, for instance, large scale cattle production can be divided into four different production systems including extensive commercial cattle production, commercial milk production, beef fattening and dual purpose production (Figure 1). Figure 1: Dairy value chain in Venezuela INFORMAL ANIMAL IMPORT AND EXPORT

SPECIALIZED PRODUCTION Holstein-Friesian; mainly in Merida and Portuguesa (10%)

DUAL PURPOSE PRODUCTION Zebu and Crossbreeds; mainly in Zulia, South of Lake Maracaibo

RAW MILK Milk collection (77%)

Cheese production on farm (23%) Wholesaler

PROCESSING PLANT CONSUMER

Retailer

Replacement animals, Discards, Male Calves WHEY Fed to backyard systems

TRADER PIGS in production

Milk from purpose

dual beef

During the workshop in Venezuela, two market chains were discussed by the stakeholders in the workshop: -Dairy value chain and -Beef cattle chain. The second step is to complete the value chains and information is to include additional information on types of products, prices or volume of transactions, or added margins of profit along the chains. From the preliminary identification of risk control points by the stakeholders in the different value chains described during the workshops, most of the risk for FMD spread in all three endemic

191

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

countries studied comes from movement of live animals, contamination from poorly regulated slaughter facilities, swill feeding, feeding of animals with milk untreated by heat, use of manure or bedding as fertilizers, spread by fomites carried by traders and transporters of live animals and to a lesser extent fomites carried by other farm service providers. Also communal grazing and sharing of breeding males in cattle or pig productions also contribute to risk of local spread. In the closing phase of this project, the value chains identified will be revisited in more detail to understand how these risks can be mitigated in a stakeholder-friendly way. Moreover, some risk mitigation measures will be proposed and evaluated with stakeholders in more detail with reference to the country-specific analyses already carried out (and supplementary information where needed) and assessed with respect to their potential impact on stakeholders within the value chains and costs associated for the final phase of this study. 3. CASE STUDY NO2 Risk assessment of FMD in Central Asian countries 3.1 Background FMD is known to be present in some of the Central Asian countries of the FAO GTFS/INT/907/ITA regional project, which encompasses Afghanistan, Pakistan, Tajikistan, Turkmenistan and Uzbekistan. The common strategy usually adopted in order to control the spreading of the disease is through vaccination targeting specific areas. Mass vaccination programs could hardly be carried out both for financial and logistics constraints and the issue of identifying specific target populations to be vaccinated appear rather important in order to optimize the usually scarce resources available. In this paper we assume that the overall risk of becoming infected with FMD virus is not evenly distributed among the different livestock productive systems that may be present in a given country and that, subsequently, it is possible to qualify and quantify such risks so that any control program could be targeted initially on the higher risk groups and thus make the biggest impact on the course to progressively reduce the load of FMD virus. The assumption is that geographical factors are not the only features that need to be taken into account in designing a disease control strategy for FMD. In this regard a collaborative study has been carried out with the Danish Technical University in the Landhi Dairy Colony located in Bin Qasim, in the suburbs of Karachi, Pakistan during 2006-2007 is of relevance. The Dairy Colony is a particular productive system and Landhi is considered the world largest buffalo colony. It was first established in 1958 in order to move existing buffalo populations out of the residential areas of Karachi which were causing problems for the disposal of dung. Initially the colony extended over an area of 752 acres but today it extends over an area of approximately 1,600 acres (approximately 6.7 km2) and with an animal population of more than 300,000 head, 95% of which are buffaloes. Currently, there are about 2,000 dairy farms in the colony. The dairy colony production system has some peculiar characteristics. The animals are kept on individual farms for milk production but only for the period of their lactation (230-300 days), so recently calved and near-calving animals are purchased and brought onto the colony farms. The calves are generally not kept with and only high-yielding milking animals are maintained in this production system. This system is dominated by the buffaloes as buffalo milk contains higher butterfat and is preferred by the customers. In order to fulfil the market demand the production level of the farm is maintained at almost constant level through an intense turn-over of animals. On average 10 to 12 per cent of animals in dairy colonies are replaced monthly. The animals are brought from livestock rich districts of Punjab and Sindh Provinces. The dairy farmers of the colonies directly purchase the animals from villages (20 to 30 %), livestock markets (55 to 65 %) and from ‘buffalo hotels’ or ‘piri’ (10 to 15 %) established in Karachi where livestock traders bring the animals for sale. There is no grazing in this production system and all animals are stall-fed using green fodder and high concentrates. Breeding activity varies from farm to farm but generally is minimal. The current trend of increase in prices of animals has resulted in enhancing the animal breeding activities at the dairy colonies (Afzal M, 2003). The study focused on the detection of FMD virus or viral RNA both from clinically and non-clinically affected animals and in the same study, a serological investigation was carried out with 180 serum samples collected between April 2006 to April 2007 from slaughtered animals that had spent at

192

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

least one lactation period into the Landhi Colony. Out of 180 samples collected 176 (98%) had detectable levels of NSP antibodies (Klein J, et al 2007, 2008). This finding was consistent with animals, sampled at any given point in time (slaughter) that had the common experience of having spent into the colony at least one lactation period (230-300 days) in Landhi. Presently no data are available on the level of NSP antibodies that may be present on the population entering for the first time the colony. Undoubtedly the level of exposure to FMDV seems to be relevant and it is assumed that transiting into the colony would increase the risk of exposure to FMD virus to a certain extent which should be reflected in the different level of antibodies between pre and post-exposure. The study carried out in the Dairy Colony has provided a good indication of how the level of exposure to FMD virus may be strictly linked with a specific productive system. It is in fact suggested (Klein J, et al 2008) that this peculiar system could be targeted for a vaccination program which in turn should contribute to reducing the overall load of the FMD virus. While for Pakistan, the Dairy Colony may be considered an important “hot spot” for FMD virus, there is the possibility that other production systems with equivalent or significant level of risk that should be identified and addressed in order to progressively reduce the overall load of FMD virus. For this purpose a surveillance program has been designed and in the present paper it is presented the serological component of such plan. 3.2 METHODOLOGY In an attempt to quantify how the risk of becoming infected with FMD virus may be distributed into different productive systems the following ones have been identified in the participating countries of GTFS/INT/907/ITA: (i) households animals; (ii) commercial sector - dairy; (iii) commercial sector - beef; (iv) genetic centers (usually government farms); (v) dairy colonies (only in Pakistan); (vi) animals seasonally moved to pastures; (vii) animals at slaughterhouses. The level of exposure to FMD virus will be measured through the detection of NSP antibodies and the target species for the survey will be cattle and buffaloes, which are considered equivalently susceptible to infection. The sample design and criteria adopted to conduct the study are a combination of what could be practically achieved while measuring the statistical robustness of the estimates The approach is the one of a cross-sectional study with a blood-sampling scheme where the criteria of eligibility for individuals to be included in the sample are category and age (Cannon R, et al 1982; Fleiss Jl, 1981). The epidemiological unit of concern may vary according to the particular category. This may be the village itself (in the case of household animals) or the farm for animals into commercial units such as dairy, beef etc.) or directly the individual animals in the case of a pre and post exposure assessment (transhumant/pastoral and dairy colony animals). The sample size required for each one of the combination age/categories identified above will not differ substantially from any survey that has to reach a certain level of statistical significance. In this particular study the proposed approach is, usually, through a multistage sampling scheme with the first stage identified with the epidemiological unit and the second stage with individuals into the unit distributed into three different age categories (0-1 yr; 1-2 yrs and more than 2 yrs). The general criteria established for estimating the sample size aimed at: (i) maintaining the standard error of the estimated proportion at an acceptable level (maximum tolerable error 11% at 95% confidence level); (ii) to be able to detect (at 95% confidence level) the presence of at least one NSP antibodies positive animal if the proportion (in each age category) of positive is equal or more than 20% (with no adjustment for test sensitivity and specificity); (iii) to be able to detect a prevalence risk ratio not exceeding 3.2 between two independent proportions with a type I and type II error of 0.05 and 0.1 respectively. In considering the criteria above, the outcome for the sample size estimation in different epidemiological units and categories is shown in table 1: Table 1: Minimum sample size in different categories in each country

Category

Househol d (with epi-unit being the village)

Dairy

Beef

Geneti c center s

Transhumant s/ Pastoralists

Slaughtere d

Colonies

193

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

N. units to be sampled N. Individual Samples

n.a.

2880

240

300 (two rounds of 150 each)

300

Max SE (95%)* Range of Min. detectabl e PRR**

3.2%

11%

8.2%

5.8%

300 (two rounds of 150 each) 8.2%

1.36 2.7

1.52 3.2

1.38 2.5

1.22 1.7

1.38 2.5

Note: * Maximum tolerable standard error at 95% confidence level ** Range of the Minimum Prevalence Risk Ratio (PRR) detectable with any given lowest group for prevalence ranging from 0.1 to 0.5 (Type I error 0.05; Type II error 0.1). Only for the categories “Transhumants” and “Colonies” the range for PRR has been estimated within group. Two additional issues are to be considered: (i) most of the large ruminants present are at the household level, thus this category is the most represented in the sampling scheme; (ii) commercial farms are rare to find in the GTFS/INT/907/ITA beneficiary countries. Live animal markets are not included into the present blood-sampling scheme. Reasons for this are that usually owners at the live markets are not keen to have their animals sampled. In order to target this system a concurrent activity which foresees clinical inspections and swabs/tissue sampling from both clinically (if any) and not-clinically affected animals will be implemented. 4. BIBLIOGRAPHY [1] Afzal, M. Consultancy report: Plan of action for the development of dairy colonies around Karachi. Food and Agriculture Organization (FAO) of the United Nations, Pakistan office, Islamabad. (2003) [2] Albu, Mike and Griffith, Alison. 2005. Mapping the Market: A framework for rural enterprise development policy and practice. Bourton on Dunsmore: Practical Action, 2005. [3] Cannon, RM, Roe, RT, Livestock Disease Surveys – A Field Manual for Veterinarians. Canberra. (1982) [4] DEFRA. DEFRA. Riskplan. [Online] http://www.defra.gov.uk/animalh/diseases/monitoring/pdf/riskplan.pdf. [5] Fleiss JL. Statistical methods for rates and proportions. 2nd edition, New York: Wiley, (1981). [6] Kaplinsky, Raphael and Morris, Mike. n.k. A Handbook for Value Chain Research. s.l.: IDRC, n.k. [7] Klein J, Hussain M, Ahmad M, Normann P, Afzal M, Alexandersen S. Genetic characterization of the recent FMD disease virus sub-type A/IRN/2005. Virol J 4:122-133 (2007). [8] Klein J, Hussain M, Ahmad M, Afzal M, Alexandersen S. Epidemiology of foot-and-mouth disease in Landhi Dairy Colony, Pakistan, the world largest buffalo colony. Virol J 5:53-68 (2008). [9] Scudamore, J.M. and Harris, D.M. 2002. Control of foot-and-mouth disease: lessons learned from the experience of the outbreak in Great Britain in 2001. 2002. pp. pp 699-710. Rev. sci. tech. Off. Int. Epiz 21 (3). [10] Taylor, N., et al. 2003. Examining the options for a livestock Disease-Free Zone in the Red River Delat of Viet Nam. 2003. VN/EC SVSV Project ALA/96/20.

194

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 31

BRIDGING THE DIVIDE BETWEEN TWO BONA FIDE GUARDIANS OF SOCIETY’S INTERESTS: THE DISEASE EXTERMINATORS AND THE SUSTAINABLE GROWTH ADVOCATES B. Perry*1, K. Rich2 1

University of Oxford, c/o P.O. Box 437, Gilgil 20116, Kenya. American University in Cairo, P.O. Box 74, New Cairo 11835, Egypt.

Clearly society must have, and indeed needs, the “we-must-control-FMD-at-all-cost” community in the world; that is how targeted disease-specific actions raise the probability of impact in their field. Society also needs the development and livelihoods perspective on FMD (often based on economic reasoning) that advises the world where (or where not) to invest and which highlights the tradeoffs between scarce resources and unlimited societal wants. Moreover, when it comes to “poverty reduction”, the war cry of national and international development agencies, the jury is still out on the ranking and merits of FMD control. Even within the broader concept of sustainable and inclusive growth (formerly called pro-poor growth), the world presents a patchwork of various regional and production system contexts in which FMD control may or may not be a priority. We would thus argue that the sustainable control and prioritization of FMD investment necessitates that the “disease exterminators” (possibly the bulk of attendees at this meeting) seek an enhanced understanding of the multidimensional implications of controlling, eradicating or reducing the impacts of FMD in different settings. We further suggest that much stronger mechanisms of communication and partnership are required between the disease exterminators and the development and livelihood economists if society is to optimise their invaluable contributions.

195

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

196

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 32

KEYNOTE: GLOBAL SURVEILLANCE FOR FMD – WHAT ARE WE DOING AND WHAT COULD BE DONE D. J. Paton*, N. P. Ferris, Y. Li, D. P. King, N. J. Knowles, J.M. Hammond and J. Bashiruddin FAO World Reference Laboratory for FMD, Institute for Animal Health, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, UK.

Foot-and-mouth disease virus (FMDV) exists as multiple serotypes and subtypes that are not randomly dispersed throughout the world but show a tendency towards association with particular ecological niches. Within regions, the distribution and prevalence of particular strains can be cyclical, and is affected by waning population immunity, viral evolution and antigenic escape, together with opportunities presented by movements of animals and their products. These factors can give rise to pandemic spread affecting new regions. Global surveillance for FMD aims to identify the current hazards and to predict heightened risk so that appropriate diagnostics and vaccines are available for their detection and control. This requires sustained effort directed towards the monitoring of FMD outbreaks and ideally also of FMDV circulation and persistence, along with collection and characterisation of FMD viruses and integration of findings with associated epidemiological intelligence. Such an extensive effort requires a team approach encompassing national and international disease control services and their laboratories along with commercial vaccine and diagnostic providers. The work of the FAO World Reference Laboratory for FMD and the OIE/FAO FMD Reference Laboratory Network is used as an example to illustrate some of what is being done and the difficulties and new opportunities associated with providing incentives, developing and sustaining capability, sharing information, exchanging materials, harmonising approaches and supporting complementary research in the field and in the laboratory.

197

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

198

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 33 KEYNOTE: GLOBAL SURVEILLANCE OF FOOT-AND-MOUTH DISEASE: CONCEPTS, CONSIDERATIONS, AND REQUIREMENTS A.Perez1, 2* and M. Thurmond1 1

FMD lab, Center for Animal Disease Modeling and Surveillance, University of California in Davis, One Shields Avenue - 1044 Haring Hall, Davis, CA 95616, USA 2 CONICET-Facultad de Ciencias Veterinarias UNR, Argentina.

ABSTRACT Introduction Prerequisite for global control and eradication of foot-and-mouth disease (FMD) is development of a global FMD surveillance program, the main mission of which would be to provide high quality, accurate, real-time FMD surveillance service on a global scale. The objective of this paper is to provoke discussions that hopefully will lead to formalizing steps leading to development of such a program. Materials and Methods A global FMD surveillance system should integrate international, regional, and country-specific efforts, all of which can be incorporated in an overarching global framework. In addition to high quality information and data, a functional real-time global surveillance system will require a) an information technology (IT) system with the ability to visualize and analyze data, and b) formal and efficient administration of the system in real time. Data should include, at a minimum, typical data available for incident cases, molecular changes in the virus, animal and human movement, international marketing and currency, and political and social changes, all of which could be surrogates for FMD risk. An IT system would map, graph, visualize, and analyze data. Analytical techniques would include regressive and auto-regressive models, geospatial analysis techniques, phylogenetic analysis, and time-space clustering techniques. Time-space Bayesian models could be applied to assess progress at a local or global scale. Results and Discussion The system should identify high risk areas that could be targeted to prevent disease spread, and to monitor evolving risk for FMD virus exposure. Ongoing analysis of risk factors would provide feedback for surveillance sampling by providing estimates of sample sizes and frequency, and targeted areas or time periods for sampling. A goal of such an integrated surveillance system is to have real-time quantitative estimates of changing and projected risks for FMD throughout the world. 1. INTRODUCTION Current FMD information systems are quite limited because they only provide access to passively collected data that can be considerably generally outdated, and because they do not necessarily provide information that can be used to prevent or control FMD spread. For example, questions that are not addressed by current information systems include: Where is FMD expected to be found today? Where will it be this time next year? What conditions are necessary for the disease to move into another country? Which emerging strains of virus might not be protected by current vaccine strains or might not be detected by current diagnostic assays? What increases the risk of a country acquiring the disease? Which countries or regions should be targeted actively for sampling and information collection? Which specific actions could be taken to control the disease and to minimize risk of FMD transmission and spread? During the last decade, there has been considerable discussion about the needs for global FMD surveillance. Activities aimed at developing, organizing, and establishing such a system, however, have not taken hold and fundamental decisions regarding the system organization and functionality have not yet been addressed on an international scale. For example, it will be necessary to 1)

199

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

identify how and by whom the system should be operated, maintained, and administered; 2) specify operational attributes required by the system; 3) identify which sources of information should be routinely collected; 4) identify which analytical tools to incorporate; and 5) determine how local decision making will be supported by, and incorporated into, the system. The objectives of this paper are to: a) Offer ideas, definitions, concepts, and considerations for a global FMD surveillance program, the main mission of which would be to provide high quality, accurate, real-time FMD surveillance service on a global scale; b) Encourage surveillance research, the results of which promotes and provokes dialogue required to move forward toward formalizing an efficient international surveillance effort; c) Propose the creation and operation of a global FMD surveillance system developed and operated through a neutral and independent international consortium; d) Discuss the architecture of a surveillance system, with emphasis on the description of analytical components that allow prediction and assessment of disease risk and virus evolution. 2. MATERIAL AND METHODS Disease surveillance has been defined as an active, ongoing, formal, and systematic process aimed at early detection of a specific disease or agent in a population or at early prediction of elevated risk of a population acquiring an infectious disease, with a pre-specified action that would follow detection of the disease, agent, or elevated risk [12]. An effective surveillance system necessarily must collect and process information rapidly, which minimizes the time between information capture and sharing the information with decision makers and stakeholders, who can act quickly to control or prevent the spread of disease [12]. A global FMD surveillance system should take advantage of the international, regional, and country-specific FMD surveillance or reporting systems that are currently in place to build the foundations for a global system. The global system should integrate formal and informal sources of information, which would be incorporated in an overarching global surveillance framework. A global surveillance system should function to identify ‘hotspots’ of risk, e.g., countries or regions that are at high risk for the disease or where strains that might not be protected by current vaccines or detected by diagnostic assays are likely to emerge. ‘Hotspots’ could be targeted to mitigate the risk for predicted outbreaks and disease spread, and monitored to assess evolving risk for FMD virus exposure. Ongoing analysis of ‘hotspot’ data could feed back into the surveillance system and provide estimates of disease frequency and trends. These estimates could be incorporated into the global system to update risk estimates at local and global levels. The ultimate goals of global surveillance system that incorporates local and regional programs would include to 1) have real-time quantitative estimates of changing and projected risks for FMD throughout the world; 2) obtain data upon which to base recommendations for where, when, and how to intervene in specific regions and how to evaluate the consequences of such interventions; and 3) use results of evaluations to improve and update projected risk estimates. Thus, the system would operate to improve its ‘intelligence’ through a process of re-evaluation, assessment of new information, and re-estimation that result in updated estimates of risk. The architecture of a functional real-time global surveillance will require: a) An information technology (IT) system that can collect, analyze, and communicate disparate sources of data in real, or near real, time; b) The ability to visualize and analyze data for the purpose of identifying, predicting and anticipating changes in disease risk; and c) A plan for formal administration of the system. 3. RESULTS AND DISCUSSION 3.1 Information technology system Current FMD reporting systems are not designed specifically to facilitate communication and data sharing among different laboratories, field veterinarians, farmers and decision makers. Surveillance

200

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

IT systems should be designed to permit this kind of communication while being able to selectively restrict confidential or sensitive information, as necessary. A real-time, web-based system, such as the FMD BioPortal (http://fmd.ucdavis.edu/bioportal/), is an example of an IT system that has been developed to retrieve and visualize publicly available data on FMD. The FMD BioPortal became operational in January 2006, and currently provides temporal and spatial visualization of FMD-related data and supports a system for capturing and disseminating FMD-related news items. An IT system like the FMD BioPortal, would link disparate sources of formal and informal data, such as ReLaIS, WAHID, ProMed, and GLEWS. It also would be able to visualize and analyze data, including estimates of spatial and temporal variation in risk, and other measures of risk. Data handled by the system would include typical incident data, as they are reported and collected now, and also data about molecular changes in the virus, animal and human movement, international marketing and currency changes, political and social changes that could be surrogates for FMD risk, and measures of disease incidence and virus circulation collected actively as part of local or regional surveillance efforts. 3.2 Techniques for data analysis A variety of analytical techniques should be included as part of an IT system in order to permit estimation of FMD risk, identification of ‘hotspots’, and prediction of virus evolution and spread. Analytical techniques should include, but not be limited to, regressive and auto-regressive models, geospatial analysis techniques, network analysis, risk analysis, time series analysis, phylogenetic analysis, and time-space clustering techniques. The space- and time-space scan statistics are among the most common analytical techniques for detection of clusters of infectious disease. The scan statistic has been used to detect areas and times of the year at high risk for FMD outbreaks in Argentina [6], Iran [8], Peru [2], and Mongolia [11].The Knox test k-Ripley’s test have been used to identify the association of time and space with clustering of FMD outbreaks in, respectively, Peru [6] and Argentina [7]. Time-space Bayesian models have been used to predict FMD risk at a global scale, using data on political, social, demographic, and climatic variables, and other surrogate information [4]. Similar models could be useful within the context of a global surveillance system for identifying areas of the world expected to be at high risk for the disease. Such areas could be targeted selectively for sample and data collection and as part of an active disease prevention program. Time-space Bayesian models were also used to assess the relation between epidemiological factors and changes in the FMD virus [10]. Autoregressive Bayesian models also could be applied to assessing disease control progress. Autoregressive Bayesian methods, such as those used to predict FMD risk in Turkey [1], could be used to monitor trends in FMD at a local or broader level, to forecast the risk of the disease and to monitor the impact of control and intervention programs. Other techniques also could be useful for predicting and assessing disease risk and evolution. Timeseries analysis has been used to assess seasonality, secular effects, trend, and influence of different intervention strategies for FMD in Colombia [3]. Geospatial analysis techniques, such as the probability co-kriging, have been used to predict risk for FMD in Pakistan using imperfect incidence data and surrogate information on herd and population demographics [9]. Quantitative models to assess the risk for FMD spread into free countries have been applied in countries such as Spain [5]. 3.3 System administration Support for the operation and development of the system requires both formal administration, which would be responsible for operation and development of the system, and an international funding mechanism. Considerable multilateral input, including broad international guidance and support for collaboration among countries, organizations, and agencies, will be required to develop the funding base and to meet the needs of participating groups. Input from various national and international agencies and organizations will be necessary to develop a conceptual plan for longterm administration, management, and funding. Possible sources of funding and indirect support (e.g., resources, personnel) from countries and agencies interested in participating in the program will need to be identified. 4. AUTHORS CONCLUSIONS

201

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

A global FMD surveillance system that provides high quality, accurate, and real-time information on FMD risk is needed to support FMD control and eradication on a global scale; Epidemiological models should be applied to identify key areas of the world to be targeted for active collection of samples and information, and for monitoring the evolution of the disease as part of the global FMD surveillance system; A prerequisite for such a global surveillance system is the creation of an international partnership of agencies, countries, organizations, and individuals to initiate conceptual development of global surveillance.

5. AUTHORS RECOMMENDATIONS

Develop an infrastructure for a global surveillance network for FMD capture, sample collection, reporting, and modelling. An international, neutral and independent consortium should explore formally the interest in, and funding for, developing a formal global FMD surveillance system and organization. Formulate specific steps necessary for exploration, design, and implementation, as necessary to move the process forward.

6. ACKNOWLEDGEMENTS The study was supported in part by the U.S. National Centre for Medical Intelligence. 7. REFERENCES [1] Branscum, A., Perez, A.M, Johnson, W.O., & Thurmond M.C. 2008. Bayesian spatiotemporal analysis of foot-and-mouth disease data from the Republic of Turkey. Epidemiol Infect 136:833-842. [2] Estrada, C., Perez, A.M., & Thurmond, M.C. 2008. Herd reproduction ratio and time-space analysis of a foot-and-mouth disease epidemic in Peru in 2004. Transbound Emerg Dis in press. [3] Gallego, M., Perez, A.M, & Thurmond, M.C. 2007. Temporal and spatial distributions of footand-mouth disease under three different strategies of control and eradication in Colombia (19822003). Vet Res Commun 31:819–834. [4] Garabed, R.B., Johnson, W.O., Gill, J., Perez, A.M., & Thurmond, M.C. 2008. Effects of politics and economics on country-level Foot-and-Mouth-Disease status. J R Stat Soc [Ser A] 171 (3):699–722 [5] Martinez-Lopez, B., Perez, A.M., De la Torre Reoyo, A., Sanchez-Vizcaino, J.M. 2008. Quantitative risk assessment of foot-and-mouth disease introduction into Spain via importation of live animals. Prev Vet Med 86(1-2):43-56. [6] Perez, A., Ward, M., & Carpenter, T. 2004a. Epidemiological investigations of the 2001 footand-mouth disease epidemic in Argentina. Vet Rec 154(25):777-782. [7] Perez, A.M., Ward, M.P., & Carpenter, T.E. 2004b. Control of a foot-and-mouth disease epidemic in Argentina. Prev Vet Med 65(3-4):217-226. [8] Perez, A.M., Thurmond, M.C., Carpenter, T.E., & Grant, P.W. 2005. Use of the scan statistic on disaggregated province-based data: Foot-and Mouth Disease in Iran. Prev Vet Med 71:197-207. [9] Perez, A.M., Thurmond, M.C., & Carpenter, T.E. 2006. Spatial distribution of foot-andmouth disease in Pakistan estimated using imperfect data. Prev Vet Med 76(3-4):280-289. [10] Perez, A.M., Konig, G., Späth, E., & Thurmond, M. 2008. Variation in the VP1 gene of serotype A foot-and-mouth disease virus associated with epidemiological characteristics of outbreaks in the 2001 epidemic in Argentina. J Vet Diagn Invest 20:433–439. [11] Shiilegdamba, E., Thurmond, M., Perez, A., & Carpenter, T. 2008. Temporal-spatial epidemiology of foot-and-mouth disease outbreaks in Mongolia, 2000 – 2002. Vet Res Commun 32:201-207 [12] Thurmond, M.C. 2003. Conceptual foundations for infectious disease surveillance. J Vet Diagn Invest 15:501-514.

202

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 34 SAMPLING, SHIPPING, ORGANISATION AND TESTING OF INACTIVATED SAMPLES TO TAKE ADVANTAGE OF MOLECULAR ANALYSIS METHODS *1+

S. Alexandersen 1

, J. Klein , T. Frederiksen and M. Hussain .

Department of Virology, National Veterinary Institute, Technical University of Denmark, Lindholm, DK-4771 Kalvehave, Denmark +

Affiliation from 1 October 2008, National Centre for Foreign Animal Disease, Canadian Food Inspection Agency, 1015 Arlington Street, Winnipeg, R3E 3M4 Manitoba, Canada

Food and Agriculture Organization of the United Nations - Pakistan, NARC, Islamabad, Park Road, PK-45500, Pakistan

ABSTRACT Introduction FMD is endemic in many countries and causes significant economic losses in livestock farming despite efforts directed towards control by vaccination. Efficient vaccination for protection against FMD in such settings is often hampered by a limited access to or excessive costs of high quality vaccines and/or the lack of a well planned vaccination strategy based on thorough epidemiological information. Limited access to or excessive costs of high quality vaccines is a matter of economic resources and may be solved by sufficient support from International Organisations. However, in such settings the methodological tools for generating thorough epidemiological information and sufficient and timely characterisation of circulating and epidemiological important strains of FMDV required for setting up a well planned vaccination strategy need a leap forward in regard to the ways epidemiological studies and strain characterisation are organised and performed. Previous experimental studies performed by us and others have looked at new sampling strategies for relatively easy estimation of e.g. prevalence of infected animals by using real-time RT-PCR analysis of swab samples collected directly into a suitable lysis buffer. The next step was to look at this approach under field conditions by using sampling of mouth swabs directly into a suitable lysis buffer for easy shipping and downstream analyses as described here. Materials and methods For this study we selected the Landhi Dairy/Cattle Colony (LCC) outside Karachi, the largest dairy colony in Pakistan and the biggest buffalo colony in the world. Due to its size and fragmented vaccination coverage, together with the Pakistani husbandry tradition of frequent transportation of animals to and from the large dairy colonies, FMDV infection is prevalent in LCC and therefore, LCC provided a suitable study population. Moreover, the information generated would also provide epidemiological information that could potentially improve the control of FMD in the colony and thus be of benefit to the many farmers enduring significant economic losses. From April 2006 to April 2007 we collected mouth swab samples of randomly selected, non-clinically affected herds as one group (group A), from non-clinically affected animals in herds with clinical evidence of a prior FMD outbreak (group B, i.e. herds with a few animals with evidence of old and healing lesions consistent with acute FMD for more than 2-3 weeks prior to sampling) and from non-clinically affected animals in a few herds in which acute FMD was obvious in at least one animal (group C). Swabs were immediately placed in Qiagen RLT buffer, which immediately stabilised the viral RNA and rendered the sample non-infectious. In addition, epithelial samples were collected from clinical FMD cases from group C. The results from analysis of these epithelial samples are described in a separate paper together with the detailed sequence and phylogeny studies of selected positive mouth swab samples from all 3 groups. The mouth swab sampling scheme included four major samplings together with smaller monthly samplings. Each sampling was accompanied by a questionnaire to gather general health status and vaccination information and the exact location of the herd was determined by GPS. The laboratory-based analysis employing real-time RT-PCR is described here

203

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 While VP1, capsid and full genome sequences derived from swabs and epithelial samples, together with the questionnaire-based information, are described in an accompanying paper. Results Sampling of mouth swabs directly into lysis buffer followed by shipment of the samples either directly to Denmark for real-time RT-PCR analysis at Lindholm, or alternatively via the National Reference Laboratory in Islamabad for RNA extraction followed by cDNA preparation and shipping of prepared cDNAs to Denmark, proved to be a reliable method for detecting FMDV RNA. Sampling in this way preserved the FMDV RNA and such samples contain no infectious FMDV and can therefore relatively easy be shipped as diagnostic samples and clearly provided valuable information on prevalence of FMDV infection in LCC over the one-year study period. Moreover, sequencing of such samples were possible provided that they contained a reasonable amount of FMDV RNA (sequencing results to be described in the accompanying paper) and in addition, preliminary in vitro results (Graham Belsham, personal communication) indicate that it may be possible to re-generate infectious FMDV from such purified RNA samples by transfection into suitable cell cultures. Discussion The approach described here has in our opinion clearly shown merit. Stabilised and inactivated swab samples can easily be collected and shipped internationally and may be used to determine the temporal-spatial distribution and prevalence of FMDV infection and further analysis of selected positive samples by sequencing can be used to characterise circulating strains of FMDV. Thus, the described method is a viable and valuable alternative to shipment of epithelial samples, containing infectious virus, for strain characterisation. 1. INTRODUCTION Pakistan has a number of large dairy buffalo and cattle colonies, FMD is endemic there and it has borders with India, Afghanistan, Iran and China where FMD is also a continued problem. Only fragmented information exists on the prevalence and extent of circulation of FMDV, including characterisation of current strains, in Pakistan. Improvement of this situation requires intensified surveillance and facilities and expertise in molecular epidemiology and may initially best be achieved by international collaboration. The project described here is such a collaboration between experts and staff in Islamabad and Karachi in Pakistan and at Lindholm in Denmark in close coordination with the FAO and the WRL for FMD in Pirbright, UK. The traditional diet in Pakistan includes buffalo milk or buffalo milk-based products and large dairy colonies are located around the major cities like e.g. Karachi to provide a continued supply of fresh non-pasteurised milk. Buffalo milk is preferred over cattle milk due to the high fat content and many of the large dairy colonies have 90-95% buffalo and only 5-10% cattle. The Landhi Dairy/Cattle Colony (LCC) is located in the eastern suburbs of Karachi, was established in 1959 and is now the largest dairy colony in Pakistan and the biggest buffalo colony worldwide. There is estimated to be a total of around 250-280 000 buffalo and 15-25 000 cattle in the colony. The number of individual farms is not known with certainty, but it has been estimated that there is between 1000-2000 farms each with around 100-200 animals on average. All of this is located on only 750 acres of land (including 250 acres for primitive roads). The farmers mainly purchase animals from the three major breeding regions in Punjab and in upper and lower Sindh and animals are primarily kept for one lactation period. After the lactation period, a few of the buffalos/cattle are used for re-breeding by the owner, some are slaughtered, and an estimated 50-75% of them go back to the regions mentioned above to be used for breeding once more. Such animals will, when they are ready to enter a new lactation period, again be transported across the country and sold to individual farmers. Many goats, sheep, cattle and buffalo from all over Pakistan are once a year transported to large markets just outside the big cities in connection with the celebration of the major Muslim festival Eid-ul-Azha. Animals that are not sold for sacrifice may, after having been mixed with many other animals from different places, then go back to where they came from or perhaps be sold and go somewhere else. The majority of the commercial farmers in LCC vaccinate their animals, either with a good quality trivalent vaccine (O, A, Asia1), but only given once due to the high cost of the vaccine, or with a monovalent type O vaccine produced in Pakistan which is of unknown quality and efficiency and apparently only contains antigen from a single, very old type O strain. This lack of an efficient and coordinated vaccination policy, together with the large number of animals kept in a relatively small area and the frequent transportation of animals in and out of the colony as well as the general lack of biosecurity procedures and awareness among LCC farmers, creates conditions that favour

204

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 continued FMDV circulation. Therefore, the LCC provides a suitable study population for establishing the methods needed for direct inactivation, stabilisation and shipment of field samples containing FMDV RNA and such studies may in turn provide useful epidemiological information to be used to improve FMD control in LCC. In order to potentially set up a future vaccination strategy that is efficient in protecting the LCC against currently circulating strains of FMDV, it is necessary to get data on the seasonal occurrence and prevalence of FMDV in the colony and to characterise the circulating strains. This may be achieved by a structured sampling strategy based on mouth swab sampling in connection with the relatively easy shipment of inactivated and stabilised samples to a laboratory with the capacity to perform large-scale RT-PCR testing to detect samples positive for FMDV which can then subsequently be further analysed by sequencing as described below and in an accompanying paper. 2. MATERIALS AND METHODS 2.1 Study Design and Sample Collection, Organisation and Shipping. The study was designed as a longitudinal and cross-sectional survey to provide information on clinical and in particular subclinical FMDV infection prevalence in LCC. We collected samples during 4 trips to Pakistan from April 2006 to April 2007 and local Veterinary Officers collected in addition a smaller number of samples each month during the same period. The monthly smaller samplings included 5 farms selected at random and on each farm 6 animals (5 buffalo and 1 cow if possible) were selected at random, clinically examined and a mouth swab taken by careful swabbing of the tongue with a sterile cotton swab (Libby Sterilin). During our 4 visits, the number of randomly selected farms was increased to 18 and the number of randomly selected animals on each herd increased to 9 buffalo as well as 1 cow if possible. When visiting the farms, the location was logged by handheld GPS equipment and the farmers thoroughly questioned for knowledge that may be of relevance for the analysis, e.g. clinical signs, vaccination status, vaccine used, age of the animal, outbreak history and any other relevant information. The randomly selected animals were checked clinically while taking samples. Farms without any clinical signs of current or relatively recent acute FMD were assigned to study group A (randomly selected, non-clinically affected animals from nonaffected herds). Animals from farms with a history and signs of recent clinical FMD were assigned to study group B (samples were taken from non-clinically affected animals in herds where the presence of a few other animals with evidence of old and healing lesions indicated that acute FMD had been present in the herd for more than 2-3 weeks before sampling) and farms with animals with current signs of acute FMD found during the initial examination were assigned to study group C (samples were taken from non-clinically affected animals in herds in which acute FMD was obvious in at least one other animal). The animals in the herds of group B and C with clinical signs of earlier or current FMD were examined in more detail, in particular concerning animals in group C in order to locate suitable vesicular lesion epithelia (lesion epithelia samples were in addition also collected during a pre-study visit in January/February of 2006; the results of the testing of collected epithelial samples are described in an accompanying paper). Mouth swabs were immediately placed, after breaking and removing the plastic handle of the swab, into 1.0 ml of RLT lysis buffer (Qiagen) in 2 ml screw-cap polypropylene tubes (Sarstedt). Such swabs in lysis buffer o

were often kept at ambient temperature for many hours before being stored at -20 C and then shipped either directly to Denmark, or alternatively via the National Reference Laboratory in Islamabad for RNA extraction followed by cDNA preparation and shipping of prepared cDNAs to Denmark. Swabs in lysis buffer and prepared cDNAs as described here contain no infectious FMDV (only stabilised RNA or cDNA) and can relatively easily be shipped as diagnostic samples both locally and internationally. 2.2 RNA Extraction, cDNA Preparation and Quantitative RT-PCR Total RNA was extracted from thawed sample aliquots using the Qiagen RNA Blood Kit (Qiagen) according to the manufacturer's instructions. The extracted RNA samples were then converted into cDNA and analysed by quantitative “real-time” RT-PCR to determine the amount of FMDV RNA present in the mouth swab samples as described in detail elsewhere, with the assumption that the swabs were initially diluted around 10-fold when placed in 1 ml of RLT lysis buffer

2-4,10-12,15

Sequencing. Selected samples with a relatively high content of FMDV cDNA based on the RT-PCR result were further analysed by sequencing essentially as described by us previously results of the sequencing are described in an accompanying paper.

6, 7

. The

205

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3. RESULTS 3.1 FMDV RNA Detected in Swab samples. During the 1 year study we collected 960 mouth swabs from 124 farms in group A (herds with no clinical signs of acute or recent FMD), 45 samples from 4 herds in group B (animals without clinical signs, but from herds where other animals had signs of healing/healed FMD) and 29 samples from 2 herds in group C (animals without clinical signs, but from herds where other animals had acute clinical signs of FMD) giving a total of 1034 swab samples analysed. The number of FMDV RNA positive swabs was 106 in group A (no clinical signs), 22 in group B (herds with relatively recent FMD) and 25 in group C (herds with acute signs of FMD), corresponding to 11%, 49% and 86% of the samples being positive in groups A, B and C, respectively. Of the 106 positive samples collected from subclinically infected animals/herds in group A, 58 contained sufficient FMDV RNA to facilitate subsequent sequencing. This selected group of 58 positive swabs from subclinically infected animals had a mean Ct value in the real-time RT-PCR of 36.0 (range 26 to 39) equivalent 4.6

3.7-7.5

) FMDV RNA target copies per ml of saliva/swab while the other 48 to around 10 (range 10 swabs from this group, not selected for sequencing, had a mean Ct of 42.5 (range 40-48) 2.7

1.0-3.4

equivalent to around 10 (range 10 ) copies per ml of saliva/swab. For comparison, the positive swabs from the farms with recent signs of FMD had a mean Ct of 37.5 (range 25-46) equivalent to around 10

4.1

1.6-7.8

(range 10

) copies per ml of saliva/swab while the positive swabs 5.8

collected in farms with acute FMD had a mean Ct of 32.0 (range 24-39) equivalent to around 10 (range 10 2

3.7-8.1

10 (range 10

) copies per ml of saliva/swab. These copy numbers usually correspond to roughly

1-5

), 10 (range 10

0-1.5

1.5

), 10

(range 10

0-5

) and 10 (range 10

1-5.5

) tissue culture

1-3,10

infectious doses (TCID50) per ml when assayed in bovine thyroid cells . For further details of seasonal prevalence of subclinical infection and sequence analysis etc please see the accompanying paper and our previously published results

6,7

4. DISCUSSION The data presented point out that carefully planned field epidemiological studies combined with collection of swab samples directly into a suitable lysis buffer for subsequent easy shipping, locally or internationally, to a laboratory for molecular testing by RT-PCR followed by sequencing of selected samples may provide a powerful tool for generating epidemiological information and valuable characterisation of circulating strains of FMDV. More than 10% of samples collected from herds with no signs of FMD turned out to be positive for FMDV RNA, while herds having animals with signs of an earlier outbreak of FMD (healed lesions) had approximately 50% of the samples from apparently unaffected neighbouring animals being positive and more than 80% of unaffected animals were positive in herds where other animals had signs of acute FMD. Taken together, these data indicate that the method described provides an accurate account of the numbers of animals being exposed to FMDV within a relatively short time period prior to sampling and in fact, may indicate that the positive animals are indeed infected. Data from our experimental studies indicate that mouth swabs are usually detected as positive within a time window of up to 2 weeks after 2, 3

exposure and are negative in carrier animals . Consequently, we consider it likely that the positive animals detected here in a field setting were exposed or infected within a 14 days window. Although FMDV RNA positive animals without clinical signs of disease may not for certain be infected, we consider the fact that more than 50% of the positive samples from farms with no 3.7-7.5

clinical signs of infection contained enough RNA sufficient for sequence analysis (a level of 10 1-5

FMDV RNA target copies corresponding to roughly 10 TCID50 per ml of saliva/swab) an indication that these animals may be subclinically infected. This notion is substantiated by the additional facts that the RT-PCR targets the 5’-untranslated region (5’-UTR) and that the sequencing PCRs involved amplification of relatively long targets, which together have been shown to be good indicators of picornavirus infectivity

5,8,9,13,14

, and that our previous studies have indicated a good correlation 1-3,10

between copy numbers and infectivity ; i.e. further evidence for these animals indeed being subclinically infected and that FMDV likely replicated in these animals to levels facilitating detailed molecular analysis. This view is further supported by the finding that non-affected animals in farms with other animals having evidence of previous, but healed, FMD lesions had a higher percentage of swabs positive for FMDV RNA, indicating an earlier, but apparently relatively widespread, circulation of virus in this group and potentially still

206

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 subclinical infection in some animals. Also, the finding of a very high percentage of positive swabs and a higher RNA content in swabs from non-affected animals from herds with ongoing acute FMD, further substantiates that such animals may truly be subclinically infected or alternatively, be in an early stage of infection before lesions develop. Consequently, although some of the animals detected as FMDV RNA positive by our methods may only have been exposed to FMDV and not necessarily actively infected, we consider it likely that a significant proportion was indeed subclinically infected. The use of on-site/in-the-field inactivated and stabilised swab samples as described here for easy shipping and subsequent analysis in advanced international laboratories, may also be combined with collection of epithelial samples from clinically affected animals as such epithelial samples also can be placed directly into a suitable solution, e.g. RNAlater (Ambion), which inactivates viral infectivity and stabilises the RNA provided that only small samples are collected and that the tissues are fully immersed in the RNAlater and kept at room temperature for at least 24 hours before shipping (Alexandersen, S. et al. unpublished). Finally, preliminary in vitro results (Graham Belsham, personal communication) indicate that, provided collected samples contain sufficient FMDV RNA, it may likely be possible to re-generate infectious FMDV from such RNA purified from inactivated and stabilised swabs or epithelia by using transfection into suitable cell cultures should the need arise to study the virus further for e.g. virulence or vaccine studies etc. 5. CONCLUSIONS

Combined field and molecular epidemiological studies using randomly selected mouth swabs collected directly into a suitable lysis buffer provide a powerful tool for generating detailed epidemiological information and molecular characterisation of circulating strains. Direct lysis of collected swab samples into a suitable lysis buffer circumvents the need for international shipment of samples as dangerous goods and provides inactivated and stable samples that without any safety problems can be shipped as diagnostic samples. Analysis of high numbers of mouth swabs by a highly sensitive, quantitative real-time RT-PCR indicates that this method may provide an efficient way of establishing the extent of virus circulation during periods of low clinical activity and for providing estimates of prevalence of infection at a given point in time. Furthermore, subsequent sequencing of selected samples can be used for detailed analysis of the relationship of circulating strains with strains circulating in other countries or with known vaccine strains in order to provide appropriate vaccine coverage. Extracted RNA may possibly be used for making infectious FMDV by using transfection into suitable cell cultures; however, this aspect needs further study.

6. RECOMMENDATIONS

7. ACKNOWLEDGMENTS We thank the many friendly and competent Veterinary Officers that have helped us in Pakistan. Drs. Giancarlo Ferrari and Keith Sumption from FAO are thanked for their continued interest in the project. Finally, we thank The National Veterinary Institute, Technical University of Denmark, the FAO Regional Project (GTFS/INT/907/ITA) and the EU Network of Excellence for Epizootic Disease Diagnosis and Control (EPIZONE, Call Identifier: FP6-2004-Food-3-A) for support. 8. REFERENCES [1] Alexandersen, S. and N. Mowat. 2005. Foot-and-mouth disease: host range and pathogenesis. Curr. Top. Microbiol. Immunol. 288:9-42.

207

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [2] Alexandersen, S., M. Quan, C. Murphy, J. Knight, and Z. Zhang. 2003. Studies of quantitative parameters of virus excretion and transmission in pigs and cattle experimentally infected with foot-and-mouth disease virus. J. Comp Pathol. 129:268-282. [3] Alexandersen, S., Z. Zhang, A. I. Donaldson, and A. J. Garland. 2003. The Pathogenesis and Diagnosis of Foot-and-Mouth Disease. J. Comp Pathol. 129:1-36. [4] Alexandersen, S., Z. Zhang, S. M. Reid, G. H. Hutchings, and A. I. Donaldson. 2002. Quantities of infectious virus and viral RNA recovered from sheep and cattle experimentally infected with foot-and-mouth disease virus O UK 2001. J. Gen. Virol. 83:1915-1923. [5] Bhattacharya, S. S., M. Kulka, K. A. Lampel, T. A. Cebula, and B. B. Goswami. 2004. Use of reverse transcription and PCR to discriminate between infectious and non-infectious hepatitis A virus. J Virol. Methods 116:181-187. [6] Klein, J., M. Hussain, M. Ahmad, M. Afzal, and S. Alexandersen. 2008. Epidemiology of foot-and-mouth disease in Landhi Dairy Colony, Pakistan, the world largest Buffalo colony. Virol. J 5:53-68. [7] Klein, J., M. Hussain, M. Ahmad, P. Normann, M. Afzal, and S. Alexandersen. 2007. Genetic characterisation of the recent foot-and-mouth disease virus subtype A/IRN/2005. Virol. J. 4:122-133. [8] Li, J. W., Z. T. Xin, X. W. Wang, J. L. Zheng, and F. H. Chao. 2002. Mechanisms of inactivation of hepatitis a virus by chlorine. Appl. Environ. Microbiol. 68:4951-4955. [9] Li, J. W., Z. T. Xin, X. W. Wang, J. L. Zheng, and F. H. Chao. 2004. Mechanisms of inactivation of hepatitis A virus in water by chlorine dioxide. Water Res. 38:1514-1519. [10] Quan, M., C. Murphy, Z. Zhang, and S. Alexandersen. 2004. Determinants of early footand-mouth disease virus dynamics in pigs. J. Comp Pathol. 131:294-307. [11] Reid, S., N. Ferris, G. Hutchings, Z. Zhang, G. Belsham, and S. Alexandersen. 2002. Detection of all seven serotypes of foot-and-mouth disease virus by real-time, fluorogenic reverse transcription polymerase chain reaction assay. J. Virol. Methods 105:67-80. [12] Reid, S. M., S. S. Grierson, N. P. Ferris, G. H. Hutchings, and S. Alexandersen. 2003. Evaluation of automated RT-PCR to accelerate the laboratory diagnosis of foot-and-mouth disease virus. J. Virol. Methods 107:129-139. [13] Simonet, J. and C. Gantzer. 2006. Degradation of the Poliovirus 1 genome by chlorine dioxide. J Appl. Microbiol. 100:862-870. [14] Simonet, J. and C. Gantzer. 2006. Inactivation of poliovirus 1 and F-specific RNA phages and degradation of their genomes by UV irradiation at 254 nanometers. Appl. Environ. Microbiol. 72:7671-7677. [15] Wernery, U., P. Nagy, C. M. Amaral-Doel, Z. Zhang, and S. Alexandersen. 2006. Lack of susceptibility of the dromedary camel (Camelus dromedarius) to foot-and-mouth disease virus serotype O. Vet. Rec. 158:201-203.

208

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 35

USE OF INFRARED THERMOGRAPHY TO DETECT SIGNS OF FOOT–AND-MOUTH DISEASE IN WILD AND DOMESTIC UNGULATES M. R. Dunbar1*, S. R. Johnson1, J. C. Rhyan2, M. McCollum2 1

U.S. Department of Agriculture, Wildlife Services, National Wildlife Research Center, 4101 LaPorte Avenue, Ft. Collins, CO 80521. 2 U.S. Department of Agriculture, Veterinary Services, National Wildlife Research Center, 4101 LaPorte Avenue, Ft. Collins, CO 80521.

Infrared thermography (IRT) measures heat emitted from a surface, displays that information as a pictorial representation, and is capable of being a remote, non-invasive technology that provides information on the health of an animal. Foot-and- mouth disease (FMD) caused by FMD virus (FMDV) is a severe highly communicable viral disease of cloven-hooved animals including both domestic and wild ruminates. Early detection of the disease may reduce economic loss and loss of susceptible wildlife. We evaluated the use of IRT to detect possible heat changes associated with FMDV infection in experimentally infected mule deer (Odocoileus hemionus). Infection occurred through either inoculation with FMDV (intraepithelial tongue inoculation with 10,000 bovine tongue infective doses of 01 Manisa FMDV) or exposure to inoculated animals. Early vesicular lesions were observed within 24 hrs post-inoculation and 48-96 hrs post-exposure on the mouth and/or feet. From internal temperature sensors in exposed animals, temperature elevated significantly from the pre-infection temperature (P ≤ 0.002) starting approximately one day before any lesions were observed. Differences in eye thermal temperatures and body temperatures of well focused images were found not to be significantly different. Therefore, eye thermal images could be used as an index to body temperature. For feet thermal images of exposed animals, the mean of the daily maximum (MMAX) foot temperature rose significantly (P= 0.017) from two days before (27.3°C ±1.9°C SE) to two days after (33.0°C ±2.0°C SE) first foot lesion occurrence. We also evaluated the use of IRT in experimentally infected pronghorn antelope (Antilocapra Americana) and found similar results. Furthermore, we evaluated IRT in naturally infected domestic cattle in a FMD outbreak in Israel. There, we found it had applicability in a field situation. These experiments and observations indicate that IRT may be a rapid, remote, and non-invasive method to screen for suspect animals to test further for FMDV infection during a FMD outbreak. It may also be possible to detect thermographic evidence of infection associated with FMDV before clinical signs are observed, thus reducing transmission of the disease.

209

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

210

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 36

SEQUENCE IDENTIFICATION AND GENETIC PROFILE OF FMDV IN A 2007 DISEASE OUTBREAK IN ISRAEL USING FULL LENGTH GENOMIC ANALYSIS L. Xu1, J. Bieker1, J. Rowland1, W. Hurtle2, C. Carrillo1, H. Yadin3, A. Perez4, T. Beckham5, M. McIntosh1 and S. Metwally1*. 1

Foreign Animal Disease Diagnostic Lab, Plum Island Animal Disease Center (PIADC), USDA, APHIS, VS, NVSL, P.O. Box 848, Greenport, NY 11944 2 Department of Homeland Security, Plum Island Animal Disease Center, Greenport, NY 11944 3 FMDV Laboratory, Kimron Veterinary Institute, 50250 Bet Dagan, Israel 4 Center for Animal Diseases Modeling and Surveillance, Department of Medicine and Epidemiology, School of Veterinary Medicine University of California, Davis, CA 95616 5 Medical Diagnostic Laboratory, Texas A&M University, College Station, TX 77845

ABSTRACT FMD is endemic and widely spread among various animal species in the Middle East. Periodically devastating epidemics occur that spread rapidly across national borders. In 2007, Israel was severely affected by serotype O epidemics despite the presence of a national compulsory vaccination program using a trivalent vaccine in cattle (containing types O, A22 and Asia1) and a monovalent vaccine in small ruminants (containing type O). Three strains of serotype O are included in the vaccine: O1 Manisa, Geshure Isr/2/85 and 3039. In addition to the unexpected nature of the outbreak, instances of mortality in affected goats were reported. To understand the genetic characteristics of the virus causing this outbreak, and analyze the possible source for these antigenic variants, a complete genome analysis of isolates obtained from geographically distinct regions affected by the outbreak was conducted. Six specimens from FMDV infected animals were selected for analysis including one probang and four tongue epithelium samples from affected cattle and one cardiac tissue sample from an affected goat. Viral genomic sequences were amplified using RT-PCR with primers that produced overlapping fragments which were subjected to automated dideoxy sequencing methods. Similarities between the nucleotide and amino acid sequences of the complete ORF among the six virus isolates demonstrated greater than 98.9% and 99.2%, respectively. Phylogenetic analysis of the P1 region of all six viruses revealed that they belong to a unique sub-lineage of the O serotype. They were also distinct from two vaccine strains (O1 Manisa and Geshur Isr2/85) used in the vaccination program, and were also distinct from other known type O FMDV isolates from Turkey between 1989 and 2000. Further analysis of the VP1 region on 446 type O viruses from GenBank revealed that the cluster of Israel isolates was more related to strains isolated from South-Asia in Bhutan, Nepal and Malaysia than those from the Middle East during 2003 and 2004, although this result may have been biased by incomplete representation of samples from the region in GenBank. Full length viral genome sequence analysis applied to FMDV outbreaks is of importance in the characterization of outbreak isolates and may facilitate identification of viral genetic attributes linked to the pathogenesis of emerging FMDV isolates. 1. INTRODUCTION. Foot-and-mouth disease (FMD) is an acute, highly contagious disease of domestic and wild clovenhoofed animal species. The etiological agent of FMDV is a small, non-enveloped, positive-sense, single stranded RNA (8.5 kb) virus belonging to the genus Aphthovirus of the family Picornaviridae (1). Foot and Mouth Disease Virus (FMDV) is currently classified as a reportable disease to the Office International des Épizooties (OIE). Due to the highly infectious nature of the virus, as well as the economic consequences resulting from constraint of international trade in animals and animal products originating from infected countries, surveillance and early disease detection and characterization are critical.

211

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 The molecular epidemiology of FMDV has been studied using the phylogenetic analysis of partial or full length sequences of the viral genome. The genome contains one open reading frame encoding Lpro and capsid proteins: VP4, VP2, VP3 to VP1, as well as the non-structural proteins 2A, 2B, 2C, 3A, 3B, 3Cpro and 3Dpol (2). VP1, the most variable capsid protein (2), has being used to genotype the seven serotypes of FMDV (A, O, C, Asia1, SAT1, SAT 2 and SAT3) into geographically distributed groups named as topotypes (3). Advances in nucleotide sequencing methods and bioinformatics makes analysis of full length FMDV genomes relatively easy. Without a doubt, comparative genomic analysis of full length FMDV genomes is more extensive than studies conducted with VP1 analysis and is used to gain further insight into FMDV biology, proteomic structure and function, virulence, virus transmission, and host range (2-8). In addition, understanding the ability of FMDV to tolerate mutations within its antigenic sites across the capsid coding P1 region could provide insight into the virus’s evasion of the host immune response and improve vaccine development and efficacy. From December 28, 2006, through June 15, 2007, nine districts in four Israeli provinces were affected by 39 serotype O FMD outbreaks. Some of the outbreaks were reported to cause mortality in infected goats. In some cases, affected animals were vaccinated using a polyvalent FMDV vaccine which included three O isolates: O1 Manisa, Geshur Isr/2/85, and 3039 (O Turkey 30/39, 1999). To assess the hypothesis that the virus may have escaped the host’s immune protection induced by the vaccine, and to analyze its origins and genetic characteristics, several samples were collected from FMDV infected animals in various regions of Israel and their full genomes were sequenced. In this paper, we present the results of the full genome sequence analysis on six field samples from the province of Hazafon. 2. MATERIALS AND METHODS FMDV isolates. A total of 6 FMDV samples were obtained from the Department of Viral Diseases at the Israeli State Veterinary Services and Animal Health, Ministry of Agriculture. They were collected from different geographic locations and species. Collection dates, vaccination status and other related information are presented in Table 1. Sample types included cardiac tissue (10% homogenate), tongue epithelium tissue (10% homogenate), and esophageal-pharyngeal fluid (probang). 2.1 RNA isolation, amplification, and sequencing Viral RNA was purified using the RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturers’ instructions. The full FMDV genome from each sample was amplified with five PCR assays including a reverse transcriptase polymerase chain reaction (RT-PCR) covering the S fragment only, and three overlapping PCR assays targeting the L fragment: two RT-PCR reactions covering from PolyC to Lpro, Lpro to 2A and one 3΄ RACE-PCR (rapid amplification of complementary ends) covering VP1 to polyA end. The RT-PCR reactions were performed using the SuperScript™ III One-Step RT-PCR System with Platinum® Taq kit or Platinum® Taq High Fidelity kit (Invitrogen, Carlsbad, CA) according to the manufacturers’ instructions. The 3΄ RACE-PCR was performed using the RNA LA PCR kit, Ver. 1.1 (Takara Bio USA, Madison, WI). After being purified with QIAquick PCR Purification kits (Qiagen, Valencia, CA), the PCR amplicons were analyzed by directly sequencing on a 3730XL DNA sequencer (Applied Biosystems) with BigDye Terminator v1.1, v3.1 5X Sequencing Buffer (Applied Biosystems). The PCR and sequencing primer information will be published separately. The individual sequences were analyzed with Sequencher 4.8 (Gene Codes Cor. Ann Arbor, MI). The multiple sequence alignments (MSA) were performed with the ClustalW 2.0.9 (9). The phylogenetic analysis on the MSA were conducted with two algorithms: Neighbor-Joining method coupled with Maximum Composition likelihood and Bootstrapping (NJ) implemented in MEGA 4 software (10) and Split Decomposition Cluster (SDC) algorithm implemented in SplitsTree 4 software (11). 3. RESULTS AND DISCUSSION The attempts to amplify the viral genome via RT-PCR across the poly C region failed for all the samples. As a result, two contigs were compiled for each sample, corresponding to S and L fragments with an approximate length about 360 bp and 7830 bp respectively. Both contigs were constructed with multiple forward and reverse sequence readings and submitted to GenBank. The accession number for these six 2007 Israel O isolates are FJ175661 to FJ175666, respectively (Table 1).

212

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Pair-wise comparisons between the open reading frames (ORF) of six samples showed that they shared greater than 98.9 % and 99.2 % similarity in nucleotide and amino acid sequence respectively. No nucleotide transversions in the VP1 gene were observed between the six isolates. In addition, there are no differences in the predicted VP1 proteins for isolates 6382, 6387, 6389 and 6391. Moreover, samples 6387 and 6389 shared the same sequences for the VP1 gene. Therefore, the six samples, although collected from different outbreaks and species, are genetically very closely related. Evaluation of the P1 region revealed that the six samples could be further divided into two sub-groups: 6378, 6380 in one sub-group and isolates 6382, 6387, 6389 and 6391 in another. The two sub-groups with respect to Israel FMD outbreaks may be polyphyletic in origin or divergent in that they may have arrived in Israel from different but related sources, or they may have arrived from the same source virus and subsequently diverged due to locality differences, host type and/or incubation period/conditions. Coincidently, epidemiological analysis of 70 FMD outbreaks reported in 2006 and 2007 in Israel and Palestine showed that the six viruses sequenced there belonged to the same time-space cluster of outbreaks, suggesting introduction from a single source (12). As a matter of fact, a wider comparison of only the VP1 coding region with a total of 446 O isolates from GenBank revealed that the cluster of the Israel isolates is more closely related to the isolates from South-Asia, including Bhutan, Nepal and Malaysia than those from Middle East in Israel, Iran, and Turkey isolated in 2003 and 2004. Note, however, that some of the six isolates analyzed here were reported to be closely related to samples from Jordan which are available at the Institute for Animal Health in Pirbright. Unfortunately, the Jordan sequences have not been submitted to GenBank and for that reason were not assessed in our analysis (12; 13). Secondly, we have compared the genomic sequences of the Israel isolates against 60 P1 full length sequences from Serotype O isolates, and confirmed that a similar tree topology is obtained based in the phylogenetic relationships of either P1, VP1 or VP3 genomic regions. In each case, results show that Israel isolates conform a unique cluster of viruses significantly different from the vaccine strains and the Turkey isolates included in the vaccine cocktail in use to immunize animals in the country (O1 Manisa (AJ251477) and Geshur (AF189157), as well as strains isolated in Turkey between 1989 and 2000. Therefore, our results show that, although not as distant as from other FMDV O strains, the genetic differences that distinguish Israel isolates and separate the group from the corresponding vaccine isolates, account for, and may have a critical role in, the antigenic properties of the new isolates that allow them to escape the immune control in vaccinated animals. Finally, we extended our analysis over a collection of 203 full length FMDV isolate sequences from GenBank and the PIADC repository, which included all 7 serotypes. The sequences were used to conduct analysis on full length genomes as well as different structures of the genome: ORF (Fig 1), Lpro, VP4, VP2, VP3, VP1, 2A, 3A, and 3D. The Phylogenetic trees using by both NJ and SDC algorithms reinforced our previous suggestion that the Israel samples share a direct ancestor but shed new light into the fact that, as has been lately described for other FMDV (2, 3), they are clustered into a distinct branch in all areas analyzed. Interestingly, despite being closely related to different O strains in the regions coding for the structural proteins, using Split decomposition and Bootscanning techniques we show that the genetic content encoding Lpro, 2A, 3A, and 3D in the Israel O sequences, shared more similarities with isolates of serotype A or Asia1 (Fig 2). These isolates possess different serotype sequences in different regions of the genome, indicating the origin of these isolates may be recombinant strains from different serotypes. However, to further explore this possibility, more sequences of recent isolates circulating in South-Asia and Middle-East are needed. Our results clearly point toward a theoretical new introduction of FMD virus into the country instead of generation of a new variant as result of the mutation/evolution of preexisting strains either circulating or preserved in the wild animal reservoir within Israel borders. In conclusion, we show that complete sequencing is able to generate critical information about FMDV circulation in endemic areas which cannot be obtained from partial sequences. Our results strongly suggest that the outbreak in Israel was caused by a single introduction of a distinct sublineage of serotype O FMDV which is distinct from the vaccine strains used in the area. The new isolates are close to the 2003 and 2004 isolates in South Asia and could be a result of recombination between different serotypes. Strategic and targeted collection of samples as part of an international surveillance system would help to understand the epidemiological dynamics of the disease in the region. 4. REFERENCES [1] Grubman, M. J., and B. Baxt. 2004. Foot-and-mouth disease. Clin Microbiol Rev 17:465-93.

213

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [2] Pereira, H.G. 1981. Foot and Mouth Disease. Academic Press Inc., London, United Kingdom. [3] Knowles, N. J., and Samuel, A. R. 2003. Molecular epidemiology of foot-and-mouth disease virus. Virus Res 91:65-80. [4] Carrillo, C., E. R. Tulman, G. Delhon, Z. Lu, A. Carreno, A. Vagnozzi, G. F. Kutish, and D. L. Rock. 2005. Comparative genomics of foot-and-mouth disease virus. J Virol 79:6487-504. [5] Cottam, E. M., D. T. Haydon, D. J. Paton, J. Gloster, J. W. Wilesmith, N. P. Ferris, G. H. Hutchings, and D. P. King. 2006. Molecular epidemiology of the foot-and-mouth disease virus outbreak in the United Kingdom in 2001. J Virol 80:11274-82 [6] Du, J., H. Chang, G. Cong, J. Shao, T. Lin, Y. Shang, Z. Liu, X. Liu, X. Cai, and Q. Xie. 2007. Complete nucleotide sequence of a Chinese serotype Asia1 vaccine strain of foot-and-mouth disease virus. Virus Genes 35:635-42. [7] Samuel, A. R., N. J. Knowles, R. P. Kitching, and S. M. Hafez. 1997. Molecular analysis of foot-and-mouth disease type O viruses isolated in Saudi Arabia between 1983 and 1995. Epidemiol Infect 119:381-9. [8] Zhang, X., Z. Liu, Q. Zhao, H. Chang, and Q. Xie. 2004. Sequencing and analysis for the full-length genome RNA of foot-and-mouth disease virus China/99. Sci China C Life Sci 47:74-81. [9] Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG. 2007. Clustal W and Clustal X version 2.0. Bioinformatics, 23, 2947-2948. [10] Tamura K, Dudley J, Nei M & Kumar S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 1596-1599. [11] Huson D. H. and Bryant D. 2006. Application of Phylogenetic Networks in Evolutionary Studies, Mol. Biol. Evol., 23(2):254-267. [12] Al Khamis M., Perez A. M., Yadin H. 2008. Tempero-spatial clustering of foot-and-mouth disease outbreaks in Israel and Palestine, 2006-2007. Transboundary and Emerging Disease, submitted. [13] IAH. 2007: Molecular epidemiology reports. FAO World Reference Laboratory for foot-andmouth Disease, http://www.wrlfmd.org/fmd_genotyping/index.html (Accessed 25/1/2008) Table 1: FMDV isolate information SampleAnimal Date Description Vaccinati Provin District Sub_Dist Epidemi Repor Popu- Number Number Number / on ce rict oting lation of of Non- of GB Acc logical Date cases infecte Outbre # Unit d ak 6378 / Capricor Jan. Cardiac No Hazafo Acco Netua Village Dec. FJ175661 n 01 tissue n 28 260 260 1 (goat) 2007 2006 520 6380 / Bovine Jan. Probang Unknow Hazafo Golan Mevho Farm Jan. 05 n n 01 FJ175662 (calf) 2007 2007 210 30 180 1 6382 / Bovine Jan. Tongue Yes Hazafo Yizreel Beit Village Jan FJ175663 (calf) 24 epithelium n Hashitha 18 2007 2007 2600 120 2480 1 6387 / Bovine Jan. Tongue Unknow Hazafo Yizreel Moledeth Village Jan. FJ175666 (adult) 27 epithelium n n 25 2007 2007 327 15 312 1 6389 / Bovine Jan Tongue Yes Hazafo Yizreel Neve Ur Village Jan FJ175664 (adult) 28 epithelium n 25 1369 3 2007 2007 1400 31 6391 / Bovine Jan. Tongue Yes Hazafo Kineret Llut Village Jan. FJ175665 (adult) 31 epithelium n 31 2007 2007 10 5 5 1

214

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 1: The split decomposition cluster analysis of multiple sequence alignment on ORF of 203 Full length FMDV isolates with SplitsTree

SAT A

Israel Asia1

215

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 2: Similarity and bootscanning plots, in upper and lower part respectively, generated by SimPlot 4 with a query of six Israel samples ORF against three reference groups after being aligned with ClustalW. The three group are serotype Asia 1, in Red, including four isolates DQ989320 to DQ989323; serotype O, in Green, including 8 isolates AF189157, AY312589, AY593811, AY593812, AY593823, AY593828 and EU140964; serotype A, in Blue, including nine isolates: AY593769, AY593789, AY593791, AY593796, AY593798, AY593799, AY593800, AY687333, EF149010. The tests were performed with the default setting of the program.

216

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 37

IDENTIFICATION OF VIRULENCE DETERMINANTS IN FMDV: POTENTIAL USE OF FUNCTIONAL GENOMICS TO PREDICTING VIRAL PATHOTYPES J. Arzt1, M.E. Piccone1,2, J. Pacheco1, M. Borca1, E. Rieder1, J. Zhu1, L. Rodriguez1* 1

Foreign Animal Disease Research Unit, Agriculture Research Service, U.S. Department of Agriculture, Plum island Animal Disease Center, Greenport, New York. 2 Department of Pathobiology and Veterinary Sciences, Univ. of Connecticut, Storrs, CT, USA.

1. INTRODUCTION Foot-and-mouth Disease virus (FMDV) causes a highly contagious, debilitating disease in clovenhoofed animals with devastating economic consequences. This member of the Picornaviridae family (genus Aphtovirus) has an 8,500 nucleotide positive stranded RNA genome organized in a single reading frame and non-coding regions at the 3’ and 5’ termini. The function and specific role in viral pathogenesis of the different regions of the viral genome remain largely undefined. 2. MATERIALS AND METHODS Viral mutants generated by random transposon mutagenesis were characterized both in-vitro, in cell culture and in-vivo by aerosol inoculation of cattle. Viruses with Tn inserts in three regions: the region between the two initiation codons and in proteins 3B1, 3B2 and 3B3 were selected for further study. 3. RESULTS AND DISCUSSION Transposon insertions in the region between the two translation initiation codons caused a delay in translation and complete attenuation in cattle. TN inserts in each of the three 3B proteins did not cause complete attenuation in cattle. 4. CONCLUSIONS The combination of functional genomics and pathogenesis in relevant hosts allows the identification of virulence determinants in FMDV. The exploration of the viral genome will generate useful knowledge that can be used to predict the virulence of naturally occurring viral strains.

217

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

218

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 38

THE DISTRIBUTION OF NSP POSITIVE ANIMALS IN REGULAR VACCINATED HERDS AS CRITERION FOR EFFECTIVENESS OF VACCINE AND VACCINATION DURING ENDEMIC YEAR H. Yadin

219

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

220

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

221

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

222

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 39

ESTIMATING THE VACCINATION COVERAGE IN AN FMD FREE ZONE-BRAZIL 2005/2006 H.V. Barbosa Jr.1*, G.M. de Moraes1, A. Mendes da Silva2, J. L. Naranjo2 1

Animal Health Department, Ministry of Agriculture, Livestock and Food Supply, Brazil. Pan American Health Organization, Regional Office of World Health Organization, UN.

1. INTRODUCTION This is a report on the investigation carried out in the Brazilian FMD free zone where vaccination is practised with the purpose of assessing the rates of vaccine coverage achieved by the FMD vaccination campaigns. 2. MATERIAL AND METHODS The region covered by the study encompassed 60% of the Brazilian territory and 90% of Brazilian cattle herd. The target population was characterised in domains, as to the size of the herds, and in sub-populations as to the age group of the animals. The analytical method used for assessing the immunity response of each individual was the ELISA-LPB standardized by PANAFTOSA for the detection of specific antibodies against proteins of the viral capsid. Each sub-population was tested for one of the three types of virus contained in the used vaccine (A, O and C). 3. RESULTS Only in one subpopulation the vaccination coverage found was less than 80%. That subpopulation showed 71% of vaccination coverage. All other subpopulations showed coverage average of 96%. 4. DISCUSSION Irrespective of all activities and procedures involved in preparation of the vaccination stages reports, the study showed compatibility of these reports with the immunization levels of the population estimated by laboratory tests. The levels of immune protection recorded reflect the country’s tradition in terms of carrying out vaccination campaigns against Foot-and-mouth Disease, which for over three decades has been one of the main strategies of the PNEFA, and the good quality of the vaccine used, particularly starting from the 1990s.

223

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

224

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 40 MANAGEMENT OF SAT-TYPE FMD IN SOUTHERN AFRICA: PRESENT CONTROL AND TRADE STANDARDS ARE INAPPROPRIATE Fred Brown debate: G R Thomson* SADC Foot-and-mouth Disease Project, Food, Agriculture and Natural Resource Directorate, Private Bag 0095, Gaborone, Botswana

ABSTRACT It is argued that international trade standards in respect of FMD in southern Africa could be strengthened to both lessen the risk of FMD being spread through export of animal products while, at the same time, enabling greater diversity of land-use options for the Region. This is important in view of the on-going establishment of transfrontier conservation areas (TFCAs) which will result in larger numbers and a wider distribution of African buffalo in southern Africa. Healthy buffalo populations maintain the SAT serotypes in southern Africa and this situation presently precludes animal products being exported to regulated markets from areas of southern Africa where infected buffalo populations occur. However, transmission of SAT serotypes from buffalo to domestic livestock can be effectively managed. This means that standards could be set that dissociate the FMD status of livestock in a country or zone from that of wildlife, as has been done for other transboundary animal diseases. Combining this with a commodity-based approach could provide the solution to a serious development conundrum facing southern African countries. 1. INTRODUCTION The three SAT serotypes of foot-and-mouth disease (FMD) virus differ on the basis of genome sequence-based phylogeny from the other 4 FMD virus serotypes (2,5). Furthermore, it is inferred from the ability of African buffalo (Syncerus caffer) and SAT-serotype populations to co-exist in a permanent near-symbiotic relationship that the SAT serotypes evolved in association with African buffalo, i.e. independently of the A, O, C and Asia 1 group which, for some hundreds of years, have been maintained by domestic livestock in Europe, Asia and north-central Africa, spreading to the Americas in the 19th Century. Eurasian serotypes of FMD virus seem not to have not established themselves in southern Africa. Some documented introductions of serotype O and perhaps A via import of livestock or their products from South America and Asia occurred in the 20th Century but these incursions either disappeared spontaneously or were eradicated. Presently, healthy African buffalo populations — which have become increasingly isolated and scattered in wildlife reserves by human expansion and agricultural development — maintain the SAT serotypes of FMD in southern Africa (5). These viruses infrequently spill over into cattle and result in outbreaks of disease in domestic livestock which are not only disruptive to livestock production but, more importantly, result in a situation whereby trade in livestock and products derived FMD-susceptible animals is severely constrained by the requirement for freedom from FMD of the country or zone of origin as the basis for access to regulated markets (Chapter 8.5 of the OIE’s Terrestrial Animal Health Code [the Code], 2007 – www.oie.int). The efforts of official veterinary services of southern Africa are to a large extent devoted to minimizing the likelihood of transmission of SAT-serotype viruses to cattle through vaccination of cattle located near to infected buffalo populations and various forms of movement control, often dependent on fencing. A fundamental consequence of this situation is that SAT-serotypes will not be eradicated from southern Africa in the foreseeable future. Over many decades the rich wildlife heritage of the southern African Region — the SAT serotypes of FMD are an accompaniment of that heritage — has been in decline for a diversity of reasons but, increasingly, attempts are being made to turn this situation around, a major objective being conservation of the bio-diversity of natural ecosystems. A major initiative in this connection is the transfrontier conservation area (TFCA) movement which aims to expand a patchwork of interconnected conservation areas across the borders of countries of sub-Saharan African countries. Thirteen transfrontier parks and transfrontier conservation areas have been identified in

225

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 the Southern African Development Community (SADC) Region. As opposed to discrete transfrontier parks, TFCAs often include national parks, neighboring game reserves, hunting areas, conservancies and communally managed land, i.e. there are usually large numbers of people living within and around TFCAs. Ultimately, more than 1 200 000 square km of sub-Saharan Africa is likely to be incorporated into TFCAs. An objective of this movement is the reestablishment of wildlife migration within and between the TFCAs (i.e. the vision of ‘Africa without fences’). A key economic driver of the TFCA movement will be eco-tourism although it is intuitively difficult to foresee how tourism alone will sustain large numbers of people. For that reason, a diversity of livelihood generating alternatives will be necessary in future. The TFCA movement, driven by organizations such as the Peace Parks Foundation (www.peaceparks.org), has significant political, financial and public support. Some TFCAs are now established and functioning so these developments are real, not simply theoretical. At the same time, other aspects of rural development such as exploitation of the abundant domestic animal resource of the southern African Region, is recognized as being vital for rural development and alleviation of poverty. There are about 70 million small-scale owners of livestock in the SADC Region and most of these are disadvantaged people who see expansion of livestockbased production as an important future economic opportunity. It also needs to be understood that for many of these people cattle are traditionally important in a sociological context. Co-incident with these developments, southern Africa is not dissociated from wider-scale events associated with the spread of transboundary animal diseases (TADs) such as avian influenza, bluetongue, classical and African swine fever and disease resulting from West Nile virus infection. These infections are maintained and potentially spread by free-living wildlife and it is becoming obvious that climate change and factors associated with globalization are the basis for their geographic expansion. It is consequently often impossible to be sure at any given time where these infections are distributed and where not, even in developed parts of the world. This creates obvious difficulties for the present international trading system for animals and animal products because the latter is based primarily on the requirement for geographic ‘freedom from disease’ as the primary mitigator of risk associated with such trade (in reality it is the presence of infection, not disease, which is vital). This was the impetus for proposals to introduce a commodity-based approach to trade in animal products, i.e. the concept that the commodity itself and not the area of production, should determine risk of spreading TADs by traded animal products (4). Of course, if geographic freedom from disease and commodity-based standards were to be combined in the management of risk of spreading human and animal infections, the system would be concomitantly more robust. Countries in sub-Saharan Africa generally and southern Africa more specifically, where livestock production is generally more commercialized, are consequently faced with a conundrum: Conservation of their unique wildlife heritage (of benefit/value to the world at large) will increasingly be penalized by limiting access to international markets for animal commodities. This includes livestock products and those that could be derived from wildlife through a system of sustainable utilization. The present basis for access to international markets for animal products, i.e. geographic freedom from FMD and other TADs is therefore inappropriate for southern Africa. Present international standards required for recognition of a country or zone (region) as being free from FMD and their effects on southern Africa It is presently possible for member countries to apply to the OIE (International Organisation for Animal Health which serves as the international standard-setting body for the World Trade Organisation [WTO] in matters related to animal health and trade) for recognition of its whole territory or part thereof (zone) as being freed from FMD (officially, the FMD status of member countries is provided in the form of a ‘list’ on the OIE website - www.oie.int). In order to achieve this status, applicant countries need to prove that no outbreak of FMD has occurred in the country or zone for the last 1-2 years (depending on whether vaccination is employed or not) and also that both the domestic livestock and wildlife populations (i.e. all susceptible species) show no evidence of ‘circulating FMD virus’ for at least a year. In this context it is important to appreciate that African buffalo very rarely, if ever, suffer clinical effects from natural infection with SAT serotype viruses. It has been demonstrated over many years that it is possible to prevent transmission of SAT viruses viruses from buffalo to cattle by separation of buffalo and cattle populations, primarily through the use of fencing (such fencing also needs also to prevent antelope crossing because they can act as intermediaries in the transmission of FMD from buffalo to cattle) (3,6). Vaccination of cattle kept adjacent to infected buffalo populations provides further protection. It has even proven possible, in the Kasane area of Botswana for example, to protect cattle populations which mix freely with infected buffalo from SAT infection over a 20 year period by vaccinating the cattle every 4 months (M. Letshwenyo, personal communication, 2008). Admittedly, this system has broken

226

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 down recently, probably due to the inability of the current vaccine to protect against a SAT2 topotype circulating in the buffalo population. It is therefore clear that the growing numbers and distribution of buffalo implicit in the development of TFCAs in southern Africa will increasingly constrain livestock production by limiting access for animal products manufactured in the region to international markets. This will constitute a significant obstacle to rural development in an area where it is desperately needed. Could a more appropriate system be devised without increasing the risk of FMD spread through trade commodities? It is increasingly accepted by the OIE that the status of domestic livestock in countries or zones in respect of TADs need not necessarily be influenced by the status of wildlife populations. So, for example, the status of a country’s poultry industry need not be negatively influenced by the presence of highly pathogenic avian influenza (HPAI) viruses in wild birds (Chapter 10.4 of the Code – www.oie.int). Similar considerations are being given to other disease where wildlife may serve as sources of infection such as classical and African swine fever. This principle is equally applicable in the context of SAT serotype infections in southern Africa. Such provision would enable countries to establish domestic animal populations demonstrably free from circulating FMD virus but without necessarily requiring infected buffalo to be excluded from the country or zone. Combining the above principle (dissociation of FMD presence in buffalo from the FMD status of domestic livestock) with a commodity-based approach (also now accepted by the OIE although specific standards, especially for processed commodities, are still largely in abeyance) for specific commonly traded products would facilitate trade in defined commodities produced in southern Africa which present a lower risk for transmission of FMD (and most other TADs-causing infections) than the present system based exclusively on geographic freedom from FMD in both domestic and wild animal populations. For example, there is good evidence that de-boned beef produced from healthy cattle by conventional methods presents an acceptably low risk of spreading FMD and other TADs (1). If such beef were produced from cattle populations proven free from FMD virus, a safer trading system than that presently in operation would prevail but, at the same time, enable wider access to international markets. As a further measure, processing of the beef (i.e. beneficiation) would provide an additional safe-guard. In this way a matrix of risk-reduction measures could be applied to ensure safe trade in products derived from animals. What needs to be done to implement necessary changes implicit in this proposal? 1. The OIE (perhaps through the Regional Commission for Africa) to be requested to amend Chapter 8.5 of the Code to accept the possibility that domestic livestock populations (cattle specifically) can be maintained free of infection with SAT viruses even when infected buffalo are present in the country or zone, i.e. the two species can be effectively sequestered from each other using physical (e.g. fencing) and/or immunological means (vaccination). 2. Development by the OIE of specific standards for commonly traded products derived from animals (such a de-boned beef and its derivatives) would ensure that products prepared from healthy animals according to processes that render the risk of FMD virus (or other TADs-causing agents) being inadvertently present, acceptably low (i.e. provide an appropriate level of protection as required by the WTO). 2. REFERENCES [1] Donaldson, A.I., Thomson, G.R. & Leyland, T.J. In press. De-boned beef - An example of a commodity for which specific standards could be developed to ensure an appropriate level of protection for international trade. Transbound. Emerging Dis. [2] Robson, K.J.H., Harris, T.J.R. & Brown, F, 1977. An assessment by competition hybridization of the sequence homology of the seven serotypes of FMDV. J. gen. Virol. 37: 271276. [3] Sutmoller, P., Thomson, G.R., Hargreaves, S.K., Foggin, C.M. & Anderson, E.C., 2000. The foot-and-mouth disease risk posed by African buffalo within wildlife conservancies to the cattle industry of Zimbabwe. Prev. Vet. Med. 44: 43-60. [4] Thomson, G.R., Tambi, E.N., Hargreaves, S.K., Leyland, T.J., Catley, A.P., van ‘t Klooster, G.C.M. & Penrith, M-L. 2004. International trade in livestock and livestock products: The need for a commodity-based approach. Vet. Rec. 155: 429-433. [5] Thomson, G.R. & Bastos, A.D.S. Foot-and-mouth disease. “In: Infectious diseases of livestock”, 2nd edn. Eds. Coetzer JAW & Tustin RC. Oxford University Press, Oxford, London, Cape Town, 2004: 1324-1365.

227

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [6] Vosloo, W., Thompson, P.N., Botha, B., Bengis, R.G. & Thomson, G.R., In press. Longitudinal study to investigate the role of impala (Aepy€ceros melampus) in foot-and-mouth disease maintenance in the Kruger National Park, South Africa. Transbound. Emerg. Dis.

228

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 41

KEYNOTE DIAGNOSTICS: A NEW ERA IN ACCESS TO HIGH PERFORMANCE DIAGNOSTICS D.P. King1* 1

Institute for Animal Health, Ash Road, Pirbright, GU24 0NF, United Kingdom.

This presentation reviews the changing face of diagnostic virology, highlighting new assay formats that have been developed over the past decade. Deployment of molecular assays into laboratories has revolutionised the way in which samples from infected animals are processed. These molecular tests are now widely used for the detection and characterisation of FMD virus: complementing, and on occasions replacing established approaches such as immunoassays and in-vitro cell culture methods. The role of these assays was recently demonstrated during the cases of FMD that occurred in the United Kingdom during the summer of 2007. In contrast to the 2001 epidemic, diagnostic support for these outbreaks was characterised by the use of real-time RT-PCR (rRT-PCR) as a principal tool for decision making. In addition to confirming the presence of FMDV on all 8 infected farms, for the first time, rRT-PCR results were also used to recognise preclinical FMD in a cattle herd. Furthermore, during the later stages of the outbreaks, the rRT-PCR assay also supported an active surveillance program within high-risk cattle herds. Full-genome sequencing approaches were also used to generate data to rapidly reconstruct the transmission histories that occurred during these outbreaks. In addition to laboratory-based tests, there are opportunities to use new technologies to provide diagnostic capability away from a centralised laboratory. Work in this area has developed simpleto-use lateral-flow devices for the detection of FMD virus, as well as new hardware platforms to allow PCR testing to be devolved to local laboratories and into the field for use by non-specialists. The driving force for these improvements has largely been influenced by the priorities of developed countries with FMD-free (without vaccination) status. In these settings, speed of the assays and the ability to detect all seven FMDV serotypes are primary considerations. However, it is important to recognise that these diagnostic approaches also show considerable promise for use in FMDendemic countries, although region-specific modifications (such as sample archiving and serotype/strain characterisation) may be required to tailor these tests for use. Using PCR-based and other antigen-detection technologies, it is now possible to imagine a completely new diagnostic paradigm for how sample collection, testing of infected animals and monitoring of the spread of FMD might be achieved.

229

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

230

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 42 FMD AND SVD COMBINED PROFICIENCY TEST STUDIES 2008 – VIROLOGICAL ASSAYS

N. P Ferris*, G. Wilsden, D. P King, G. H Hutchings, S. M Reid, K. Ebert, Y. Li and D. J Paton BBSRC Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey GU24 0NF

ABSTRACT The aim is to complete a combined proficiency study for FMD and SVD during 2008 for virology and serology. Particular virological tests are not specified but laboratories are invited to select tests as if samples arose from suspected outbreak cases and to report which samples are positive with which viruses and tests and overall which cases are positive with which virus. Panel 1 consisted of 12 coded samples: six prepared from vesicular epithelia of two cattle herds (1a and 1b), three prepared from vesicular epithelia from pigs in one herd (1c) and three faecal suspensions from pigs in another herd (1d), each herd being affected with vesicular conditions. Samples were derived from cases: 1a, two from a type O FMDV positive epithelium (one sample in duplicate) and one from naïve cattle; 1b, one from a type A FMDV positive epithelium and two from naïve cattle; 1c: two samples from SVDV positive epithelia and one from a naïve pig; 1d, three faecal suspensions from naïve pigs. Panel 2 consisted of 16 coded samples prepared from inactivated, cell culture supernatant fluids: eight (2a) derived from cattle and eight (2b) from pigs, each from a different case of a herd with a vesicular condition. Samples were derived from the propagation of four different FMD virus serotypes and SVD virus or from uninoculated cell cultures arising from cases: 2a, from two dilutions of each of FMDV type A, SAT 2 and Asia 1 and negative antigens; 2b, from two dilutions of each of FMDV type O and SVDV and two negative antigens. Packages are being delivered to laboratories around the world together with result templates for data return to the FMDWRL. These are being collated and a summary will be presented at the meeting. 1. INTRODUCTION The aim is to complete a combined proficiency study for FMD and SVD during 2008 for virology and serology to ascertain whether individual laboratory procedures are suitable for diagnostic use. Particular virological tests were not specified but laboratories were invited to select tests as if samples arose from suspected outbreak cases and to report which samples were positive with which viruses and tests and overall which cases were positive with which virus. 2. MATERIAL AND METHODS Panel 1 consisted of 12 coded samples: six prepared from vesicular epithelia of two cattle herds (1a and 1b), three prepared from vesicular epithelia from pigs in one herd (1c) and three faecal suspensions from pigs in another herd (1d), each herd being affected with vesicular conditions. Samples were derived from cases: 1a, two from a type O FMDV positive epithelium (one sample in duplicate) and one from naïve cattle; 1b, one from a type A FMDV positive epithelium and two from naïve cattle; 1c: two samples from SVDV positive epithelia and one from a naïve pig; 1d, three faecal suspensions from naïve pigs. Approximate 10% (w/v) suspensions of the epithelia were prepared in 0.04 M phosphate buffer (Ferris and Dawson, 1988), mixed with an equal volume of sterile glycerol, aliquoted into cryotubes in 1 ml amounts and stored at -20oC until dispatch. Panel 2 consisted of 16 coded samples prepared from inactivated, cell culture supernatant fluids: eight (2a) derived from cattle and eight (2b) from pigs, each from a different case of a herd with a vesicular condition. Samples were derived from the propagation of four different FMD virus serotypes and SVD virus or from uninoculated cell cultures arising from cases: 2a, from two dilutions of each of FMDV type A, SAT 2 and Asia 1 and negative antigens; 2b, from two dilutions of

231

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 each of FMDV type O and SVDV and two negative antigens. The samples were aliquoted in 5 ml amounts in glass bottles and stored at -80oC prior to distribution. The requisite sample sets were distributed to each of the participating laboratories by airfreight in dry ice. Templates for the return of the results from the proficiency testing were also supplied and which included prompts designed to elicit information on individual laboratory test methodology. The identities of the samples and virus serotypes are listed in Figs 1 and 2. 3. RESULTS AND DISCUSSION The results achieved to date for Panel 1 are summarised in Fig. 1. All laboratories used their inhouse procedures for carrying out virus isolation (VI; Fig. 1a) and amongst the group included primary bovine thyroid and kidney cells, primary and secondary lamb kidney cells, foetal goat tongue and BHK cells and other cells of porcine origin including primary and secondary kidney cells, PK-15 and SK-6 cells plus those of the IB-RS-2 cell line. Additionally, there were differences in both how the cell monolayers used for sample analysis were grown (e.g. in tubes, flasks or plates of different sizes) and the manner in which VI was performed. It can be seen that the majority of laboratories had adequate cell culture systems to detect infectious virus. The results of RT-PCR analysis are shown in Fig. 1b and illustrate that the majority of RT-PCR test procedures were specific and that laboratories had sufficiently useful procedures to compensate for any deficiency in VI as the overall interpretation of individual sample and herd disease status was generally good (Fig. 1c). The results achieved to date for Panel 2 are summarised in Fig. 2 and indicate that workable antigen ELISAs exist in most laboratories (Fig. 2a), although there was several instances of misdiagnosis of sample virus serotype. One laboratory used monoclonal antibodies while the remainder used polyclonal antisera either produced in-house or else obtained from Pirbright. The majority of laboratories also examined the samples by RT-PCR (Fig. 2b) and the results confirm the general adequacy of the laboratory RT-PCR test procedures. The overall interpretation of sample status (Fig.2c) was reasonably accurate with certain exceptions. A more detailed analysis will be carried when the full set of results is received from all the participating laboratories. 4. ACKNOWLEDGEMENTS We thank all the laboratories for participating in the study. This work was funded by the UK Department of the Environment, Food and Rural Affairs and by FAO. 5. REFERENCE [1] Ferris, N.P. and Dawson, M. (1988). Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Vet. Microbiol. 16, 201-209.

232

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 a)

233

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Fig.1: Results achieved by participating laboratories for Panel 1 from: a) virus isolation and ELISA; b) RT-PCR plus c) overall case interpretation a)

234

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig.2: Results achieved by participating laboratories for Panel 2 from: a) antigen ELISA; b) RT-PCR plus c) individual sample interpretation

235

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

236

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 43

FMD AND SVD COMBINED PROFICIENCY TEST SCHEME STUDIES 2008 (PHASE XXI) – SEROLOGY Y. Li*, P. Keel, G. Wilsden, P. Hamblin, N. Ferris, K. Swabey, B. Statham, J. Hammond and D. Paton Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey, GU24 0NF, UK.

ABSTRACT The aim of this exercise was to complete a proficiency testing study for virological and serological diagnosis for FMD and SVD in 2008 to assist National FMD Laboratories in developing and maintaining accurate and reproducible FMD diagnostic tests. This presentation is focused on the FMD and SVD serology studies; for FMD the priority serotypes were O and A. Particular tests were not specified, but labs were invited to select tests and interpret results as if the samples were from post-outbreak serosurveys conducted after either vaccination or non-vaccination (according to specification given with panel). The FMD serology panel consisted of 10 coded cattle serum samples from 4 hypothetical cases. Two cases were from post type O PanAsia outbreaks and the other two were from post type A Iran 96 outbreaks. The SVD serology panel consisted of 10 coded serum samples from different herds in a surveillance zone surrounding a SVD outbreak. 67 laboratories were invited to participate. Panels were dispatched to participants on request along with a template for results and a set of instructions. The participants were requested to interpret the results for each sample and an overall interpretation for each case. They were also asked to provide information on their tests to evaluate the homogeneity of methods and to enable some analysis of possible causes for discrepant results. Test results have been submitted from different laboratories and they are being collated and analysed. All results received before 10th October 2008 will be presented at the meeting. There has been a delay in sending out some of the packages of samples due to changes in the process and administration of our biosecurity regulators following the outbreak of FMD in UK last year. It is anticipated that not all participants will have sent their results back before this meeting. 1. INTRODUCTION In order to assist National Reference Laboratories (NRL) of Member States of the European Commission for the Control of FMD (EUFMD) and elsewhere to develop and maintain accurate and reproducible FMD diagnostic tests, also to harmonise testing between national and international reference laboratories. FAO FMD World Reference Laboratory (WRL) has conducted a series of proficient testing studies for FMD diagnosis known as the “Phase” studies (Paton et al., 2002, 2003; Li et al. 2006). It is currently at Phase XXI and has been jointly organised by the WRL and the European Community Reference Laboratories (CRL) for FMD and SVD in the Pirbright laboratory of the Institute for Animal Health (IAH-P). The current study incorporates both serological and virological assays for both FMD and SVD diagnosis. This report is focused on the FMD and SVD serology studies; for FMD the priority serotypes were O and A. Particular tests were not specified, but labs were invited to select tests and interpret results as if the samples were from post-outbreak serosurveys conducted after either vaccination or non-vaccination (according to specification given with panel). 2. MATERIALS AND METHODS As shown in Table 2a, the FMD serology panel 3 consisted of 10 coded cattle serum samples from 4 hypothetical cases as the following:

237

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Case 3a: 2 sera from cattle in different herds that had been vaccinated against O Manisa, 30 days before, in a vaccination zone surrounding an outbreak due to O PanAsia strain. Case 3b: 3 sera from cattle in different herds that had not been vaccinated, in a surveillance zone surrounding an outbreak due to O Pan Asia strain. Case 3c: 2 sera from cattle in different herds that had been vaccinated against A Iran 96, 30 days before, in a vaccination zone surrounding an outbreak due to an A Iran 96 strain. Case 3d: 3 sera from cattle in different herds that had not been vaccinated, in a surveillance zone surrounding an outbreak due to an A Iran 96 strain. The SVD serology panel 4 consisted of 10 coded serum samples from different herds in a surveillance zone surrounding a SVD outbreak (Table 2b). All samples in each panel were repeatedly tested 10 times to establish the status of the samples with confidence and also to set up the criteria to assess the performance. An invitation letter for Phase XXI was sent to the potential participants in early this year. Panels were dispatched to laboratories on request along with a template for result reporting and a set of instructions. The participants were requested to interpret the results for each sample and an overall interpretation for each case. They were also asked to provide information on their tests to evaluate the homogeneity of methods and to enable some analysis of possible causes for discrepant results. On receipt, the results were sent back to the participant for their confirmation and then decoded. The results from all participants for each serology assay have been collated and analysed. 3. RESULTS Participants A total of 67 laboratories from all EU member state countries, EUFMD but not EU countries, and countries in other regions were invited to take part in Phase XXI studies. Out of them, 37 agreed to participate in this study with 73% from EU and 17% from Non-EU countries. Up to 10th October 2008, results for panel 3 and 4 have been received from 20 and 20 laboratories with 75% from EU and 15% from Non-EU countries, respectively (Table 1). FMD serology panel 3 NSP tests All 20 laboratories tested the panel for antibodies against the FMDV non-structural protein (NSP). All but one labs used commercially available kits and 19 out of 20 labs used the Cedi/PrioCHECK FMDV-NS kit (Prionics Lelystad B.V.) either solely or in conjunction with other commercial kits including SVANOVIR FMDV 3ABC-Ab ELISA (Svanova) and CHEKIT-FMD-3ABC (Bommeli) or inhouse ELISA. Results obtained for NSP antibody tests showed a good level of consistency except that one lab gave a negative result from a positive sample; while one lab gave a negative result from a positive sample and a positive result from a negative sample (Table 3). Solid Phase Competitive ELISA (SPCE) Eight Labs used the SPCE. All of them tested samples against serotypes O and A, 3 out of 9 labs also tested samples against type Asia1. Almost all 8 labs correctly scored all samples except that one lab gave a negative result from a positive sample. Cross-reactivity between serotype O and A samples was observed in 3 labs (Table 4). 8 labs have tested the samples using the PrioCHECK Type O ELISA kit (Prionics Lelystad B.V.). All of them showed consistent results. Nevertheless, a cross-reactivity with A Iran 96 samples was observed in 5 labs (Table 5). Liquid Phase Blocking ELISA (LPBE) 9 labs carried out the LPBE. All of them tested the serology panel for serotype O and A antibodies, and 2, 3 and 1 out of them also tested for antibodies against FMDV serotypes Asia1, C and SAT 13, respectively. All labs have given expected results to all samples except that one lab showed negative result for a type A sample. Cross-serotype reactions between type O and A were observed in 5 labs for 2-6 positive samples. Reactivity cross serotypes O, A, SAT1 and O, A, Asia1 were also found in one lab, respectively (Table 6). Virus Neutralisation Test (VNT)

238

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 12 labs tested samples using VNT against serotypes O and A (all 12 labs), Asia1 (2 out of 12 labs) and C (5 out of 12 labs). Out of them, 8 labs gave expected results to all samples. One lab has only tested 2 out of 10 samples. One lab gave inconclusive results from 2 out of 4 negative samples, one lab scored type A positive for a negative sample, and one lab showed type Asia1 positive from one and inconclusive results from the rest 3 negative samples (Table 7). Overall interpretation for cases Out of 20 labs, 8 interpreted all 4 cases as expected. 11 labs scored the correct type of antibodies but provided no information on the infection status for each case. One lab gave the right type of antibodies for case 3a and 3c but failed to test and interpret case 3b and 3d (Table 8). SVD serology panel 4 Eighteen labs tested SVD serology panel using ELISAs including 5B7 MAC ELISA (7 labs), PrioCHECK/Cedi ELISA (Prionics Lelystad B.V.) (10 labs), Liquid phase competition ELISA (1 lab) and in-house ELISA (3 labs with 2 used in conjunction with five B7 MAC ELISA and 1 used it solely). 14 labs tested the panel using SVD VNT. All results obtained from all 20 labs either by ELISA or VNT were all consistent and scored samples correctly (Table 9 and 10). Quality Accreditation About 61.4% of FMD and 52.9% of SVD serology tests used in participating laboratories were accredited as ISO 17025 QA standard. 10.5% of FMD tests were accredited as ISO 9001. About 28.1% of FMD and 47.1% of SVD serology tests were either adhered to a local national system or declared no quality system or provided no information (Table 11a and 11b). 4. DISCUSSION It is different from the previous phase studies for serology which tried to achieve multiple aims, the current proficiency test for FMD serology is focused on the post-outbreak serosurveys conducted after either vaccination or non-vaccination. An advisory board is constituted this year and Dr Kris De Clercq (Belgium), Dr Aldo Dekker (Netherland), Dr Emiliana Brocchi (Italy) and Dr Sandra Blome (Germany) are members of the board. They will advise on the scope and design of the studies and on the feedback provided to participants. All samples were repeatedly tested 10 times to establish the status of the samples with confidence and also to set up the criteria to assess the performance of laboratories. As each case was clearly defined in this study, most participating laboratories tested the FMD serology panel against both serotype O and A as expected. The strains used for structural protein (SP) tests were in a good level of consistence. 8 out of 8, 8 out of 9 and 11 out of 12 labs used O1 Manisa strain and A22 or A Iran 96/97 strain(s) for SPCE, LPBE and VNT tests, respectively. All of 20 labs selected NSP ELISA to detect past and present FMDV infection in vaccinated and unvaccinated animals. Only 2 labs used NSP ELISA solely. 6 labs tested samples using NSP ELISA and one SP test. 9 labs performed 3 or 4 assays. Two labs performed all 5 tests including SPCE, LPBE, VNT, PrioCHECK Tyep O and PrioCHECK NSP ELISA. These results reflected the fact that SP tests were good at evaluation of the immunity level at individual level, so NSP and SP tests should be used jointly to examine the infection status and antibody level for a suspected case. There has been a delay in sending out some of the packages of samples due to changes in the process and administration of our biosecurity regulators following the 2007 outbreak of FMD in UK. It is anticipated that not all participants will have sent their results back before this meeting. 5. AUTHOR’S CONCLUSIONS All of 20 labs tested the panel for antibodies to viral non-structural proteins. All but one labs used commercial kits which led to harmonised results for a given test kit. 8, 9 and 12 out of 20 labs selected SPCE, LPBE and VNT to test samples, respectively. A good level of consistency was observed in the qualitative results obtained using tests for antibodies to both structural and non structural proteins of the virus among participating laboratories. Similar virus strains were used in the structural protein tests by almost all but one laboratories. The cross serotype reactivity between type O and A was observed in 3 out of 8 labs , 5 out of 9

239

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 labs and none out of 12 labs for SPCE, LPBE and VNT tests, respectively. A cross-reactivity with A Iran 96 samples was observed in 5 out of 8 labs using PrioCHECK Type O kit. 6. AUTHOR’S RECOMMENDATIONS Phase XXII – FMD and SVD PTS 2009 serology The priorities will be type O and /or A and/or Asia1 for FMD serology To include one or two replicates in a panel to test the repeatability of a test in a laboratory. 7. ACKNOWLEDGEMENTS This work was supported by EC, FAO and Defra, UK. The authors would like to acknowledge all the participants for Phase XXI. 8. REFERENCES [1] Li Y., Ferris N., Hamblin P., Goris N., Keel P., Hutchings G., Statham B. and Paton D. 2006. FAO Phase XIX FMD Inter-laboratory Comparative Test Exercise on Serology. The Report of the Session of the Research Group of the Standing Technical Committee on Control of Foot-andMouth Disease 2006. Appendix 57, page 352-362. [2] Paton D., Armstrong R, Turner L., Hamblin P., Corteyn M. and Anderson J. 2002. FAO Collaborative Study Phase XVII: Standardisation of FMD Antibody Detection. European Commission for the Control of foot-and Mouth Disease, Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of foot-and-Mouth Disease. Cesme, Izmir, Turkey, Sept 2002, pp226-234. [3] Paton D., Armstrong R. and Anderson J. 2003. An update on progress with the FAO Collaborative Studies for FMD Serology Standardisation, Phases XVII and XVIII. European Commission for the Control of foot-and Mouth Disease, Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease. Gerzensee, Switzerland, Sept 2003, pp102-115. Table 1: Participants for Phase XXI

Phase XXI

Non-EU

Country invited

No. of lab participated

No. of lab required panel 3

No. of lab required panel 4

No. of lab panel 3 (up No. of lab panel 4 (up

sent results for to 10th Oct.) sent results for to 10th Oct.)

Table 2a: Phase XXI FMD serology panel 3* Sample Animal Vaccine Infection Bled Dilution PrioCHECK-O NSP O1 O1 3a-1 VS78 35dpc 1:2 Pos Pos Manisa UKG34/2001 O1 3a-2 VT62 n/a 21dpv neat Pos Neg Manisa O1 3b-1 VS83 n/a 35dpc 1:2 Pos Pos UKG34/2001 3b-2 ABS n/a n/a n/a neat Neg Neg 3b-3 ABS n/a n/a n/a neat Neg Neg 3c-1

240

UZ63

A Iran96 A Iran96

33dpc 1:2

WP/Neg

SPCE LPBE

VNT

O,A

275.4

154.8

O,A

708.7;A(90) 284.8

O,A

375.8

211.9

Neg Neg

≥1473.1

≥761.5

Pos/Neg A

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

3c-2

UZ70

A Iran96 n/a

3d-1 3d-2 3d-3

UZ68/69 n/a ABS n/a ABS n/a

A Iran96 n/a n/a

21dpv neat

Neg

33dpc 1:10 n/a neat n/a neat

Neg Neg Neg

Pos Neg Neg

A/Neg 476.5 Neg Neg Neg Neg

Results presented were obtained by CRL for FMD.

dpc/v;

n/a: Not available Pos: Positive Neg: Negative Table 2b: Phase XXI SVD serology panel 4*

days

post

≥1516

226.6 134.2 Neg Neg

challenge/vaccination.

ABS: Adult Bovine Serum. WP: Weak positive (50≤PI≤60)

Sample

Animal

Details

Dilution

VNT

ELISA (5B7)

1 2 3 4 5 6 7 8 9 10

n/a n/a SL98 SL82 n/a SI-0 SI-2 VO49 VO50 n/a

NPS(PAA) NPS (local) ITL/8/94 6dpi ITL/1/92 9dpi NPS (Sigma) UKG/27/72 28dpi UKG/27/72 28dpi POR/1/2003 22dpi POR/1/2003 22dpi NPS(Invitrogen)

neat neat neat neat neat neat neat neat neat neat

Neg Neg Pos Pos Neg Pos Pos Pos Pos Neg

Neg Neg 2.26 1.59 Neg Pos Pos Pos Pos Neg

Results presented were obtained by CRL for SVD. Pos: Positive Neg: Negative

NPS: Normal Porcine Serum n/a: Not available

Table 3: Results of NSP ELISA for FMD serology panel 3-Phase XXI P: Positive N: Negative I: In conclusive

Sample

L 1

a 2

b 5

C 8

P N

N P

3 a - 1

3 a - 2

3 b - 1

3 b - 2

3 b - 3

3 c - 1

3 c - 2

d 26

3 d - 1

3 d - 2

3 d - 3

Table 4: SPCE results for FMD serology panel 3-Phase XXI*

241

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

3a-1 3a-2 3b-1

Lab Code 1 2 O O O O O O

3 O O, A O

4 O O O

5 O O O

17 O O O

31 O O O

34 O O O

3b-2

NEG

3b-3

NEG

3c-1 3c-2 3d-1

A NEG NEG

A A A

A, O A, O A

A A A

A ,O A A

A A A

A A NEG

3d-2

NEG

3d-3

NEG

Sample

*: The dominant serotype was expressed in bold if there is cross-serotype reactivity observed. NEG: Negative Table 5: Results of PrioCHECK Type O ELISA for FMD serology panel 3-Phase XXI Lab Code

Sample

3a-1

3a-2

3b-1

3b-2

N N

3b-3

N N

3c-1

3c-2

3d-1

3d-2

3d-3

P: Positive

N: Negative

Table 6: LPBE results for FMD serology panel 3-Phase XXI*

Sample 3a-1 3a-2

O, A

3b-1

O, A

3b-2

Neg

3b-3

Neg

3c-1 3c-2

242

Lab Code

A A

3d-1

3d-2

Neg

6 O O, A, SAT1

O, A

Neg

Neg A, O, Asia1 A, O, Asia1

Neg

A, O

A, C

A, O

A, C

A, O

Neg

A, O Neg

A Neg

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3d-3

Neg

*: The dominant serotype was expressed in bold if there is cross-serotype reactivity observed. Neg: Negative Table 7: VNT results for FMD serology panel 3-Phase XXI

Sample

1 O O O N N A A Inc N N

3a-1 3a-2 3b-1 3b-2 3b-3 3c-1 3c-2 3d-1 3d-2 3d-3

2 O O O Inc Inc A A A N N

3 O O O N N A A A N N

4 O O O N N A A A N N

5 O O O N N A A Inc N N

Lab code 7 10 O ND O ND O O Inc ND Inc ND A ND A, Asia1 ND A A Inc ND Asia1 ND

11 O O O N N A A A N N

16 O O O N N A A Inc N N

17 O O O N A A A A N N

13 O O O N N A A A N N

34 O O O N N A A A N N

The

dominant serotype was expressed in bold if there is cross-serotype reactivity observed. N: Negative Inc: Inconclusive ND: Not done Table 8: Interpretation for each case for FMD serology panel 3 – Phase XXI Case Lab Code 3a

1 2 3 O OO

4 5 6 7 8 10 O Inc O* O* O* O* O*

11 13 14 O* O* O*

15 16 17 19 22 26 O* O* O* O* O O*

31 O

34 O

O OO

O* O* O*

O* ND O* O* O

A* A A* A Inc A* A* A* A* A* Inc A* A* A* Inc NI A* A* A* NI O Inc A Inc A

A A

O* O* O* O* O*

A* A* A* A* A*

*: Antibody positive. NI: No infection

A* A* A*

A* ND A* A* A

O A

Inc: Inconclusive ND: Not done

Table 9: ELISA results for SVD serology Panel 4-Phase XXI

Lab Code 2 3 4 5 N N N N

6 N

9 N

10 N

13 N

15 N

16 N

18 N

19 N

21 N

24 N

25 N

28 N

29 N

31 N

N N N N N: Negative

Sample

10 P: Positive

Table 10: VNT results for SVD serology Panel 4-Phase XXI

243

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Sample

Lab Code 2

10 P: Positive

N N N N: Negative

N N ND: Not done

Table 11a: Quality accreditation for FMD serology tests-Phase XXI Tests Accreditation

Total SPCE

LPBE

VNT

NSP

PrioCHECK-O

ISO 17025

ISO 9001

Local or no or Unknown

Total

Table 11b: Quality accreditation for SVD serology tests-PhaseXXI Tests 5B7 MACELISA

PrioCHECK Test

In House

VNT

ISO:17025

Local or no or unknown

Total

Accreditation

244

Total

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 44 KEYNOTE: DRIVING UP GLOBAL STANDARDS FOR FMD DIAGNOSTIC: A KEY ROLE FOR PROFICIENCY TESTING AND INTERNATIONAL ORGANISATIONS. K. De Clercq1*, K. Luyten1, D. Paton2, D. King2, K. Sumption3 and N. Goris1 1

Epizootic Diseases Section, Virology Department, Epizootic Diseases Section, Veterinary and Agrochemical Research Centre, Groeselenberg 99, 1180 Brussels, Belgium 2 Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Surrey GU24 ONF, UK 3 Secretary, EuFMD, FAO HQ, Room C-518, Viale delle Terme di Caracalla, Rome 00153, Italy

ABSTRACT Laboratory diagnostics are used for foot-and-mouth disease (FMD) outbreak confirmation, postoutbreak and post-vaccination surveillance, vaccine matching, and import/export control. Participation in a national quality assurance (QA) system and international proficiency testing (PT) schemes is essential to maintain confidence in results between different laboratories. A workshop on PT, with the National Reference Laboratories (NRL) for FMD and classical swine fever in the European Union, recommended: the yearly organisation of PT, the use of the ISO 17043/43 standard to guide the development of proficiency testing schemes for the organisers and participants; the primary goal of PT schemes to be the evaluation of laboratories with well defined criteria for lab conformity prior to the start; the establishment of an Advisory Board; a clear feedback with a draft report prior to the annual meeting including results, conclusions on nonconformities and recommendations; corrective actions to address non-conformities agreed between Community Reference Laboratory and NRL and communicated to DG SANCO. Regional Reference Laboratories organising PTs should also be evaluated via a PT schemes supervised by international organisations such as FAO or OIE. The availability of reference standards is fundamental in a QA system but the production and evaluation through collaborative studies is highly expensive, time consuming and requires a serious international financial investment. Commercial ELISAs and (real time) RT-PCR kits are currently available and should be fully validated and licensed. Recently the OIE implemented a system with different levels of assay validation giving private companies the opportunity to submit a dossier for international recognition. However, the question remains what to do if an assay without this international recognition is used to support a country’s international disease status recognition. Moreover, an assay recognition or marketing authorisation does not provide a sufficient guarantee for the quality of the test kit batches and therefore serial release testing on each serial of FMD kits that will be used in routine testing is highly recommended. 1. INTRODUCTION International trade in animals and animal products is liberated and confidence in this global trade can increase only if appropriate control measures are applied. As foot-and-mouth disease (FMD) diagnostics play an essential role in this respect, the Food and Agriculture Organisation European Commission for the Control of Foot-and-Mouth Disease (EUFMD) co-ordinates, in collaboration with the European Commission, several programmes to increase the quality of FMD diagnostics. A quality assurance (QA) system is deemed essential for laboratories using FMD diagnostics aiming to (i) confirm an outbreak; (ii) perform a post-outbreak (sero)surveillance; (iii) test vaccine efficacy; (iv) perform a post-vaccination surveillance; (v) check vaccine matching; control import / export. The performance of laboratories participating in a National Accreditation System must be evaluated at least yearly by an Independent EQC body. The latter should be integrated in an international system such as the Organisation for Economic Co-operation and Development (OECD). Key elements in a world-wide quality system are first of all the possibility for all Veterinary Services to have access to FMD diagnosis through a National Reference Laboratory (NRL) or through an agreement with an OIE-FAO Regional Reference Laboratories (RRLs). Moreover, this is a prerequisite with a quality system for Veterinary Services. Furthermore, NRLs should implement an accreditation system with External Quality Control (EQC) and should take part in a Proficiency Testing (PT) scheme. A World PT should also be organised for OIE/FAO RRLs in which International

245

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 and Regional Organisations should play an important role as supervisors and financing source. The availability of International or regional reference material is essential for controlling and comparing test results. The series of devastating epizootics in the last decade increased the commercialisation of FMD diagnostics and therefore procedures for kit and batch control should be developed. 2. PROFICIENCY TESTING TO EVALUATE THE LABORATORY PERFORMANCE As there is a considerable variation in the organisation of PTs, a workshop was organised for the European NRLs for FMD and CSF with the participation of EC, FAO and OIE. The objective of the Workshop on Proficiency Testing was to make recommendations for Good Laboratory Practice concerning proficiency testing, in order to improve QA/QC of laboratory tests for FMD/SVD and CSF. Representatives of FMD/SVD and CSF National Reference Laboratories from the following countries were invited and have participated: Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxemburg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, United Kingdom (the EU member states) and Bosnia Herzegovina, Iceland, Norway, Israel, Turkey, Switzerland, Croatia, Albania, FYR of Macedonia, Serbia, Montenegro (non-EU but EUFMD member states). The workshop consisted of a number of presentations outlining the problems with regard to the organisation, participation, reporting, follow-up and feedback of proficiency testing followed by a discussion session and a conclusions and recommendations session. Following recommendations were formulated:

246

Need for Guidelines: 1. The Community Reference Laboratories for CSF and FMD/SVD will use the proposed ISO 17043/43 standard to guide the development of proficiency testing schemes for the organisers and participants. Use Guide 43 in conjunction with ILAC G13, until ISO17043 is formally published. The Community Reference Laboratories for CSF and FMD/SVD will develop a Standard Operating Procedure that outlines the proficiency testing scheme. This will not be too prescriptive and they will try to harmonise between FMD and CSF in first instance. 2. This document will be provided to the Community Reference Laboratories for other exotic viral diseases. Evalutation of Lab or Method: 1. The primary goal of Proficiency Testing schemes is to evaluate lab (individual test and/or test system) performance against assigned values (qualitative or quantitative); 2. Secondary purposes must be clearly separated in design and reporting; 3. The scope and purpose of the Proficiency Testing exercises need to be clearly stated in advance. Scope of Proficiency Test: 1. The scope of the Proficiency Test needs to be fit for purpose and the purpose should be clearly defined in advance; 2. The scope of the Proficiency Test (outbreak, surveillance, etc.) should be agreed and clearly communicated to the National Reference Laboratories prior to the start of the Proficiency Test. Statistical analysis: 1. Statistical guidelines need to be followed where relevant; 2. The design of PTS should be done so as to maximize the power of statistical analysis; 3. Both trueness and precision should, if possible, be addressed. Criteria for acceptance: 1. Criteria for lab conformity must be set PRIOR to the start of the Proficiency Test; 2. The scope will define whether the individual results and/or the complete test system is evaluated; 3. DG-Sanco should provide a list of contacts for the National Reference Laboratories. Advisory board: 1. An Advisory Board should be established at the beginning of each Proficiency Testing exercise 2. Membership (~4 persons) – but this should not be too prescriptive: Has to be appointed at the beginning of the Proficiency Test and communicated to all participants;

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Representatives should mainly be from participating labs; Statistical representation o participants; The Community Reference Laboratory should represent the EU opinion; A representative from regulatory bodies is not essential. Reporting: 1. Feedback: The Community Reference Laboratories should provide a draft report prior to the annual meeting; The final report should include feedback from Proficiency Test participants (questionnaire?); The report should include details of the scope, criteria, statistical approaches, findings, conclusions on non-conformities and recommendations. 2. Confidentiality: The consensus from the workshop was to maintain confidentiality between participants; Decoded results will be provided to DG Sanco (for EU Member States). 3. Follow up: Corrective actions to address non-conformities will be agreed between Community Reference Laboratory and National Reference Laboratory and communicated to DG SANCO. 4. To whom? To all the Proficiency Test participants and DG Sanco; Proficiency Test reports could be placed on a restricted access website.

3. PRODUCTION AND VALIDATION OF REFERENCE MATERIAL/REFERENCE STANDARDS It is clear that the key element in the validation of tests and in the comparison of test results among laboratories used for the control of transboundary diseases or for trade purposes is the availability of reference standards. At presence only 3 sera from cattle against FMD are international recognised by the OIE as reference standards. Weak positive and strong positive reference standards, as well serological as virological, against all serotypes (O, A, C, Asia1, SAT1, SAT2, SAT3) and within the serotypes against some important topotypes are urgently needed. Therefore several standards must be produced, tested by different laboratories and a final selection must be made. This is a huge task, which can only be performed by a network of laboratories through a ‘Collaborative Study’ with the specific aim of establishing international standards. Trying to get agreements on standards as a secondary goal of proficiency tests is not recommended. The standards agreed upon in the Collaborative study can then be considered as primary standards and distributed among other laboratories to calibrate their in house tests. Based on this calibrated test the laboratories should make their own secondary standards to use in routine diagnosis. The main difficulty in this respect is the fact that a lot of laboratories have not the animal facilities to produce secondary standards. Especially producing standards from infected animals to serve as secondary standards in e.g. the NSP tests is impossible for most European laboratories. Thus large quantities of secondary standards should be produced and made available to FMD laboratories. As well standards from vaccinated as from infected animals must be produced. To be recognised as international reference standards the production must fulfil international recognised criteria and specific information must be available as specified by the OIE. The standards should be similar in nature to those routinely tested by participants (fit for purpose). The importance of one species (cattle, pigs, sheep, goats) versus another depends on the region. For some laboratories it will be very important to have secondary standards from pigs while others need sheep sera. The development of standards from different species has, therefore, to be considered. Special attention is needed in providing evidence of the homogeneity and stability of the newly produced standards. Adequate packaging and transport methods to protect the stability and characteristics are essential taking into account the biosecurity hazards. 4. VALIDATION AND CERTIFICATION OF DIAGNOSTIC ASSAYS Recently, the OIE established an evaluation process for the Validation and Certification of Diagnostic Assays. A Standard Operating Procedure (SOP) was developed and is available to all kit producers. Producers have to enter their application form and the application will be subjected to formal evaluation by an expert panel appointed by the OIE, as detailed in the SOP. The applicant is highly recommended to consult the SOP and the following chapters in the OIE Manual (2008):

247

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Chapter 1.1.3., “Principles of validation of diagnostic assays for infectious diseases” and Chapter 1.1.4., “Validation and quality control of polymerase chain reaction methods used for the diagnosis of infectious diseases”. Although this OIE procedure is a major step forward in the harmonisation of a world-wide quality control system, some serious questions remain without answer such. Is the OIE certificate now internationally recognised or will individual countries maintain their own system of kit certification? If producers will have to certify their kits in all countries then there is no interest for them to go centrally to OIE. What will happen if a manufacturer does not enter an application to OIE for the certification of its kits and those kits are used for the recognition of a change in disease status? Is a validation by a NRL or RRL sufficient? In any case an independent evaluation is essential! 5. BATCH CONTROL In some parts of the world large difference in quality between batches of the same FMD diagnostic kit could occur. To guarantee a high grade of reliability, it is necessary to test (if possible partly by the producer) each serial (lot - batch) and to describe its capacities with regard of sensitivity and specificity. As a consequence, some countries established governmental regulations for batch quality control and require manufacturers to submit a report of the test results for each batch or lot of kits produced. Batches with satisfactory test results are approved for marketing. This licensing authority, however, may select random samples for confirmatory testing. In other countries, marketing is allowed upon certification that a batch was produced and monitored in accordance with a ‘good manufacturing practice’ marketing authorisation. Animal Health authorities may require additional testing of kits used in disease control programmes sponsored by the government. Generally, the purpose of such testing is to ensure that the performance of the kit is appropriate for the proposed use. Most countries require retention of samples from each batch of kits for future examination should problems arise.

248

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 45

AN ADVANCED FIELD DEPLOYABLE “PEN SIDE” SAMPLE PREPARATION AND PCR SYSTEM Doug Green1, C. Volpe2*, John Czajka1, Jason Betley2 and Jay Lewington2 1

Smiths Detection, 2202 Lakeside Blvd. Edgewood, MD 21040, USA Smiths Detection, 459 Park Avenue, Bushey, Watford, WD23 2BW, United Kingdom.

INTRODUCTION Exotic and Zoonotic diseases pose a threat to the world’s wildlife, commercial livestock and the population in general. Efforts to control the spread of naturally occurring highly infectious pathogens, for example Highly Pathogenic Avian Influenza, have been complicated by the need to transport samples to the lab for ultimate identification. This is especially problematic in remote locations. This abstract describes a briefcase sized portable sample preparation and PCR system, which has been designed, from the outset, with the field veterinarians needs and mode of operation in mind, including the ability to sanitise the unit with disinfectant. The system automatically purifies nucleic acids from a wide range of sample types and carries out PCR analysis, reporting either a positive or negative result or a strain level identification where applicable. The system uses a number of novel technologies and approaches to provide a fully automated portable on-site identification capability in a wide range of weather conditions, by a person with no knowledge of PCR. MATERIAL & METHODS Operation of the device is extremely simple. A veterinarian suspecting the presence of disease takes a sample from the animal. The nature of the sample is dependant on the disease under suspicion and is not limited by the instrument. For example they may take blood samples or vesicular tissue as appropriate. The veterinarian then places that sample in to a single use sample preparation device and places the device on the instrument. At this point the assay to be performed is automatically selected and the automated sample preparation process begins. The purified nucleic acid is automatically mixed with PCR reagents and the PCR process begins. The instrument then performs the appropriate data analysis and reports the result as a positive and negative test result for the test carried out, and reports the strain level identification of the pathogen where applicable. This entire process is performed with no user intervention and is designed to be performed by a person with no knowledge of molecular biology techniques. RESULTS The overall performance of the system depends on the individual performance of the instrument, sample preparation and assays. Recent analysis of the performance of the PCR instrument has shown a good correlation with selected lab based PCR instruments, indicating a similar level of performance can be expected in the field. The sample preparation device was designed to take the lab based process into the field and automate it. Initial results indicate that the device performs as well as the original bench process. There are currently 2 assays being developed and validated on this platform, one to detect all seven serotypes of Foot and Mouth Viruses in a single tube, and one to differentiate high and low pathogenic H5N1 Avian Influenza Virus. These assays are in development/validation using real samples in collaborating labs to determine their efficiency in a lab setting prior to their transfer to our field deployable platform (see abstracts in this volume). The platform uses a novel PCR chemistry called LATE-PCR which is able to identify many pathogens or strains of a pathogen in a single tube. The highly multiplexable nature of LATE-PCR means that a wide range of pathogens can be tested for simultaneously resulting in a simplified mode of operation. The system is currently being prepared for field based trials. DISCUSSIONS AND CONCLUSIONS

249

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 “Pen side” detection systems offer the promise of rapid detection of disease allowing a more resilient response and more effective outbreak management. The challenge in deploying PCR based analysis systems in the field is the requirement for a simple, sample preparation system capable of producing good quality nucleic acid from a wide range of animal samples. Systems such as Smiths Detection’s “Pen Side” testing system may provide the means to rapidly detection disease outbreaks. For more information please visit www.smithsdetection.com/vet or e-mail vets@smithsdetection.com

250

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 46 VALIDATION OF REAL-TIME RT-PCR: MATRIX EFFECT, UNCERTAINTY OF MEASUREMENT AND PRECISION N. Goris*,1, F. Vandenbussche1, J. Villers1, C. Herr2, Y. Van der Sted

and K. De Clercq1

Epizootic Diseases Section, Virology Department, Epizootic Diseases Section, Veterinary and Agrochemical Research Centre, Groeselenberg 99, 1180 Brussels, Belgium 2 Coordination Centre for Veterinary Diagnostics, Veterinary and Agrochemical Research Centre, Groeselenberg 99, 1180 Brussels, Belgium

ABSTRACT Real-time RT-PCRs are routinely being used for diagnosing foot-and-mouth disease virus (FMDV) [3]. Although most laboratories determine the analytical/diagnostic sensitivity and specificity, a thorough validation of the assay in terms of matrix effect, uncertainty of measurement and precision is generally not performed or reported. In this study, different FMDV-negative matrices were spiked with a 1:10 dilution series of FMDV SAT 1 ZIM 27/89. The effect of the matrix on the detection limit was assessed. Compared to cell-culture spiked viral control samples, no matrix effect of blood and foot epithelium on the detection limit was observed. An approximate 1 log10 reduction in detection limit was noted for faecal and tongue epithelium samples, whereas a 3 log10 decrease in detection signal was observed for spleen samples. By testing the same dilution series in duplicate on ten different occasions, an estimation of uncertainty of measurement and precision was obtained using blood as matrix. 1. INTRODUCTION Rapid and accurate diagnosis of veterinary diseases such foot-and-mouth disease (FMD) is of utmost importance given the extreme contagiousness of the causative virus (FMDV) and the severe socio-economic consequences related to FMD outbreaks in disease-free and disease-endemic areas alike [9]. Several FMDV serotype-independent, real-time RT-PCR (rRT-PCR) methods have been developed for this purpose [4, 13, 17-19, 22]. According to the World Organisation for Animal Health (OIE) recommendations on test validation and quality control of PCR methods for the diagnosis of infectious diseases [25], assay validation involves a four-step procedure. Part 1 consists of (i) optimising and standardising reagents and methods, (ii) estimating the intra- en inter-run variability (i.e. repeatability or precision), and (iii) determining the analytical sensitivity (ASn; detection limit) and specificity (ASp; cross-reactivity with similar pathogens) of the assay. During Part 2, the performance characteristics in terms of diagnostic sensitivity (DSn; proportion of known infected animals that test positive in the assay) and specificity (DSp; proportion of known uninfected animals that test negative in the assay) need to be investigated. In Part 3, the interlaboratory reproducibility of the assay is assessed by using an identical assay in at least three different laboratories on a fixed panel of minimum 20 samples. Part 4 is aimed at monitoring the validity of the assay performance using internal quality control (QC) and external proficiency testing schemes (PTS) [25]. In literature, validation reports for FMDV real-time RT-PCR methods are mostly restricted to Part 1 and Part 2 and more specifically to estimating (comparative) ASn, ASp, DSn and DSp. Rarely are parameters such as precision (component of Part 1), inter-laboratory reproducibility [7] (Part 3) and internal QC (element of Part 4) reported. Uncertainty of measurement (i.e. a parameter associated with the result of a measurement that characterises the dispersion of the values that could reasonably be attributed to the measurand [12]) [23] and the effect of the matrix are usually also not investigated, which are nonetheless requirements under ISO/IEC 17025. Therefore, this study is aimed at (i) assessing the influence of different matrices on the detection limit of the assay, (ii) estimating the uncertainty of measurement and (iii) quantifying the precision using the coefficient of variation (CV) method for two pan-reactive FMDV two-step real-time RT-PCR methods.

251

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 2. MATERIALS AND METHODS 2.1. Viruses and matrices The FMDV strain SAT1 ZIM 27/89 was kindly provided by the FAO FMD World Reference Laboratory (Pirbright, United Kingdom) and further cultured on swine kidney (SK6) cells. In total, six different FMDV-negative matrices were used. They include SK6 cell culture samples (for comparative purposes), bovine blood samples, porcine foot epithelium and tongue epithelium specimens, bovine spleen and porcine faecal samples. Each matrix was spiked with a 1:10 serial dilution series of FMDV strain SAT 1 ZIM 27/89 ranging from pure to 10-6. 2.2. Extraction, reverse transcription and PCR methods 2.2.1. Extraction Extraction was performed using the commercially available NucleoSpin RNA extraction columns according to the manufacturer’s instructions (Filter Service, Belgium). Spleen and faecal specimens were also pre-treated with 1,1,2-trichlorotrifluoroethan (Freon) to eliminate possible inhibitory and fatty substances. 2.2.2. Reverse transcription Reverse transcription (RT) was random-hexamer based and is described in Goris et al. [8]. The generated cDNA was subsequently used in three PCR reactions (FMDV 5’UTR, FMDV 3D and β-actin as internal control), each performed in a different well of a 96-well plate. 2.2.3. Real-time PCR FMDV 5’UTR The FMDV 5’UTR rRT-PCR was a modified version of the assay described by Reid et al. [18] as described in Ferris et al. [7]. The cut-off was set at a Ct-value of 45. 2.2.4. Real-time PCR FMDV 3D The FMDV 3D rRT-PCR was a modified version of the assay described by Callahan et al. [4]. The cut-off was set at a Ct-value of 45. 2.2.5. Real-time PCR internal β-actin control The β-actin rRT-PCR conditions were as previously described [24]. 2.3. Matrix effect The effect of the matrix on the detection limit or analytical sensitivity of both rRT-PCRs was determined by testing each matrix singly by the same technician on three different days. Mean Ctvalues and standard deviations were calculated. 2.4. Uncertainty of measurement and precision Extraction, RT and PCR on blood specimens spiked with a 1:10 serial dilution series of FMDV SAT1 ZIM 27/89 (100 – 10-6) were performed in duplo on 10 different days by three different technicians to include all likely sources of laboratory variation. Hence, 20 individual Ct-values were obtained for each viral dilution. The methods for measuring uncertainty and assessing intra- and inter-run variability (i.e. precision) of both FMDV rRT-PCRs were as previously described by Toussaint et al. [23] for competitive and indirect ELISAs. Briefly, the assays’ precision was determined for each viral dilution by random effect analysis of variance and expressed as CV using the procedure “NESTED” in SAS statistical software (SAS Institute Inc v9.2) [23]. The CV was calculated as follows: CV =

s tan dard _ deviation _ of _ replicates mean _ of _ replicates

(1)

The normality of the residuals was checked using the Shapiro-Wilk test [11]. The standard errors of the individual predicted Ct-values were subsequently transformed into confidence intervals around the corresponding observed Ct-values and into probabilities of being above or below the cut-off value using the normal law. Logistic regression models were finally deployed to interpolate a probability value for any possible result falling within the detection range of the assay The uncertainty about a positive test result was defined as the probability of not observing the same

252

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 qualitative test results (i.e. positive in this case) when retesting the same sample a second time. Ct-values above the cut-off Ct-value of 45 were arbitrarily set at 50. 3. RESULTS 3.1. Matrix effect Figures 1 and 2 depict the effect of the six different matrices on the detection limit of the assay. Compared to cell-culture spiked viral control samples, little or no matrix effect of blood and foot epithelium specimens on the detection limit was observed. An approximate 1 log10 reduction in detection limit was noted for faecal and tongue epithelium samples, whereas a 3 log10 decrease in detection signal was found for spleen samples. Freon treatment of spleen and faecal samples led to a 1 log10 increase in detection limit for the spleen samples, but no effect was observed for faecal specimens.

Ct value

40 35 30 25 20 10-1

10-2

10-3

10-4

10-5

10-6

Serial dilution series

Figure 1: The effect of different matrices on the detection limit of the FMDV 5’UTR rRT-PCR. The results obtained with spiked cell culture specimens are represented by open diamonds; those for spiked blood samples are given by filled diamonds; results of tongue and foot epithelium specimens are depicted by open and filled squares, respectively; Spiked spleen sample results with or without Freon treatment are given by open and filled circles, respectively, whereas, spiked faecal specimen results treated or not with Freon are shown by open and filled triangles, respectively.

Ct value

40 35 30 25 20 10-1

10-2

10-3

10-4

10-5

10-6

Serial dilution series

Figure 2: The effect of different matrices on the detection limit of the FMDV 3D rRT-PCR. The results obtained with spiked cell culture specimens are represented by open diamonds; those for spiked blood samples are given by filled diamonds; results of tongue and foot epithelium specimens are depicted by open and filled squares, respectively; Spiked spleen sample results with or without Freon treatment are given by open and filled circles, respectively, whereas, spiked faecal specimen results treated or not with Freon are shown by open and filled triangles, respectively.

253

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3.2. Uncertainty of measurement The uncertainty of measurement was assessed using spiked blood samples. The levels corresponding to the lowest concentrations of viral material (i.e. dilution 10-6), was non-normally distributed for both rRT-PCR methods. Individual probabilities of scoring positive were deduced from standard errors of the individual predicted values and the logistic model was used to interpolate the probabilities from the dataset to the entire range of possible test results. The probability estimations generated for both rRT-PCR methods are summarised in Table 1. Table 1: Results associated with probabilities of positive test result for the FMDV 5’UTR and FMDV 3D rRT-PCR method using blood as matrix. p value

FMDV 5’UTR assay positive result (Ct-value)

FMDV 3D assay positive result (Ct-value)

0.99 0.95 0.90 0.80 0.70 0.60 0.50 0.40

37.8 40.4 41.6 42.9 43.8 44.5 45.1 45.8

37.7 40.2 41.4 42.7 43.6 44.2 44.9 45.5

Every sample with a Ct-value smaller than 37.8 and 37.7 has a probability of at least 99% of scoring positive upon retest in the FMDV 5’UTR and 3D rRT-PCR assay, respectively (certainty). In other words, every sample with a Ct-value of 41.6 in the FMDV 5’UTR assay or 41.4 in the FMDV 3D assay has a probability of retesting negative of 10% (uncertainty) Interestingly, a sample scoring around the cut-off value of 45 Ct has exactly 50% probability of resulting in a positive test result if it were tested a second time. 3.3. Precision Table 2: represents the precision in terms of intra- and inter-run variability for the FMDV 5’UTR and 3D rRT-PCR. All four different CV% values obtained are below 5. The values of intra-run repeatability are slightly lower than those observed for inter-run repeatability. Table 2: The precision of the FMDV 5’UTR and 3D rRT-PCR assays. Precision Intra-run variability Inter-run variability

FMDV (CV%) 1.89 2.71

5’UTR

assay

FMDV 3D assay (CV %) 4.58 3.49

4. DISCUSSION Molecular techniques, such as real-time PCR, are replacing conventional techniques such as virus isolation (VI) and antigen-detection ELISA (Ag-ELISA) for FMDV diagnosis for several reasons, among which are the ease of automation [19-22] and rapidity of the results [4, 17, 19, 21-22]. Moreover, ASn of FMDV rRT-PCR exceeds that of conventional RT-PCRs [10] and Ag-ELISAs [18] and has been reported to be 1000 times higher than that of conventional virus culture methods (up to 1 tissue culture infective dose 50% per ml can be readily detected) [15, 17]. The superior ASn does not compromise the ASp of the assay which was estimated to be 100% as assessed by using specimens of viruses causing similar vesicular diseases such as bovine viral diarrhoea virus, swine vesicular disease virus and vesicular stomatitis virs [4, 17]. Furthermore, when necessary measures are taken to avoid cross-contamination and false positive results, 100% DSp can be obtained using rRT-PCR for diagnosing FMDV [15]. However, validation of molecular techniques, in terms of the OIE guidelines and accreditation to ISO/IEC 17025, is a relatively new concept. Consequently, the matrix effects, uncertainty of measurement and precision have not yet been reported for any of the assays described above.

254

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 From the data obtained using six different spiked matrices, it is concluded that spleen specimens are not the preferred matrix for FMDV diagnostic purposes given its negative effect on the assay’s detection limit. Even if both rRT-PCR assays detect the FMDV in the spiked spleen specimens, the ASn is nonetheless highly affected. Pre-treatment using Freon could, however, potentially increase the detection limit with 1 log10. A similar increase in ASn was not observed for spiked faecal samples, which shows that extraction procedures have to be standardised for each particular matrix. Ideally, this study should be performed using field samples or samples obtained from animals experimentally infected with FMDV, instead of using spiked materials. However, in the absence of such specimens, an initial estimate of the matrix effect can and should be obtained using spiked materials. Furthermore, the study could be extended to include other potentially interesting matrices such as milk [20] and probang samples [4, 17, 19]. In this study we have demonstrated that two pan-reactive rRT-PCR methods targeting different conserved regions of the FMDV genome are highly precise both within and between runs with CV% below 5. These values are comparable to highly normalised ELISAs as reported by Toussaint et al. [23] based on data generated by the solid-phase competition ELISA to detect antibodies against FMDV. This characteristic highlights the value of using rRT-PCR for FMDV diagnosis as the assay is capable of producing highly reproducible results. In contrast, Toussaint et al. [23] reported CV% of 13.4% and 21.7% for, respectively, within and between-run variability for indirect ELISA such as the Ag-ELISA. The method to quantify the uncertainty of measurement as developed for ELISA [23] is valid and applicable to rRT-PCR as well. The method provides a quantitative probability measure of getting the same outcome in qualitative terms of positive and negative on repeated test (e.g. 60% probability of classifying a test samples with a Ct-value of 44.2% in the FMDV 3D assay as positive upon retest using the same assay). As such, it is a useful tool for deciding on whether or not there is added value in repeating the test. It might in the above-given example be more desirable to select a second test for retesting or to request an additional sample before declaring the sample as positive for FMDV RNA. This decision will inevitably be risk based [6]. Depending on the circumstances and the risk associated with making a wrong decision (e.g. consequences of false declaring or not declaring a FMDV outbreak in a disease-free country or zone), the probability of an incorrect decision may be or may not be sufficiently small to justify the decision. This is particularly relevant for values just below or above the cut-off level or the limit of detection of the assay. Without further information (e.g. inclusion of positive and negative control samples to monitor the test performance; clinical symptoms etc), it is not possible to use the results to make a decision on compliance. It should, however, be understood that fixing the cut-off level of any assay should be done independently from measuring its uncertainty. A representative reference pool of known diseased and non-diseased animals will have to be tested for this purpose which shall result in a level of DSn and DSp [25]. If uncertainty were to be used to adapt the cut-off level (e.g. lower the cut-off until 99.9% certainty is obtained), a new uncertainty of measurement around this cut-off level will inevitable arise necessitating further adaptation. This would lead to a vicious circle of cut-off adaptation. Uncertainty of measurement is a tool to evaluate the risk associated with declaring a test sample as positive or negative and should not be seen as a means to standardise the standard operating procedure of the assay. As stated in the OIE guidelines for validation and quality control of PCR methods for the diagnosis of infectious diseases [25], assay standardisation is the very first step in the validation procedure and measuring of uncertainty should not be estimated prior to completing at least Parts 1 and 2 of the validation process. In conclusion, matrix effect, uncertainty of measurement and precision should also be estimated for other state-of-the art molecular and immunological techniques developed for FMDV detection such as nucleic acid sequence-based amplification (NASBA) [14], loop-mediated isothermal amplification (LAMP) [5], micro-arrays [1-2], and ligation assays proximity (PLA) [16]. 5. ACKNOWLEDGEMENTS The study was partly funded by the FMD_ImproCon project (www.fmdimprocon.org) under the Sixth Framework Programme (grant SSPE-CT-2003-503603). The authors would like to thank Dr. John Bashiruddin for useful comments on an earlier version of the paper. 6. REFERENCES

255

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [1] Banér, J., Gyarmati, P., Yacoub, A., Hakhverdyan, M., Stenberg, J., Ericsson, O., Nilsson, M., Landegren, U., Belák, S. 2007. Microarray-based molecular detection of foot-andmouth disease, vesicular stomatitis and swine vesicular disease viruses, using padlock probes. J Virol Methods 143(2): 200-206. [2] Baxi, M.K., Baxi, S., Clavijo, A., Burton, K.M., Deregt, D. 2006. Microarray-based detection and typing of foot-and-mouth disease virus. Vet J 172(3):473-481. [3] Belák, S. 2007. Experiences of an OIE Collaborating Centre in molecular diagnosis of transboundary animal diseases: a review. Dev Biol (Basel) 128: 103-112. [4] Callahan, J.D., Brown, F., Osorio, F.A., Sur, J.H., Kramer, E., Long, G.W., Lubroth, J., Ellis, S.J., Shoulars, K.S., Gaffney, K.L., Rock, D.L., Nelson, W.M. 2002. Use of a portable real-time reverse transcriptase-polymerase chain reaction assay for rapid detection of foot-andmouth disease virus. J Am Vet Med Assoc 220(11): 1636-1642. [5] Dukes, J.P., King, D.P., Alexandersen, S. 2006. Novel reverse transcription loop-mediated isothermal amplification for rapid detection of foot-and-mouth disease virus. Arch Virol 151(6): 1093-1106. [6] EUROCHEM / CITAC Guide. Use of uncertainty information in compliance assessment. First edition, 2007. [7] Ferris, N.P., King, D.P., Reid, S.M., Hutchings, G.H., Shaw, A.E., Paton, D.J., Goris, N., Haas, B., Hoffmann, B., Brocchi, E., Bugnetti, M., Dekker, A., De Clercq, K. 2006. Foot-andmouth disease virus: a first inter-laboratory comparison trial to evaluate virus isolation and RT-PCR detection methods. Vet Microbiol 31; 117(2-4):130-140. [8] Goris, N., De Palma, A., Toussaint, J.F., Musch, I., Neyts, J., De Clercq, K. 2007. 2'-Cmethylcytidine as a potent and selective inhibitor of the replication of foot-and-mouth disease virus. Antiviral Res 73(3):161-168. [9] Goris, N., Vandenbussche, F., De Clercq, K. 2008. Potential of antiviral therapy and prophylaxis for controlling RNA viral infections of livestock. Antiviral Res 78(1): 170-178. [10] Hearps, A., Zhang, Z., Alexandersen, S. 2002. Evaluation of the portable Cepheid SmartCycler real-time PCR machine for the rapid diagnosis of foot-and-mouth disease. Vet Rec 150(20): 625-628. [11] Henderson, A.R. 2006. Testing experimental data for univariate normality. Clin Chim Acta 366(1-2): 112-129. [12] ILAC-G17. Introducing the concept of uncertainty of measurement in testing in association with the application of the standard ISO/IEC 17025. Silverwater, 2002. [13] King, D.P., Ferris, N.P., Shaw, A.E., Reid, S.M., Hutchings, G.H., Giuffre, A.C., Robida, J.M., Callahan, J.D., Nelson, W.M., Beckham, T.R. 2006 Detection of foot-and-mouth disease virus: comparative diagnostic sensitivity of two independent real-time reverse transcription-polymerase chain reaction assays. J Vet Diagn Invest 18(1): 93-97. [14] Lau, L.T., Reid, S.M., King, D.P., Lau, A.M., Shaw, A.E., Ferris, N.P., Yu, A.C. 2008. Detection of foot-and-mouth disease virus by nucleic acid sequence-based amplification (NASBA). Vet Microbiol 126(1-3): 101-110. [15] Moonen, P., Boonstra, J., van der Honing, R.H., Leendertse, C.B., Jacobs, L., Dekker, A. 2003. Validation of a LightCycler-based reverse transcription polymerase chain reaction for the detection of foot-and-mouth disease virus. J Virol Methods 113(1): 35-41. [16] Nordengrahn, A., Gustafsdottir, S.M., Ebert, K., Reid, S.M., King, D.P., Ferris, N.P., Brocchi, E., Grazioli, S., Landegren, U., Merza, M. 2008. Evaluation of a novel proximity ligation assay for the sensitive and rapid detection of foot-and-mouth disease virus. Vet Microbiol 127(3-4): 227-236. [17] Oem, J.K., Kye, S.J., Lee, K.N., Kim, Y.J., Park, J.Y., Park, J.H., Joo, Y.S., Song, H.J. 2005. Development of a Lightcycler-based reverse transcription polymerase chain reaction for the detection of foot-and-mouth disease virus. J Vet Sci 6(3): 207-212. [18] Reid, S.M., Ferris, N.P., Hutchings, G.H., Zhang, Z., Belsham, G.J., Alexandersen, S. 2002. Detection of all seven serotypes of foot-and-mouth disease virus by real-time, fluorogenic reverse transcription polymerase chain reaction assay. J Virol Methods 105(1): 67-80. [19] Reid, S.M., Grierson, S.S., Ferris, N.P., Hutchings, G.H., Alexandersen, S. 2003. Evaluation of automated RT-PCR to accelerate the laboratory diagnosis of foot-and-mouth disease virus. J Virol Methods 107(2):129-139. [20] Reid, S.M., Parida, S., King, D.P., Hutchings, G.H., Shaw, A.E., Ferris, N.P., Zhang, Z., Hillerton, J.E., Paton, D.J. 2006. Utility of automated real-time RT-PCR for the detection of footand-mouth disease virus excreted in milk. Vet Res 37(1): 121-132. [21] Shaw, A.E., Reid, S.M., King, D.P., Hutchings, G.H., Ferris, N.P. 2004. Enhanced laboratory diagnosis of foot-and-mouth disease by real-time polymerase chain reaction. Rev Sci Tech 23(3): 1003-1009.

256

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [22] Shaw, A.E., Reid, S.M., Ebert, K., Hutchings, G.H., Ferris, N.P., King, D.P. 2007. Implementation of a one-step real-time RT-PCR protocol for diagnosis of foot-and-mouth disease. J Virol Methods 143(1): 81-85. [23] Toussaint, J.F., Assam, P., Caij, B., Dekeyser, F., Knapen, K., Imberechts, H., Goris, N., Molenberghs, G., Mintiens, K., De Clercq, K. 2007. Uncertainty of measurement for competitive and indirect ELISAs. Rev Sci Tech 26(3): 649-656. [24] Vandenbussche, F., Vanbinst, T., Verheyden, B., Van Dessel, W., Demeestere, L., Houdart, P., Bertels, G., Praet, N., Berkvens, D., Mintiens, K., Goris, N/, De Clercq, K. 2008. Evaluation of antibody-ELISA and real-time RT-PCR for the diagnosis and profiling of bluetongue virus serotype 8 during the epidemic in Belgium in 2006. Vet Microbiol 129(1-2): 1527. [25] World Organisation for Animal Health. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Chapter 1.1.5, Paris, 2008.

257

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

258

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 47

DEVELOPMENT OF SOLID PHASE COMPETITIVE ELISAS BASED ON MONOCLONAL ANTIBODIES FOR THE SEROLOGY OF FMDV SEROTYPES SAT1 AND SAT2 S. Grazioli1, E. Brocchi1*, V. Tranquillo1, S. Parida2 and D. Paton2 1

Istituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia Romagna, Brescia, Italy Pirbright Laboratory, Institute for Animal Health, Ash Rd, Woking, Surrey, GU24 0NF, UK

ABSTRACT In the spectrum of diagnostic tools for Foot-and-Mouth Disease (FMD) standardisation and validation of immunoassays for the serology of SAT serotypes has been relatively neglected. Because of the endemic presence of SAT strains in Africa and the risks due to market globalization, filling this gap should be a priority. Panels of characterised monoclonal antibodies (MAbs) specific for the serotypes SAT 1 and SAT 2 are available and there is much evidence to indicate that diagnostic ELISAs may benefit from the use of MAbs in terms of standardisation and specificity. Within the programme to complete and improve the sets of reagents and diagnostics for FMD, Solid Phase Competitive ELISAs (SPCE) for the assessment of antibodies to FMDV types SAT1 and SAT2 were developed. In these assays, specific antibodies in serum are detected due to their ability to compete with peroxidase-labelled neutralising MAbs, specific for FMDV SAT 1 (MAb 4C5) and SAT 2 (MAb 2H6) respectively and directed against linear sites of VP1. A large panel of known negative samples and a smaller panel of positive experimental sera were used for calibration of the assays. Specificity of 100% and 97.7% were recorded for the SAT 1 and SAT 2 antibody-detection ELISA respectively. A satisfactory type-specificity was also demonstrated by examining sera positive against heterologous FMDV serotypes. The diagnostic performances of the new ELISAs were then evaluated in comparison with results of VNT, using a wide range of samples collected in six herds with a known history of vaccination and infection with FMDV types SAT 1 and SAT 2. An optimal concordance between results of VNT and ELISA was found for the detection of anti-SAT 2 antibodies (k value = 0.87), while higher detection rates, a likely index of better sensitivity, were provided by SPCE compared with VNT for the detection of anti-SAT 1 antibodies. The results suggest that the developed assays have potential applications as rapid, simple and inexpensive tests in the serodiagnosis of FMD due to serotypes SAT 1 and SAT 2 and in serosurveillance programmes. 1. INTRODUCTION The desired global eradication of Foot-and-Mouth Disease (FMD) will require implementation of control measures in endemic areas and a better understanding of epidemiology. Until global eradication is achieved, countries ability to trade in livestock and animal products greatly relies on demonstration and maintenance of their FMD-free status. FMD virus exists in seven serotypes that are not uniformly distributed in the regions of the world where the disease still occurs. While serotypes O, A and C have occurred almost worldwide, the serotype Asia 1 is restricted to Asia (with an unique incursion into Europe in the year 2000) and the three serotypes SAT 1, 2 and 3 occur almost exclusively in Sub-Saharan Africa, with sporadic incursions of SAT 1 and SAT 2 into the Middle East and Maghreb countries (Rweyemamu et al., 2008). However, the presence and incidence of the disease is likely to be underestimated, especially in undeveloped regions, where surveillance intensity may not be sufficient to identify emergent infections and to prevent FMD epidemics. An intensification of surveillance in endemic countries would contribute to the reduction of the global risk, and the availability of reliable diagnostic tools, easily usable and accessible, plays a key role for the achievement of this objective. Then, improvement of diagnostic assays and filling of gaps still existing in the complex spectrum of diagnostic tools for FMD continue to be a priority; one major need is the availability of standardised and validated laboratory tests for SAT serotypes, with particular focus for serological assays.

259

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Currently, the principal methods used are the virus neutralisation test (VNT) and the liquid phase blocking ELISA (LPBE), which rely on specialised materials and are not simple tests to perform (OIE Manual, Office International des Epizooties, 2008a). A polyclonal antibody-based SPCE has been described (Sammin et al., 2007). Taking this need into account, this study describes the development and evaluation of Solid Phase Competitive ELISAs (SPCE) for the assessment of antibodies to FMDV types SAT1 and SAT2. Diagnostic ELISAs may benefit from the use of MAbs in terms of standardisation and specificity and panels of characterised monoclonal antibodies (MAbs) specific for these serotypes were recently produced (Grazioli et al., 2006). In the developed assays, specific antibodies in serum are detected due to their ability to compete with pre-selected, neutralising MAbs, labelled with peroxidase. A SPCE system for the detection of serotype-specific antibodies was chosen as it has replaced the LPBE for screening and surveillance programs, due to the evidence of a higher specificity combined with an equivalent sensitivity and a better robustness (MacKay et al., 2001, Anderson et al., 2003, Paiba et al., 2004). 2. MATERIALS AND METHODS 2.1 Serum samples All sera were of bovine origin. Negative sera A total of 1723 negative sera from healthy cattle were randomly selected from sera collected for other national serosurveillance programs. These included 80 samples from Belgium and 80 from Denmark. Experimental sera positive for FMDV type SAT 1. These included eight sera from four vaccinated cattle, collected after the first and the second vaccination, and samples from eight cattle infected with three different SAT 1 strains, sampled from 12 to 32 dpi. Experimental sera positive for FMDV type SAT 2. These included: 29 samples from 15 cattle used for SAT 2 vaccine potency testing, collected sequentially on two occasions (14 and 21 dpv); eight sera from four vaccinated cattle, collected after the first and the second vaccination; 26 samples from infected cattle, 15 of them challenged after vaccination. Four different SAT 2 strains were represented in this serum panel. Sera positive to other FMDV serotypes The type-specificity of the assay was evaluated using convalescent sera from either experimental or field animals, infected by FMDV serotypes O, A, C, Asia 1 and the heterologous SAT type. Post-outbreak sera Field, post-outbreak samples used in this study originated from 403 cattle sampled in six herds with varying vaccination status in Zimbabwe, from 1 to 5 months after outbreaks of type SAT 1 or SAT 2 (Table 3). Herd status, description and results of virological and serological investigations in these herds were previously published (Sammin et al., 2007). 2.2 Monoclonal antibodies (MAbs) MAbs used for preliminary feasibility studies and finally selected for the design of the developed ELISAs belong to panels produced at IZSLER, Brescia laboratory, The MAbs were previously characterized by epitope mapping (Grazioli et al., 2006). Monoclonal antibody-based Solid Phase Competitive ELISA (SPCE) The principle and the procedure of the MAb-based competitive ELISA, firstly described for the quantification of antibodies specific to the European subtypes O1, A5 and C1 of FMDV (Brocchi et al., 1990) and later applied to the measurement of anti-Asia 1 antibodies (Brocchi et al., 2004), was adapted to the detection of antibodies specific for FMDV serotypes SAT 1 and SAT 2. The MAbs selected are reported in the paragraph of results. Strains SAT1 Botswana (BOT) 1/68 and SAT2 Zimbabwe (ZIM) 5/81 were used as antigens in ELISA. Viruses were propagated in swine kidney cells (IBRS-2), culture supernatants were harvested when cytopathic effect was maximum (24h post-infection) and clarified at 3000 g for 20 min.; viruses were inactivated with 0.001 M binary ethyleneimine (BEI) for 48 h at 26°C and stored at -80°C. Working dilution of antigens and MAbs were predetermined by checkerboard titration; screening dilution for sera was 1/10. Results were

260

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 expressed as percentage inhibition generated by sera with respect to the non-inhibited reaction between the immune-captured antigen and a peroxidase-conjugated MAb. 2.3 OTHER SEROLOGICAL ASSAYS Virus Neutralisation Test (VNT) and Solid-phase Competitive ELISA (SPCE) using polyclonal sera were performed following the procedure described in the OIE Manual (Office International des Epizooties, 2008a). FMD virus strains used for VNT and SPCE were as previously described (Sammin et al. 2007). Titres of ≥ 1/45 in VNT and percentage inhibition values ≥ 60 in SPCE were regarded as positive results. 3. RESULTS FEASIBILITY STUDIES: TEST DESIGN AND SELECTION OF BEST SUITED MABS The assay design corresponds to a SPCE and implies the use of a capture MAb, coated to ELISA plates, to immune-purify the viral antigen from supernatants of infected cells and to present the trapped antigen to a second MAb, directly conjugated with peroxidase. Binding of the second antibody to the trapped virus is inhibited by positive sera, so it is essential that the conjugated MAb is directed against an immunogenic domain. Usually, this requirement is satisfied by neutralising MAbs, as epitopes involved in neutralisation of viral infectivity are surface-exposed and thus expected to be more immunogenic. For FMDV SAT 1, six neutralising MAbs representative of the different antigenic sites previously mapped (Grazioli et al., 2006) and one non-neutralising MAb were evaluated in all possible combinations as catching and conjugated MAbs; those providing strong and type-specific reactions with the homologous virus were subsequently evaluated in SPCE assays using a small panel of known negative and positive samples. Figure 1 shows the results provided by all sandwich combinations of the seven MAbs tested, while table 1 reports results of the prototype SPCEs performed with twelve selected MAbs pairs. These were obtained with six different catching MAbs combined with two different conjugated MAbs used as competitor; the conjugated MAb 4D3 was excluded from this analysis, yet providing a strong signal, since it does not neutralise virus infectivity. In spite of the low number of samples examined (34 negative and 4 positive sera), results indicated that a SPCE based on these MAbs is feasible. A clear differentiation in the distribution of negative and positive samples was achieved with most of the MAb combinations evaluated. In general, in all tests using MAb 3E4 as competitor percentage inhibition values generated by negative sera were below 30%, whilst in all tests using MAb 4C5 as competitor the distribution of negative sera was broader. In contrast, percentage inhibitions provided by positive sera were proportionally higher in SPCEs based on MAb 4C5 than in tests based on MAb 3E4. Best separation between positive and negative samples was achieved when using MAb 4C5 for both functions of catching and peroxidase-labelled conjugate; the combination obtained with the unique MAb 4C5 was then selected for further validation. Table 1: Feasibility studies for MAb-based SAT 1-SPCE: frequency distribution of percentage inhibitions generated by known sera in prototype SPCEs performed with different MAbs anti-SAT 1 MAbs combinations - catching MAb / conjugated MAb 4C5/ 4D3/ 4E3/ 4B1/ 4H8/ 3E4/ 4E3/ 4H8/ 3E4/3 4D3/ 4C5 4C5 4C5 4C5 4C5 4C5 3E4 3E4 E4 3E4 % N inhib. 2 0-10 3 114 20 213 30 311 40 4150

1 7

1 8

2 4

1 1

N 3 0

3 0

2 6

3 0

2 7

4B1/ 3E4

4C5/ 3E4

3 0

261

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 5160 6170 7180 8190 91100 TOTA 3 L 1 N = naive; I

1 1

3 3 4 2 1 = Infected 4

3 3 1

3 4

3 1

Feasibility studies of a SPCE for detection of antibody to FMDV SAT 2 followed a similar approach; however, since the available panel of MAbs specific for this serotype is less heterogeneous, only four MAbs were used for preliminary studies: these included the neutralising MAbs 3C5 and 2H6 recognising the G-H loop and the C-terminus of VP1 respectively (Grazioli et al., 2006) and two non-neutralising MAbs (1F5 and 2A8). A series of six different combinations of catching and conjugated MAbs were evaluated by testing a panel of naive cattle sera and positive samples from 22 experimentally infected cattle. Table 2 shows the frequency distribution of percentage inhibitions generated by negative and positive sera in the SPCEs performed with the different MAbs, as indicated. The MAb combination that best discriminated between negative and positive samples was obtained with MAb 1F5 as catching antibody and MAb 2H6, peroxidase-labelled, for the competitive step. Table 2: Feasibility studies for MAb-based SAT 2-SPCE: frequency distribution of percentage inhibitions generated by known sera in prototype SPCEs performed with different MAbs anti-SAT 2 MAbs combinations - catching MAb / conjugated MAb 1F5/2H6 1F5/3C5 2H6/1F5 2A8/2H6 3C5/1F5 2H6/2A8 % inhib. 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91100 TOTAL

N 3 19 35 24 15 5 1 1

N 5 3 20 15 15 8 3

1 1

20 103

20 69

N 3 3 6 15 14 10 6 8 4

14 19 37 24 18 39 33 38 22

1 1

1 1 20

22 69 22 245 22 N = naive; I = Infected

N 1 2 1 4 13 11 18 12 6

1 1

N 2 2 2 6 8 7 14 13 8

1 1 1

Definition of cut-off and diagnostic performances of SAT 1- and SAT 2-SPCEs using experimental samples Based on results of feasibility studies, SPCEs performed with MAb combinations 4C5/4C5 for antibody anti-SAT 1 and 1F5/2H6 for antibody anti-SAT 2 were further evaluated for the definition of optimal cut-off and preliminary estimates of diagnostic performances by testing increased numbers of negative and positive samples. The following sample categories were tested: For SAT 1-SPCE, 780 naive cattle sera representing a negative population and 16 positive samples, of which eight were from experimentally vaccinated cattle and eight from experimentally infected cattle; For SAT 2-SPCE , 1723 naïve cattle sera (most from Italy and a smaller proportion from two other EU countries, 37 sera from experimentally vaccinated animals (19 cattle each sampled two times) and 26 sera from experimentally infected cattle (20 of which challenged after vaccination).

262

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Figures 2 and 3 show frequency distribution of percentage inhibitions generated by the different categories of sera, examined at the screening dilution 1/10. Seventy percent inhibition represented the threshold that best discriminated between negative and positive populations in both tests. Setting the threshold at 70%, the SAT 1-SPCE scored diagnostic specificity of 100%, combined with a correct detection of the 16 positive sera. For the SAT 2-SPCE a specificity of 97.7% was estimated, with 40 false positives out of 1723 sera tested, mostly showing inhibition percentage just above the threshold. Sensitivity in infected cattle reached 100% but dropped to 84% in vaccinated cattle. However, sensitivity in vaccinated cattle is underestimated if compared with results of VNT: in fact SPCE detected 31 out of 37 samples from vaccinated cattle, whilst VNT identified only 28 samples positive, the majority of which showed titres close to the VNT threshold of 1/45 (not shown). Analytical specificity Infection with one FMDV serotype does not confer protection against the other six virus types; however, epitopes common to all serotypes do exist that may elicit cross-reactive antibodies. Cattle sera with high antibody titres against heterologous types were examined in the SAT 1 and SAT 2 SPCEs to evaluate the type-specificity of the assays developed. Results showed that in the SAT 1-ELISA 16 out of 85 heterologous sera reacted positive at the first screening dilution, with the majority just borderline; in the SAT 2-SPCE, 11 out of 106 heterologous sera produced a crossreactivity. In end-point titration analyses, differences found in heterologous and homologous titres were strongly significant. Comparative performances of MAb-based SPCE, VNT and polyclonal SPCE in post-outbreak surveillance As field animals may represent the target population for serological investigations, the performance of the new MAb-based ELISAs was evaluated on post-outbreak samples, originated from six herds, the FMD status of which is summarised in table 3, according to previously published results (Sammin et al., 2007). Table 3: Summary information on post-outbreak sera sampled in Zimbabwe Herd location

Herd code A Zimbabwe B north F C Zimbabwe D south E

N. sera 131 41 60 65 41 42

Virus detected SAT 2 SAT 2 ND SAT 1 SAT 1 SAT 1

Recent vaccination YES ? YES NO NO YES

Clinical disease YES YES NO YES YES YES

NSP seroprevalence + (91%) + (71%) + (20%) + (97%) + (41%) + (46%)

FMDV isolated from OP fluids Trivalent SAT1, SAT2, SAT 3 vaccine Outbreak reported 1 to 5 months earlier Results of MAb-based SPCE and VNT for both SAT 1 and SAT 2 FMDV serotypes were comparatively analysed on 403 field sera (table 4). For both serotypes MAb-based ELISA provided an overall higher detection rate (65% versus 61% for SAT 2 and 78% versus 63% for SAT 1). The analysis of concordant/discordant results showed a high level of concordance in the SAT 2 tests, with a k-value of 0.87 and 94% of samples (379 out of 403) providing concordant positive or concordant negative outcomes. When also results of the polyclonal SPCE were included in the comparison, the level of concordance among the three assays reached 92.3% (proportion of sera positive in three tests or negative in three tests). Differently, only 72% of sera gave concordant results in the MAb-based ELISA and VNT for SAT 1 antibodies: the majority of discordant cases were scored positive by ELISA and negative by VNT. The polyclonal SPCE detected much less positive sera than the other two tests, so that concordant results in the three assays further decreased to 53%.

263

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Table 4: Comparative results of MAbs-based SPCE and VNT in post-outbreak sera VNT

ELISA

402 samples Concordant 228 (72%)

Discordant 114 (28%)

VNT +

+ 227 87

SAT 2

ELISA

SAT 1

403 samples -

+ 243

Concordant 379 (94%)

136

Discordant 24 (6%)

To explore the origin of this discrepancy, the distribution of seroprevalence rates detected by MAbbased ELISA and VNT assays in each herd was analysed individually (figure 4). For SAT 2 antibodies, the concordance between the two assays was confirmed high in each herd. For SAT 1 antibodies, ELISA and VNT provided highly concordant results (similar seroprevalence rates) in herds C and D, in contrast to the remaining four herds (A, B, F and E), where detection rates recorded by ELISA were significantly higher compared to those found by VNT. The higher SAT 1 seroprevalence recorded by ELISA in herd E is consistent with demonstration of SAT 1 infection in the herd; however, in herds C and D, also infected with a SAT 1 virus, VNT and ELISA showed similar performances. Considering that VNT is likely to be more influenced by antigenicity of the strain used in the test than is ELISA, a possible explanation for this observation is that an inadequate matching between the virus strain used in the VNT and antibodies present in herd E may have affected VNT sensitivity. Interestingly, both VNT and ELISA showed a consistent type-specificity when applied to herds D and E; therefore, the uniform high seroprevalence for both SAT 1 and SAT 2 serotype recorded by either ELISA and VNT in herd C is suggestive of a double infection. Taking into consideration history of repeated vaccination in herds A, F, and possibly B, antibodies to SAT 1 detected in these herds may be the result of immune response to vaccines. Again, the apparently higher sensitivity of ELISA compared to VNT could be explained by a better matching between antigens used to perform tests and vaccine-induced antibodies present in the target population. 4. DISCUSSION The objective of this study was to develop new immunoassays for the measurement of antibodies to the FMDV serotypes SAT1 and SAT2 as part of the program to complete the spectrum of standardized and validated assays for FMD diagnosis. The model of tests developed is that of the solid phase competitive ELISA (MacKay et al., 2001, Anderson et al., 2003, Paiba et al., 2004) and the novelty of the approach was the substitution of two polyclonal sera (rabbit and guinea-pig) as capture and detector antibody with type-specific MAbs presenting desired properties. For both SAT1 and SAT 2 ELISAs, the MAbs selected for the competitive phase are directed against surfaceexposed, linear epitopes located along VP1 and able to elicit virus neutralising antibodies (Grazioli et al., 2006). Results reported in this study proved that the developed tests are reliable as single dilutionscreening assays; in addition, testing of serial dilution of all positive sera provided evidence that MAb-based SPCEs are also suited for antibody end-point titrations (not shown). According to the OIE guidelines for the validation of serological assays (Office International des Epizooties, 2008b), the evaluation of diagnostic performances should require testing of a large number of known positive and negative sera; however, the limited availability of experimental positive sera for the SATs serotypes needs consideration, as it may represent a weakness for the compliance with stringent validation criteria. Once optimal conditions were established for the SAT1- and SAT 2-SPCE, in particular the selection of the best MAb combinations and the screening dilution for test sera, the examination of a large population of negative sera and of a representative sample of experimental positive sera enabled the definition of the optimal cut-off as well as estimates of diagnostic specificity: with the cut-off fixed at 70% inhibition, specificity achieved 100% for the SAT 1-SPCE and 97.7% for the SAT 2SPCE. Serotype-specificity is an additional desired quality when serology against viral capsid proteins is adopted; on the other hand, cross-reactive antibodies elicited by epitopes common to all virus serotypes may affect this property. In the two MAb-based ELISAs, heterologous antibodies produced some borderline cross-reactivity in only a minority of cases and major differences between homologous and heterologous titres were observed. In the SPCEs, antigens are immune-

264

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 captured onto the solid-phase before reacting with sera: the binding of virus to a catching MAb should limit alteration of the capsid structure, so that mainly surface-exposed epitopes, known to be type-specific, are accessible for the reaction with serum antibodies (McCullough et al., 1985). The performance of the new ELISAs on testing of field samples was evaluated in comparison with the results of VNT, using a wide range of samples collected in six herds with known history of vaccination and infection with FMDV types SAT 1 and SAT 2 (Sammin et al., 2007). An optimal concordance between results of VNT and ELISA was found for the detection of anti-SAT 2 antibodies, either as an overall estimate on all samples derived from the six herds (k value = 0.87) irrespective of the serotype involved, and in the analysis of each herd individually. Furthermore, the absence of cross-reactivity of anti-SAT 1 positive samples (recorded in herds D and E ) in the SAT 2 test provided further evidence of the acceptable level of type-specificity. In contrast, the higher seroprevalence rate recorded by the SAT 1-SPCE in comparison to VNT in four of the six herds evaluated was unexpected. Sensitivity of the SPCE may exceed that of VNT in certain circumstances, such as when a considerable antigenic divergence occurs between the virus strain used to conduct the test and antibodies in tested samples. VNT performances maybe more affected by antigen-antibody mismatching than are ELISA performances. Further cases should be analysed to investigate this hypothesis. The VNT cut-off used here ensures maximal specificity for herd-based serosurveillance; a less stringent cut-off is used for certification of individual animals (International des Epizooties, 2008a). This study reports for the first time on the development and validation of serological assays for the SAT serotypes. The results suggest that the MAb-based SPCEs developed for the detection of antibodies to FMDV types SAT 1 and SAT 2 have potential applications as rapid, simple and inexpensive tests in the serodiagnosis of FMD and in serosurveillance programmes. Benefits conferred by the use of MAbs include homogenicity of the immunological reagents and potentially unlimited supply, improvement of specificity and standardization. In contrast, polyclonal antisera are heterogeneous, resulting in lack of standardisation; furthermore they are produced in limited volumes and may have unexpected cross-reactivity. Another major advantage over polyclonal-based ELISA is the replacement of three antisera (rabbit, guinea pig and anti-species immunoglobulins) with a unique or a combination of two MAbs. This greatly simplifies test performance and trouble-shooting. An initial validation process of the developed ELISAs provided evidence of adequate diagnostic performances, with substantial concordance and similar sensitivity with respect to VNT for the detection of anti-SAT 2 antibodies and a likely better sensitivity of the SPCE compared with VNT for the detection of anti-SAT 1 antibodies. Given the absence of validated tests for the serology of SAT serotypes to be used as gold standard, diagnostic sensitivity and specificity of the new ELISAs will be further evaluated also by Bayesian analyses conducted on results obtained with field samples. Further progress in the validation process of the tests developed should include the production of International Standard sera for FMDV SAT serotypes, the study of the influence that strains with different antigenicity may have on serological assays performances and the organisation of interlaboratories trials. 5. REFERENCES [1] Anderson, J., Corteyn, M., Hamblin, P. & Paton, D. 2003. Further validation of the solidphase competitive ELISA for FMDV types a, C & Asia 1. Report of Sess. Res. Gr. St. Tech. Committee of the European Commission for the Control of Foot-and-Mouth Disease. Berne, Switzerland, September 16-19 2003. Appendix 24: 157-165. [2] Brocchi E, Grazioli S, Yadin H, De Simone F. Validation of a Solid Phase Competitive ELISA (SPBE) based on the use a single neutralising monoclonal antibody for the measurement of antibodies to FMDV type Asia 1. Report of Sess. Res. Gr. St. Tech. Committee of the European Commission for the Control of Foot-and-Mouth Diseas, Chania, Crete, Greece, 11-15 October 2004. Appendix 45, 288-297. [3] Brocchi, E., De Simone, F., Bugnetti, M., Gamba, D. &Capucci, L. 1990. Application of a monoclonal antibody-based competition ELISA to the measurement of anti-FMDV antibodies in animal sera. Report of Sess. Res. Gr. St. Tech. Committee of the European Commission for the Control of Foot-and-Mouth Disease, Lindholm, Denmark, 25-29 June 1990. p 83-88. [4] Grazioli, S., Moretti, M., Barbieri I., Crosatti ML., & Brocchi, E. 2006. Use of monoclonal antibodies to identify and map new antigenic determinants involved in neutralization on FMD viruses type SAT1 and SAT2. Report of Sess. Research Group St. Tech. Committee of the European

265

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Commission for the Control of Foot-and-Mouth Disease, Cyprus 17-20 October 2006. Appendix 43, p 289-300. [5] MacKay, D.K.J., Naci Bulut, A., Rendle, T., Davidson, F. & Ferris, N.P. 2001. A solidphase competition ELISA for measuring antibody to a foot-and-mouth disease virus. Journal of Virological Methods 97:33-48. [6] McCullough, K.C., Crowther, J.R., Butcher, R.N. 1985. Alteration in antibody reactivity with foot-and-mouth disease virus (FMDV) 146S antigen before and after binding to a solid phase or complexing with specific antibody. J Immunol Methods 3: 91-100. [7] Office International des Epizooties. 2008a. Foot-and-mouth Disease. In: OIE Terrestrial Manual, 2008 sixth Edition. OIE, Paris, France, p 190-216. [8] Office International des Epizooties. 2008b. Principles of validation of diagnostic assays for infectious diseases. In: OIE Terrestrial Manual, 2008 sixth Edition. OIE, Paris, France, p 34-45. [9] Paiba, G.A., Anderson, J. ,Paton, D.J., Soldan, A.W., Alexandersen, S., Corteyn, M., Wilsden, G., Hamblin, P., MacKay, D.K.J. & Donalson, A.I. 2004. Validation of a foot-andmouth disease antibody screening solid-phase competition ELISA (SPCE). Journal of Immunological Methods 115: 145-158. [10] Rweyemamu M., Roeder P., Mackay D., Sumption K., Brownlie J., Leforban Y., Valarcher JF., Knowles NJ. & Saraiva V. 2008. Epidemiological patterns of foot-and-mouth disease worldwide. Transbound Emerg Dis. 55 (1): 57-72. [11] Sammin, D.J., Paton, D.J., Parida, S., Ferris, N.P., Hutchings, G.H., Reid, S.M., Shaw, A.E., Holmes, C., Gibson, D., Corteyn, M., Knowles, N.J., Valarcher, J-F., Hamblin, P.A., Fleming, L., Gwaze, G., Sumption, K.J. Evaluation of laboratory tests for SAT serotypes of footand-mouth disease virus with specimens collected from convalescent cattle in Zimbabwe. 2007. Veterinary Record 160: 647-654. 6. ACKNOWLEDGMENTS The study was supported by the EU project LABONSITE, grant SSP-513645 and by National Grant from Ministry of Health PRC2003/015. The authors wish to thank Donal Sammin for collection of field sera, Bernd Haas and Wilna Vosloo for provision of some experimental samples, P. Hamblin, D. Gamba and M. Scaramuzza for their reliable and precious technical contribution. Figure 1: Selection of ideal combinations of catching and conjugated MAbs for the SAT 1-SPCE

anti-SAT 1 - catching MAb

anti-SAT 1 - conjugated MAb Epitope location

MAb

VP1 146-148

4C5

VP2 72+VP1 181

3E4

VP3 135

4B1

VP1 111

4H8

VP3 76/135+VP1 179/181

4E3

VP1 154

4B9

Non neutral

4D3

4C5

3E4

4B1

Black cells: OD ≥ 2; grey cells: OD >1 - <2; white cells: OD ≤ 1

266

4E3

4B9

4H8

4D3

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 2: SAT 1-SPCE: frequency distribution of percentage inhibitions generated by naive and infected samples

Figure 3: SAT 2-SPCE: frequency distribution of percentage inhibitions generated by different sample categories

348

349 284

234

400 179

166 123

N. sera

300

200 2

100 1

Naive

2 9

Inf. Vacc.

% inhibition

267

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 4: Distribution of seroprevalence rates detected by MAb-based SPCE and VNT in six herds (A, B, C, D, E, and F) after outbreaks of FMDV types SAT 1 and/or SAT 2.

Seroprevalence rates (%)

120 100

SAT

80 60

VNT 40 ELISA

20 0 A

Seroprevalence rates (%)

120 100

SAT 2

80 60

VNT

ELISA

20 0 A

268

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 48

FULL GENOME SEQUENCING TO SUPPORT TRACING OF UK OUTBREAKS OF FM N. J. Knowles1*, E. M. Cottam1, J. Wadsworth1, K. Ebert1, D. P. King1 and D. J. Paton 1

Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, United Kingdom.

INTRODUCTION Identifying sources of Foot-and-mouth Disease (FMD) outbreaks is often confounded by incomplete epidemiological evidence and the numerous routes by which virus can spread (movements of infected animals or their products, contaminated persons, objects and aerosols). The 2007 FMD epidemic in the United Kingdom was characterised by the emergence of two temporally and spatially distinct outbreak clusters. Eight infected premises were identified, two in the west of the county of Surrey in August and six in the north of Surrey in September. Nucleotide sequencing of viruses recovered from field cases played an important role during these outbreaks. MATERIALS AND METHODS Complete genomes were amplified from RNA extracted from suspensions of epithelial lesion material using 24 sets of FMDV O1 BFS 1860-specific primers which had M13 sequences added to their 5′ ends. Each overlapping amplicon was directly sequenced using both FMDV-specific and M13 sequencing primers. RESULTS Within 24 hours of the first case, FMDV sequence data were obtained showing that the outbreak virus had a VP1 gene-identity of 99.8% to FMDV O1 British Field Sample 1860, a widely used reference and vaccine strain, originally derived from an outbreak of FMD in the United Kingdom in 1967. Complete genome sequencing of laboratory viruses and representative isolates from all eight infected premises enabled a phylogeny to be reconstructed. This demonstrated a genetic link between the two epidemiological outbreak clusters suggesting virus transfer from a farm in the Normandy area to north Surrey, rather than a second laboratory escape. DISCUSSION RT-PCR protocols were developed to allow the complete genomes of FMDV to be sequenced within 24-48 hours of sample receipt for the subsequent cases. These data were used to determine the most likely source of the outbreak and to track FMDV movement from farm-to-farm in real-time, assisting the field epidemiological investigations.

269

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

270

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 49

SERO-SURVEILLANCE AGAINST FOOT-AND-MOUTH DISEASE VIRUS (FMDV) NONSTRUCTURAL PROTEIN ANTIBODIES IN SHEEP, GOATS AND CATTLE IN JORDAN AFTER 2006 OUTBREAK S. Amareen1, 2, P. Grainger1, L. Fleming1, M. Mahapatra1, H. Khalil2, I. Bani Younis2, A. Tahaineh3, D. Paton, F. Aldomy2 and S. Parida1* 1

Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey, GU24 0NF, UK. 2 Jordan Bio-Industries Centre (JOVAC), Amman, 11941, Jordan 3 Veterinary Department, Ministry of agriculture, Amman, Jordan.

1. INTRODUCTION A field study during the month of February-March, 2007 was conducted in Jordan under a collaborative project between IAH and Jovac. The primary aim of this study was to evaluate the performance of non-structural serological tests (Cedi, Bommeli and UBI) in field for the detection of O and A serotype of foot-and-mouth disease (FMD) virus infection, particularly in vaccinated and infected sheep and goats and specifically in unnoticed FMDV infected sheep and goat farms. Secondary aims were to examine NSP seroconversion rates in sheep, goat and cattle that had been exposed to infection and to find out NSP tests specificity in field situations. 2. MATERIALS AND METHODS Clinical specimens were collected from 409 sheep and lambs, 91 goats and 258 cattle in 15 herds in different provinces of Jordan, in which the history of FMD vaccination was known. Four vaccinated sheep and goat herds, three vaccinated cattle herds and one vaccinated mixed herds for sheep, goat and cattle had reported outbreaks of disease two to three months previously. A 5th vaccinated herd containing sheep and goats (n=107) where lesions were not found were also sampled. 122 sheep and goats were also randomly sampled from a sixth farm where vaccination was not applied for several years. Out of total 258 cattle, 160 cattle were sampled from 4 vaccinated farms where clinical lesions were observed. Further, 98 cattle from five vaccinated herds from no out break area were also sampled to find out the tests specificity. A monovalent vaccine (O1 Manisa) had been applied in above mentioned 14 farms. 3. RESULTS AND CONCLUSION NSP antibody tests provided evidence of FMDV infection in all eight known infected herds as well as in one herd where FMDV infection was unnoticed. The overall sero-prevalence for sheep and goats varied with different NSP tests from 13% to 30% and 73% to 94% in vaccinated and unvaccinated animals, respectively. Similarly, the overall sero-prevalence for vaccinated cattle varied with different NSP tests from 30% to 82% with a tests specificity of 92% to 97%. 4. INTRODUCTION Though available NSP antibody tests were validated in Brescia, Italy under the scope of an EU funded International research group, a consortium of European reference laboratories (Brocchi et al., 2006), it was mainly based on serum samples obtained from vaccinated and infected cattle. Very few serum samples from sheep, goats and pigs were used in this validation process. It has been advocated that another NSP antibody test validation is necessary for sheep as well as for Pigs. Recently we have reported the efficiency of NSP tests in vaccinated challenged sheep and pigs in experimental condition (Parida et al., 2007 and 2008).Therefore the primary aim of this study was to evaluate the performance of non-structural serological tests (Cedi, Bommeli and UBI) in field for the detection of O and A serotype of foot-and-mouth disease (FMD) virus infection, particularly in vaccinated and infected sheep and goats and specifically in unnoticed FMDV infected sheep and goat farms. Secondary aims were to examine NSP seroconversion rates in sheep, goat and cattle that had been exposed to infection and to find out NSP tests specificity in field situations.

271

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 5. MATERIALS AND METHODS During February and March 2007, a serosurveillance was conducted in different provinces (Fig. 1) with the help of a Jordanian vaccine company, Jovac, based in Amman, Jordan. Clinical specimens were collected from 409 sheep and lambs, 91 goats and 258 cattle in 15 herds in different provinces of Jordan, in which the history of FMD vaccination was known. Four vaccinated sheep and goat herds, three vaccinated cattle herds and one vaccinated mixed herds for sheep, goat and cattle had reported outbreaks of disease two to three months previously. A 5th vaccinated herd containing sheep and goats (n=107) where lesions were not found were also sampled. 122 sheep and goats were also randomly sampled from a sixth farm where vaccination was not applied for several years. Out of total 258 cattle, 160 cattle were sampled from 4 vaccinated farms where clinical lesions were observed whereas 98 cattle from five vaccinated herds were sampled from no out break area to find out the tests specificity. A monovalent vaccine (O1 Manisa) had been applied in above mentioned 14 farms. Blood, saliva and probang samples were collected from each of the animals. Due to a delay at the airport, probang samples received were unsuitable for virus isolation, but are being examined by RT-PCR. Serum and saliva samples were analysed for NSP antibodies by Cedi NSP test (Ceditest® FMDV-NS originally from Cedi Diagnostics, now Prionics B.V. Lelystad, The Netherlands) and IgA test respectively and this work was described in the 2007 report. During 2008, further analyses of serum samples were conducted for detection of NSP antibodies by Bommeli (CHEKIT-FMD-3ABC, Bommeli Diagnostics, Bern, Switzerlandand) and UBI (UBI® FMDV NS ELISA, United Biomedical Inc., New York, USA) tests and the results were compared to the original findings from use of the Cedi test. 6. RESULTS AND DISCUSSION: The main objectives of this work were to evaluate the performance of available commercial NSP tests to detect infection with O and A serotypes of FMD in cattle, sheep and goats in the field and to establish seroconversion rates in vaccinated sheep and goats. Manifestations of FMD in sheep and goats are often mild and there is considerable danger that infection will be missed, even where serology is carried out, if only few animals are infected with limited virus replication. Detection of NSP antibodies in 160 cattle serum samples from 4 vaccinated infected farms revealed that 82%, 31% and 54% of the samples were scored positive by Cedi, Bommeli and UBI tests respectively. The proportion of seropositive animals found in apparently infected flocks was lower in vaccinated sheep and goats than in vaccinated cattle herds. Analysing 271 sheep and goat serum samples from 5 vaccinated and presumed infected farms, the proportion of seropositive animals was 30%, 20% and 13% by Cedi, Bommeli and UBI tests. As seen earlier in unvaccinated cattle, the proportion of NSP antibody seropositive animals in unvaccinated sheep and goat flocks was higher than amongst vaccinated animals. Out of 122 random serum samples from sheep and goats from an unvaccinated infected herd 95%, 73% and 75% animals were seropositive for NSP antibodies and were detected by Cedi, Bommeli and UBI tests. Detection of NSP antibodies in 107 sheep and goat serum samples from one vaccinated farm where FMD infection was not noticed revealed 49%, 6% and 4% NSP serocopositive animals in Cedi, Bommeli and UBI tests respectively. During this study in Jordan, NSP test specificity was also estimated from analsysi of sera collected from 98 vaccinated cattle from 5 vaccinated and apparently uninfected farms in areas where disease had not been reported. This revealed 92%, 97% and 96% specificity in Cedi, Bommeli and UBI tests respectively. For specificity estimation of these NSP tests in sheep and goats, 100 vaccinated sheep and goats were sampled from unaffected areas of Jordan during October 2008, and analysis of these samples is ongoing.

272

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 1: Arrows showed the place of sample collection in Jordan after 2006 FMD outbreak 7. CONCLUSION

NSP antibody tests provided evidence of FMDV infection in all eight known vaccinated infected herds, one unvaccinated herd and in one herd where FMDV infection has not been noticed. The overall sero-prevalence for sheep and goats varied with different NSP tests from 13% to 30% and 73% to 95% in vaccinated and unvaccinated animals, respectively. Similarly, the overall sero-prevalence for vaccinated cattle varied with different NSP tests from 31% to 82% with specificity ranging from 92% to 97%.

8. RECOMMENDATION

NSP antibody tests are useful to detect infection in vaccinated sheep.

9. ACKNOWLEDGEMENT Authors like to thank Dr Faisal A. Dayim, Head, Jordanian Vaccine Centre and Prof. Martin Shirley, Director, IAH for facilitating the collaborative work. Authors also like to thank Dr.Wafaa Ramadneh and Dr Oh D Saddam Faig Abdalghani BinTarif for helping in sample collection at Jordan. Sample collection has been funded by Jordanian Vaccine Centre (JOVAC) and Sample analysis work has been funded by the UK Department for Environment Food and Rural Affairs through grant SE 1122 and FMD Improcon project of the EU 6th Framework Programme, SSPE-CT-2003-503603. DP is a Jenner fellow and SP is an adjunct professor to Murdoch University, Australia.

10. REFERENCES [1] Brocchi, E., Bergmann, I. E., Dekker, A., Paton, D. J., Sammin, D. J., Greiner, M., Grazioli, S., De Simone, F., Yadin, H., Haas, B., Bulut, N., Malirat, V., Neitzert, E., Goris, N., Parida, S., Sorensen, K. & De Clercq, K. 2006. Comparative evaluation of six ELISAs for the detection of antibodies to the non-structural proteins of foot-and-mouth disease virus. Vaccine 24: 6966-6979. [2] Parida s, Fleming l, Oh Y, Mahapatra M, Hamblin P, Gloster J, Doel C, Gubbins S and Paton D J (2007). Reduction of foot-and-mouth disease (FMD) virus load in nasal excretions,

273

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 saliva and exhaled air of vaccinated pigs following direct contact challenge. Vaccine, 25 (45), 78067817. [3] Parida S , Fleming L, Oh Y, Mahapatra M, Hamblin P, Gloster J and Paton D J. (2008)Emergency vaccination of sheep against foot-and-mouth disease: significance and detection of subsequent sub-clinical infection. Vaccine 26, 3469-3479.

274

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 50

ANTICIPATORY CONTROL MEASURES: GEOGRAPHICAL INFORMATION SYSTEMS-BASED IDENTIFICATION OF TOPOGRAPHIC FACTORS ACTING AS OBSTACLES OR DISSEMINATORS IN THE 2001 URUGUAYAN FMD EPIDEMICS A. L. Rivas1, 2, A. L. Hoogesteyn3, S. J. Schwager2 and K.L. Anderson1 1 Population Health & Pathobiology, College of Veterinary Medicine, NCSU, Raleigh, NC, USA; Biological Statistics and Computational Biology, College of Agriculture & Life Sciences, Cornell University, Ithaca, NY, USA; 3Centro de Investigación de Estudios Avanzados, Mérida, Mexico E-mail: alrivas@ncsu.edu or alr4@cornell.edu

ABSTRACT The individual and collective roles of spatial variables (rivers, roads, and farms) as facilitators or obstacles of FMD dissemination were explored. Based on geo-referenced data of the first 11 weeks of the 2001 Uruguayan FMD epidemic and using simple descriptive statistical analysis, local territorial units (counties or municipalities, n=275) were categorized as either putative epidemic facilitator (F) or non-facilitator (NF). In the first 9 of the 11 epidemic weeks, F counties displayed a higher number of case density/sq km than NF counties. Case density/sq km peaked at week 2 in F counties (3 weeks before the national epidemic peak), when these counties displayed 238% more cases/sq km than NF counties. The epidemic ceased soon after F counties showed a lower case density than NF counties. Because the possible role as epidemic facilitator or non-facilitator of local geographic units can be explored prior to the emergence of epidemics, anticipatory assessment of, and preventive planning based on, F-like sites is recommended. 1. INTRODUCTION Geography has long been recognized to play a role in epidemics. For instance, islands have been recognized, for several centuries, to act as a “barrier” against the introduction of epidemics. If “barriers” can prevent disease dispersal, then geographical “facilitators” should have the opposite effect. If geographical “barriers” or “facilitators” could be identified, preventive measures could focus not only on the susceptible population but, in addition, on the geographical factors that promote (or block) disease spread. Hence, the question of interest is: can “obstacles” and “facilitators” of disease dissemination be identified before FMD outbreaks occur? Because epidemiology is not an experimental science, only retrospective data can provide tentative answers. One such opportunity may be provided by the 2001 Uruguayan FMD epidemics. That may be so because (a) the bovine was the species primarily affected in that epidemic (unlike other epidemics where the host was not symptomatic, the 2001Uruguayan epidemic was observable since its outset), and (b) in 2001 Uruguay had already geo-referenced all its rural parcels (Rivas et al., 2003). Therefore, the geo-temporal progression of this epidemic can be analyzed. However, there is no known information on whether a particular geographical entity acts as an obstacle or facilitator of epidemic dispersal. It is plausible to postulate that a given factor (e.g., rivers) may possess either role (if not both roles). For instance, because water is an obligatory requirement for any agricultural activity, it follows that the presence of rivers in a territorial unit may be associated with epidemics. However, if the magnitude of river density is too high and/or if the boundaries of a territorial unit are predominantly composed of rivers, rivers may act as obstacles to epidemic dispersal. In the most extreme case (when 100% of the borders of a territorial unit are composed of rivers), such territory becomes insulated. Therefore, no universal hypothesis can be postulated on whether a given geographical factor is a facilitator (or obstacle) of epidemic dispersal. In addition, the critical values within which a geographical variable may perform a certain role need to be determined. To determine both the role of a specific geographical variable (if any) and the critical values that influence epidemic dispersal, specific geographical contexts need to be explored. Here we asked whether 3 spatial variables (rivers, roads, and farm density), alone or together acted as either “obstacles” or “facilitators” of the 2001 Uruguayan FMD epidemic. This study

275

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 pursued two goals: 1) to generate data-based hypotheses on the possible role of specific Uruguayan municipalities (counties) on FMD dispersal, and 2) to assess, over time, the case density (FMD+ farms/sq km) of counties regarded as epidemic facilitators (“F”) or non-facilitators (“NF”). 2. MATERIALS AND METHODS Correlations among county spatial variables (county area, county number of farms, road density and river density [km of length/county area], and percentage of county perimeter composed of roads or rivers) were assessed in 275 Uruguayan counties affected, in 2001, by a FMD epidemic (Chowell et al., 2006). Relationships including 2 and 3 variables were explored where inflection points, if they were observed, were suspected to indicate different effects on epidemic dispersal. Data on rivers, roads, and farms (Fig. 1) were collected from public sources, and geo-referenced using Geographical Information Systems (Rivas et al., 2003). FMD+ cases were reported as case density (FMD+ farms/sq km). Based on descriptive analysis, territorial units (counties) were classified as either “facilitators” (F) or “non-facilitators” (NF) of FMD diffusion. Fig. 1: Spatial variables investigated for their possible role as epidemic facilitators or no facilitators. County river length/county perimeter (km/km). County road density (km/sq km). County farm density: count of farms/county area.

3. RESULTS Positive correlations were found between county farm density and county road density, county farm density and county road length, and county road density vs. county road length (r ≥ 0.34, P < 0.01), as well as between river density and river length (r=0.13, P=0.03). Negative correlations were found between county area vs. farm density, and road length vs. river length (r =–0.37, P <0.01). Hence, farm density, road density, and road length were considered as possible facilitators (F), whereas river density and river length were suspected to act as non-facilitators (NF). Supported by three-dimensional plots (Fig. 2), the tentative cut-off points that differentiated F (n=31) from NF (n=244) counties were determined (Table 2). This information was then applied to explore the FMD case density in counties regarded as either F or NF (Fig. 2).

276

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Facilitators Non-facilitators

Farm density

0.75 0.50 0.25 0.00 0.0

River length

0.5 1.0 0.15

0.00

0.30

0.45

Road density

Fig. 2: Relationships among county-based spatial variables. Farm density (number of farms/sq km), river length as percent of county perimeter (km of rivers/km of county perimeter), and road density (km of roads/sq km) were determined for every county (n= 275). Counties suspected to act as epidemic facilitators were characterized by (a) road density > 0.1, (b) > 0.28 but < .65 county perimeters composed of rivers, and (c) farm density >0.12. In the first 9 epidemic weeks, facilitator counties displayed a higher number of case density, which peaked at week 2 when such counties displayed 2.4 times more cases/sq km than NF counties (Fig. 3). Once F counties showed a lower case density than NF ones, the epidemic ceased.

0.006

Facilitators Non-facilitators

County cases/sq km

0.005 0.004 0.003 0.002 0.001 0.000 1

Epidemic week Fig. 3: Temporal number of FMD+ infected farms (cases/sq km) in counties suspected to be epidemic “facilitators” and “non-facilitators.”

277

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 4. DISCUSSION Because the epidemic ceased soon after the case density of F counties approached zero, the hypothesis that the variables here assumed to act as either F or NF was not negated. Because the role of geographic units as possible epidemic facilitators or obstacles can be identified in the absence of epidemics, anticipatory assessments of geographical units are recommended. 5. REFERENCES [1] Rivas AL, Smith S, Sullivan PJ et al. Amer J Vet Res 64: 1519-1527, 2003. [2] Chowell G, Rivas AL, Smith SD et al. Amer J Vet Res 67: 102-113, 2006.

278

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 51

NETWORK ANALYSIS OF LIVESTOCK MOVEMENTS TO ESTIMATE POTENTIAL SILENT SPREAD OF FOOT-AND-MOUTH DISEASE C. Dubé1,*, C. Ribble 2, D. Kelton3, B. McNab 4, S. Javier1 and A. Rivera5 2

1 Canadian Food Inspection Agency, 59 Camelot, Ottawa, Ontario, K1A 0Y9. Department of Ecosystem and Public Health Faculty of Veterinary Medicine, University of Calgary 3 Department of Population Medicine, Ontario Veterinary College, University of Guelph 4 Ontario Ministry of Agriculture, Food and Rural Affairs 5 Servicio Agrícola y Ganadero – SAG, Ministerio de Agricultura, Chile

ABSTRACT Introduction Recognizing the importance of livestock movements in the spread of contagious diseases, various countries in the world have developed livestock movement databases. Social network analysis techniques have recently been applied to study such databases as this technique has the advantage of allowing the study of the interactions among all pairs of livestock holdings that are formed following the movement of livestock. As a result, important holdings, which are central in the flow of animals, may be identified. The objective of this paper was to show two examples of uses of network analysis: (1) to estimate potential epidemic size following routine livestock movements during the silent spread phase of a highly contagious disease, and (2) to generate production types to specify the contact structure among livestock holdings in disease simulation models. Materials and methods We used network analysis techniques using monthly networks of adult dairy cow movements among dairy farms in Ontario (years 2004-2006) and beef cattle movements among all livestock holdings in Region XI, Chile (year 2007). Potential epidemic size was calculated in Ontario using an approach called “infection chain”. We used the degree distributions to classify beef farms into production types required in Chile. Results The monthly networks of livestock movements were highly fragmented throughout the year in Canada (mean=0.997, sd=0.001) and in Chile (mean=0.996, sd=0.002). The median monthly maximal potential epidemic size in Ontario included 13-15 farms. Four production types were created to simulate the spread of FMD in Chile: non-sellers, non-buyers, buyers-sellers and markets. Discussion The infection chain provided a biologically plausible estimate of potential epidemic size as it accounted for the direction of shipments and the time sequence of these shipments. Using the degree distributions also allowed modellers to classify farms according to their movement patterns and according to their management practices. 1. INTRODUCTION One of the most important transmission pathways for the spread of highly infectious diseases like foot-and-mouth disease (FMD) is the movement of animals among livestock holdings in a country or region. This is particularly true during the silent spread phase, when the virus is spreading undetected among these holdings. Various examples of such undetected spread have been

279

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 reported in the literature; the most recent and memorable was the movement of infected sheep through the United Kingdom in 2001, through the use of livestock markets (Mansley et al., 2003). It is therefore recognized that there is a need to record these movements of animals to enable tracing activities when a case of FMD is detected. As a result, several countries have implemented livestock movement databases. Another indirect use of these data would be to describe the frequency and patterns of livestock movements in order to inform contingency planning and disease modelling. The large amounts of data and the fact that it is possible to link source and recipient farms of animal movements have made it possible to study them by applying social network analysis techniques (Kiss et al., 2006; Robinson et al., 2007). In a population of livestock holdings, each holding represents a node in the network and the movement of animals among nodes represent arcs, or directed links: from the source to the recipient node. Following the movements of all animals during a period of time, a web of connections among holdings emerges and it becomes possible to study the interactions among pairs of holdings in the network. Whereas classical livestock movement studies focussed on obtaining the frequency of movements to and from each holding (Sanson et al., 2005), it now becomes possible to consider the relationships among all holdings in the network and to identify those that might be central in the flow of animals in the network, or those that are highly connected and therefore at risk of becoming infected or transmitting infection to a high number of other holdings. The objective of this paper was to show two useful applications of network analysis to study livestock movements: (1) to provide plausible estimates of potential epidemic size at first detection of FMD in a country like Canada, more specifically the Province of Ontario, and (2) to develop a structured analytical approach to classify livestock holdings in Chile into production types according to real movement data, required to simulate FMD in the North American Animal Disease Spread Model (NAADSM; Harvey et al., 2007). 2. MATERIALS AND METHODS 2.1 Estimating potential epidemic size in Canada Livestock movement information is scarce in Canada. One potential source of information is the Dairy Herd Improvement (DHI) program which stores lactation information of adult milking cows from member herds across the country. Milk testing occurs monthly in every herd at which time herd inventories are obtained. When cows are sold among DHI herds, their lactation records are also transferred to the recipient herd which enables the tracing of those cows. We obtained all movement information for individual adult milking cows moving among DHI farms in Ontario in 2004-2006. These movements were grouped into shipments, defined as the movement of ≥1 cow, on a single day from a single source farm to a single recipient farm. Monthly networks (n=36) of shipments among farms were created to represent the longest plausible duration of undetected FMD spread in Ontario. The average number of shipments per farm, per year was calculated by dividing the number of shipments by the total number of DHI farms enrolled that year. Network fragmentation was obtained in UCINet (Borgatti et al., 1999). Fragmentation is defined as the proportion of pairs of farms in a network that are unreachable, that is, the proportion of farms that do not have direct or indirect links to join them (Wasserman and Faust, 1994). In order to estimate potential epidemic size, we developed a network measure called the “infection chain” which is a modified breadth-first search (Knuth, 1997). The breadth-first search, used in graph theory, calculates the number of holdings that can be exposed by an infected holding either directly, by receiving infected animals from that holding, or indirectly by receiving infected animals from an intermediate holding. It accounts for the direction of the links among the farms in the network and we modified it so that it would also account for the order of shipments in time. Therefore if shipments took place from farm A to B to C, the shipment from A to B would have had to occur prior to the shipment of B to C for it to be considered an infection chain of size 3. These two features, time sequence and direction of shipments make the infection chain represent the possible spread of an infectious agent. Stata 8 was used to obtained descriptive statistics in this study (StataCorp. 2005; Stata Statiscal Software: Release 8.0; College Station, TX 77845, USA: StataCorp LP).

280

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 2.2 Creating production types based on movement patterns As part of a large counter-terrorism and capacity building project (CTCB), the North American Animal Disease Spread Model (NAADSM) is being applied in three countries in South America to simulate the spread of FMD: Brazil, Chile and Colombia. In Chile, Region XI was selected as it includes mandatory traceability of livestock movements (Figure 1). The database used in this study included movements of all bovines for year 2007, originating in Region XI. All markets, abattoirs and farms involved in movements, either as sources or destinations, were included in the study. In order to simulate the spread of infectious agents in the NAADSM, the population at risk must be specified. The NAADSM requires the following information: latitude and longitude of the holding, the number of susceptible animals in the holding and the production type of holding. This last characteristic, the type of holding, is defined by the user and can range from simply representing species (cattle, sheep, swine) to more elaborate production systems (dairy, cow-calf, feedlot). In most livestock holding databases the number of animals by species and age class are available. Therefore rules must be developed to classify livestock holdings into types. We first identified sheep only and cattle only farms. All units which declared swine also had sheep or cattle on premises in majority and were classified as sheep or cattle. It is assumed that these swine animals were used for personal consumption rather than being raised for commercial production. In the case of herds where both sheep and cattle were present, if >50% of animals in the unit were cattle, the unit would be classified as cattle and vice-versa. In the case of cattle herds, we decided to use the movement patterns to classify these units into further production types. We first extracted all movements among farms and markets to represent movements at risk of transmitting FMD through direct movement of animals. The movements of individual animals were grouped into shipments. We used the in- and out-degree values of each holding in the database to classify units into the following types: non-seller, non-buyer, buyer-seller and market. The outdegree is defined as the number of individual recipients per seller in a defined time period while the in-degree is defined as the number of individual sellers per buyer in a given time period (Wasserman and Faust, 1994). Units were classified as non-seller if they had out-degree=0, units were classified as non-buyer if they had an in-degree=0 and units were classified as buyer-sellers if their out- and in-degree values were different than 0. Markets were identified as having very large in- and out- degree values compared to the other units in the database. In order to represent the contact structure in NAADSM, the proportion of movements from cattle farms to markets was calculated. Production type combinations, required in NAADSM to specify who can infect who, were then defined and entered into NAADSM. For example, if non-buyers sold animals to markets 50% of the time, then they sold to non-sellers and buyer-sellers for the remaining 50% of the time. Expert opinion was obtained to further describe the proportion that would be sold to non-sellers and to buyer-sellers. 3. RESULTS 3.1 Estimating potential epidemic size in Canada According to the 2006 census of Agriculture, the Province of Ontario has close to 33% of the dairy cows in Canada. Approximately 50% of these farms are located in SW Ontario. A total of 77% (4060/5282), 72% (3601/5013), and 76% (3583/4695) of Ontario dairy farms were enrolled on DHI in 2004, 2005 and 2006, respectively. The mean farm size of DHI farms in Ontario was 61 cows (sd=43.3) over the three years. The number of shipments by month varied throughout the year and across years. On average, a DHI farm had a 5.5% (sd=7%) chance of shipping a cow or group of cows in a given month (Figure 1). The months of September to November and March were associated with higher numbers of shipments, while summer months, June to August, were less active. The size of shipment did not vary over the three years with an average of 1.39 cows (sd=1.77) in 2004, 1.41 cows (sd=1.93) in 2005 and 1.43 cows (sd=1.82) in 2006. The average fragmentation of monthly networks of farms linked by cow shipments was 0.997 (sd=0.001) in 2004-2006. This means that 99.7% of pairs of farms in the networks were unreachable, indicating that the monthly networks were very highly fragmented. In year 2005, the mean potential epidemic size, based on the average infection chain, was 0.1 farms (range 0-36 farms; Table 1). 3.2 Creating production types based on movement patterns There were a total of 1,398 livestock premises with latitude and longitude information from the traceability database of Region XI in Chile. Three of these were markets and the remaining

281

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 included either cattle, ovine or porcine as shown in Table 2. Of the 727 mixed farms, 376 were classified as ovine and 351 were classified as bovine. An example of the network observed on a monthly basis is shown in Figure 2. The monthly networks were highly fragmented, on average, 99.6% of pairs in the network were unreachable. There was an average of 420 holdings involved in shipments in the monthly networks of movements. The out-degree and in-degree distributions of all farms sending animals either to other farms or to market are shown in Figure 3. The final classification of farms into production types and daily contact rate based on movement patterns are shown in Table 3. Close to 50% of cattle farms were classified as non-buyers. These farms would typically sell throughout the year to only 1 other holding, although a maximum of 5 holdings was reported. Another 33% of farms were classified as non-sellers. Although these farms would purchase from one source only in a year up to 75% if the time, one farm purchased animals from 12 different sources. Seventeen percent of farms were classified as buyer-seller. Up to 99% of these farms sold animals to up to 6 recipients while one farm sold to 10 different recipients. In the case of in-degree, up to 50% of these buyer-seller farms would purchase from 1 farm, up to 99% of farms would purchase from 18 different sources and one farm purchased from 45 different sources. The production type combinations used in simulations in Chile are shown in Table 4. 4. DISCUSSION The use of network analysis to study livestock movements can provide very useful information to evaluate the impact of various livestock holdings in the spread of highly contagious diseases. The infection chain procedure, used to study the DHI networks, provided plausible estimates of epidemic size following direct movements of livestock as it accounted for the direction and the time sequence of the shipments among farms. Other measures have been used for this purpose, to estimate maximal potential epidemic size, however these estimates do not account for the time sequence of shipments, only considering the network as cumulative with regards to the movement of animals in a defined time period (Robinson et al., 2007). The high fragmentation found in the DHI networks reflects how the networks were created: they covered only a one month time period and only movements of adult milking cows among DHI farms in the Province were used. Therefore we likely underestimated the potential epidemic size at first detection. Considering movement networks of longer duration as well as adding movements of other age classes of dairy cattle and movements involving non-DHI farms to the networks would increase resulting epidemic size estimates and the level of fragmentation. These movements are not presently tracked in any Canadian database. In addition, indirect contacts such as the movement of people and fomites among herds and airborne spread, if applicable, could increase the estimates of potential epidemic size of FMD at first detection. This study represented a first step in trying to represent dairy cattle movements in Canada. As more information becomes available, the infection chain approach will be used to determine the possible extent of spread of a disease like FMD. The values obtained to date however can be used to parameterize disease spread models. The livestock movement database obtained in Chile included the movements of cattle of all age classes, through all types of livestock holdings. Region XI is a beef and sheep region. For the first time in NAADSM’s existence, we used network analysis measures, such as the in- and out-degree values of holdings in the Region to characterize production types. Rather than trying to classify holdings into feedlot or cow-calf based on the age structure of animals in each holding, we opted to classify them according to their movement characteristics: were they sellers, buyers or both? Unless there is a reason to differentiate cow-calf farms from feedlot farms, to represent varying detection efficacy, or the biology of FMD in those farms for example, then only using movement patterns can be a very useful and quick way to specify the contact structure in FMD disease simulation models such as NAADSM or Interspread (Sanson, 1993). The monthly networks of cattle movements in Region XI were highly fragmented. This can be explained by the fact of low number of movements which are very well marked throughout the year. The study of the out-degree and in-degree distributions provided a quick and efficient analytical approach to identify the potential risk posed by the different farm types based on their real movement patterns. For example, the majority of cattle farms would not represent a risk to other farms or markets as 50% of them did not report selling animals in year 2007. However, a few farms exhibited dealer-like behaviour by buying and selling to and from various different holdings. This dealer-like behaviour, much like that of markets, places these operations at higher

282

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 risk of becoming infected and of infecting a large number of recipients. These behaviours are very important to capture when modelling the spread of infectious diseases. Network analysis has only recently been used to study the movements of livestock. It has multiple applications and provides various approaches to understand the behaviour of producers, dealers, markets and abattoirs in the network. Making the maximal use of this information is critical for the proper representation of the impact of the disease spread and its consequences in the population. The approach developed in this study will help to produce input parameters for disease spread models. The two examples in this study should encourage government organizations to develop movement databases and make the data accessible to researchers for modelling and policy decision making. 5. REFERENCES [1] Borgatti, S.P., Everett, M.G. & Freeman, L.C. 1999. UCINET 6.0 Version 6.17. Natick: Analytic Technologies. [2] Garland, A.J.M. & Donaldson, A.I. 1990. Foot and Mouth Disease. Surveillance 17: 6-8. [3] Harvey, N., Reeves A., Schoenbaum M.A., Zagmutt-Vergara, F.J., Dubé, C., Hill, A.E., Corso, B.A., McNab, W.B., Cartwright, C.I. & Salman, M.D. 2007. The North American Animal Disease Spread Model: A simulation model to assist decision making in evaluating animal disease incursions. Prev. Vet. Med. 82: 176-197. [4] Kiss I.Z., Green D.M. & Kao R.R. 2006. The network of sheep movements within Great Britain: network properties and their implications for infectious disease spread. J. R. Soc. Interface, 3: 669-677. [5] Knuth, D.E. 1997: The art of computer programming Vol 1: Fundamental algorithms, 3rd Edition. Reading, Massachussetts: Addison-Wesley. [6] Mansley, L.M., Dunlop, P.J., Whiteside, S.M. & Smith, R.G.H. 2003. Early dissemination of foot-and-mouth disease virus through sheep marketing in February 2001. Vet. Rec. 153: 43-50. [7] Mclaws, M., Ribble, C., Martin, S.W. & Wilesmith, J. 2005. Factors associated with the early detection of Foot-and-Mouth Disease during hte 2001 epidemic in the UK. Proc. Soc. Vet. Epidemiol. Prev. Med., p.211-221. [8] Robinson S.E., Everett M.G., Christley R.M. 2007. Recent network evolution increases the potential for large epidemics in the British cattle population. J. R. Soc. Interface, 4(15), 587-762. [9] Sanson, R.L., 1993: The development of a decision support system for an animal disease emergency, Unpublished PhD thesis, Massey University, Palmerston North, New Zealand. [10] Wasserman S., Faust K., 1994. Social network analysis: methods and applications. Cambridge University Press, New York, NY.

283

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Table 1: Infection chain distributions by month in year 2005 based on the network of adult milking cow movements in DHI member farms in Ontario. Similar results were observed in 2004-2006. Infection chain Average 25% infection chain

Month

January February March April May June July August September October November December

0.1 0.1 0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1

0 0 0 0 0 0 0 0 0 0 0 0

50%

75%

95%

99%

Max infection chain

0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 0 1 1 1 1 1 1

2 2 3 2 2 2 2 2 2 3 2 2

9 9 14 14 9 13 25 15 8 30 36 22

Table 2: Description of species by farm in Region XI, Chile. ProductionType

Total

Bovine only

630

Ovine only

Bovine and ovine

580

Bovine and swine

Ovine and swine

Bovine + ovine + swine

147

Total

1,395

Table 3: Description of the production types and their contact rate (number of recipients / day) created for FMD simulations in Region XI, Chile. Production type Non-seller Non-buyer Buyer-seller Market Ovine TOTAL

284

Frequency 72 699 234 3 390 1,398

Max rate/year 0 28 20 78 28

Median rate/year 0 1 3 36 1

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Table 4: Contact matrix showing the production type combinations used in FMD simulations in Region XI, Chile. The proportion of movements sold directly to a market for cattle was obtained from the livestock movement database. The remaining proportions were derived from expert opinion.

Source Markets Cattle non-buyer

Recipients Markets Cattle seller 0 20% 61% 12%

Cattle buyer-sellers Ovine

45% 10%

Cattle sellers 70% 27%

17% 0

Year 2004

0.080

non-

buyer-

Ovine 10% 0%

39% 0%

Year 2005

0% 90%

Year 2006

0.070

Probability of shipment

0.060

0.050

0.040

0.030

0.020

0.010

December

October

November

September

July

August

May

June

April

March

January

February

December

October

November

September

July

August

May

June

April

March

January

February

December

October

November

September

July

August

May

June

April

March

January

February

0.000

Month of year

Figure 1: Probability of direct shipments by DHI farm in Ontario, by month in 2004-2006 (does not include movements through markets).

285

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 2: Representation of the Regions of Chile. Region XI was selected for NAADSM simualtions.

286

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

InDegree 100 150

200

250

Figure 3: Network diagram of the movements of cattle among farms, markets and abattoirs in May 2007, Region XI, Chile. Circles represent cattle farms, triangles represent markets, squares represent abattoirs. This network includes movements from all livestock holdings in Region XI to farms, auctions or abattoirs located inside or outside of Region XI.

Markets

20 OutDegree

Figure 4: Scatterplot showing the in- and out-degree of all farms and markets in Region XI, Chile. The three markets are represented by high in- and out-degree.

287

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

288

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 52 MODELLING OF FMD OUTBREAKS IN THE NETHERLANDS: VACCINATION AND REGAINING THE STATUS ‘FREEDOM OF INFECTION’ J. A. Backer, T. J. Hagenaars, G. A. Nodelijk and H.J.W. van Roermund* Quantitative Veterinary Epidemiology and Risk Analysis, Division Virology, Central Veterinary Institute of Wageningen UR, P.O. Box 65, 8200 AB Lelystad, the Netherlands

ABSTRACT To limit the impact of Foot and Mouth Disease (FMD) epidemics on animal welfare and economics, control measures and end screening should be applied as effectively as possible. Here we study this for The Netherlands using an individual-based stochastic model. It describes virus transmission between animals and between farms, and takes differences between animal species into account (cattle, sheep, pigs). The effect of vaccination is included at the individual level, making a comparison between control strategies possible at the livestock area level. The results for individual animals indicate how many infected animals escape clinical detection during the epidemic (i.e. undetected minor outbreaks), enabling a comparison between end screening scenarios. Our model results show that the minimal control measures required by the EU suffice in sparsely populated livestock areas (ca 2 farms per km2), but ring culling or vaccination is required in addition to curb epidemics in densely populated areas (>3 farms per km2). According to the model, 2 km ring vaccination is less effective than 1 km ring culling in terms of size and duration of the epidemic, but the difference is small when comparing with the minimal control strategy. 5 km vaccination and 1 km culling are equally effective, although the vaccination strategy yields higher and later epidemic peaks. Excluding pig farms from vaccination does not drastically reduce the effectivity of vaccination strategies (for the studied virus strain O/NET/2001), because of the relatively low pig farm numbers (compared to cattle) and their low susceptibility for FMD. The infected farms that escape clinical detection during the epidemic are mainly vaccinated cattle and sheep farms and unvaccinated sheep farms. Therefore, compared to the screening required by the EU, the relative risks are not markedly reduced in screening strategies in which more effort is placed on unvaccinated cattle and pig farms and, likewise, not markedly enhanced when less effort is placed on vaccinated pig farms. 1. INTRODUCTION Outbreaks of Foot and Mouth Disease (FMD) in the Netherlands represent a risk of major importance to the Dutch farming industry, as around 17 million cattle, pigs and sheep are at risk of being infected by the virus. To control an ensuing epidemic as quickly as possible, emergency vaccination is preferred in The Netherlands (see Dutch contingency plan FMD, 2005). Concerns exist though that vaccination might not be as effective in controlling the epidemic as preemptive ring culling, because vaccinated animals are not instantaneously protected against infection. Another concern is that vaccination increases subclinical infections. These subclinically infected animals might escape clinical detection during the outbreak, and need to be detected serologically in the end screening. When they also escape the end screening they might pose a risk to a new outbreak and to the export position of the country (when detected later). The Dutch contingency plan as well as the EU screening regulations for vaccinated animals do not make a distinction between animal species or farm sectors. However, virus transmission and the effect of vaccination differs considerably between species. Cattle are highly susceptible to infection, and they can become carriers of the disease. Infected sheep can also become carriers and show fewer clinical symptoms than cattle. Infected pigs excrete large amounts of virus, but they are not easily infected. Vaccination might not work as effectively on them as on cattle or sheep. Furthermore, the socio-economic impact of an FMD epidemic can vary for different farm sectors.

289

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Pig and veal farmers will most probably encounter difficulties in marketing vaccinated meat, once the epidemic is under control. The hobby farm sector (i.e. small herds held for non-commercial purposes, in the Netherlands mainly small sheep flocks) on the other hand is largely unknown due to incomplete registration, but the social impact of an epidemic will be considerable here as well. Here we will evaluate whether vaccination can be an effective and safe strategy to control an FMD epidemic, while taking the differences between four farm types (see below) into account. For this purpose we developed a mathematical model that describes the within-herd and between-herd dynamics at two distinct levels. Results of transmission experiments and data of the FMD outbreak that occurred in The Netherlands in 2001 (virus strain O/Net/2001) serve to estimate the model parameters. The model is applied to the farm density situation of 2006, involving 36000 cattle farms, 18000 sheep farms, 9000 pig farms and 20000 hobby farms (here small sheep flocks physically separated from commercial farms, held for non-commercial purposes), of varying farm sizes. Location coordinates and number of animals of each of these farms are taken into account. We did not distinguish between different commercial herd/flock types of the same animal species. With this model the outbreak size and duration is calculated for hypothetical epidemics of FMD, starting in different area’s of The Netherlands on a cattle or pig farm. Also the number of infected farms and animals that escape clinical detection, is predicted. This result is used as input for a model that describes the serological testing in the end screening. For different test characteristics and end screening strategies it predicts the number of seropositive animals (i.e. animals with a detectable anti-body response) that will remain when the country has been declared free of infection. 2. MATERIAL & METHODS We developed a mathematical model that captures the key differences in the epidemiology of the four different farm types. It consists of two modules that describe the within-herd and betweenherd transmission dynamics of FMD, as shown schematically in Figure 1. For the within-herd model, that is formulated in terms of individual animals, parameters are estimated for each species from literature on transmission and vaccination experiments. In the between-herd model the farm itself is the smallest unit. The transmission at this level is modelled by distant-dependent probabilities, calibrated by the outbreak data of 2001 in the Netherlands and other available data. This transmission kernel describes the between-herd transmission as occurred in the country after the high risk period (so during stand still). To apply the model to the 2006 situation, it needs the locations and type of all farms in the Netherlands, which are available in databases. The withinherd module produces simulated farm outbreaks; from these the profile of the infection pressure (i.e. number of infectious animals as a function of time) enters into the between-herd module, that determines which herds are infected by the source herd. The within-herd module also produces a detection time of the within-herd outbreak (if applicable), at which the between-herd module determines which herds need to be culled or vaccinated, depending on the control strategy. The information on infection, vaccination and culling times is returned to the within-herd module as input for calculating the simulated outbreak on the next infected farm. The result – after the last infected farm-outbreak is simulated – is the total course of the hypothetical epidemic. This model structure allows for the extrapolation of the effects of vaccinating individual animals to the level of an area with many farms.

290

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 control strategy

infection pressure over time detection time

within-herd module

infection time vaccination time (preemptive) culling time

for each species: - transmission rate - infectious period - clinical symptoms - vaccination effect - detection limit literature on cattle, sheep, pigs: - transmission experiments - vaccination experiments data FMD outbreak 2001

between-herd module

locations farm types

I&R database (cattle, sheep) GD database (pigs) data hobby farm surveys

course of hypothetical epidemic

transmission kernel

data FMD outbreak 2001 literature on cattle, sheep, pigs: - infectiousness - susceptibility

Figure 1: Schematic representation of two-level transmission model for FMD 3. RESULTS AND DISCUSSION

number of infected herds (95%)

We evaluated several control strategies, by simulating 1000 hypothetical epidemics for each control strategy. Figure 2 shows the results of some important control strategies, when the epidemic starts on a cattle farm in the Gelderse Vallei (i.e. a cattle- and pig-dense area of 4.2 farms per km2) and 10 farms have been infected by the time that the first detection of an infected herd occurs.

1 km ring culling 2 km ring vaccination 2 km ring vaccination except pig farms 5 km ring vaccination 5 km ring vaccination except pig farms

Time (days since first detection) Figure 2: Number of infected herds (95% percentile) during an epidemic in a densely populated livestock area under different control strategies From the model-evaluation of the control strategies it is concluded that: Additional measures such as preemptive culling or vaccination are necessary to control the epidemic in densely populated livestock areas (> 3 farms per km2), but in sparsely populated livestock areas (of ca 2 farms per km2) the minimal control strategy suffices (as required by the EU: culling of detected infected herds, tracing of their dangerous contacts and regulation of transport). In densely populated livestock areas 2 km ring vaccination is less effective than 1 km ring culling in terms of size and duration of the epidemic, but that difference is small when comparing with the minimal strategy as required by the EU. In densely populated livestock areas 5 km ring vaccination and 1 km ring culling are equally effective (when vaccination capacity is not limiting).

291

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

More than 75% of the infected farms are cattle farms, even in pig-dense areas, regardless of control strategy. Excluding hobby farms (here small sheep flocks of 10 animals, physically separated from commercial farms) from preemptive culling has a negligible effect on epidemic control. In the ring culling strategy, it reduces the number of farms to be culled by 20% (and the number of animals to be culled by 3%). Excluding pig farms from vaccination causes a significant but limited increase of the size and duration of the epidemic (for the virus strain under study). The number of animals to be vaccinated is more than halved. Whether this strategy is economically beneficial despite the increased risk, is to be evaluated in an economic analysis. For non-vaccination strategies around 5% of the infected farms remains clinically undetected during the epidemic, consisting mainly of unvaccinated sheep farms. This percentage is between 11% and 20% for vaccination strategies, involving mainly vaccinated cattle and sheep farms.

Before the country can be declared free of infection, the EU requires all animals on all vaccinated farms to be serologically tested, as well as a sample of sheep on unvaccinated farms. We studied the results for this end screening strategy for three basic control strategies (for epidemics that started on a cattle farm in the Gelderse Vallei). From the model-evaluation of the end screening it is concluded that: In non-vaccination strategies 1000-5000 farms need to be tested in the end screening, while vaccination strategies require twice as many farms (2000-11000) to be tested. About half of the tested farms must be retested to exclude or confirm infection. Before the end screening, vaccination yields approximately 5 times as many seropositive animals as ring culling. After the end screening the number of seropositive animals is similar for 1 km ring culling and 5 km ring vaccination, and slightly higher for 2 km vaccination. Compared to the screening required by the EU, screening strategies in which more effort is placed on unvaccinated cattle and pig farms (testing a sample of animals instead of none) do not provide added value. Compared to the screening required by the EU, screening strategies in which less effort is placed on vaccinated pig farms (testing a sample of animals instead of all) can be safely implemented. In conclusion, vaccination is an effective control strategy to mitigate FMD epidemics, provided it can be applied on a large scale, especially in densely populated livestock areas. Control measures should primarily target cattle farms, as these are predicted to play the largest role in the epidemic (for the virus strain in this study). After the epidemic, most seropositive animals are expected on sheep farms and vaccinated cattle farms. An effective end screening strategy should focus on these farms. The simulation results are now being used in a socio-economical analysis of vaccination against FMD. 4. ACKNOWLEDGMENTS The research was funded by the Ministry of Agriculture, Nature and Food Quality (LNV), the Netherlands (CVI Project nr. 1691026700, LNV-theme BO-08-010 Animal Health, 2008). We would like to thank Gert Jan Boender (CVI) for estimating the between-herd transmission kernel and Bas Engel, Aldo Dekker, Phaedra Eblé, Clazien de Vos (CVI), Mart de Jong (WUR), Ron Bergevoet (LEI), Stephanie Wiessenhaan, Wim Pelgrim, Eric van der Sommen and Huibert Maurice (LNV) for the discussions during the project. 5. REFERENCES [1] Anonymous. 2005. Beleidsdraaiboek Mond- en Klauwzeer, versie 2.0, september 2005. Report from the Ministry of Agriculture, Nature and Food Quality (LNV), 138p (in Dutch).

292

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 53

DEFINING THE PERIOD OF INFECTIOUSNESS IN CATTLE NATURALLY INFECTED WITH FOOT-AND-MOUTH DISEASE VIRUS. B. M. Bankowski1, R. Howey2, N. Juleff1, D. Gibson1, S. J. Cox1, P. V. Barnett1*, M. E.J. Woolhouse2 and B. Charleston1 1

Institute for Animal Health, Pirbright Laboratory, Ash Rd, Woking, Surrey, GU24 0NF, UK. Epidemiology Group, Centre For Infectious Diseases, Ashworth Laboratories, Kings Buildings, Mains Road, University of Edinburgh, EH9 3JF, UK.

ABSTRACT Objectives To broaden our understanding of FMDV transmission in cattle; quantify variability in infectiousness at different time points and establish the peak of infectiousness; pin-point predictors of infectiousness and provide accurate data for mathematical modelling of FMD. Materials and methods For each of four experiments, two naïve animals were inoculated intradermolingually with O UKG 34/2001 and two days post inoculation used to directly challenge two further naïve cattle for 24 hrs. At different days post challenge these direct contact challenged ‘donors’ were used to challenge the remaining naïve animals (recipients) either by direct or indirect contact, which were clinically monitored and sampled for 14 days. Clinical assessment included rectal temperature and detailed scoring of clinical signs. Samples included oropharyngeal and nasal fluid, and blood, and air samples, taken pre- and at challenge and assayed for the presence of virus. Results All ‘donors’ developed clinical signs of FMD, although the incubation period to onset of clinical signs varied between 2 and 8 days post challenge. Transmission only occurred by direct contact (8 transmission events out of 28 challenges) and the incubation period similarly varied between 2 and 8 days. Significant amounts of infectious virus were recovered from air samples during challenge. Virus was recovered from all donors and clinically diseased ‘recipients’. Discussion Infected cattle were shown to be infectious up to a maximum of 4 days. However, estimates indicate a range of values (42.53 to 87.97 hours) for the duration of the infectious period, with an average of 61.85 hours. A good indicator of infectiousness is the presence of lesions in the oral cavity, mouth, tongue or snout. No transmission was evident by indirect contact which indicates that airborne transmission may not readily occur between cattle. 1. OBJECTIVES: After an initial pilot study, four animal experiments were performed to broaden our understanding of FMDV transmission between individual cattle. Further inferences to within herd or ‘over the fence’ transmission could be made with subsequent quantification of variability in infectiousness at different time points, and establishing the peak of infectiousness. Using clinical data we aimed to identify and pin-point the predictors of infectiousness in cattle. These will be used to provide further data for the mathematical modellers, to further clarify the window of infectiousness in cattle and effect of prophylactic intervention. 2. MATERIALS AND METHODS

293

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 2.1 Animals Experiments 1 and 2 have been described and preliminary data presented [2]. Two further, but modified, experiments were performed, each using twenty Holstein Friesian calves weighing 100150 kg and 6-9 months of age. Two animals were randomly selected and injected intradermolingually [11] with 0.2 ml of a diluted stock of cattle adapted virus (1x105 TCID50, 50% bovine thyroid tissue culture infectious dose). Following inoculation these animals were kept in a separate box in the bio-secure isolation compound and two days later two naïve animals (referred to as donors) were introduced to these inoculated cattle and challenged by direct contact for 24 hrs. The inoculated animals were then removed from the study and the ‘naturally’ challenged donors left in the box. These donors were then used to challenge, in separate rooms, the remaining cattle both directly and indirectly at 2, 4, and 6 days (experiments 1 & 2) and directly only at 2, 4, 6 and 8 days (experiments 3 & 4) for a period of 8 hrs each time. In the two challenge rooms, a wooden mesh partition was used to separate indirect recipients from donors and direct contact recipients (Diag. 1). At each time point the relevant animals were moved to two separate challenge boxes in the following order: indirect contact recipient (IDC) were moved to each challenge room first, followed by direct contact recipient (DC) and then the donor animal. During the challenge period the ventilation in the challenge boxes was turned off and the relative humidity was close to 99%, (measured with Fisher Scientific, Thermometer/hygrometer, Traceable FB70257). In addition, a wall mounted fan (Diag.1) was used to mix the air evenly during challenge. Airflow was directed from donor side to IDC side (smoke tests were carried out before these experiments to ensure that appropriate air movement occurred). After challenge the recipients were moved back to individual rooms, keeping the recipient animals separate. In Experiments 3 and 4 IDC animals were substituted by indirect contact companion animals (IDCA) to maintain the challenge conditions, but were not subsequently monitored or sampled during the experiments. Each IDCA was kept in the box with one DC separated by sealed solid metal partition to ensure that the development of the disease could only be attributed to the direct 8 hour challenge. Thus, the experimental design allowed for examination of virus transmission, by both the direct and indirect route [5]. All the animals were clinically examined and rectal temperatures were recorded daily. The recipient animals were euthanized fourteen days post challenge. 2.2 Clinical Scoring The animals were monitored for clinical signs of FMD and rectal temperatures were recorded daily. Quan’s [14] subjective scoring system was modified and used to evaluate the development of the disease in cattle. The modified scheme incorporated changes in rectal temperature and discriminated between new lesions and healing lesions. Hence, the profile of the clinical score plots reflected the induction and resolution of clinical signs. Cattle therefore could score a maximum of 22 points (the sum of scores from all feet was divided by 2 to prevent the score from being dominated by foot lesions). 2.3 Sampling regime Blood samples, nasal swabs and nasal fluid samples were taken daily for the first week following challenge and thereafter every other day to termination. The samples were transferred immediately to the laboratory and stored appropriately [2] for subsequent virus detection using cell culture and real-time quantitative RT-PCR. 2.4 Air sampling Air samples from the animal rooms were collected using an all-glass Cyclone sampler operated for 5 min at a flow rate around 390 l/min, and an all glass Porton impinger sampler operated for 5 min at a flow rate of 11 l/min; these sampling periods were the optimal sampling configurations for the instruments [4, 8, 12]. Several air samples were collected simultaneously, at one hour intervals, during the challenge periods on days 2, 4 and 6. Each box that was sampled contained one IDC/IDCA, one DC, and one donor animal. The collecting media employed in the air samplers was Modified Eagle Medium (MEM) -HEPES with antibiotics and 0.1% (w/v) BSA [6, 10]. The concentration of virus per litre of air was determined by endpoint titration, which was multiplied by the volume of the collecting fluid and the flow of the sampler. The amount of infectivity recovered was expressed as the total amount (TCID50) of airborne FMDV per cattle per challenge period. In addition RNA was extracted for reverse transcription PCR and quantitative real-time PCR (QPCR).

294

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2.5 Virus strain The O UKG 34/2001 FMD isolate used was derived from bovine vesicular lesions. The titre of the stock virus was 1x108.1 TCID50/ml by primary bovine thyroid (BTY) cells [15] and the animals received 0.2 ml of this stock diluted to provide a final dose of 1x105 TCID50 per animal. 3. RESULTS 3.1 Clinical Signs and Scores All donor animals developed typical clinical signs of the disease within 3 to 8 days after challenge (Table 2). Observed signs included pyrexia, nasal discharge, vesicles on feet, mouth area or snout (for summary scores see Table 1, detailed scores to be described elsewhere). Specimen samples were assayed for viral genome and live virus (data not shown) confirming infection with FMD virus. The only exception was animal VR57 that was not viraemic although it seroconverted at 12dpc. Twenty eight cattle were used as direct contact recipients and out of these only 8 developed clinical signs of the disease as described above, or were viraemic and/or seroconverted (data not published) as a result of direct challenge. Most of the successful transmissions occurred during the onset of clinical signs in donor animals and coincided with presence of lesions in the mouth, oral cavity and snout or tongue. None of the indirect contact animals (12 in total) developed clinical or subclinical disease nor did they develop detectable viraemia or seroconvert , even though high levels of virus were recovered during the challenge period (Table2) sufficient to cause the disease in cattle [1, 7]. 3.2 Infectious Period of Animals To estimate the infectious period of the donors we firstly make the assumption that the donors will transmit during some fixed time window after challenge. The second assumption we make is that the infectious period is the same duration across all donors. Although we do not expect this to be true, we expect them to be close enough in order that we can use the transmission data collectively to estimate a typical infectious duration. Many of the successful challenges are either the last or first challenge (see figure 1.) resulting in an inaccurate estimate of the duration of infectiousness. Therefore, by considering this small quantity of data together we strive to obtain a better estimate of the duration of infection. Considering the transmission data for donor VR57 it is not possible to estimate when it was the most infectious since it did not transmit. There are a number of explanations: (i) it was not infectious enough to transmit, or (ii) it was infectious but failed to transmit, or (iii) it was only infectious between challenges. Reasons (i) and (ii) add no further information that is of any use to estimate the duration of the infectious period. Reason (iii) is a direct contradiction to the assumption that the infectious period is of equal length, if this is relaxed to include individual variation then the infectious period would be around 40 hours in order to incorporate the two transmissions of donor VN89 and none of donor VR57. Here, we do not consider donor VR57, which failed to transmit at any time, in our estimate of the length of the infectious period. The probability density function, pdf, of the duration (hours), l, of the infectious period was constructed by considering the possible timings of infectious periods for each donor that reproduces the observed transmissions:

l < 40.08; 0,  pdf (l ) = k (l − 40.08)a (l )b(l )(88 − l )(105.17 − l ), 40.08 ≤ l < 88; 0, l ≥ 88.  Where k is a constant ensuring that otherwise; and

b(l ) = 48

∫ pdf (l )dl = 1;

l < 47.42 and 95.42 − l

a (l ) = 8 + l

l < 40.27 and 88.25 − l

otherwise.

If our second assumption is changed so that transmission is possible if, and only if, the infectious period covers the entire challenge period then the probability density function,

pdf e , is:

295

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

l < 56.08; 0,  pdf e (l ) = h(l − 56.08)c(l )(103.42 − l )(104 − l )d (l ), 56.08 ≤ l < 103.42; 0, l ≥ 103.42.  Where h is a constant ensuring that otherwise; and

d (l ) = 48.08

∫ pdf

(l )dl = 1 ; c(l ) = l − 8

l < 56.25 and 104.33 − l

l < 56.27 and 104.25 − l

otherwise. Figure 2 illustrates when a

challenge is successful under the two different conditions for a successful transmission. From the probability density functions (see figure 3) we can calculate estimates of the infectious period duration with 95% confidence limits, firstly when infectious period only needs to overlap the challenge period to transmit, and secondly when the entire challenge period must be covered:

Overlap period Entire period

2.5% Percentile

Median

97.5% Percentile

42.35

53.33

71.86

58.15

68.75

86.80

As can be seen, the requirement that the infectious period covers the entire challenge period implies that the duration of the infectious period must be longer. These two scenarios are the extreme cases and we may expect a true value between these two predictions. Moreover, the infectious period is unlikely to be exactly described by a step function, but considering the small data set and reliable transmissions over short periods near the onset of clinical signs, this was considered the most appropriate function to use. Alternative functions were fitted to the data but provided no further insights. From these estimates we observe a set of plausible values (based on the 95% confidences intervals) for the duration of the infectious period from 42.35 hours to 86.80 hours. 4. DISCUSSION These experiments were designed to identify when cattle will transmit FMDV to naïve sentinels. The FMDV donor animals were exposed to cattle that had been directly inoculated with FMDV two days previously. After these exposures the onset of clinical signs were variable, occurring 3 to 8 days after exposure for the 8 animals studied. Transmission from these donor animals to naïve recipients were also variable, occurring on days 2, 4 and 6 (Exp 1 & 2) and 4, 6 and 8 (exp 3 & 4) after the donor animals had been exposed to needle challenged animals. In the first study the transmission occurred to direct contact recipients on day 4 (to VN97 and VN98) and 6 (to RZ76), in the second study the transmission and the onset of the disease took place on day 2 (to VO79) and on day 6 (to VO88). The transmission in experiment 3 took place on day 4 (VQ05 to VQ13) and 8 (VQ06 to VQ23) and in experiment 4 only on day 6 (VR56 to VR66). In the latest experiment donor VR57 did not transmit the disease on any of the challenge days, despite showing clinical signs of the disease. Further analysis of the specimen samples from this animal revealed that although it had virus in its nasal and OPF and had seroconverted at 12 dpc, it did not become viraemic at any time point (virus isolation and qRT-PCR data not shown). Preliminary analysis of the donor animals reported in Bankowski et al. [2] showed that these animals transmitted the virus from 24 hours before to 72 hours after the onset of the clinical signs, suggesting duration of infectious period up to 4 days. Further analysis showed that the infectious period in cattle is likely to last from 42.35 hours to 86.80 hours. Moreover, we analysed the clinical signs present at a time of successful challenges and concluded that presence of lesions in mouth/oral area (oral cavity, mouth, snout or tongue) coincided with transmission events. Those were present more frequently than high temperature or lesions on feet. This is in line with previous observations (oropharynx was identified as a primary site of infection – [3, 9, 13]) that lesions in the oral area appear first and then the disease is generalised to the feet. Therefore lesions in the oral area were suggested as indicators of infectiousness in cattle infected with FMDV. These indicators will be used in subsequent studies with animals at different time points after vaccination to allow challenge during the peak of infectiousness from the donor animals. Interestingly, no transmission to the indirect contact animals occurred. The minimum dose of airborne FMD virus required to infect cattle during a 24-hour exposure has been estimated to be 10 TCID50 [7]. The concentration of virus measured in the air samples in the current experiments was

296

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 approximately 104 TCID50/8hrs. This quantity of virus in the air samples should therefore have been satisfactory for infecting the indirect contact recipients. 5. AUTHORS CONCLUSIONS Transmission occurred between inoculated to donor animals and the incubation period to the onset of clinical signs in the donor animals varied between 3 and 8 days post challenge. Following challenge of the donors, transmission events occurred at days 2, 4, 6 and 8. In our model system, FMDV infected cattle have been shown to be infectious from 42.35 hours to 86.80 hours. Presence of lesions in oral cavity, mouth, tongue and/or snout correlates with transmission occurrence and appear to be good indicators of infectiousness. Airborne transmission within pen was not evident between cattle, highlighting a lesser importance of this route epidemiologically with this virus. 6. AUTHORS RECOMMENDATIONS To undertake further experiments and expand and validate the data that could be used for developing mathematical models for ‘in farm’ and ‘between-farm’ spread of the disease as well as to help in the design of FMDV vaccination programmes. To liaise with mathematical modellers and establish the optimal time point for infectiousness in cattle. 5. ACKNOWLEDGEMENTS The authors would like to thank the staff of animal isolation units for their useful and valued assistance in the care of the animals used in this study; John Gloster and Eoin Ryan for assistance with air sampling. This work was supported financially by BBSRC, UK (Grant Ref BBSB00549). 6. REFERENCES [1] Alexandersen, S., Zhang, Z., Donaldson, A. I., Garland, A. J. M. (2003) The Pathogenesis and Diagnosis of Foot-and-Mouth Disease. Journal of Comparative Pathology. 129: 1–36. [2] Bankowski, B., Juleff, N., Gibson, D., Gloster, J., Doel, C., Barnett, P., Woolehouse, M., Charleston, B. (2006) Understanding the FMDV transmission between cattle: Preliminary data from animal experiments. In: Proceedings of an Open Session of the Research Group of the European Commission for the Control of Foot-and-Mouth Disease (EUFMD), Cyprus, 17 to 20 October 2006, Appendix 25: 165 – 170. [3] Burrows, R., Mann, J. A., Garland, A. J., Greig, A. and Goodridge, D. (1981). The pathogenesis of natural and simulated natural foot-and-mouth disease infection in cattle. Journal of Comparative Pathology, 91, 599–609. [4] Doel, C.M.F.A., Gloster, J., Valarcher, J-F. (Available online 30 October 2007) Airborne transmission of foot-and-mouth disease in pigs: Evaluation and optimisation of instrumentation and techniques. The Veterinary Journal, In Press, Corrected Proof. [5] Donaldson, A. I. (1972) The influence of relative humidity on the aerosol stability of different strains of foot-and-mouth disease virus suspended in saliva. Journal of General Virology. 15: 2533. [6] Donaldson, A. I., Gibson, C. F., Oliver, R., Hamblin, C. & Kitching, R. P. (1987) Infection of cattle by airborne foot-and-mouth disease virus: minimal doses with O1 and SAT 2 strains. Research in Veterinary Science, 43: 339-346. [7] Donaldson, A.I. and Alexandersen, S. (2002) Predicting the spread of foot-and-mouth disease by airborne virus. Revue Scientifique et Technique Office International des Epizooties. 21, 569-575. [8] Errington, F. P., Powell, E. O. (1969) A cyclone separator for aerosol sampling in the field. Journal of Hygiene (Cambridge). 67, 387–399. [9] Garland, A.J., (1974). The inhibitory activity of secretions in cattle against FMDV. PhD Thesis, University of London. [10] Gibson, C. F. and Donaldson, A. I. (1986) Exposure of sheep to natural aerosols of footand-mouth disease virus. Research in Veterinary Science. 41, 45-49. [11] Henderson, W. M. (1952). A comparison of different routes of inoculation of cattle for detection of the virus of foot-and-mouth disease. Journal of Hygiene (London). 50, 182–194.

297

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [12] May, K. R., Harper, G. J. (1957) The efficiency of various liquid impinger samplers in bacterial aerosols. British Journal Industrial Medicine. 14, 287–297. [13] McVicar, J. W. and Sutmoller, P. (1976). Growth of foot-and- mouth disease virus in the upper respiratory tract of non-immunized, vaccinated, and recovered cattle after intranasal inoculation. Journal of Hygiene (London) 76, 467–481. [14] Quan, M., Murphy, C. M., Zhang, Z. and Alexandersen, S. (2004) Determinants of early foot-and-mouth disease virus dynamics in pigs. Journal of Comparative Pathology. 131, 294-307 [15] Snowdon, W. A., (1966) Growth of foot-and-mouth disease virus in monolayer cultures of calf thyroid cells. Nature. 210, 1079-1080.

Experiment 3

Experiment 2

Experiment 1

VN89

VN90

VO7 5

VO7 6

VQ0 5

VQ0 6

Experiment 4

VR56

VR57

Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n Clinical score Transmissio n

Days post challenge 0 1 2 3 4

4. 5

Table 1: Appearance of clinical signs in donor animals and transmission events to recipient animals: -transmission failure, - transmission success.

DPC 2 4 6 8

1st Experiment VN89 VN90 0 0 3* 4.5*/4** 3* 0

2nd Experiment VO75 VO76 0 0 3* 0 0 4*/5**

3rd Experiment VQ05 VQ06 0 0 0 0 0 0 0 0

4th Experiment VR56 VR57 0 0 4* 0 3.3* 0 0 0

Table 2: Recovery and estimated release of airborne virus over 8hrs by the donor animals; * Cyclone sampler, ** Porton sampler. DPC – days post challenge of the donor animals.

298

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Diagram: Challenge box layout, location of air sampling equipment and wall mounted fan=WMF; location of IDC=indirect contact for experiments 1 & 2, IDCA=Indirect contact companion animal for experiments 3 & 4, Donor and DC=direct contact animals for experiments 1-4.

Fig. 1: Chart of the donor challenges and successful transmissions. The dashed line shows the end of the 24 hour donor challenge. Boxes indicate a challenge with another cow with the black boxes showing a successful challenge.

Fig. 2: Diagram showing when a challenge is successful in the model under the two different assumptions: firstly when the intervals only need to overlap and secondly when the entire interval must be covered.

299

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig. 3: Probability density functions of the duration of the infectious period. Solid line: challenge and infectious periods only need to overlap for transmission. Dashed line: the infectious period needs to cover the entire challenge period.

300

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 54

UK 2001 FOOT-AND-MOUTH DISEASE EPIDEMIC: SEQUENCE DATA AND POSSIBLE AIRBORNE SPREAD G. A. König1 2 *, E. M. Cottam1, S. Upadhyaya1, J. Gloster3, L. M. Mansley4, D. T. Haydon5 and D. P. King1 2

1 Institute for Animal Health, Pirbright, Surrey GU24 0NF, UK Dr. König is based at Instituto de Biotecnología, CONICET-INTA, N. Repetto y de los Reseros, Hurlingham, Buenos Aires, 1686, Argentina. 3 Met Office, FitzRoy Road, Exeter, EX1 3PB, UK 4 Animal Health Divisional Office, Lamberkine Drive, Perth PH1 1RZ, UK 5 Department of Ecology and Evolutionary Biology, University of Glasgow, G12 8QQ, UK

The United Kingdom 2001 epizootic of FMD (Foot–and-mouth Disease) was caused by the PanAsia O strain of FMD virus. Full-genome sequencing methods have been developed to help reconstruct the transmission pathways between the IPs (infected premises). A sub-set of eight samples were sequenced in order to investigate possible airborne transmission of FMDV from IP27 (Smalmstown). Of the tested samples, four farms were found to be descendents (with 1, 4, 4, and 7 nucleotide substitutions respectively) from Smalmstown and two others differs by only 1 and 2 nucleotides respectively to the sequenced consensus existed on IP27. Only IP 227 can be excluded from a direct transmission of IP27. Consensus sequence data cannot provide information regarding the route of transmission between animals. However, the sequence data clearly shows that the FMD viruses from six of these farms are very closely related to IP27, consistent with a direct route of transmission including the airborne route. 1. INTRODUCTION Much has been written about the 2001 UK epidemic of foot-and-mouth disease and its impact upon the farming community and tourist industry [2]. Particular attention has been paid to the early phases of the epidemic and how, after it commenced at a pig farm near the Scottish Borders during mid February [1], it spread to many centres across England, Wales, Scotland, Northern Ireland and also to France and the Netherlands [7]. Many of these introductions have been attributed to the movement of animals, particularly sheep, through a series of livestock auction markets e.g. Hexham, Longtown, Northampton, Ross, and numerous animal dealers [5]. Spread from these markets was not only confined to long distances but also seeded further outbreaks in the immediate area. One such example of local FMD transmission was the spread around Longtown (Cumbria), where it is believed that virus passed via farm personnel who visited the market (on 15 February) to Smalmstown Farm (IP27) to affect approximately 300 cattle (FMD confirmed on 1 March). Interestingly, it has been suggested that subsequent disease spread from Smalmstown Farm to nine farms in the area was consistent with airborne spread of FMD virus involving distances from 1 to 16 km, many of which were over the Solway Firth [6]. The aim of this study was to investigate whether full-genome sequence data could be used to provide support for the airborne route of transmission from Smalmstown Farm. 2. MATERIAL AND METHODS Full genome sequences of FMD viruses were obtained from epithelium samples (Cottam and others 2006) collected from Smalmstown Farm as well as seven out of the nine premises referred to by Gloster and others (2005); no samples were available from the two other premises (Table 1 and Figure 1A). The samples were originally submitted to the Institute for Animal Health for laboratory investigation and were stored for the interim period at -70 ºCs. The method used to generate full genome sequences for these samples has been described previously [4]. Briefly, approximately 1.5 g of vesicular epithelium was ground by using a pestle and mortar and resuspended to a 10% suspension with 0.04 M phosphate buffer. Sample suspensions were placed into TRIzol (Invitrogen) from which RNA was subsequently extracted. The complete genomes were amplified in five

301

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 fragments using RT-PCR as described in Cottam and others (2006). Eighty four sequencing were performed in a 96-well Beckman sequencing plate using a master mix (CEQ dye terminator cycle sequencing with quick start kit; Beckman Coulter), The plate was run on a program of 30 cycles of 96°C for 20 s, 50°C for 20 s, and 60°C for 4 min. Following thermocycling, the reactions were cleaned up by ethanol precipitation before being run on the Beckman Coulter sequencing machine. The data was analyzed and assembled using SeqMan (DNAStar) and BioEdit. The genealogical relationships shown in Figure 1B were based on statistical parsimony as implemented in the software package TCS [3]. Table 1: Samples included in this study Infected premises (IP) (Farm name)

WRL ref

Estimated emission period*

Estimated infection window*

Livestock Cattle

Sheep

IP27 Smalmstown

417

22/2 to 2/3 [Peak emission 28/2 to 2/3**]

334

700

1450

9/2 to 1/3

312

280

1558

19/2 to 5/3

274

1734

20/2 to 6/3

337

197

2000

22/2 to 8/3

114

625

2085

23/2 to 9/3

143

403

2526

25/2 to 11/3

382

621

2640

28/2 to 14/3

108

IP104 The Grange IP133 West End IP191 Bowness Hall IP201 Northview IP227 Blackrigg IP342 Burnfoot IP348 Old Sandsfield

* Dates as estimated by Gloster and others (2005). ** Ideal meteorological conditions for airborne transmission of FMD virus to the farms involved in this study occurred on the night of February 28 to March 1 [6]. 3. RESULTS The sequence analysis (Figure 1B) showed that FMD virus sequences from four premises (IP104, 201, 348 and 342) were genetic descendants of IP27 with 1, 4, 4 and 7 nucleotide substitutions respectively. In terms of nucleotide substitutions, the closest of these farms was IP104, a coastal farm some 13 km from IP27 suggesting that a direct virus transmission is a strong possibility in this case. In contrast, the FMD virus sequenced from IP342 is the most genetic distant of the direct descendents from Smalmstown Farm, although the clinical material (collected on 19/3) tested was collected 11 days after the material from IP104; therefore the large number of substitutions may have been generated by on-farm replication of the virus (either on Smalmstown or Burnfoot farm) or, alternatively, by an unknown intermediate infected farm.

302

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

B A74 (M)

IP100

IP227

Longtown Mkt?

IP27

GRETNA

5 Km

IP342 IP227

IP191

SOLWAY FIRTH

IP133

IP203

IP27 IP104

IP191

IP348 IP201

IP104

IP201

IP348

IP342

IP133

CARLISLE

By the time disease was confirmed on IP27, it is probable that there had been several cycles of viral replication on the farm that may have been available to form the aerosol plume between 26 February and 2 March [6]. Therefore, while the data clearly shows that FMD virus on IP227 is not derived from IP27, it is possible that IP191 could have been infected by FMD virus released from animals infected during the early stages of disease incursion on IP27 sharing a common ancestor only one nucleotide substitution away from the virus sampled on 28 February. A similar argument can be proposed for IP133 which has a common ancestor with IP27 that differs in only 2 places with the sequenced virus on IP27, although this is less likely since this virus lineage is upstream of the main cluster of viruses recovered from the 2001 outbreak [4]. Figure 1A: Geographical locations of the farms involved in this study. Red dot = Smalmstown Farm/ IP27, green dots = infected premises, yellow dots = no samples were available from these premises, but were considered as possible for airborne transmission by Gloster and others (2005). 1B: Genetic relationships between the samples. Statistical parsimony analysis by TCS [3] of complete genome sequences of 8 FMDVs samples during this study. The connecting lines represent a nucleotide substitution, while the white circles indicate putative ancestral viruses. The dotted box is drawn to indicate the extent of lineages may have arisen from Longtown Market and the thick black line shows the location of the genetic root for the majority of samples collected during the 2001 outbreak. It is also possible to speculate about the origin of FMD on other farms where the sequences are not direct descendents of the virus recovered from Smalmstown Farm. In particular, the presence of intermediate (and un-sampled) infected farms cannot be ruled out for the majority of these transmission events. For example, IP201 and IP348 share two nucleotide substitutions from IP27, indicating that they have a common transmission history and as a result direct transmission from IP27 is perfectly feasible. As an alternative to direct transmission, infection could have spread to IP201 and IP348 via a single untyped premise (possibly IP203 where disease was detected in around 100 out of 2000 sheep grazing in the field directly across the River Eden). 4. DISCUSSION The sequence data clearly shows that the FMD viruses from at least 5 of these farms are very closely related, consistent with a direct route of transmission including the airborne route. Further work has been proposed in a follow-on project to improve our understanding of the variability that arises between sequences of FMD viruses that infect different animals within a herd, as well as the mechanisms (such as population bottleneck) that drive consensus sequences between farms. 5. ACKNOWLEDGEMENTS The authors wish to thank Defra for funding the work (SE2936 and SE4205) and INTA to support Dr. König laboratory work on Pirbright.

303

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 6. REFERENCES [1] Alexandersen, S., Kitching, R. P., Mansley, L. M., Donaldson, A. I. 2003. Clinical and laboratory investigations of five outbreaks of foot-and-mouth disease during the 2001 epidemic in the United Kingdom. Vet. Rec. 152: 489-496. [2] Anonymous. Royal Society (2002) Infectious diseases in livestock. Royal Society, Policy document 15/02. 2002. [3] Clement, M., Posada, D., Crandall, K.A. 2000. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 9: 1657–1659. [4] Cottam, E. M., Haydon, D. T., Paton, D. J., Gloster, J., Wilesmith, J. W., Ferris, N. P., Hutchins, G. H., King, D. P. (2006) Molecular Epidemiology of the foot-and-mouth disease outbreak in the United Kingdom. J. Virol. 80: 11274–11282. [5] Gibbens, J.C., Sharpe, C.E., Wilesmith, J.W., Mansley, L.M., Michalopoulou, E., Ryan, J.B.M. and Hudson, M. 2001. Descriptive epidemiology of the 2001 FMD epidemic in Great Britain: the first five months. Vet. Rec. 149: 729-743. [6] Gloster, J., Champion, H. J., Mansley, L. M., Romero, P., Brought, T., Ramirez, A. 2005. The 2001 epidemic of foot-and-mouth disease in the United Kingdom: epidemiological and meteorological case studies. Vet. Rec.156: 793–803. [7] Mansley, L. M., Dunlop, P. J., Whiteside, S. M., Smith, R. G. H. 2003. Early dissemination of FMD virus through sheep markets in February 2001. Vet. Rec. 153: 43–50.

304

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 55 LOCALISATION OF FOOT-AND-MOUTH DISEASE VIRUS AFTER ACUTE INFECTION IN CATTLE; A NOVEL, IMMUNOLOGICALLY SIGNIFICANT SITE. N. Juleff1, 2, M. Windsor1, E. Reid1, J. Seago1, Z. Zhang1, P. Monaghan1, I. W. Morrison2 and B. Charleston1 2

1 Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey GU24 0NF, UK. Centre for Tropical Veterinary Medicine, University of Edinburgh, Easter Bush Veterinary Centre, Roslin, Midlothian EH25 9RG, UK.

Control of foot-and-mouth disease involves both vaccination and controversial “stamping out” policies. Two fundamental problems remain to be understood before more effective control measures can be put in place. These problems are the foot-and-mouth disease virus [FMDV] “carrier state” and the short term duration of protection and serum neutralising antibody titres after vaccination which contrasts with the prolonged duration of immunity after natural infection. We have shown by laser capture microdissection in combination with quantitative real-time reverse transcription polymerase chain reaction, immunohistochemical analysis and corroborate by in situ hybridization that FMDV locates rapidly to, and is maintained in, the light zone of germinal centres following primary infection of naïve cattle. We propose that maintenance of non-replicating FMDV in these sites represents a source of persisting infectious virus and also contributes to the generation of long-lasting antibody responses against neutralising epitopes of the virus. 1. INTRODUCTION A strong B-cell response, characterised by high-affinity circulating neutralising antibodies, is a crucial component of the immune response against foot-and-mouth disease virus [FMDV]. Virus is cleared rapidly from blood during the acute stage of foot-and-mouth disease [FMD], coinciding closely with the emergence of an antiviral antibody response. Viral RNA is detected in the blood of infected cattle, using real-time reverse transcription polymerase chain reaction [rRT-PCR], but becomes undetectable from as early as 3 to 5 days after onset of clinical signs. This is in contrast to pharyngeal tissue including the soft palate, nasopharynx, oropharynx, palatine tonsil and mandibular lymph node which have been shown to contain viral RNA for up to 72 days after infection (Zhang and Alexandersen, 2004). The significance of continued detection of viral RNA has not been clear since FMDV proteins have not been detected, in previous studies in these tissues, following the resolution of vesicular lesions. Importantly, FMDV proteins have not been detected previously in lymphoid tissue in vivo at any stage of infection and viral proteins have not been detected in any tissue following resolution of vesicular lesions. A number of different pathologically relevant proteins, organisms and their products [including retroviruses like HIV, FIV and SIV, tetanus, and prion protein] have been shown to be retained on specialised cells called follicular dendritic cells [FDCs] in lymphoid tissue (Kosco-Vilbois, 2003; McGovern and Jeffrey, 2007). FDCs are non-phagocytic, non-dividing, radio-resistant cells that express surface receptors to bind immune complexes. The ability of FDCs to trap and retain antigen and infectious virus in a stable conformational state in the form of immune complexes for months or even years within germinal centres [GCs] and their intimate association with B-cells is a crucial component of the humoral response. FDCs are important for affinity maturation and memory B-cell development either through the presentation of surface-retained antigen to B-cells or by supporting B-cell proliferation and differentiation in a non specific manner (Haberman and Shlomchik, 2003). Additionally, the slow release of antigen from the surface of FDC is thought play a role in maintaining serum titres of specific antibody and studies have shown that the levels of retained antigen can regulate serum immunoglobulin levels (Tew et al., 1980; Szakal et al., 1989; Szakal et al., 1992). 2. MATERIALS AND METHODS 2.1. Challenge

305

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 The cattle examined in this study were infected by close contact with cattle infected with FMDV isolates O/UKG/34/2001 or O1BFS1860. 2.2. Laser capture microdissection Laser capture microdissection [LCM] was performed as described previously (Allen et al., 2004), expect that ethanol fixed frozen sections were stained with a 1% solution of toluidine blue [Fluka] for 3 minutes, washed for 30 seconds in nuclease free water, dehydrated by a graded ethanol series and allowed to air dry for 10 minutes. Three replicates of the different tissues regions, each containing six microdissected samples collected in the caps of RNase-free PCR tubes [Ambion], were collected from each tissue for RNA isolation with the RNeasy Micro Kit [Qiagen] and processed by quantitative rRT-PCR. 2.3. Quantitative rRT-PCR Reverse transcription of isolated RNA was performed concurrently with 10-fold dilution series of FMDV and bovine 28s standard RNA [Applied Biosystems]. FMDV primer/probe sets and detection were as previously described; samples with no detectable fluorescence above threshold after 50 cycles were taken to be negative (Quan et al., 2004). 28s primer/probe sets and detection were adapted from methods previously described (Valarcher et al., 2003). 2.4. Selection of monoclonal antibodies specific for conformational, non-neutralising epitopes of the FMDV capsid Mouse monoclonal antibodies [MAbs] IB11, FC6, AD10 and BF8 raised against 146 S FMDV type O1 antigen were selected by ELISA and virus neutralising antibody test as described in the Office International des Epizooties [OIE] Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, 5th edition, 2004. Immunoprecipitation analysis was performed as previously described by Rouiller et al (Rouiller et al., 1998). MAbs were subsequently screened by western blotting analysis and on FMDV vesicular lesions, non-infected tissue and on infected and mock-infected cells by immunofluorescence microscopy. 2.5. Immunofluorescence confocal microscopy Samples collected at post-mortem were snap frozen in O.C.T. compound [Tissue-Tek] and stored at − 80o C until processing. Acetone fixed frozen sections of acute tissue and tissue from 29 to 38 days post contact infection were labelled in duplicate with FMDV capsid MAbs, with consecutive sections labelled with MAbs 2C2, 3C1 (De Diego et al., 1997; Brocchi et al., 1998), kindly provided by E Brocchi, and isotype control MAbs TRT1, TRT3 (Cook et al., 1993) and AV29, a MAb directed against a chicken antigen provided by F Davison, IAH. Additional sections were labelled with MAb D46 [specific for dark zone FDCs], MAb CNA.42 [specific for light zone FDCs, kindly provided by G Delsol, Toulouse, CHU Purpan, Laboratoire d’anatomie et cytologie pathologiques, France] (Lefevre et al., 2007), MAb CC51 [anti-bovine CD21] (Howard and Morrison, 1991) and control MAb AV48, a MAb directed against a chicken antigen provided by F Davison, IAH. Anti-bovine CD32 MAbs CCG36 and CCG37 were kindly provided by C Howard, IAH. Goat anti-mouse Molecular Probes Alexa-Fluorconjugated secondary MAbs [Invitrogen] were used and nuclei were stained with DAPI [Sigma]. All data were collected sequentially using a Leica SP2 scanning laser confocal microscope. 2.6. In situ hybridization An optimised In situ hybridization method for the detection of FMDV was developed using digoxigenin-labelled RNA probes based on the protocol described by Prato Murphy et al (Prato Murphy et al., 1999) and optimised for frozen sections incorporating pre-hybridization blocking steps, tyramide signal amplification and alkaline phosphatase based visualization as described by Yang et al (Yang et al., 1999). Consecutive frozen sections were simultaneously probed with; 500 nt 3D antisense and sense RNA probes specific for serotype O/UKG/34/2001 non-structural protein 3D coding sequence, IgG1 antisense RNA probe [686 nt] for the CH2, CH3 and hinge region of bovine IgG1 mRNA and a 615 nt swine vesicular disease virus probe specific for structural proteins 1C and 1D coding sequence (Prato Murphy et al., 1999). 3. RESULTS 3.1. Laser capture microdissection GC and non-GC regions of the dorsal surface of the palatum molle [dorsal soft palates], pharyngeal tonsils (Liebler-Tenorio and Pabst, 2006), palatine tonsils, lateral retropharyngeal lymph nodes and mandibular lymph nodes obtained from four cattle 38 days post contact exposure to FMDV

306

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 serotype O were selected for LCM. FMDV genome was detected consistently by quantitative rRTPCR within the GC samples obtained by LCM. No FMDV genome was detected in the epithelium of the dorsal soft palates and pharyngeal tonsils. No FMDV genome was detected in the crypt epithelium, glandular epithelium and interfollicular regions of the palatine tonsils or the interfollicular regions of the mandibular lymph nodes and lateral retropharyngeal lymph nodes. No FMDV genome could be detected in GC samples obtained by LCM from non-infected control animals. 3.2. In situ hybridization An optimised detection protocol with tyramide signal amplification was compared to conventional chromagenic detection (Prato Murphy et al., 1999). Tyramide signal amplification combined with endogenous peroxidase and alkaline phosphatase quenching enhanced the specific hybridization signal and reduced background signal compared to conventional detection. The FMDV 3D antisense RNA probes, control 3D sense probes and control SVD RNA probes were validated on FMDV infected and mock infected BHK-21 cells and on tissue samples collected from non-infected and FMDV infected animals. Despite the obvious signal obtained when detecting positive strand viral RNA in infected cells, it was difficult to detect negative strand viral RNA by in situ hybridization. In situ hybridization of dorsal soft palates, mandibular lymph nodes, palatine tonsils, pharyngeal tonsils and lateral retropharyngeal lymph nodes from ten animals 14 to 38 days post contact infection supported the LCM results. FMDV 3D RNA was identified in GCs of mandibular lymph node, palatine tonsil and lateral retropharyngeal lymph node sections but not in other compartments of these tissues. 3.3. Immunofluorescence confocal microscopy To determine whether viral RNA was associated with viral structural proteins; frozen sections from the dorsal soft palates, pharyngeal tonsils, palatine tonsils, lateral retropharyngeal lymph nodes and mandibular lymph nodes collected 29 to 38 days post contact infection were analysed using a new set of virus-specific MAbs shown to be specific for conformational, non-neutralising epitopes of the FMDV capsid. These new MAbs immunoprecipitated FMDV capsid proteins, consistent with previously published antibodies, yet did not detect FMDV proteins by western blot and were nonneutralising. Also, the MAbs readily detected virus in bovine tongue during acute infection, virus infected cells and rare infected cells in lymphoid tissue during acute infection. The ability of the MAbs specific for FMDV capsid to detect immune complexed virus on the surface of cells in culture was also confirmed. The anti-FMDV capsid MAbs gave a diffuse punctate pattern of positive labelling which was restricted to GCs within lymphoid tissue and confined to the light zone within the GC from 29 days post infection. By contrast, the FMDV non-structural proteins 3A and 3C could not be detected in any of the tissues from animals after 28 days post contact infection. Antibodies specific for 3A and 3C readily detected infected cells in FMDV vesicles and lymph tissue during the acute phase of infection and in infected BHK-21 cells, co-localising with FMDV capsid. The diffuse punctate pattern of labelled viral capsid was shown to be localised to the light zone FDC network by co-labelling with an antibody specific for light zone FDCs. Detailed analysis of in situ hybridization and immunohistochemistry showed a consistent punctate pattern. The punctate labelling pattern observed was consistent with the distribution pattern of iccosomes on FDCs (Szakal et al., 1988). This pattern is in contrast to the diffuse cytoplasmic labelling pattern of cells observed during acute infection in vivo and in infected cells in vitro. 4. DISCUSSION We have shown that FMDV genome, using LCM and quantitative rRT-PCR, can be detected consistently in GCs within the dorsal soft palate, pharyngeal tonsil, palatine tonsil, lateral retropharyngeal lymph node and mandibular lymph node at 38 days post contact infection. Also, FMDV genome in these tissues was restricted to the GC. These findings were confirmed with in situ hybridization studies, which revealed FMDV 3D RNA in GCs of lymphoid tissues but not in other compartments of these tissues. Using MAbs specific for conformational, non-neutralising epitopes of the FMDV capsid, we identified viral structural proteins restricted to the light zone FDC network of GCs within mandibular lymph nodes, lateral retropharyngeal lymph nodes and palatine tonsils up to 38 days post contact infection, but not in the dorsal soft palates or pharyngeal tonsils. FMDV capsid was detected in mandibular lymph node GCs of all animals examined between 29 to 38 days post contact infection [n = 22], including five animals where FMDV could not be recovered by virus isolation or detected by rRT-PCR analysis of oropharyngeal scrapings collected at post-mortem 29

307

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 to 34 days post infection using probang sampling cups (Alexandersen et al., 2002). These results indicate that virus is likely to persist in all cattle to some degree following infection. Although MAbs specific for FMDV non-structural proteins 3A and 3C could detect infected cells in vitro and in vivo during the acute phase of infection, no FMDV non-structural proteins were detected in any of the tissues examined from 29 days post contact infection. The absence of detectable FMDV nonstructural proteins indicates that the presence of viral RNA is not associated with active viral replication. The finding of close co-localisation of viral RNA and capsid conformational epitopes, in the absence of non-structural proteins, supports the hypothesis that FMD viral particles or immune complexes are maintained in GC light zones in a non-replicating state. The results of these studies have important implications for understanding both the mechanism of viral persistence and the ability of FMDV infection to stimulate long-lasting antibody responses. FDCs are known to be non-endocytic cells capable of capturing and retaining antigen in the form of immune complexes for long periods of time (Mandel et al., 1980; Haberman and Shlomchik, 2003). Retention of immune complexed FMDV particles within lymphoid tissue represents a possible source of the infectious material detected by pharyngeal sampling of infected cattle either by direct harvesting of mucosal associated lymphoid tissue GCs or sampling of secondary cells, for example macrophages, dendritic cells or B-cells, able to support a low level virus replication cycle in the presence of high levels of neutralising antibodies (Mason et al., 1993; Rigden et al., 2002). FDCs are notoriously difficult cells to work with, so far we have been unable to rescue infectious or immune complexed FMDV from lymphoid tissue most likely due to technical difficulties working with the bovine system. Retention of other viruses such as HIV in a replication-competent state within the light zone of GCs has been reported and the next step will require the development and interrogation of murine model systems (Smith et al., 2001). FDC-trapped HIV has been shown to represent a significant reservoir of infectious and highly diverse HIV, demonstrating greater genetic diversity than most other tissues, providing drug-resistant and immune-escape quasispecies that contribute to virus transmission, persistence and diversification (Keele et al., 2008). Retention of intact FMDV particles on the FDC network would therefore provide an ideal mechanism of maintaining a highly cytopathic and lytic virus like FMDV extracellularly in a non-replicating, native, stable nondegraded state (Tew and Mandel, 1979; Smith et al., 2001). This reservoir could serve as a source of genetically diverse viral mutants [quasispecies] able to infect susceptible cells that come into contact with the FDC network (Vosloo et al., 1996; Domingo et al., 2002). FMDV infection in ruminants elicits an immune response that can provide protection for several years (Cunliffe, 1964) and the level of protection correlates well with specific serum neutralising antibody titres [SNTs] (Alexandersen et al., 2003). This is in contrast to vaccination, with current FMDV vaccines prepared with inactivated virus and adjuvants, providing short term duration of SNTs and protection. Long-term maintenance of elevated, specific antibody levels in mice following acute vesicular stomatitis virus [VSV] infection has been shown to be associated with the colocalisation of antigen with specific memory B-cells within long-lived GCs (Bachmann et al., 1996). VSV is a cytolytic virus that does not persist in an infectious form in mice, thus highlighting the function of FDC trapping and retention serving as a long-term repository of immunogenic antigen for maintenance of SNTs. Hence, efficient retention within the GCs of intact viral capsids, as opposed to the constituent viral proteins, may be a requirement for sustaining antibody responses relevant for providing protection against challenge. Indeed, in a recent review of the functional significance of antigen retained on FDC, Kosco-Vilbois suggests the observation that B-cell responses are independent of FDC-associated antigen is only valid in mice that are immunised with forms of antigen that leave persistent depots (Kosco-Vilbois, 2003). Therefore, we believe that long-term antibody responses detectable after FMDV infection are maintained in part by antigen persisting on FDCs. Based on the evidence presented here we suggest the persistence of FMDV after acute infection is both a consequence of the host immune response and a requirement for the long-term maintenance of protective virus-specific antibody responses. The data described above is a summary of a manuscript currently in press for publication in PLoS ONE 2008, with the title Foot-and-Mouth Disease Virus persists in the Light Zone of Germinal Centres. 5. ACKNOWLEDGEMENTS We thank B Bankowski and D Gibson for assisting with the in vivo procedures. P Hamblin for performing the virus neutralisation tests. K Ebert for assisting with screening of probang samples by rRT-PCR. P Hawes and J Simpson for assistance with confocal microscopy and R Aitken for

308

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 providing bovine IgG1 cDNA. The work was funded by the Biotechnology and Biological Sciences Research Council. 6. REFERENCES [1] Alexandersen, S., Zhang, Z. & Donaldson, A.I. 2002. Aspects of the persistence of footand-mouth disease virus in animals-the carrier problem. Microbes. Inf. 4: 1099-1110. [2] Alexandersen, S., Zhang, Z., Donaldson, A.I. & Garland, A.J.M. 2003. The pathogenesis and diagnosis of foot-and-mouth disease. J. Comp. Path. 129: 1-36. [3] Allen, C.D.C., Ansel, K.M., Low, C., Lesley, R., Tamamura, H., Fujii, N. & Cyster, J.G. 2004. Germinal center dark and light zone organization is mediated by CXCR4 and CXCR5. Nat. Immunol. 5: 943-952. [4] Bachmann, M.F., Odermatt, B., Hengartner, H. & Zinkernagel, R.M. 1996. Induction of long-lived germinal centers associated with persisting antigen after viral infection. J. Exp. Med. 183: 2259-2269. [5] Brocchi, E., De Diego, M., Berlinzani, A., Gamba, D. & De Simone, F. 1998. Diagnostic potential of Mab-based ELISAs for antibodies to non-structural proteins of foot-and-mouth disease virus to differentiate infection from vaccination. Vet. Quart. 20: S20-24. [6] Cook, J.K.A., Jones, B.V., Ellis, M.M., Jing, L. & Cavanagh, D. 1993. Antigenic differentiation of strains of turkey rhinotracheitis virus using monoclonal antibodies. Avian Path. 22: 257-273. [7] Cunliffe, H.R. 1964. Observations on the duration of immunity in cattle after experimental infection with foot-and-mouth disease virus. Cornell Vet. 54: 501-510. [8] De Diego, M., Brocchi, E., Mackay, D. & De Simone, F. 1997. The non-structural polyprotein 3ABC of foot-and-mouth disease virus as a diagnostic antigen in ELISA to differentiate infected from vaccinated cattle. Arch. Virol. 142: 2021-2033. [9] Domingo, E., Baranowski, E., Escarmís, C. & Sobrino, F. 2002. Foot-and-mouth disease virus. Comp. Immunol. Microbiol. Infect. Dis. 25: 297-308. [10] Haberman, A.M. & Shlomchik, M.J. 2003. Reassessing the function of immune-complex retention by follicular dendritic cells. Nat. Rev. Immunol. 3: 757-764. [11] Howard, C.J. & Morrison, W.I. 1991. Leukocyte antigens of cattle, sheep and goats. Vet. Immunol. Immunopath. 27: 1-94. [12] Keele, B.F., Tazi, L., Gartner, S., Liu, Y., Burgon, T.B., Estes, J.D., Thacker, T.C., Crandall, K.A., McArthur, J.C. & Burton, G.F. 2008. Characterization of the follicular dendritic cell reservoir of human immunodeficiency virus type 1. J. Virol. 82: 5548-5561. [13] Kosco-Vilbois, M.H. 2003. Are follicular dendritic cells really good for nothing? Nat. Rev. Immunol. 3: 764-769. [14] Lefevre, E.A., Hein, W.R., Stamataki, Z., Brackenbury, L.S., Supple, E.A., Hunt, L.G., Monaghan, P., Borhis, G., Richard, Y. & Charleston, B. 2007. Fibrinogen is localized on dark zone follicular dendritic cells in vivo and enhances the proliferation and survival of a centroblastic cell line in vitro. J. Leuk. Biol. 82: 666-677. [15] Liebler-Tenorio, E.M. & Pabst, R. 2006. MALT structure and function in farm animals. Vet. Res. 37: 257-280. [16] Mandel, T.E., Phipps, R.P., Abbot, A. & Tew, J.G. 1980. The follicular dendritic cell: long term antigen retention during immunity. Immunol. Rev. 53: 29-59. [17] Mason, P.W., Baxt, B., Brown, F., Harber, J., Murdin, A. & Wimmer, E. 1993. Antibodycomplexed foot-and-mouth disease virus, but not poliovirus, can infect normally insusceptible cells via the Fc receptor. Virology 192: 568-577. [18] McGovern, G. & Jeffrey, M. 2007. Scrapie-specific pathology of sheep lymphoid tissues. PLoS ONE 2: e1304. [19] Prato Murphy, M.L., Forsyth, M.A., Belsham, G.J. & Salt, J.S. 1999. Localization of footand-mouth disease virus RNA by in situ hybridization within bovine tissues. Virus Res. 62: 67-76. [20] Quan, M., Murphy, C.M., Zhang, Z. & Alexandersen, S. 2004. Determinants of early footand-mouth disease virus dynamics in pigs. J. Comp. Path. 131: 294-307. [21] Rigden, R.C., Carrasco, C.P., Summerfield, A. & McCullough, K. 2002. Macrophage phagocytosis of foot-and-mouth disease virus may create infectious carriers. Immunology 106: 537-548. [22] Rouiller, I., Brookes, S.M., Hyatt, A.D., Windsor, M. & Wileman, T. 1998. African swine fever virus is wrapped by the endoplasmic reticulum. J. Virol. 72: 2373-2387. [23] Smith, B.A., Gartner, S., Liu, Y., Perelson, A.S., Stilianakis, N.I., Keele, B.F., Kerkering, T.M., Ferreira-Gonzalez, A., Szakal, A.K., Tew, J.G. & Burton, G.F. 2001. Persistence of infectious HIV on follicular dendritic cells. J. Immunol. 166: 690-696.

309

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [24] Szakal, A.K., Kapasi, Z.F., Masuda, A. & Tew, J.G. 1992. Follicular dendritic cells in the alternative antigen transport pathway: microenvironment, cellular events, age and retrovirus related alterations. Sem. Immunol. 4: 257-265. [25] Szakal, A.K., Kosco, M.H. & Tew, J.G. 1988. A novel in vivo follicular dendritic celldependent iccosome-mediated mechanism for delivery of antigen to antigen-processing cells. J. Immunol. 140: 341-353. [26] Szakal, A.K., Kosco, M.H. & Tew, J.G. 1989. Microanatomy of lymphoid tissue during humoral immune responses: structure function relationships. Ann. Rev. Immunol. 7: 91-109. [27] Tew, J.G. & Mandel, T.E. 1979. Prolonged antigen half-life in the lymphoid follicles of specifically immunized mice. Immunology 37: 69-76. [28] Tew, J.G., Phipps, R.P. & Mandel, T.E. 1980. The maintenance and regulation of the humoral immune response: persisting antigen and the role of follicular antigen-binding dendritic cells as accessory cells. Immunol. Rev. 53: 175-201. [29] Valarcher, J.-F., Furze, J., Wyld, S., Cook, R., Conzelmann, K.-K. & Taylor, G. 2003. Role of alpha/beta interferons in the attenuation and immunogenicity of recombinant bovine respiratory syncytial viruses lacking NS proteins. J. Virol. 77: 8426-8439. [30] Vosloo, W., Bastos, A.D., Kirkbride, E., Esterhuysen, J.J., van Rensburg, D.J., Bengis, R.G., Keet, D.W. & Thomson, G.R. 1996. Persistent infection of African buffalo (Syncerus caffer) with SAT-type foot-and-mouth disease viruses: rate of fixation of mutations, antigenic change and interspecies transmission. J. Gen. Virol. 77: 1457-1467. [31] Yang, H., Wanner, I.B., Roper, S.D. & Chaudhari, N. 1999. An optimized method for in situ hybridization with signal amplification that allows the detection of rare mRNAs. J. Histochem. Cytochem. 47: 431-446. [32] Zhang, Z. & Alexandersen, S. 2004. Quantitative analysis of foot-and-mouth disease virus RNA loads in bovine tissues: implications for the site of viral persistence. J. Gen. Virol. 85: 25672575.

310

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 56

INVESTIGATIONS INTO FINDINGS OF FMD SEROPOSITIVE SHEEP AND GOATS IN CYPRUS, AN FMD-FREE COUNTRY D. J Paton1*, N. Ferris1, G. Hutchings1, Y. Li1, K. Swabey1, P. Keel1, P. Hamblin1, D.P. King1, S. Reid1, K. Ebert1, S. Savva2, K.Georgiou2 and C. Kakoyiannis2 1

European Community Reference Laboratory for FMD, Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, United Kingdom. 2 Veterinary Services, 1417 Nicosia, Republic of Cyprus.

1. INTRODUCTION Foot-and-mouth Disease (FMD) has not been recorded in Cyprus since 1964 and there has been no vaccination programme since 1984. In October 2007, clinical suspicion of Bluetongue led to the testing and detection of FMD specific antibodies in sheep and goat flocks. According to the OIE Terrestrial Animal Health Code, infection with FMD virus is confirmed by findings of specific antibodies in association with clinical signs of the disease.

2. MATERIALS AND METHODS Clinical histories, age profiles and large numbers of samples for virus and antibody detection were collected from farms close to the index case where antibodies to FMD virus were found. Wider clinical and serological surveillance was conducted throughout the areas which are under the effective control of the Republic of Cyprus.

3. RESULTS Antibodies to structural (serotype O) and non-structural proteins of FMD were found in several flocks of sheep and goats in south east Cyprus, but not in other parts of the country. Only one seropositive flock showed clinical signs of FMD. Virus could not be recovered or detected from large numbers of samples collected from serologically positive flocks. Nearby cattle and pigs did not show clinical signs and were seronegative. The seropositive animals were homebred and at least three years old.

4. DISCUSSION It was concluded that the most likely cause of these findings was an unreported and self-limiting incursion of FMD virus infection into south east Cyprus more than three years ago.

311

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

312

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 57

RESULTS OF MOLECULAR TESTING OF SAMPLES FROM PAKISTAN *1+

S. Alexandersen 1

, J. Klein , and M. Hussain

Department of Virology, National Veterinary Institute, Technical University of Denmark, Lindholm, DK-4771 Kalvehave, Denmark +

Affiliation from 1 October 2008, National Centre for Foreign Animal Disease, Canadian Food Inspection Agency, 1015 Arlington Street, Winnipeg, R3E 3M4 Manitoba, Canada

Food and Agriculture Organization of the United Nations - Pakistan, NARC, Islamabad, Park Road, PK-45500, Pakistan

ABSTRACT Introduction FMD is endemic in Pakistan and causes significant economic losses in livestock farming, in particular in dairy buffalo and cattle. In an accompanying paper we describe the sampling, shipping and real-time RT-PCR results of samples collected at the Landhi Dairy/Cattle Colony (LCC) outside Karachi in Pakistan and in the present paper we describe the results of the sequence analysis performed on the same samples. Materials and methods From April 2006 to April 2007 we collected mouth swabs and epithelial samples from a large number of animals in LCC. Samples were shipped to Lindholm, Denmark, as either stabilised and inactivated swab samples in RNA stabilising buffer (i.e. Qiagen RLT buffer), as cDNA prepared at the National Reference Laboratory in Islamabad, Pakistan, or in a few cases as vesicular epithelium containing live virus and shipped as dangerous goods via the WRL at Pirbright Laboratory, UK (kindly assisted by David Paton, Nigel Ferris and Geoff Hutchings). PCR-based sequencing of the VP1 and capsid region as well as full genome sequencing was performed using standard techniques. Results The results of the sequencing and phylogenetic analyses are presented. A high number of samples contained serotype O of the Pan Asia II lineage or serotype A of the A Iran 2005 lineage. Discussion Sequencing of FMDV RNA present in mouth swabs rendered stable and non-infectious by a suitable stabilising buffer provides a valuable and useful alternative to sequencing based on infectious epithelial samples and provides reliable sequence information for characterisation of circulating strains of FMDV in an endemic situation as observed in the LCC in Pakistan. Preliminary studies indicate (data not shown) that epithelial samples may also be stabilised and rendered noninfectious “on site” by using e.g. Ambion RNAlater, and consequently, shipment of samples containing infectious FMDV, which is both cumbersome, expensive and may potentially cause infection if accidentally released, appears no longer to be required for prevalence studies and molecular analysis including sequencing and strain characterisation. Moreover, as samples shipped in a stabilised form may after RNA extraction be used for re-generating infectious virus by transfection into susceptible cell cultures (Graham Belsham, personal communication), this approach may truly provide a significant leap forward for a thorough knowledge-based vaccination approach for FMD control in difficult endemic settings. INTRODUCTION The Landhi Dairy/Cattle Colony (LCC) is located in the eastern suburbs of Karachi and is the largest dairy colony in Pakistan and the biggest buffalo colony worldwide. There is estimated to be a total

313

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 of around 250-280 000 buffalo and 15-25 000 cattle in the colony and that there are between 1000-2000 farms each with around 100-200 animals on average. All of this is located on only 750 acres of land (including 250 acres for primitive roads). Additional background information regarding the LCC, the collaborative project involved, the collection and shipping of samples and the results of real-time RT-PCR analysis is described in an accompanying paper. 2. MATERIALS AND METHODS 2.1 Virus isolation and sequencing. The study combines both field and molecular epidemiology and was planned as a longitudinal and cross-sectional survey in the Landhi Dairy/Cattle Colony (LCC). Samples were collected during field trips to Pakistan in January/February, April and September 2006 and January and April 2007. In addition, a smaller number of samples were collected by local Veterinary Officers each month. Mouth swab samples were collected from randomly selected animals while epithelial samples were collected from animals showing clinical signs of acute FMD. A few epithelial samples were also collected from a small farm just outside LCC and from the Nagori Cattle Colony, a smaller dairy colony about 50 km from Karachi with around 10 000 animals (mainly buffalo) and with less crowded conditions for the animals and with better animal husbandry than LCC. The location of all farms being sampled was logged by handheld GPS equipment and the farmers were thoroughly questioned for knowledge that may be of relevance for the analysis. Epithelial samples were collected using standard procedures in phosphate-buffered 50% glycerol and these samples were o

stored at –20 C until shipped on dry ice. Epithelial samples were initially sent to the World Reference Laboratory (WRL) at Pirbright, UK, to allow sharing of samples and results. From October 2006 to March 2007 we collected monthly serum samples from 30 randomly selected buffaloes immediately after slaughter at the local slaughterhouse in LCC. Virus isolation, ELISA, real-time RT-PCR, sequencing and phylogenetic analysis were done using standard methods as 3, 4

described in the accompanying paper and in previous publications . For additional details including how mouth swab sampling was organized and handled please see the accompanying paper. 3. RESULTS 3.1 Infection prevalence The prevalence of samples positive for FMDV RNA by real-time RT-PCR analysis of mouth swab samples collected from randomly selected animals in randomly selected farms in LCC is described in the accompanying paper. Among animals from farms where no animals with clinical signs of FMD were present, the overall average percentage of positive swabs was above 10% while the percentage of positive swabs in farms with other animals having signs of healing/healed FMD was almost 50% and more than 80% of swabs were positive in farms with other animals having signs of acute FMD. Figure 1 displays the FMDV infection prevalence at aggregate level from April 2006 to April 2007 among animals from farms where no clinical signs of FMD were present, based on the number of FMDV RNA positive animals found in a two-stage sampling scheme. The farm-level (herd-level) prevalence reflects the number of farms with positive animals, calculated as the proportion of farms with infected animals per month, and the animal-level prevalence reflects the number of FMDV RNA positive animals within the sampled population. Over the one-year period the mean animal-level prevalence for farms with PCR-positive animals was nearly 20% and the mean farm level prevalence was around 35% meaning that around one third of the 134 visited farms over the 13 months study period had an average of one in five animals positive for FMDV RNA. The prevalence of FMDV RNA positive animals was very high in August 2006 where samples from all 9 sampled farms were positive for FMDV RNA (farm-level prevalence of 100%) and 17 of 27 individual samples were positive (animal level prevalence around 60%) possibly due to the rainy season in July and August. The moving average analysis (see Figure 1), which removes random variations within the point estimates, showed an appreciable increase in both farm-level and animal level prevalence from December 2006 to March 2007 (farm-level prevalence approaching 100% and animal-level prevalence at 50% or higher), likely expressing the cumulative effect of the second rainy season, the additional movements of animals caused by the Eid ul-Azza festival (which that year fell on 31 December 2006) and possibly the slightly cooler temperature during this period (Figure 1).

314

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3.2 Participatory information During sampling the owners/managers of the farms were questioned about their FMD vaccine practice. Of the 127 farmers included in the questionnaire, 88% vaccinated their animals. Of those, 79% used trivalent Aftovax vaccine (Merial) and 9% the local monovalent type O vaccine. Another 4% mentioned that they did not know whether they had vaccinated against FMD, because they had vaccinated, but the vaccine used was unknown, while 8% did not vaccinate against FMD. All farmers that vaccinated their animals, administered the vaccine only once and not as recommended with an additional booster vaccination two to six weeks after the initial vaccination and moreover, only 4% were vaccinating their animals on a regular basis twice a year, whereas the majority of the farmers only vaccinated the new animals after they had arrived onto the farm. In regard to the likely cause of new outbreaks of FMD on their farm, the owners/managers indicated that they believed that introduction of new animals was most important (43%) while introduction from the surroundings was rated as the second most likely cause (5%) while more than half of the replies indicated that they did not know (unknown/no reply 52%). Thus, if the unknown/no reply answers are not considered, around 90% of the farmers replying indicated that the most likely cause of new outbreaks was introduction of new animals while 10% considered introduction from the surroundings most likely. 3.3 Serology The results of the antibody ELISA testing of the 180 samples collected (30 samples per month for 6 months) from the local slaughterhouse in LCC are shown in Figure 2 and Figure 3. The results indicated that the animals had a high level of antibodies against serotype O and serotype A during the whole period. Antibodies against serotype Asia 1 and serotype C (the latter likely due to vaccination or cross-reaction) were more variable although many animals were apparently positive. The levels of antibodies against these two serotypes were particularly variable, and low in several animals, in the November 2006 sampling. The reason for this is unknown, but it may correlate with the apparent low infection rate observed in LCC during the months of September, October and November of 2006 (Figure 1) although the circulating FMDV were clearly of serotype O and A (see later). All 180 samples were also tested for antibodies against the non-structural proteins of FMDV and 176 were positive and only 4 samples negative (Figure 3). The calculated mean ODP for all 180 tested serum samples for antibodies to serotypes O, A and Asia 1 (at a 1:5 serum-dilution) were 9 ± 4, 6 ± 1 and, 8 ± 6, respectively, corresponding to an estimated endpoint-titre of 1:320 and with a range of one twofold dilution step (i.e. endpoint titres of 1:160 to 1:640). For serotype C the mean ODP was significantly higher (20 ± 2), corresponding to an endpoint-titre of 1:40 (range 1:20 to 1:80) and as stated above likely due to vaccination or possibly cross-reactions to the other serotypes. To further substantiate these findings, 10 serum samples were randomly selected to determine the highest serum dilution that gave a positive signal in the ELISA for each serotype (Figure 4). The Median antibody titre for serotypes O, A and Asia 1 was 1:320 while the Median titre for serotype C was only 1:40. We also determined the endpoint-titres of these 10 samples in the virus neutralisation test (Figure 5). Generally, the virus neutralisation titres were consistent with the results of the ELISA titration with the Median titres for serotypes O, A and Asia 1 equal to or above 1:100 with the titres for serotype A and Asia 1 being more variable than for type O. The Median virus neutralisation titre for antibodies against serotype C was less than 1:50 and with very little variability. 3.4 Sequencing and phylogenetic analysis We sequenced the partial or full 1D coding region from the cDNA preparations from 58 of the 106 FMDV RNA positive swab-samples from animals without clinical signs (see accompanying paper). In addition, we sequenced the full 1D region from cDNA preparations from 17 epithelium samples collected during 2006. From all sequenced samples, 19 belonged to serotype O, including 9 swabs and 10 epithelium samples, and 56 to serotype A, including 49 swabs and 7 epithelium samples. Figure 6 shows the unrooted phylogenetic tree of the serotype O isolates from Pakistan in relation to similar serotype O sequences published in Genbank. The serotype O isolates from the Pakistan cluster are monophyletic and thus share a common ancestor. The most related isolates originate from Bhutan/Nepal, collected between 2003 and 2004 and belonging to a new Pan Asia lineage described by WRL at Pirbright in 2007 (http://www.wrlfmd.org) and designated Pan Asia II. Figure 7 shows a subtree of serotype O, containing only sequences from Pakistan, Bhutan, Nepal and Malaysia and demonstrates the close relationship between the isolates from Bhutan/Nepal and Pakistan. Noticing the small branch lengths, it is highly surprising that the sequence derived from the locally produced monovalent type O vaccine produced in Lahore is very closely related to samples derived from infected animals.

315

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Figure 8 displays the deduced amino acid sequence of the partial VP1 sequence of the serotype O isolates. There is very high amino acid conservation, even though samples were collected from 2003 to 2006. However, the isolates from Pakistan are clearly distinct from the related isolates from Malaysia, Bhutan and Nepal at residues 143 and 200. Residue 143, located only 4 amino acids before the RGD motif in the GH-loop, contains histidine in the isolates from Pakistan whereas the others, like the majority of other published serotype O sequences, have proline at this position likely leading to a change in secondary structure. Figure 9 shows the phylogram of the serotype A isolates. All the isolates from Pakistan belong to the recently discovered A/Iran/2005 lineage although the branch lengths, as typical for serotype A, are larger than those of serotype O. 3.5 Virulence and host species In our study the majority of clinically affected animals infected with the FMDV A/Iran/2005 lineage were cattle (7 cattle and only 2 buffalo) while for FMDV type O, 13 out of 18 epithelium samples from clinically affected animals were from buffaloes and only 5 from cattle. In regard to the nonclinically affected animals, the results from the sequencing of FMDV RNA from 58 positive mouth swab samples showed that around 20% were positive for serotype O and the other approximately 80% positive for type A and approximately 90% came from buffalo and only around 10% from cattle. Taking these results together, and taking into consideration that the LCC has a large population of buffaloes and a minor population of cattle and that we took approximately 10 times more mouth swab samples from buffalo than from cattle, suggest that serotype O caused clinical FMD in both cattle and buffalo while in contrast, the A/Iran/2005 isolates described here, mainly caused clinical disease in cattle but subclinical infection in buffalo. 3.6 Sequencing and analysis of the complete coding sequence of selected A/IRN/2005 isolates and the potential for a recombination event The complete coding sequence (CDS) of 3 A/IRN/2005-like isolates from epithelial samples collected in LCC (Pakistan1, Pakistan3 and Pakistan5) in early 2006 together with an isolate from the first recognized outbreak of the A/IRN/2005 lineage in Turkey in 2005 (sample kindly provided by WRL) were sequenced and compared with sequences available in Genbank. Phylogenies were then inferred for the complete CDS, as well as for each protein coding region of the FMDV genome. Figure 10 displays the phylogeny of the complete CDS of selected serotype O, Asia1 and A isolates and indicates a close relationship of the A/IRN/2005 subtype to the A22 and A28 subtypes circulating several years ago in the Middle East region. The A/IRN/2005 subtype appears to share a common ancestor with A Iran105 which may have originated in Iran in 1998. The relationships were further analysed by inferring phylogenies of the individual genome regions encoding for the structural proteins and subsequently the non-structural proteins. With the exception of 1A (VP4) which is not surface exposed, distinct, serotype-specific clustering was observed in the structural region and the A/IRN/2005 sublineage obviously shares a common ancestor with A Iran105 from 1998 and is also related to the A22 lineage (Figure 11). However, when analysing the 1D genome region, encoding the VP1 protein, in more detail, the A/IRN/2005 sublineage clusters together with the A/IRN/99 sublineage while A Iran105 (A Iran105 not to be confused with the A/IRN/05 lineage) clusters together with the A/IRN/96 sublineage and the A22 sublineage is in this region well separated from the A/IRN/ 2005 sublineage (Figure 12). In contrast to the situation in the structural protein coding region, the non-structural protein coding regions displayed a much more complex phylogenetic picture. The phylogeny of the 2B region places the A/IRN/2005 sublineage in close relation to an A15 lineage from Thailand in 1960, A16 Belem from 1959, A12 Valle from 1932 and O5 India from 1962 while the inferred phylogeny of the 2C region indicates a clear relationship between the A/IRN/2005 sublineage and an Asia1 lineage from Lebanon, an Asia1 India97 vaccine strain as well as O1 Manisa. In both phylogenies non-serotype specific grouping can be observed between some Asia1, A and O sublineages, however the Pan Asia sublineage of serotype O and the A22, A23 and A28 sublineage of serotype A are monophyletic, i.e. each has apparently an inferred common ancestor well separated from A/IRN/2005 (Figure 13). The phylogeny of the coding sequence of the 3AB non-structural region indicated that the A/IRN/2005 sublineage related to a group of old serotype O, A and C isolates that are also related to the A/IRN/2005 sublineage in the 2B region while the phylogeny of the 3C protease region indicated that the A/IRN/2005 sublineage may share a common ancestor with the Pan Asia sublineage of serotype O and the Asia1 India97 vaccine strain. The 3C phylogeny indicated non-serotype specific clustering while the phylogeny of the 3D region indicated a potential relationship with the Pan Asia lineage of serotype O and with the previously mentioned group of old isolates and the A Iran105 and Asia1 India01 isolates. The phylogeny of the Leader protease region indicates that the A/IRN/2005 sublineage is related to the A22/A28 lineages, but still with a clear evolutionary distance (data not shown).

316

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 4. DISCUSSION The results presented indicate that mouth swab samples collected into a suitable inactivating and stabilising buffer represent an easy way to ship samples to international laboratories and can be used to generate valid epidemiological information regarding the temporal prevalence of virus circulation and characterisation of circulating strains of FMDV by RT-PCR and sequencing. The apparent prevalence of subclinical infection in the LCC varied over the one year study period with low levels in April to July and September to November and with peaks in August 2006 and December 2006 to March 2007 coinciding with the periods with more precipitation and for the period stretching to March 2007 with the Eid ul-Azza festival that likely increases transmission due to the high number of animals transported across the country. The average percentage of FMDV RNA positive mouth swabs collected in randomly selected farms without clinical evidence of FMD was above 10% and assuming that these animals are infected and the detection window for FMDV 1,2

RNA is around 14 days , then the accumulated “prevalence” of infection over a one-year period, the time period most animals stay in LCC before being transported for breeding or slaughter, would be between 200 and 300%, indicating that an individual animal may be exposed to circulating FMDV on average 2 to 3 times during a single lactation period in LCC. The rate of FMDV RNA positive animals was much higher in farms with clinical signs of healing/healed or with acute FMD where FMDV RNA positive mouth swabs reached 50% and 80% respectively, indicating that under the conditions observed in LCC all/most animals in an affected herd are exposed/infected to/with FMDV within a period of 3-4 weeks, however, more data is needed before any definitive conclusions can be drawn. The indication of animals in LCC being exposed to FMDV on average 2 to 3 times a year corresponds well with our findings of significant circulation of FMDV serotype O and A together with other evidence pointing to the potential presence of serotype Asia 1 as reported by WRL in the period from 2002 to 2005. This significant circulation of FMDV is occurring despite widespread, but apparently non-effective, vaccination reported by the farmers, and although the effect of vaccination can not completely be ruled out, the serological results also support widespread circulation as indicated by 176 of 180 serum samples collected from the LCC slaughterhouse being positive for antibodies to FMDV non-structural proteins and with high titres of antibodies against serotype O and A. The titres of antibodies to serotype Asia 1 were also relatively high but more variable and this finding may be explained by a more limited circulation of this virus, consistent with us not finding any virus samples positive for type Asia 1, or with likely circulation of Asia 1 in previous years. The finding of low titres of antibodies to serotype C likely correlates with sporadic vaccination as the vaccines used by farmers are non-controlled, or could alternatively be due to cross-reactions caused by animals being infected and vaccinated multiple times taking into consideration that most animals slaughtered in LCC are at least 4 years of age and may have endured several lactation “cycles” in and out of the colony. Antibody titres against serotype Asia 1 and C were particularly variable and low in many animals sampled in November 2006 which may correlate to low circulation of FMDV in the preceding months and potentially supporting our hypothesis that at least part of the titres detected against Asia 1 and C is caused by cross-reacting antibodies generated by multiple infections and vaccinations together with recent infection with e.g. FMDV serotype O and A. However, at this point the exact explanation for this finding is currently unknown. Sequencing of RNA extracted from 58 mouth swabs and 17 epithelium samples indicated widespread circulation of the Pan Asia II lineage of serotype O FMDV and the A/IRN/05 lineage of serotype A. Surprisingly, the sequence derived from the locally produced monovalent type O vaccine from Lahore was very closely related to the type O sequences from the field. The Lahore vaccine is reportedly based on a 30-year old vaccine strain which is clearly contradicted by the sequencing results; however, whether the vaccine has been recently updated is unknown. The RNA extracted from the vaccine was very difficult to RT-PCR amplify and sequence and we could thus only sequence a very small part of the genome and could therefore not determine whether the detected sequence has the hallmarks of cell culture propagated FMDV or alternatively may be a contaminant and not at all representative of the Lahore type O vaccine strain. Nevertheless, we consider the sequences derived from the field samples to be representative of the circulating strains which is further supported by the fact that although the type O sequences were related to sequences from isolates from Bhutan, Nepal and Malaysia, they had distinct differences, e.g. in the VP1 coding sequence. Additional sequencing of the complete coding sequence of 3 of the A/IRN/05 isolates from Pakistan suggested a complex evolutionary path for this sublineage involving recombination events possibly involving an A22-like ancestor for the structural protein region of the genome and an Asia 1-like ancestor for the non-structural protein region (Figure 14). This recombination event may have

317

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 occurred in the buffaloes which, in contrast to the situation for the circulating serotype O virus that apparently causes clinical FMD in both buffalo and cattle, appear to mainly become subclinically affected with the circulating serotype A virus possibly making concurrent infection with another FMDV, e.g. of serotype Asia 1, more feasible. This is currently only a working hypothesis, but nevertheless, it suggests that detailed phylogenetic analysis of the individual coding regions of FMDV may provide significant additional information not obtained by the more routinely used analysis of the VP1 region only. Taking all the results into consideration, the vaccination practices in LCC should be significantly updated and enhanced in order to become effective. First of all, the vaccines used should be of high quality and contain the appropriate vaccine strains, currently including antigens covering the Pan Asia II lineage of serotype O, the A/IRN/05 lineage of serotype A and an appropriate antigen for serotype Asia 1. Moreover, the vaccinations will have to be applied in a more strategic way, if possible including vaccination of animals at origin once or better twice in the weeks before transportation into the colony occurs, combined with a twice annual mass vaccination scheme commencing in June and October-November before the rainy periods and thus providing herd immunity before the observed peaks in virus circulation. This advice may have to be modified to incorporate the fact that the timing of the Eid-ul-Azza festival, and the associated increased animal transportation across the country, moves forward a couple a weeks each year, and most likely one of the vaccination rounds should always occurs prior to commencement of these animal movements. 5. CONCLUSIONS

Sampling and testing of directly inactivated and stabilised mouth swabs, in combination with epithelial samples that may possibly also be inactivated and stabilised before shipment, have shown feasibility and strength in combined field and molecular epidemiological studies. The prevalence of circulating FMDV in LCC varied over the one-year study period and showed peaks associated with the rainy periods and the Eid-ul-Azza festival. Sequencing revealed circulation of the Pan Asia II lineage of serotype O and the so-called A Iran 2005 (A/IRN/05) lineage. The circulating serotype O FMDV apparently caused clinical FMD in both cattle and buffalo while the circulating serotype A FMDV mainly caused clinical FMD in cattle and subclinical infections in buffalo. Detailed analysis of the coding region of the A/IRN/05 FMDV indicated that this lineage may have evolved by recombination of a type A and a type Asia 1 FMDV, potentially by concurrent dual infection of buffaloes, possibly facilitated by the relative modest or absent clinical manifestations, but relatively high rate of subclinical infection, observed in buffalo.

6. RECOMMENDATIONS

Collaborative studies using inactivated and stabilised samples and involving the relevant authorities and laboratories of countries with endemic FMD, together with International Organisations such as FAO, the WRL for FMD and a National Reference Laboratory from a resource-rich country may provide an efficient avenue for strengthening FMD control programs. Further studies should be supported to further establish the suitability of the proposed methods, to provide more epidemiological data and knowledge of circulating strains of FMDV in various settings and to establish the potential for using such samples to re-generate infectious FMDV by optimising the methods used for lysis and stabilisation, RNA extraction and in particular the methods for maximising the efficiency of transfecting susceptible cells cultures. An improved vaccination strategy for the LCC is proposed and if provided with international support, such a strategy could be established as a starting point for improved control of FMD in this setting. This would in turn provide improved outcomes for the many farmers and workers in LCC and in addition potentially provide a success story on which to base improved FMD control in other regions of Pakistan as well as in other countries with endemic FMD.

7. ACKNOWLEDGMENTS We thank the several friendly and competent Veterinary Officers that have helped us in Pakistan. Drs. Giancarlo Ferrari and Keith Sumption from FAO are thanked for their continued interest in the project. Finally, we thank The National Veterinary Institute, Technical University of Denmark, the FAO Regional Project (GTFS/INT/907/ITA) and the EU Network of Excellence for Epizootic Disease Diagnosis and Control (EPIZONE, Call Identifier: FP6-2004-Food-3-A) for support.

318

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 8. REFERENCES [1] Alexandersen, S., M. Quan, C. Murphy, J. Knight, and Z. Zhang. 2003. Studies of quantitative parameters of virus excretion and transmission in pigs and cattle experimentally infected with foot-and-mouth disease virus. J. Comp Pathol. 129:268-282. [2] Alexandersen, S., Z. Zhang, A. I. Donaldson, and A. J. Garland. 2003. The Pathogenesis and Diagnosis of Foot-and-Mouth Disease. J. Comp Pathol. 129:1-36. [3] Klein, J., M. Hussain, M. Ahmad, M. Afzal, and S. Alexandersen. 2008. Epidemiology of foot-and-mouth disease in Landhi Dairy Colony, Pakistan, the world largest Buffalo colony. Virol. J 5:53-68. [4] Klein, J., M. Hussain, M. Ahmad, P. Normann, M. Afzal, and S. Alexandersen. 2007. Genetic characterisation of the recent foot-and-mouth disease virus subtype A/IRN/2005. Virol. J. 4:122-133.

319

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

320

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

321

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

322

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

323

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

324

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

325

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

326

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

327

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

328

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 58 TRACING 2007-2008 EMERGENCY EPISODES OF FOOT-AND-MOUTH DISEASE VIRUS IN SOUTH AMERICA: PHYLOGENETIC ANALYSIS V. Malirat1, I. E. Bergmann1*, R. de Mendonça Campos1, E. Neitzert1, M. Villamil2, J. L. Quiroga Civera3, F. Conde4 and G. Salgado Jijón5 1

Pan American Foot-and-Mouth Disease Center (PAHO/WHO), Rio de Janeiro, Brazil P.O. Box 589, 20.001, Rio de Janeiro, Brazil. 2 Instituto Colombiano Agropecuário –ICA, Bogotá, Colombia. 3 Laboratorio de Investigación y Diagnóstico Veterinario - LIDIVET, Santa Cruz, Bolívia. 4 Instituto Nacional de Investigaciones Agrícolas, Maracay, Venezuela. 5 Laboratorios Veterianarios, Instituto de Medicina e Higiene Tropical “Izquieta Perez”, Quito, Ecuador. * Correspondence to I.E. Bergmann: Pan American Foot-and-Mouth Disease Center PAHO/WHO. Av. Presidente Kennedy, 7778 – Duque de Caxias, Rio de Janeiro, Brazil. CEP 25040-004 Tel: (55) (21) 3661.9056 Fax: (55) (21) 3661.9001 E-mail: ingrid.bergmann@hotmail.com

ABSTRACT Since the implementation in 1988 of the foot-and-mouth disease eradication program in South America, a reduction of over 95% of clinical cases was attained, with endemic areas restricted to Ecuador and Venezuela. Thus, molecular characterization of strains circulating in endemic regions as well as those appearing during eventual emergencies is critical to determine genetic links. This work reports the molecular epidemiology studies performed to accompany the tracing of viruses recorded in 2007-2008 in controlled areas, closed to or within buffer zones, in Santa Cruz, Bolivia (type O) and in Norte de Santander, Colombia (types O and A). Complete VP1 sequencing was performed after RT-PCR in RNA extracted from epithelium or OP samples. Sequences were determined for the viruses causing the emergencies in Bolivia and Colombia and for viruses circulating in endemic areas of the continent in 2001-2008. Phylogenetic trees, generated by the neighbor-joining method, also included vaccine and epidemiologically relevant strains. The Bolivian type O episode showed the closest relationship (87%) with viruses responsible for outbreaks in 2000-2002 in Beni, Pando and La Paz (Bolivia). Colombian isolates of both types O and A, were closely related to one of the various lineages within each type, circulating lately in Venezuela. Phylogenetic analysis provided evidence to support the ecosystem dynamics in the region. Emergencies were linked to risk areas reinforcing the need to strengthen surveillance. All isolates belonged to endogenous topotypes and were not derived from vaccine strains.Independent lineages were identified in Venezuela, Ecuador and Bolivia, all of them different from the one causing emergencies in the Southern Cone (2000-2006). Within the Andean region, different lineages co-circulate for each type, illustrating the potential for FMD diversification under limited selective pressure (poor vaccine coverage). Although up to 19% sequence divergence was established with the vaccine strains, vaccine matching studies suggest that currently in use vaccines have wide coverage. 1. INTRODUCTION Since the implementation in 1988 of the foot-and-mouth disease (FMD) eradication program in South America (PHEFA), significant advances were made in the control of the disease. Great progress has been attained in the Southern Cone as well as in some of the Andean countries. In overall outbreaks are mainly concentrated in Ecuador and Venezuela where 46 episodes took place in the year 2007 and 29 in 2008 (until September). This represents an overall reduction of over 95% in terms of clinical cases of the disease in the region, since the plan was implemented. In this

329

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 context and taking into account the progress of the PHEFA, nucleotide sequencing and molecular epidemiological studies become essential tools for identifying genetic links among circulating strains (PANAFTOSA, 2008). These kind of studies are particularly important to trace the origin of outbreaks in already free regions, with or without vaccination, or in regions with advanced eradication plans, where the knowledge of the putative origin of viruses emerging or re-emerging is necessary not only for control measures but also to give confidence to the sanitary plans (Bergmann et al., 2005). In 2007-2008, outbreaks of foot-and-mouth disease were registered in Bolivia (type O) and in Colombia (types O and A) in controlled areas, close to or within buffer zones (OIE, 2007; OIE 2008). This paper reports the application of complete VP1 nucleotide sequencing to perform a phylogenetic study of FMDV samples recovered from these episodes. Also determined and included in the analysis were the sequences of viruses responsible for previous outbreaks in endemic regions in the continent during the past 6 years. The sequences were also compared with selected South American field and vaccine strains, as well as with exogenous viruses. 2. MATERIALS AND METHODS 2.1 Viruses and Cells The designation and origin of viruses examined in this study, as well as the geographic locations of FMD Types O and A outbreaks are listed in Table 1. All viral RNAs were directly extracted from epithelium samples or from oesophageal-pharyngeal (OP) fluid. 2.2 Oligonucleotide Primers Three oligonucleotide primers were used in this study to amplify and sequence the complete VP1coding region, rendering amplification fragments of 790 bp and 795 bp for types O and A viruses, respectively. Oligonucleotide sequences are: for the Type O forward primer: 5' AATTACACATGGCAAGGCCGACGG 3'; for the Type A forward primer: 5' TACCAAATTACACACGGGAA 3'; and for the reverse one (used for both types): 5’-GAAGGGCCCAGGGTTGGACTC-3’. 2.3 Viral RNA extraction and Reverse Transcription-Polymerase Chain Reaction (RT-PCR) The general protocol was as previously described (Malirat and Bergmann, 2003). Total RNA was extracted from epithelium samples or from OP fluids using Trizol reagent (Invitrogen) according to the manufacturer’s protocol. Reverse transcription of the viral RNAs (5 µl, 3-5 µg RNA) was carried out using 50 ng of random primers and 50 units of Superscript II reverse transcriptase (Invitrogen) and incubating at 42 oC for 60 min, followed by extension at 70°C, 15 min., in a 25 µl reaction mix containing 20 mM TrisHCl (pH 8.4), 50 mM KCl, 2.5 mM MgCl2, 10 mM dithyothreitol and 0.6 mM of each dNTPs. In vitro amplification was carried out with a programmable thermocycler GeneAmp PCR system 9700 (Applied Biosystems). Each reaction was performed in a final volume of 50 µl of a reaction mixture containing: 5 µl cDNA, 0.5 µM of each primer, 2.5 units of Taq Polymerase (Invitrogen), 0.2 mM each dNTP (dATP, dCTP, dGTP, dTTP) and 1.5 mM MgCl2 in 20 mM Tris-HCl, (pH 8.4), 50 mM KCl, 0.1% Triton X-100. After denaturing 5 min. at 94 °C, for type O viruses, 30 cycles were performed in the reaction mixtures consisting of incubation at 94 °C for 1 min., followed by cooling to, and then holding at 60°C for 45 sec., to allow primer annealing. Each cycle ended with a chainelongation step at 72 °C for 2 min. After the last cycle, polymerization was continued at 72 °C for 5 min. Thermocycling conditions for type A viruses were: 5 min. at 94 °C to denature DNA, followed by 30 cycles (94 °C for 1 min., 55°C for 1 min. and 72 °C for 1.5 min), and a last hold at 72 °C for 5 min. The amplified products were purified by band excision from 1% agarose gel electrophoresis, followed by chromatography in affinity columns (Promega) and the recovered material was quantified by band intensity comparison with DNA mass and molecular weight marker (Invitrogen) in 1% agarose gel electrophoresis.

330

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 2.4 Nucleotide sequence determination and analysis The nucleotide sequences were determined from 20-60 ng of the purified amplicon, using the Big Dye Terminator kit 3.1 (Applied Biosystems), according to the manufacturer’s procedure, in a thermocycler, performing 40 cycles of 94°C, 45 sec.; 50°C, 30 sec.; 60°C, 4 min. After cycles, the reaction product was purified by exclusion chromatography (CentriSep columns, Princeton Separations); the recovered material was dyed. For reading, the dyed samples were ressuspended in formamide 10%, as recommended for use in an ABI Prism 3100 Avant Genetic Analyzer sequencing machine. Chromatograms obtained for each individual reaction were edited manually to avoid misreading of peak dyes. Nucleotide sequences were analyzed on an IBM compatible personal computer using for editing and alignment the program BioEdit, version 5.0.2.1. All pairwise comparisons were performed by giving each base substitution equal statistical weight. An unrooted tree was constructed according to sequence relatedness across the interval of nucleotides coding for the complete VP1 gene, using the neighbor–joining method as implemented in the computer program MEGA, version 4 (Tamura et al., 2007). Bootstrap resampling analysis was performed with 1,000 replicates, as implemented in the program. 3. RESULTS 3.1 Type O isolates To address the genetic relatedness between the isolates responsible for the episodes in Santa Cruz, Bolivia and in Norte de Santander, Colombia with other type O FMDVs, a phylogenetic tree was constructed. Comparisons included sequences from epidemiologically relevant South American strains, type O viruses circulating lately in endemic areas, viruses responsible for the emergencies recorded sporadically between the years 2000-2006 in the Southern Cone, strains used for vaccine production, as well as those from type O worldwide representative viruses. All viruses isolated in South America are endogenous and placed within the Euro-SA topotype, including the Bolivian and Colombian emergencies. Virus O/Santa Cruz/Bol/07 clusters with viruses isolated from previous episodes in other Bolivian departments (Beni, Pando and La Paz), occurring between the years 2000-2002, although showing divergence values of up to 13% (figure 1). This cluster can be distinguished from the lineage in which all 2000-2006 Southern Cone type O emergencies are placed (Malirat et al 2007). Between these groups differences of about 18% are registered. Moreover the Bolivian lineage is independent from any of the various lineages circulating in the Andean endemic countries (divergence values of about 18-20%). When compared with the O1 Campos vaccine strain, a genetic diversity of almost 19% was recorded. The genetic links established between the isolates responsible for the episode in Cúcuta, Norte de Santander, Colombia, and other type O FMDVs is also depicted in Figure 1. Isolates recovered from OP fluids in two affected premises shared 99% homology in the VP1 sequence. The VP1 sequence of the isolates recorded in Cúcuta, showed close relationship with that of one of the genetic lineages of circulating viruses recorded in Venezuela during 2005-2007, with homology values of about 93-95%. When compared with the O1 Campos vaccine strain, the virus responsible for the Colombian episode in 2008 shows a genetic diversity of 9%. 3.2 Type A isolates Phylogenetic analysis of type A viruses recorded in Sardinata, Norte de Santander, is shown in Figure 2. The six isolates recovered from OP fluids show that they are endogenous from the continent and they present homology of over 99.5% among them. Likewise, and as reported for virus type O in Cúcuta, Norte de Santander, Colombia, 2008, they show a close relationship to one of the various lineages viral genetic lineages circulating in Venezuela in the years 2004-2006 (about 95% homology) and differed significantly from the one responsible for the emergency that occurred in Tibú, Colombia in the year 2004, with which they show a divergence of about 20%. When compared with the A24 vaccine strain a divergence of 15% was recorded. 4. DISCUSSION The molecular characterization of FMDV strains is critical to determine genetic links, particularly during emergencies, baring in mind the epidemiological situation of South America. The phylogenetic analysis performed in this work to trace the emergencies recorded in Bolivia and

331

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Colombia in the years 2007-2008 provided evidence to support the ecosystem dynamics for FMD in South America. All South American isolates belonged to endogenous topotypes and the viruses responsible for the referred emergencies were not derived from vaccine strains. Independent lineages were identified in Venezuela, Ecuador and Bolivia which are also different from the one causing emergencies in the Southern Cone (2000-2006). Molecular tracing of the episodes indicate a link to high risk areas, supporting the need for strengthening control measures, not only within the country, but also in the hot spots of border areas. Within the Andean region, different lineages could be defined for each type, even within the same time bracket whereas emergencies in the Southern Cone were placed in a unique lineage within each type. This observation illustrates the potential for FMD diversification under limited selective pressure (poor vaccine coverage). Although up to 19% sequence divergence was established with the vaccine strains, vaccine matching studies suggest that currently in use vaccines under immunization programs in South America would protect. (PANAFTOSA, 2008) 5. CONCLUSIONS Phylogenetic analysis provided evidence to support the ecosystem dynamics for FMD in South America. Genetic divergence with vaccine strains is not necessarily associated with lack of protection Further discussion on topotypes is needed 6. RECOMMENDATIONS Harmonize reference strains representative of each topotype Promote studies to correlate genetic and immunogenic changes 7. REFERENCES [1] Bergmann, I.E.; Malirat, V.; Neitzert, E. & Correa Melo E. 2005. Evaluation of diagnostic tools for epidemiological purposes - application to FMD In: Makkar HPS, Vilijoen GJ, eds. Applications of gene-based technologies for improving animal production and health in developing countries, pp. 335-340. Dordrecht, The Netherlands, Springer. 794 pp. [2] Tamura, K.; Dudley, J.; Nei, M.; Kumar, S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0. Mol. Biol. and Evol, 24: 1596-1599. [3] Malirat, V. & Bergmann, I.E. 2003. Fiebre aftosa: instrumentos moleculares para la caracterización viral. RT/PCR y secuenciamiento para estudios de epidemiología molecular del virus de la fiebre aftosa. Centro Panamericano de Fiebre Aftosa: Serie de Manuales Didácticos: 17. [4] Malirat V, Barros JJF, Bergmann IE, Campos RM, Neitzert E, Costa EV, Silva EE, Falczuk AJ, Pinheiro DSB, Vergara N, Quiroga JL, Maradei E, Di Landro R. 2007. Phylogenetic analysis of Foot-and-Mouth Disease Virus type O re-emerging in free areas of South America. Vir. Res., 124(1-2): 22-28. [5] OIE. World Organization for Animal Health-OIE. 2007. Summary of Immediate notifications and Follow-ups – 2007. Foot-and-mouth Disease. http://www.oie.int/wahis/public.php?page=disease_immediate_summary&disease_type=Terrestrial&d isease_id=1&empty=999999) [6] OIE. World Organization for Animal Health-OIE. 2008. Summary of Immediate notifications and Follow-ups – 2007. Foot-and Mouth Disease. In Wahid Interface. (Available at: http://www.oie.int/wahis/public.php?page=disease_immediate_summary&selected_year=2008 [7] PANAFTOSA OPS/OMS. 2008. Laboratorio: informe anual, 2007. Centro Panamericano de Fiebre Aftosa, PANAFTOSA, OPS/OMS. Rio de Janeiro, Brasil. PANAFTOSA

332

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Table 1: Designation and origin of Foot-and-mouth Disease types O and A viruses from the emergencies in Bolivia and Colombia Virus designation

Type County

Geographical Location State

Date collected day/month/year

Material

Country

O/Santa_Cruz/Bol/07 O/Santa_Cruz/Bol/07(a)

Ñuflo de Chávez Ñuflo de Chávez

Santa Cruz Santa Cruz

Bolivia Bolivia

23/01/2007 23/01/2008

Mouth epithelium (bovine)

O/Santa_Cruz/Bol/07(6133)

Ñuflo de Chávez

Santa Cruz

Bolivia

23/01/2008

O/Cúcuta/Norte de Santander/Col/08(1)

Cúcuta

Norte de Santander

Colombia

05/06/2008

Epithelium (swine) OP a

O/Cúcuta/Norte de Santander/Col/08(2)

Cúcuta

Norte de Santander

Colombia

05/06/2008

OP a

O/Cúcuta/Norte de Santander/Col/08(3)

Cúcuta

Norte de Santander

Colombia

05/06/2008

OP a

A/Sardinata/Norte de Santander/Col/08(542106)

Sardinata

Norte de Santander

Colombia

12/07/2008

OP a

A/Sardinata/Norte de Santander/Col/08(541171)

Sardinata

Norte de Santander

Colombia

12/07/2008

OP a

A/Sardinata/Norte de Santander/Col/08(542114)

Sardinata

Norte de Santander

Colombia

12/07/2008

OP a

A/Sardinata/Norte de Santander/Col/08(541101)

Sardinata

Norte de Santander

Colombia

12/07/2008

OP a

A/Sardinata/Norte de Santander/Col/08(541172)

Sardinata

Norte de Santander

Colombia

12/07/2008

A/Sardinata/Norte de Santander/Col/08(541180)

Sardinata

Norte de Santander

Colombia

12/07/2008

OPa

Epithelium (bovine)

(a): bovine oesophageal faringeal fluid

Figure 1: Phylogenetic tree showing the genetic relationships of FMDV type O isolates in South America. Also included are representative strains of the other O topotypes. P-distances were calculates based on the comparison of the 633 nucleotides of the VP1 gene. The tree was constructed using the MEGA software, version 4. Viruses of the Bolivian and Colombian emergencies are highlighted inside squares. Black triangles represent different groups of viruses from the locations and isolation years indicated.

0.03 0.02

Ecuador2004-2005 0.04

O/Pichincha/Ecu/06(17-06) Ecuador2002-2004 and Peru 2004 Colombia2000 0.01 Colombia1995-1999 Ecuador2002-2007 andColombia2002 0.02 0.02 Ecuador2005-2006 0.05 Ecuador2005 O/Cauca/Col/94(a) 0.03 O/Genova/Narino/Col/94 0.03 O/Cauca/Col/94(b) 0.05 O/Ecuador/02-GI O/Trujillo/Ven/05 O/Zulia/Ven/05 0.02 O/Trujillo/Ven/0a 0.01 0.01 O/Trujillo/Ven/0b O/Cucuta/Col/08 (Carmen de Santa Cruz) 0.03 0.02 O/Cucuta/Col/08 (El Buchugo) 0.04 O/Merida/Ven/04 0.04 O/Casanare/Col/94 0.07 O/Yaracuy/Ven/03 O1 Campos Vaccine S’train 0.07 O/Guayas/Ecu/05(013-05) 0.03 O/Cundinamarca/Col/00 0.05 0.01 Colombia1994-1995 0.03

0.01

0.02

0.07 0.08 0.01 0.04 0.05 0.07

0.02 0.01

0.06 0.01

O/Caseros/Arg/67 (Tamiet al.1997) 675 b 0.02 O/Irigoyen/Arg/82 (Saiz et al. 1993) 675 O1/Lombardy/Italy/46 (Krebset al. 1991) O/La Paz/Bol/02(b) O/xx/Los Andes/La Paz 0.01 O/La Paz/Bol/02(c) O/Chonchocoro/Ingavi/ O/La Paz/Bol/02(a) O/Kallutaca/Los Andes 0.01 O/Beni/Bol/01 O/Magallanes/Yacuma/Beni/0 O/Pando/Bol/00 O/Santa Cruz/ Bol /07(a) O/Santa Cruz/ Bol /07(b) O/Santa Cruz/ Bol /07(c) O/Irigoyen/Arg/83 SouthernCone Emergencies

0.09 0.09 0.12 0.08 0.06 0.09 0.08 0.07 0.06

O/JAV/5/72 (Java) ISA -2 AJ303509 O11/ISA/1/62 Indonesia ( ) ISA-1 AY593813O/Yunlin/Taiwan/97 CATHAY AF308157 -CATHA O/MAL/1/98 Malawi ( ) EA-2 DQ165074 O1/Manisa/Tur/69 (Turkey) ME SA AY593823 O/CAM/3/98 Cambodia ( ) SEA AJ303488 O/GHA/5/93 Ghana ( ) WA AJ303488 O/K83/79 Kenya ( ) EA-1 AJ303501 O/SUD/2/86 Sudan ( ) EA-3 DQ165075

0.02

333

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Figure 2: Phylogenetic tree showing the genetic relationships of FMDV type A isolates in South America. Also included are representative strains of the other A topotypes. P-distances were calculates based on the comparison of the 633 nucleotides of the VP1 gene. The tree was constructed using the MEGA software, version 4. Viruses of the Colombian emergency are highlighted inside the square. The black triangle represent a group of viruses from the locations and isolation years indicated.

A/Tachira /Ven/04(a) A/G de Hevia /Tachira A/Tachira Hevia /Ven/04(d) A/G de /Tachira A/Tachira Hevia /Ven/04(b) A/G de /Tachira A/Tachira Hevia /Ven/04(c) A/G de /Tachira A/Merida /Ven/04 A/Alberto Adriani /Merida 0.01 A/Apure/ Ven/04 A/Paez/Apure/ Ven/04 (2130 A/Barinas /Ven/04(a) A/E Zamora /Barinas /V 0.01 A/Barinas /Ven/04(b) A/ Barinas /Barinas /Ve /Ven/05 A/S.de Maldo /Tachira /Ve 0.01 A/Tachira A/Merida Ven/05(a) A/ Jaji/Merida Ven/05 / / 0.02 A/Apure/ Ven/06 (21571 original 21461) )2 0.02 A/Yaracuy /Ven/04 A/ Veroes /Yaracuy /Ven/04 A/Sardinata /N.Santander//Col/08(541180) 0.04 A/Sardinata /N.Santander//Col/08(541171) A/Sardinata /N.Santander//Col/08(542114) A/Sardinata /N.Santander//Col/08(541172) A/Sardinata 08(542106) 1 /N.Santander/Col/ A/Sardinata /N.Santander//Col/08(541101) 0.02 A/Barinas /Ven/03 A/ Torunos /Barinas /Ven/0 0.01 0.01 A/Falcon 0.02 /Ven/04 A/M. Iturriza /Falcon /Ven/ 0.03 A/Apure/ Ven/05 A/ Mantecal /Apure/ Ven/05 ( 0.04 A/Venezuela/89 A/Bolivar Chien/Bolivar /V /Ven/01 A/Padre 0.04 0.02 A/Bolivar Chien/Bolivar /V /Ven/03 A/Padre 0.07 A/Portuguesa/ Ven/07 (21540) 21540 OR 215 0.06 A/Sabana /Col/85 747 -Ediatada -VECOL -127R5 A/ Merida Ven /05(b) A/ Obispo Ramos de L./ / 0.04 0.03 A/Merida ObispoRamos de L./ /Ven/05(c) A/ 0.03 A/Merida Ven/03(b) A/Alberto Adriani Mer / / 0.03 A32/Venezuela/70 0.04 A/Cordoba /Col/97-Colombia -Cordoba15 -9-9 0.02 0.01 A/Antioquia/ Col/97 15 -9-97 0.02 0.01 A/Cundinamarca Col/97 15 97 / 9 0.04 A16/Belem /Bra/59 0.02 0.02 A24/Cruzeiro/ Bra/55(PANAFTOSA) 0.04 A/Ecuador /79 0.02 0.04 A/Ecuador /02 128.798 0.03 0.02 A/Peru/99 0.02 0.02 A/Peru/00 0.09 A/SaoCarlos /Bra/84 0.01 0.07 A26/Argentina/66 0.06 A/BahiaBlanca /Arg/71Konig-2001 AJ30869 0.01 A 79 0.02 0.02 A 76 0.05 A/RS/ Bra/81 0.03 A24/Bra/70 (PANAFTOSA) 0.04 0.04 A76/Argentina/76 0.05 A29/Peru/69 0.01 A81 (include Southern Coneemergencies 2000-2001 0.07 A25/Argentina/59 0.06 A10/Argentina/61 - Boothroyd -1982 675 ba 0.03 A12/119/Kent/UK/32 M10975 0.06 A19/Suipacha /Arg/62 0.02 A13/Brasil/58 0.04 0.02 A17/Guarulhos/ Bra/59 0.03 A/SaoPaulo /Bra/94(A3) 0.03 0.03 A/SaoPaulo /Bra/94(A5) 0.05 A/SanMartin/Per/75 EDITADA 0.04 A/Beni/ Bol/02 A/ Yucumo /Beni/02 /Ballivian 0.02 0.03 A/SaoPaulo /Bra/95(A1) 0.03 A/StaCruz/Bol/00 A/ Pailon /Chiquitos /Sta 0.03 A/RS/ Bra/97 Br(6562)5 -8-97 0.02 Bolivia2000-20010.02 A/Merida Merida /Ven/02 A/Alberto Adrian/ / 0.01 A/Tachira /Ven/01 A/ Panamericano /Tachira / 0.02 0.05 A/Tibu/Col/04 369b( A-Tibu) 0.02 0.01 A/Merida Cordero /Ven/03(a) A/Febres /Meri 0.05 A/Colombia /84 0.05 A/Ecuador /75 0.07 A18/Zulia/Ven/62 0.02 A27/Cundinamarca /Col/76 0.03 0.02 A27/Colombia /67 0.09 A22/IRQ/24/64 ASIA 0.10 A23/Kitale/Ken/64 AFRICA AY593766 - a23ke 0.08 A21/Lumbwa /Ken/3/64 AFRICA AY593761 -a21k

EURO SA

0.01

0.02

0.01

0.02

334

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Appendix 59

SERO-PREVALENCE OF FOOT-AND-MOUTH DISEASE IN SMALL RUMINANTS UNDER CONTRASTING HUSBANDRY PRACTICES IN UGANDA S. N. Balinda1*, C. Masembe1, K. Tjørnehøj3, A. Sangula1, F. Mwiine2, C. Ayebazibwe2, R. Ademun2, S. Alexandersen3, H. Siegismund4, V. Muwanika1 1

Makerere University, Institute of Environment and Natural Resources P.O.Box 7298, Kampala, Uganda 2 Ministry of Agriculture Animal Industry and Fisheries, P.O. Box 102, Entebbe, Uganda 3 National Veterinary Institute, Technical University of Denmark, Lindholm, DK-4771 Kalvehave, Denmark 4 Department of Biology, Universitetsparken 15, DK-2100, Copenhagen, Denmark

ABSTRACT Foot-and-mouth disease is endemic in Uganda with control strategies focusing on the disease in cattle. The role of small ruminants in its epidemiology is largely ignored. In order for Uganda to achieve the desired situation of establishing FMD free zones to allow for exportation of animal products, it is crucial that the epidemiology of the disease is fully understood. In this study, we determined the prevalence of FMD in small ruminants under two different husbandry practices. Sero-prevalence of foot-and-mouth disease virus (FMDV) in 346 asymptomatic animals (goats = 272; sheep = 74) randomly selected from two areas with contrasting husbandry practices was estimated using both non structural protein (NSP) and structural protein (SP) antibody Cedi Elisa method. Samples positive on either NSP or SP Elisa were subsequently screened and titrated using a serotype specific Elisa. Serotypes identified in this animal category included type O and the three Southern African territories serotypes (SAT 1, 2, and 3). A sero-prevalence of 13% and 21% in goats and sheep, respectively, was observed in the area that practices communal grazing, while 0% prevalence was recorded for paddock system. Our results suggest that these species could be playing an important role in the epidemiology of the disease and husbandry practices may influence disease distribution. 1. INTRODUCTION Foot-and-mouth disease (FMD) is endemic in Uganda with six of the seven known FMD virus serotypes having been recorded viz; O, A, C, South African Territories serotypes (SAT 1, 2 and 3).Type O and SAT 2 are the most prevalent serotypes, while C and SAT 3 were last recorded in the 1970s with SAT 3 isolated only from the African buffalo (Vosloo et al., 2002). Most of the reported outbreaks involve cattle and like elsewhere in the sub-Saharan region, in Uganda, despite the mixed animal husbandry practise, sheep and goats are still largely ignored even in epidemic situations. Consequently vaccinations in the country are performed mainly on cattle exposing the control strategy to complications from infected small ruminants (Cox et al., 1999). Sheep have been shown to play a significant role in the spread of FMD during outbreaks because of the subclinical nature of the disease and possibility of a carrier status, with sheep and goats estimated to maintain the live virus for 9 and 4 months respectively (Leforban, 1999; Zhang and Kitching, 2001; Blanco et al., 2002; Kitching, 2002). Uganda’s desire to promote livestock production earnings from exports hinge on among other factors the control of animal diseases including FMD (http//www.agriculture.go.ug/policies.htm). In the case of FMD, scanty information is available for a complete understanding of the epidemiological dynamics of the disease and yet it is necessary for a successful control strategy. For instance most outbreaks are concentrated in the western region which has high livestock populations and also large wildlife reservoirs. At the same time, husbandry practises are varied with some areas practising agro-pastoralism tending towards fencing while others practise pastoralism characterised by communal grazing (Rweyemamu et al., 2008).

335

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 In this study, we determine the prevalence of FMD in small ruminants under two different husbandry practices in two districts of Western Uganda namely, Bushenyi (mainly agro-pastrol) and Kasese (mainly pastrol).

2. MATERIALS AND METHODS 2.1. Sampling A two stage random sampling strategy was adopted assuming a 10% herd level FMD seroprevalence for both goats and sheep. Using the Survey Toolbox program (http://www.ausvet.com.au/surveillance/toolbox.htm) and taking the number of herds per district as; 12,000 goats and 3,750 sheep (Bushenyi); 3,545 goats and 349 sheep (Kasese); cost ratio of village to farm of 5:1; a 90% confidence with an accuracy of +/-10%; a within farm/herd variance and between farm/herd variance of 0.2 and 0.05 respectively; 17 villages/6 farms (goats) and 15 villages/6 farms (sheep) were selected for Bushenyi and Kasese respectively. List of farmers per village were drawn with the help of the local veterinary officer and 5 farms randomly selected without replacement. In total 346 asymptomatic animals (goats = 272; sheep = 74) were sampled as summarized in the table 1. The study area is as identified on the map of Uganda (Figure1). 2.2. Laboratory Methods Three different serological Enzyme linked immunosorbent assay (ELISA) tests were used to detect antibodies to FMDV in the sheep and goat sera. One commercially available non-structural protein ELISA (NSPE) test-kit was used to assay antibodies to non-structural proteins (NSP) of FMD virus; the Ceditest® FMDV-NS, (Cedi Diagnostics B.V.; Sorensen et al., 1998; Chung et al., 2002). Antibodies to structural proteins were tested using a commercial test-kit; Ceditest® FMDV type O, (Cedi Diagnostics B.V., Lelystad, the Netherlands) and an in-house serotype specific SPCE test-kit from the Danish Technical University, National Veterinary Institute, Lindholm, Denmark. Each ELISA was conducted in accordance with the manufacturer’s instructions. Only sera considered as positive by the Cedi tests were further screened for serotype identification in a two stage process of screening and titration using the serotype specific test-kit of Lindholm. 3. RESULTS AND DISCUSSION No animal was reactive for antibodies to non-structural proteins in the 13 villages of Bushenyi district as shown in Figure 2 below. One goat from Katunda village was reactive for antibodies to structural proteins specific to SAT1 with a titer of 160. This animal was suspected to have been introduced through purchase by the farmer from an unknown source. The prevalence estimate based on the NSP antibodies was therefore assumed to be approximately zero for this district.

Figure 2: NSP and SP ELISA results for small ruminants in Bushenyi district In Kasese district, most animals sampled from the villages practicing communal grazing were reactive for antibodies to non-structural proteins as shown in Figure 3.

336

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 3: NSP and SP ELISA results for small ruminants in Kasese district Prevalence estimates for NSP antibodies among the goats in the villages of the District were; Busunga (60%), Kabaka (15.4 -27%), Kayanja (8.3-9%) and Kisasa (37.5-66.7%) with an overall district seroprevalence estimate of 13%. Sero-prevalence among the sheep in the district was estimated to be; Kahendero (20%), Busunga (100%), Kabaka (23%) and Kisasa (60%) with the overall estimated being 21%. Although communal grazing is the most common practice, some fenced farms do exist. Goats and sheep sampled in these farms were sero-negative (data not shown). Inclusion of these animals in estimating the overall prevalence in the district resulted in the observed low value. The FMD prevalence among the cattle population are similar to those observed here (data not shown). The serotypes identified among the sheep and goats are shown in table 2 below. Although most of these serotypes have been recorded in cattle, SAT 3 was previously only found in the African buffalo (Bronsvoort et al., 2008). This serotype could be circulating among the livestock as shown by our results. As an area with extensive communal grazing, it is easy for FMD virus to easily spread among the herds. Our results suggest that these species could be playing an important role in the epidemiology of the disease and farming practices influence disease distribution. Table 2: FMD Serotypes in animals with high level of FMDV antibodies Serotype O SAT 1 SAT 2 SAT 3

No. of Animals Goats Sheep 5 7 4 2 9 2 4 5

4. CONCLUSIONS Although ignored in FMD control strategies, goats and sheep are sero-positive for FMD in Uganda Serotypes identified in this animal category are O, SAT 1, 2 and 3 Husbandry practices may have an effect on the sero-prevalence of FMD in endemic situations. 5. ACKNOWLEDGEMENTS The Researchers would like to thank DANIDA-ENRECA for funding this project. 6. REFERENCES

337

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [1] Blanco E., R.L.J., Harrachb M.E., Sanchez-Vizcaino J. M. 2002. Serological evidence of FMD subclinical infection in sheep population during the 1999 epidemic in Morocco. Veterinary Microbiology 85, 13-21. [2] Bronsvoort, B. M. D. C., Parida, S., Handel, I., McFarland, S., Fleming, L., Hamblin, P., Kock, R. 2008. Serological Survey for Foot-and-Mouth Disease Virus in Wildlife in Eastern Africa and Estimation of Test Parameters of a Nonstructural Protein Enzyme-Linked Immunosorbent Assay for Buffalo. Clinical and Vaccine Immunology, 15 (6), 1003–1011. [3] Chung WB, Sorensen KJ, Liao PC, Yang PC., Jong MH.2002. Differentiation of foot-andmouth disease virus-infected from vaccinated pigs by enzyme-linked immunosorbent assay using nonstructural protein 3AB as the antigen and application to an eradication program. Journal of clinical Microbiology 40:2843-8. [4] Cox, S. J., Barnett, P. V., Dani, P., Salt, J. S. 1999. Emergency vaccination of sheep against Foot-and- mouth disease: protection against disease and reduction in contact transmission. Vaccine 17: 1858 -1868. [5] Kitching, R.P. 2002. Identification of foot-and-mouth disease virus carrier and subclinically infected animals and differentiation from vaccinated animals. Revue Scientifique et Technique de L’Office International des Epizooties 21, 531- 538. [6] Leforban, Y. 1999. Prevention measures against foot-and-mouth disease in Europe. Vaccine 17: 1755-1759. [7] Ryeyemamu, M., Roeder, P, Mackay,D., Sumption, K., Brownlie, J., Leforban,Y., Valarcher,J.F., Knowles, N.J., Saraiva V. 2008. Epidemiological patterns of foot-and-mouth disease worldwide. Transboundary and Emerging Diseases 55: 57-72. [8] Sorensen, K.J., Madsen, K.G., Madsen, E.S., Salt, J.S., Nqindi,J., Mackay, D.K.J. 1998. Differentiation of infection from vaccination in foot-and-mouth disease by the detection of antibodies to the non-structural proteins 3D, 3AB and 3ABC in ELISA using antigens expressed in baculovirus- Archives of Virology 143, 1461-1476. [9] Vosloo, W., Bastos A. D. S., Sangare, O., Hargreaves, S. K., Thomson G. R. 2002. Review of the status and control of foot-and-mouth disease in sub-Saharan Africa. Revue Scientifique et Technique de L’Office International des Epizooties 21 (3): 437-449. [10] Zhang, Z.D. & Kitching, R.P. 2001. The localisation of persistent foot-and-mouth disease virus in the epithelial cells of the soft palate and pharynx. Journal of Comparative pathology 124, 89-94

338

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 31°

33°

4°

35° 4°

SU DAN

Distric t boundary International boundary Open water

2°

Kasese

Lake Ky og a

DEM O R EP CRA T IC UB LIC O FC

ON G O

2°

KEN YA

KAS ES E

0°

KIR UHU RA BUS HENYI MBA RARA

Lak e

V ict oria

KAB INGO

Loc ation of U ga nda in Africa

TANZA NIA

Bushenyi 0 50 Kilometers 31°

33°

35°

Ug an da

Figure 1: Map of Uganda showing the location of the districts in the study in the western region Table 1: Samples collected from the respective villages District/Village Bushenyi Katara Kyamakoba BuzengaII Kajwiga Kyobukyeera Komondo II Mutojo Ngoma Kihunda Nyakahita Katunda Katooma MugogoIII Kabushaho Kasese Kahendero Busunga Rwentutu Kabaka Kayanja Kisasa Rwembyo Total

Samples collected Goats Sheep 6 6 11 10 7 18 6 11 10 4 21 8 1 3 3 21 19 24 43 17 16 272

2 3 5 9 13 14 15 74

339

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

340

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 60

THE STATUS OF FOOT-AND-MOUTH DISEASE (FMD) IN ETHIOPIA

G. Ayelet*, E. Gelaye, J. Guitian, M. Sahle, N. J. Knowles and M. Mahapatra National Veterinary Institute, Debre Zeit, Ethiopia

The study was designed to determine the serotypes and subtypes of FMD virus circulating in the country, and to generate epidemiological information. Out of 241 submitted samples FMD virus was isolated from 89.2% (n= 215) of samples in cell culture and identification of the serotype was made using Complement Fixation Test (CFT) revealed presence of five different serotypes of the virus; O (72.9%), A (19.7%), C (1.4%), South African Territories (SAT) 1 (1.8%) and SAT 2 (4.1 %). SAT 2 was identified for the first time in 1989 from a bovine sample collected from Leben Ranch, Borena area, southern Ethiopia and SAT 1 in 2007 from Mezan Teferi area, however SAT 3 has never been reported in Ethiopia. The records of ministry of agriculture and rural development (MOARD) from 1997 to 2006 also indicated that FMD outbreaks occurred every year with the highest in 1999. Also serological study was conducted using serum samples collected from cattle from the whole part of the country. From the total 4465 sera tested 10.5% (n=467) were positive for FMD. From 1876 samples collected from pastoral areas of the country 14.8% (n=325) prevalence was obtained. The epidemiology of FMD in sub-Saharan Africa is probably more complicated than in any other region of the world. Not only have six of the seven serotypes occurred in Africa (only Asia 1 has never been recorded), but also marked regional differences in the distribution and prevalence of serotypes and intratypic variants occur (Vosloo et al., 2002; Sahle et al., 2004). Foot-and-mouth disease was first recorded in Ethiopia in 1957 when serotypes O and C were found (Martel, 1974; Martel and Gallon, 1975). Type A and SAT 2 were not identified until 1969 and 1989, respectively (Martel and Gallon, 1975). During the period 1988 to 1991 samples from 16 foot-and-mouth disease outbreaks in Ethiopia were examined at the National Veterinary Institute, Ethiopia, and at the FAO World Reference Laboratory for Foot-and-Mouth Disease, UK. Typing of the virus responsible was possible in 13 of these outbreaks representing 10 separate disease events; 8 of these were caused by serotype O and 2 by serotype SAT2. This is the first record of the presence of serotype SAT 2 foot-and-mouth disease virus in Ethiopia. In contrast to earlier studies serotypes A and C were not detected (Roeder et al., 1994). The molecular epidemiology of FMDV has been studied in some detail for world using nucleotide sequencing of the main antigenic determinant of the virus and phylogenetic analysis. However, sufficient genetic information of viruses from Ethiopia has not been available to determine the number of viral lineages and genotypes and to investigate whether certain patterns of spread in the country have occurred in the past. This study also describes an initial attempt to describe the status of FMD using sero-surveillance through the detection of antibodies in serum samples collected from cattle and using FMD outbreak reports, from the whole parts of the country to address the need and to indicate the possible areas for disease free zone establishment Therefore the objectives of this study are:

To isolate and identify the circulating serotypes and subtypes of FMD Virus To determine the strains appropriate for vaccine production To generate epidemological information of FMD that helps to design control measures.

2. MATERIALS AND METHODS 2.1 Samples collection and processing Serum samples collected over from 2003 to 2006 period for sero-surveillance of rinderpest from cattle between 1 and 3 years of age were supplied by Sebeta laboratory and used for this study. During rinderpest sero-surveillance 20 sera were collected from each of randomly selected village.

341

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 The sera were processed with the Bommeli Diagnostics FMD virus non-structural enzyme-linked immunosorbent assay (3ABC ELISA with 95% sensitivity and 97% specificity) for identifying infected animals from none infected ones (Egel, et al., 2008). The records of ministry of agriculture and rural development (MOARD) from 1999 to 2006 were also used to determine the status of FMD in the country. Tissue samples were collected from veterinary clinics, regional veterinary laboratories, animal health and production research centers and farms and submitted to the National Veterinary Institute, Foot and Mouth Disease Laboratory, between 1981 and 2007. A total of 133 tissue culture FMD virus samples were submitted to Institute for Animal Health (IAH), Perbright, UK for further molecular characterization and phylogenetic analysis. The VP1 gene characterization was used to study phylogenetic relationships between serotypes of foot-and-mouth disease (FMD) viruses in Ethiopia as well as with other serotype isolates from East, South and West Africa, the Middle East, Asia and Europe. 2.2 Data Analysis A homologous region of 639 nucleotides corresponding to the whole VP1 gene was used for all phylogenetic analysis. Nucleotide sequences of serotype O isolates from other African countries were included to deduce the phylogeny of this serotype on the African continent as well as isolates from the Middle East, Asia and South America to ensure that all previously identified lineages and genotypes were represented (Knowles and Samuel, 2003). Phylogenetic trees were constructed using methods of analysis included in MEGA version 4.0 (Tamura et al., 2007) and confidence levels were assessed by 1000 boot-strap replications. Serotypes were distinguished on the basis of nucleotide sequence differences of 30-50% and high boot-strap support (> 90%) while a divergence of > 15% distinguished Topotypes (Knowles and Samuel, 2003). For serological result; the 3ABC test results for each animal, origin (village, woreda, zone and administrative region) were recorded in an Excel (Microsoft Corp.) spreadsheet 3. Descriptive statistical analysis was carried out using Stata software version 9 (State Corp., College Station, TX, USA), while spatial regression was analyzed with GeoDA 0.9-I (Beta). The maps were generated using ArcGISv9.0 (ESRI, Redlands, CA, USA) 3. RESULTS 3.1 Retrospective result The records of ministry of agriculture and rural development (MOARD) from 1997 to 2006 indicated that FMD outbreak occurred every where through out the country with the highest in central part particularly in North showa 128 outbreaks reported during the indicated period of time (figure 1). Figure 1: Map showing No FMD outbreaks recorded in different part of Ethiopia (1999-06)

342

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

3.2 Serology result From the total 4465 sera tested 10.5% (n=467) were positive for FMD with 3ABC ELISA test. The highest seroprevalence was detected from samples of Oromia (20.7%), but from Gambella and Benshagul FMD virus specific antibody was not detected. In zonal administration level the highest sero-positivity was obtained from Eastern zone of Tigray with 41.5 % followed by Guji and Yeka sub-city of Addis Ababa with 32.7% and 30 % respectively. From 1876 samples collected from pastoral areas of the country 14.8% (n=325) prevalence was obtained (P<0.05) (figure 2). Figure 2: Sero-prevalence of FMD in different parts of Ethiopia

Figure 3: Status of FMD in Ethiopia

343

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Key: PER_P = percent positive, INC= incidence, MEAN_OB= mean number of outbreak in general the mean number of outbreaks, incidence rate and sero-prevalence of FMD showed that Tigray, central and southern part of Ethiopia are highly affected areas. 3.3 Viruses isolated Of the total 241 outbreak samples examined, cytopathic effect (CPE) was observed in 215 samples in primary calf kidney or IBRS 2 cell culture for the FMD virus. The CPE was characterized by a fast destruction of the cell monolayer and infected cells were round and formed singly. Complete destruction of the cell sheet was mostly seen with in 48 hours of inoculation. On those samples that showed CPE further examination were done to identify the type of the virus with complement fixation test; thus serotype O, A, C, SAT1 and SAT 2 were recorded. Type O was the dominant serotypes identified with 72.9% rate, while type A (19.7%), C (1.4%) SAT 2 (4.1%) and SAT1 (1.8 %) rate were detected. SAT 2 was recorded for the first time in Ethiopia in 1989 from a Bovine sample collected from Borena area (Third Livestock Development Project), southern Ethiopia, while SAT1 from Mizan Teferi area recently in 2007 (figure 5). Figure 4: The map indicating areas where different FMD serotypes isolated

344

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3.4 Phylogenetic analysis Phylogenetic trees were constructed using the Neighbour-joining method (MEGA4) based on the comparison of complete VP1 gene. Pairwise distance matrix was used to study phylogenetic relationships of foot-and-mouth disease (FMD) virus isolates of Ethiopia as well as with other isolates from the rest of the world (Tamura et al., 2007). Most of the O serotype isolates of Ethiopia lies on East Africa Lineage III (figure3). A new topotype was identified for the first time from serotype O and it is designated as East Africa 4 (EA-4) from Samples of Mizan Teferi located South West of Addis Ababa bordering Kenya and Sudan and also two National Parks (Omo and Mago) are adjacent to this location and also SAT1 was isolated from the same area for the first time from samples collected in 2007. 4. DISCUSSION Foot -and-mouth disease is enzootic as in most parts of Africa and only few countries in the south and north of the continent have managed to control the disease and have access to lucrative export markets for live animals and animal products (Sahle et al., 2004, Voslo et al., 2002). In Ethiopia, factors such as the presence of high numbers of susceptible animals, wild and domestic animals sharing common grazing pastures and watering points in areas where wildlife occur, as well as lack of control of animal movement contribute to the frequent occurrence of FMD outbreaks and to the difficulty in controlling the disease (Sahle et al., 2004). The results of this study indicated that the occurrence of Foot and Mouth Disease outbreak has been serious challenge every year in Ethiopia with the highest in 1999 with 821 outbreaks and this statement agrees with report of Asfaw (2000). Out of the seven serotypes of FMD virus the existence of serotype O, A, C, SAT1 and SAT 2 were recorded between 1981 and 2007 from Bovine and Swine samples collected from outbreak areas of the country. Most of the outbreaks were occurred by serotype O followed by A, SAT 2, C and SAT1 (figure 5). This shows that type O has highly prevalent and a dominant serotype causing an outbreak in Ethiopia and this observation agrees with the survey result (Martel, 1974; Martel and Gallon, 1975) that there is a tendency for type O strain to occur most frequently in the outbreak area. The first isolation of SAT 2 was in 1989 a sample collected from Bovine that were reared in Leben Ranch, Borena Zone southern Ethiopia operated by the Third Livestock Development Project (TLDP); These animals were purchased from an area called Wachle, which is around 100 km far from the border with Kenya. SAT1 and SAT2 were isolated recently from Mezan Teferi and Benshagul-Gumuz areas bordering Kenya and Sudan respectively from 2007 collected samples. This suggests that SAT1 and SAT 2 might be introduced from Kenya and Sudan along with cattle movement since SAT 1 and SAT 2 are endemic in those countries (OIE, 2002; Vosloo, et al., 2002). These findings are also similar to the previous report by Martel (1974) that SAT 1 SAT 2 were not isolated in Ethiopia. In a country such as Ethiopia where FMD is endemic, and where large numbers of susceptible domestic and wild ruminants exist with limited vaccination on some dairy farms, serological surveys are a pre-requisite to understand the epidemiology of FMD. To delineate the epidemiological profiles of the endemic occurrence of FMD in Ethiopia, 4465 sera of cattle from different regions were investigated using 3ABC ELISA serological test and 10.5 % prevalence was obtained. This result is lower than the study of Sahle (2004) who reported 26.25%, this might be due to the decrease of FMD outbreak for instance it was 176 in 2001 but decrease to 26 in 2005 ( MOARD report). In zonal administration level the highest sero-positivity was obtained from Eastern zone of Tigray with 41.5 % followed by Guji, Yeka sub-city of Addis Ababa and Borena zones with 32.7 % 30 % and 26.7 % respectively. Rufael and others (2008) reported 21 % sero-prevalence from Borena zone which comparable with the current finding.From 1876 samples collected from pastoral areas of the country 14.8% (n=325) prevalence was obtained (P<0.05). 5. ACKNOWLEDGEMENTS The researchers would like to thank Agricultural Research Fund (ARFCG-2003) programme, Ethiopian Agriculture Research Institute, for funding this project. Molecular analyses were supported by the Department for Environment, Food and Rural Affairs (DEFRA), UK (Grant numbers SE291 and SE2935). We need to thank also National Animal health diagnostic and Investigation centre (NAHDIC-Sebeta) for their supply of serum samples for this study. The eight months stay of Mr. Ayelet in IAH and Royal Veterinary College was supported by Rothamsted

345

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 International African Fellowship Programme. EUFMD/FAO for funding the current FMD study project and my participation in this meeting 6. REFERENCE [1] Engel B , Buist W., Orsel K., Dekker A., Clercq K, Grazioli S., Roermund H., (2008). A Bayesian evaluation of six diagnostic tests for foot-and-mouth disease for vaccinated and nonvaccinated cattle. Prev Vet Med. doi: 10.1016/j. prevetmed.2008.03.009. [2] Ferris, N.P.and Donaldson, A.I. (1992). The World Reference Laboratory for Foot and Mouth Disease: a review of thirty-three years of activity (1958-1991). Rev. sci. tech. Off. int. Epiz. 11: 657-684. [3] Knowles N.J., Samuel A.R., (2003). Molecular epidemiology of foot-and-mouth disease virus. Virus Research; 91 (2003) 65-/80. [4] Martel J. L., (1974). Foot-and-mouth disease in Ethiopia. Distribution of Foot and Mouth Disease Virus Serotypes. Revue Elev. Med. Vet. Pays Trop., 27 (2), 169-175. [5] Martel J.L. and Gallon C. (1975). Comparative serological study of the principal foot-andmouth disease virus strains isolated in Ethiopia, 1969-1974. Revue d'Elevage et de Medecine Veterinaire des Pays Tropicaux, 28(3): 287-295. [6] OIE (2002). World Animal Health in 2001. Reports on the Animal Health Status and Disease Control Methods. Office International des Epizooties (OIE), Paris, France, 131-132. [7] Roeder, P.L., Abraham, G., Mebratu, G.Y. and Kitching, R.P. (1994). Foot-and-mouth disease in Ethiopia from 1988 to 1991. Trop. Anim. Health. Prod. 26: 163-167. [8] Rufael T., Catley A., Bogale A., Sahle M.and Shiferaw Y., (2008). Foot-and-mouth disease in the Borana pastoral system, southern Ethiopia and implications for livelihoods and international trade. Trop Anim Health Prod (2008) 40:29–38 [10] Sahle M., Venter E.H., Dwarka R.M. and Vosloo W. Molecular epidemiology of serotype O foot-and-mouth disease virus isolated from cattle in Ethiopia between 1979-2001. Onderstepoort J Vet Res. 2004; 71: 129-138. [11] Vosloo W., Bastos A.D.S., Sangare O., Hargreaves S.K. and Thomson G.R., (2002). Review of the status and control of foot-and-mouth disease in sub-Saharan Africa. OIE Scientific and technical Review, 21(3): 437-447

346

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 61

FACTORS INFLUENCING GLOBAL FMD REPORTING AND RISK R. Garabed1, 2*, W. Johnson3, A. Perez1, 4 and M. Thurmond1 1

Center for Animal Disease Modeling and Surveillance, School of Veterinary Medicine, University of California, One Shields Avenue, Davis, CA 95616, USA 2 Present Address: Department of Veterinary Preventive Medicine, The Ohio State University, A100G Sisson Hall, 1920 Coffey Rd, Columbus, OH 43210, USA 3 Department of Statistics, Donald Bren School of Information and Computer Sciences, University of California Irvine, Bren Hall 2019, Irvine, CA 92697, USA 4 CONICET and Facultad de Ciencias Veterinarias UNR, Ov. Lagos y Ruta 33, Casilda, 2170, Argentina

ABSTRACT The quality of FMD surveillance and reporting varies globally and over time. Though information about FMD risk varies, harmonious measures are needed for active surveillance programs and development of global disease transmission models. As an alternative to the use of small regional studies and expert opinion estimates, we present models that use available and incomplete data to predict global risk and explore factors related to FMD risk and reporting. Our global models are used to predict true FMD risk and to compare the prediction to reported FMD risk. Maps of the models’ two FMD risk estimates represent differences in perceived FMD risk based on reporting. Traits associated with both FMD reporting and FMD presence varied by geographic region and might provide unconventional targets for intervention. The different prediction model formulae suggest traits of countries and local areas that might contribute to differences in FMD reporting and presence. 1. INTRODUCTION As is evident in regional FMD situation updates published by FAO EMPRES and EUFMD (2007) and in incidence reports voluntarily submitted to OIE (2008), the quality of FMD surveillance and reporting varies globally and over time. Because information about FMD risk varies, harmonious measures are needed for active surveillance programs. Knowing the number of expected cases in an area is critical for planning surveillance sampling and vaccination. In addition, consistent measures of baseline risk are necessary to develop global transmission models and to measure the progress of control programs. To derive consistent estimates in the face of inconsistent reporting, designers of vaccination and surveillance programs and developers of trade policy have necessarily 1) asked ‘experts’ to make recommendations extrapolated from their knowledge (Wint and Sumption, 2005), 2) had researchers collect data on FMD risk in small regions, or 3) assumed a worst-case scenario (ECHCP, 2007). Though the second technique (collecting data) is the most accurate method, time, expense, international politics, privacy issues, and possible danger to research teams argue against global use of this method. Techniques one and three can be sufficient for trade purposes, but their accuracy may be insufficient for active global surveillance and disease eradication. As an alternative, we present models that use available and incomplete data to provide a standardized approach to predicting global risk. In addition, these models have been used to explore for factors related to FMD risk and reporting. 2. METHODS The global case-control models were fit using expert opinion, data on FMD presence and absence, and publically available predictor data. The population at risk for FMD in each month in these

347

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 models was the land area of the world divided into 2500 km2 grid cells. Thus, the “risk of FMD” estimated here was the “probability of at least one case of FMD per cell per month.” The first model controlled for differences in reporting by using 1) OIE-certified FMD-free regions as controls for reported outbreaks and 2) expert opinion about differences in reporting among geographic subregions. This first model provided a best prediction of true FMD risk based on extrapolation from the biological mechanisms behind reported FMD cases and controls. The second model did not control for reporting using method 1, mentioned above, and, instead, used a random subset of all non-case cells as controls. Thus, the second model provided estimates of risk of reported FMD while accounting for general sub-regional differences in reporting. 2.1 The Models The basic form of both models was a case-control Bayesian logistic regression (Formulae 1 and 2) that used expert opinion to adjust for sub-regional differences in reporting and surveillance (Johnson et al., 2009). Experts estimated different reporting and surveillance proportions for the following six sub-regions: 1) Africa, 2) the Americas and Australia, 3) Europe, 4) the Middle East, 5) Central and South Asia, and 6) East and Southeast Asia. Experts also estimated their uncertainty about their assessments. FMDim~ Bernoulli (rim) logit (rim) = log (ρs) + ximβ + zjyθ RFMDim~ Bernoulli (rrim) logit(rrim) = log(ρs) + ximβ + zjyθ

(1) (2)

FMD – case-control data on FMD presence (1) or absence (0) for 1998 controls are from OIE certified FMD-free areas matched to the cases by the month and region of occurrence RFMD - case-control data on FMD presence (1) or absence (0) for 1998 controls are matched to the cases based on the month and region of occurrence r – risk: a probability of FMD presence in the given cell for the given month rr – reported risk: a probability of presence of reported FMD in the given cell for the given month ρ – ratio of probability of reporting in cases versus controls x – vector of cell-level predictor data z – vector of country-level predictor data β – vector of regression coefficients for the cell-level predictors (varies by region) θ – vector of regression coefficient for the country-level predictors (varies by region) i – indicator for cell m – indicator for month j – indicator for country (contains a subset of cells) y – indicator for year (contains all months) s – indicator for sub-region (contains a subset of cells and countries) region ≥ sub-region > country ≥ cell In addition, the predictor variables (listed in section 2.2) and their coefficients, used by the models to predict FMD risk, were selected independently by model, by region, and by year based on the value of the Bayes Factor (Kass and Raftery, 1995) calculated for each candidate model. Four regions (some of which contained multiple sub-regions with differences in reporting) were specified for the purposes of selecting different predictor variables: 1) Africa, 2) the Americas and Australia, 3) Europe and the Middle East, and 4) Asia. Selecting different predictor variables for these regions assumed that the factors influencing FMD occurrence, persistence, and reporting were the same within these regions. 2.2 The Data Cases of FMD voluntarily reported to OIE for the years 1997 and 1998 (FMD BioPortal, 2008) that could be geocoded at the second administrative unit or better based on a reported location name were used to fit both models. For the first, true FMD, model, a subset of cells in areas certified as free-of-FMD by OIE (OIE, 1997-1998) were used as controls. For the second, reported FMD, model, a random subset of non-case cells occurring in the same region and in the same month were used as controls. Data from 1997 were used to construct informed Bayesian priors to fit the models in 1998. Predictor variables available for the models to use were: Bovine density (FAOSTAT), Buffalo density (FAOSTAT), Small ruminant density (FAOSTAT), Pig density (FAOSTAT), Cell-level water borders (calculated), Cell-level water borders (as a probability, calculated), Human density (ORNL), Bovine meat deficit in the previous year (calculated using ORNL and FAOSTAT), Pig meat deficit in the previous year (calculated using ORNL and FAOSTAT), Sheep and goat meat deficit in the previous

348

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 year (calculated using ORNL and FAOSTAT), Distance to case this month in the previous year (calculated), Distance to case any time in the preceding year (calculated), Distance to case in the previous month (calculated), Month of Eid ul-Adha, Border length (CIA Factbook), Voice and accountability previous year (Kaufmann et al., 2007), Political stability previous year (Kaufmann et al., 2007), Government effectiveness previous year (Kaufmann et al., 2007), Regulatory quality previous year (Kaufmann et al., 2007), Rule of law previous year (Kaufmann et al., 2007), Control of corruption previous year (Kaufmann et al., 2007), Total literacy rate previous year (CIA Factbook), Female literacy rate previous year (CIA Factbook), Gross domestic product per capita previous year (CIA Factbook), FMD-positive borders proportion previous year (calculated), FMDnot-free borders proportion previous year (calculated), Country reported FMD previous year (OIE, 2008), Country not certified free of FMD previous year (OIE, 1997-1998). A subset of these available predictors was chosen for each region within each model based on a step-wise process that added and subtracted variables from the model if the resulting model improved the Bayes factor by at least ten-fold (Kass and Raftery, 1995). The final models’ predictions were internally validated and found to accurately distinguish between known case and control cells using ROC curves. 3. RESULTS AND DISCUSSION Maps of the two FMD risk estimates for January 1998 (Figures 1 and 2) represented differences in perceived FMD risk based on reporting. In general, these maps showed that areas of high predicted FMD-risk were more extensive than would be expected based on reported FMD alone. Especially in Africa and Asia, it appears that the conditions for FMD reporting were lacking over nearly half of the area at risk.

Figure 1: Estimated risk of FMD for January 1998

Figure 2: Estimated risk of reported FMD for January 1998 More interesting than the risk estimates themselves were the factors associated with predicted FMD risk and reported FMD risk in the different regions for 1998. In Africa, predicted FMD risk increased with decreasing bovine and small ruminant densities, increasing human density, decreasing pig

349

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 meat deficit, less distance from FMD outbreaks any time during the previous year and during the same month in the previous year, poorer voice and accountability in the government, poorer literacy rate, lower GDP per capita, more borders with countries reporting FMD, fewer borders with countries not free of FMD, reports of FMD in the country in the previous year, and FMD-free status for the country in the previous year. Reported FMD risk in Africa was predicted to increase with increasing bovine and small ruminant densities, increasing human density, decreasing pig meat deficit, greater distance from cases in the same month in the previous year, less distance from FMD outbreaks any time during the previous year or in the previous month, shorter total land border length, better voice and accountability in the government, poorer literacy rate, lower GDP per capita, more borders with countries reporting FMD, reports of FMD in the country in the previous year, and FMD-free status for the country in the previous year. In the Americas and Australia, the model predicted that FMD risk increased with decreasing bovine and swine densities, increasing human density, decreasing bovine and pig meat deficits, proximity to FMD outbreaks in the same month in the previous year, poorer control of government corruption, improved literacy rate, low GDP per capita, more borders with countries reporting FMD or not free of FMD, reports of FMD in the country in the previous year, and no FMD-free status for the country in the previous year. Reported FMD risk in the Americas and Australia was predicted to increase with decreasing bovine density, increasing pig and human density, decreasing bovine meat deficit, increasing pig meat deficit, proximity to FMD outbreaks in the same month in the previous year or at any time in the previous year, shorter total land border length, poorer control of government corruption, improved literacy rate, low GDP per capita, more borders with countries reporting FMD or not free of FMD, reports of FMD in the country in the previous year, and no FMDfree status for the country in the previous year. In Europe and the Middle East, the model predicted that FMD risk increased with decreasing buffalo and small ruminant densities, decreasing human density, more land borders (as opposed to water borders) at the cell level, greater sheep and goat meat deficit, proximity to FMD outbreaks any time in the preceding year, poorer total and female literacy rate, increasing GDP per capita, and reports of FMD in the country in the previous year. Reported FMD risk in Europe and the Middle East was predicted to increase with decreasing buffalo density, increasing small ruminant and human densities, more water borders at the cell level, decreasing bovine meat deficit, increasing sheep and goat meat deficit, proximity to FMD cases in the same month in the previous year or anytime in the previous year, shorter land border length, poorer political stability, improved total literacy rate, increasing GPD per capita, and no FMD-free status for the country in the previous year. Finally in Asia, the model predicted that FMD risk increased with decreasing buffalo density, increasing pig density, more land borders at the cell level, higher deficit of sheep and goat meat, proximity to FMD cases any time in the preceding year or month, months other than the month of Eid ul-Adha, increasing land border length, improved rule of law, poorer literacy rate, reports of FMD in the country in the previous year. Reported FMD risk in Asia was predicted to increase with increasing buffalo and human densities, decreasing small ruminant density, more water borders at the cell level, higher bovine and sheep and goat meat deficits, proximity to FMD cases in the previous month, anytime in the previous year or in the same month in the previous year, the month of Eid ul-Adha, shorter land borders, and improved rule of law. Though not all of these effects were significant, several differences between the models for reported FMD and true FMD were significant and provided insight into the mechanisms behind FMD risk and reporting in the different geographic regions (Table 1). 4. AUTHORS’ CONCLUSIONS

350

Factors influencing FMD risk and FMD reporting are different and vary by geographic region. Predicted targets for intervention: Africa – education, political voice and accountability, geographic coverage of surveillance programs (access) Americas and Australia – regional inertia Europe and Middle East – education, funding for reporting, and seasonal consistency in surveillance Asia – surveillance before animal movement and border control

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 5. AUTHORS’ RECOMMENDATIONS

Target interventions by region using data-based evaluations of the biological and human factors driving FMD and FMD reporting in that region. Continue using statistical models to evaluate the global FMD situation, changes in factors influencing FMD risk, and success of FMD control programs.

6. ACKNOWLEDGEMENTS Thank you to the staff and students at the UC Davis FMD Modeling and Surveillance Laboratory for assistance in collecting the FMD presence data; to Paul Kitching and Julio Pinto for providing their expert opinions; and to Caesar Orozco, Moetapele Letshwenyo, Miryam Gallego, Bernardo Cosentino, Conrad Estrada, Willie Ungerer, and Cleopas Bamhare for providing maps of FMD-free areas. Support was provided by funding from the National Center for Medical Intelligence and the UC Davis School of Veterinary Medicine. 7. REFERENCES [1] Christensen, R, Johnson, W, Branscum, A and Hanson, T. Bayesian Ideas and Data Analysis: An Introduction for Scientists and Statisticians. Accepted for publication. [2] CIA Factbook. Central Intelligence Agency, Washington, D.C., USA, available at https://www.cia.gov/library/publications/the-world-factbook/index.html. [3] EC-HCP, 2007. General guidance on EU import and transit rules for live animals and animal products from third countries. EC Health and Consumer Protection Directorate General, Brussels, Belgium, available at http://ec.europa.eu/food/international/trade/guide_thirdcountries2006_en.pdf. [4] FAO EMPRES and EUFMD. Focus on foot-and-mouth disease situation worldwide and major epidemiological events in 2005-2006. FAO, Rome, 2007 available at http://www.fao.org/docs/eims/upload//225050/Focus_ON_1_07_en.pdf. [5] FAOSTAT. http://www.fao.org/waicent/portal/statistics_en.asp. [6] FMD BioPortal. FMD Modeling and Surveillance Laboratory, Davis, CA, USA, 2008. Available at http://fmd.ucdavis.edu/bioportal. [7] Kass, R.E. and Raftery, A.E. 1995. Bayes Factors. JASA. 90(430): 773-795. [8] Kaufmann, D., Kraay, A., Mastruzzi, M., Governance matters VI: governance indicators for 1996-2006. World Bank Policy Research Working Paper 4280, World Bank Institute, Washington, D.C., 2007, available at SSRN: http://ssrn.com/abstract=999979. [9] OIE. Recognition of the foot-and-mouth disease status of member countries. Resolutions adopted by the international committee of the OIE during its 65th [to 66th] general session, World Animal Health Organization, Paris, 1997-1998. [10] OIE. World animal health information database (WAHID) interface.OIE, Paris, France, 2008, available at http://www.oie.int/wahid-prod/public.php?page=home. [11] ORNL. Landscan 2004. Oak Ridge National Laboratory, Tennessee, USA, 2004. [12] Wint, W. and Sumption, K. Mapping the FMD homelands: An exploratory look at global ruminant production systems associated animal movements, and FMD risk. Consultancy report, EUFMD Commission, Food and Agriculture Organization of the United Nations, Rome, Italy, 2007, presented to the General Session of the European Union Commission on Foot and Mouth Disease. Table 1: Direction of effect of and possible rationale for selected factors that appear to be associated with FMD risk and FMD reporting in different regions Region Africa

Evidence "True" Risk decreases increasing density

Interpretation with bovine

"Reported" Risk decreases with decreasing human density

decreases with improved voice and accountability increases if the country reported outbreaks in the previous year decreases with

increases with improved voice and accountability decreases if the country was not free of FMD in the previous year not significantly

- areas with high bovine density are better at controlling FMD and cases are reported more often in areas with lots of people due to access and high value of cattle - communication between the people and the government improves FMD reporting and control - countries that have FMD are likely to have FMD again and countries that have not reported FMD are not likely to report in the next year - status quo - education and economic health are related

351

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The Americas and Australia

Europe and the Middle East

improved literacy rate and GDP per capita

related to literacy and GDP per capita

not significantly related to distance from cases in the previous month decreases with increasing bovine density: pig density insignificant greater if country was not declared free-of-FMD in the previous year increases with more borders with countries not free of FMD not significantly related to pig meat deficit not significantly related to distance from cases in the previous year decreases with increasing small ruminant density

decreases farther from cases in the previous month

decreases with increasing human density decreases with improved total and female-specific literacy rates not significantly related to seasonal distance from cases Asia

increases with increasing pig density: buffalo density insignificant not significantly related to human density increases with longer land border length decreases during the month of the Eid ul-Adha not significantly related to sheep and goat meat deficit not significantly related to seasonal distance from cases

352

increases with increasing pig density: effect of bovine density insignificant effect smaller and less significant

to better FMD control but do not appear to improve reporting significantly - control may be occurring without the help of outside sources - reporting of cases but not occurrence of cases is more likely in the month after and close to the location of outbreaks - possible indication of limited surveillance - areas with high bovine density have lower risk of FMD possibly due to high value of cattle while areas with high pig density are more likely to report FMD - regional inertia of infection across borders

increases with increasing pig meat deficit decreases farther from cases anytime in the previous year

- areas that import pig meat are more likely to report FMD possibly due to recognition at slaughter or better funding - reporting (not occurrence) of cases is somewhat consistent over the year, but may be localized

increases with increasing small ruminant density

- increased reporting and decreased risk are associated with small ruminant density - a possible indication of successful control programs in small ruminants - indicates better control and reporting programs where more people are present

increases with increasing human density not significantly related to literacy

decreases farther from cases in the same month in the previous year increases with increasing buffalo density: pig density insignificant increases with increasing human density decreases with longer land border length

increases during the month of the Eid ulAdha increases in areas with high sheep and goat meat deficits decreases farther from cases in the same month in the previous year

- education, and specifically in women, is associated with reduced FMD risk but not reporting - a possible indication of control without outside assistance - proximity to locations that had FMD outbreaks a year ago increases the chances of reporting an outbreak - surveillance may be seasonal - areas with higher pig density have a higher FMD risk possibly due to greater transmission in pigs, while reporting appears to be better in buffalo - reporting of FMD is more likely in areas with high density of humans - possibly due to better surveillance - countries with longer land borders have poorer FMD control and poorer FMD surveillance - likely due to a lack of resources for border control and surveillance - areas that import sheep and goats, specifically in the month of the Eid, are likely to report cases of FMD, but cases are less likely to occur overall in the month of the Eid possibly because the animals have been shipped and slaughtered in that month - proximity to locations that had FMD outbreaks a year ago increases the chances of reporting an outbreak - surveillance may be seasonal, possibly related to the Eid

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 62 PARTICIPATORY EPIDEMIOLOGY AS COMPARED TO CONVENTIONAL FOOT AND MOUTH DISEASE SURVEILLANCE TOOL T. Rufael1*, A. Catley2, A. Bogale3, M. Sahle1 and Y. Shiferaw4 1

National Animal Health Diagnostic and investigation Center, Ethiopian, P.O. Box 04, Sebeta, Ethiopia e-mail: rufaelc@yahoo.com 2 Feinstein International Center, Friedman School of Nutrition Science and Policy, Tufts University, P.O. Box 1078, Addis Ababa 3 Faculty of Veterinary Medicine, Addis Ababa University, P.O. Box 34, Debre Zeit 4 Ethiopian Institute of Agricultural Research, P.O. Box 2003, Addis Ababa

ABSTRACT Participatory epidemiology (PE) was used on the Borana plateau of southern Ethiopia to understand pastoralist’s perceptions of the clinical and epidemiological features of foot-and-mouth disease (FMD) in cattle. This paper describes the use of Participatory epidemiology to collect information on the basic epidemiology of FMD and compare the data from Participatory epidemiology with estimate of conventional seroprevalence result of FMD in Borana pastoral herds. Participatory appraisal and conventional veterinary investigation methods were applied to generate information on FMD in Borana pastoral system. The participatory appraisal methods used were Matrix scoring, Clinical observation, Proportional piling, and Seasonal calendar. The validity of participatory diagnosis of disease and their perceptions of proportions of animal affected by FMD can be cross checked by modern laboratory diagnosis. Serum samples were collected from randomly selected Borana cattle herds and screening by 3ABC ELISA for Non structural protein. Positive serum samples were serotyped for specific FMDV using liquid phase blocking ELISA (Hamblin et al, 1986a, b). Matrix scoring showed good agreement between informant groups on the clinical signs of acute and chronic FMD, and findings were cross-checked by clinical examination of cattle and assessment of previous clinical FMD at herd level by detection of antibody to non structural proteins of FMD virus. The positive predictive value of pastoralist’s diagnosis of FMD at herd level was 93.1%. The annual age-specific incidence and mortality of acute FMD in 50 herds was estimated using proportional piling. The estimated mean incidence of acute FMD varied from 18.5% in cattle less than two years of age to 14.0% in cattle three to four years of age. The estimated mean mortality due to acute FMD varied from 2.8% in cattle less than two years of age to 0.3% in cattle three of age or older. Pearson correlation coefficients for acute FMD by age group were −0.12 (p>0.05) for incidence and −0.59 (p<0.001) for mortality. Estimates of the annual incidence of chronic FMD varied from 0.2% in cattle less than two years of age to 1.8% in cattle three to four years of age. The Pearson correlation coefficient for the incidence of chronic FMD by age group was 0.47 (p<0.001). Outbreaks of FMD peaked in Borana cattle during the two dry seasons and were attributed to increased cattle movement to dry season grazing areas. The mean seroprevalence of FMD was estimated at 21% (n=920) and 55.2% of herds (n=116) tested seropositive. Serotyping of 120 seropositive samples indicated serotypes O (99.2%), A (95.8%), SAT 2 (80%) and C (67.5%). The observed agreement between pastoralist perception on FMD diagnosis and conventional laboratory diagnosis from this study has proven that Borana pastoralists have an enormous wealth of knowledge on diagnosis of cattle diseases. Therefore, Participatory Appraisal methods complements FMD surveillance to gain better understanding of FMD dynamics in pastoral areas and help to formulate appropriate disease control strategies. 1. INTRODUCTION Foot-and-mouth disease (FMD) is highly contagious viral disease of cloven-hoofed domestic and wild animals. It is widely distributed and occurs most commonly in Asia, Africa, the Middle East, and parts of South America (Kitching 1999). In pastoral areas of Africa outbreaks of FMD are reported frequently but the disease remains largely uncontrolled using conventional methods (Thomson and Bastos 2005). Providing veterinary services to the communities according to the western model has proven difficult due to lack of infrastructure and the veterinarian has limited

353

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 experience in harsh environments of pastoral system. Participatory Appraisals (PA) is a systematic collection and analysis of data for diseases like FMD that are often under reported by conventional veterinary services due to comparatively high tolerance of local breeds to the clinical episodes of the disease (Leforban, 2005), However, at certain times of year pastoralists rely heavily on milk for food and therefore, they often prioritize FMD due to its impact on milk supply. They also associate FMD with mortality in calves and ‘chronic FMD’ cases showing heat intolerance, reduced fertility and other signs (Catley et al. 2004). Participatory epidemiology is the application of participatory methods in a number of animal health epidemiological disease surveillance. The participatory methods offer input in addition to laboratory based epidemiology, that pastoralists often possess detailed indigenous knowledge on livestock diseases and in terms of clinical diagnosis, describe diseases in similar ways to veterinarians (Catley, 2006). Veterinarians and livestock workers have used and are presently using a variety of PA methods to investigate animal health problems (Catley, 1999). The tools include interviewing, scoring and ranking, and visualization such as seasonal calendars, maps, Venn diagrams, and flow charts. Although PE studies can be purely qualitative in nature, standardization and repetition of PE methods can produce quantitative data leading to estimates of disease incidence, mortality or other variables. The process of triangulation in Participatory epidemiology is used to improve the validity of findings and involves cross checking information from different sources or methods, including conventional veterinary methods. The global rinderpest eradication program adopted Participatory epidemiology as a surveillance tool for controlling rinderpest. This approach was subsequently used in both rural and urban settings in Africa and Asia, for foot-and-mouth disease, peste des petitis ruminantus and highly pathogenic avian influenza (Jost et. al, 2007). Participatory disease surveillance approach is now recognized in world organization for animal health (OIE) guidelines for rinderpest surveillance and as an important approach for general surveillance (OIE, 2007). This paper describes the use of Participatory epidemiology to collect quantitative and qualitative information on the basic epidemiology of FMD and compare the data from Participatory epidemiology with estimate of conventional seroprevalence result of FMD in Borana pastoral herds. 2. METHODOLOGY The research was conducted in three districts (Yabello, Dirre and Moyale) of Borana pastoral system of Oromia Regional State, located between 03037' 23.8" to 050 02' 52.4" North and 370 56' 49.4" to 390 01' 10.1"East, in the Southern part of Ethiopia. The Borana pastoral system represents a vast lowland area, covering about 95,000 km2 (Coppok, 1994) and bordering northern Kenya. Borana pastoralists are mainly cattle-keepers and they manage their cattle using traditional pastoral systems. The methodology involved both participatory appraisal and conventional veterinary investigation methods were applied to generate information on FMD in the study areas. The participatory appraisal methods used were Matrix scoring, Clinical observation, Proportional piling, and Seasonal calendar was adapted from method described by Catley, (Catley, 2005; Catley et al., 2001). The validity of participatory diagnosis of disease and their perceptions of proportions of animal affected by FMD can be cross checked by modern laboratory by using antibody detection. 920 Serum samples were collected from randomly selected Borana cattle herds and examined for antibodies to 3ABC non structural proteins of FMD virus by using 3ABC ELISA (chekit-FMD 3ABC, Intervet). 120 Samples, 3ABC ELISA positive samples were further tested to determine FMD virus serotypes O, A, C and SAT2 by using the liquid phase blocking ELISA according to the manufacturers manual (Ferris, 2004). 3. RESULTS The results of matrix scoring for FMD and other diseases are shown in Table 1. There was good agreement between the 12 informant groups for all 17 disease indicators (W=0.54 to 0.96; p<0.001). In all the informants groups the informants scored FMD by considering both acute and chronic forms of the disease called hoyaale and gaandile respectively. Hoyaale was associated with salivation, lameness, skin lesions, reduced appetite, abortion, decreased milk yield, loss of body condition and mortality (in young animals). In comparison, gaandile affected animals that recovered from hoyaale but later developed signs of hair overgrowth, panting, shade-seeking behaviour and infertility. Informants also mentioned that if these cases produced calves, the calves were weak or stillborn and the dam’s milk was watery for the first two week. Matrix scoring also indicated that hoyaale/FMD was transmitted by direct contact and affected wildlife such as oryx,

354

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 kudu, and gazelle. Informants noted that other domestic animals such as sheep and goats could also suffer from hoyaale/FMD. Eight chronic FMD (gaandile) case identified by pastoralists were examined clinically by researcher for the triangulation. Table 1: Summarized Matrix scoring of disease indicators in three districts of Borana pastoral area (Dec. 2004-Nov. 2005).

Indictors Coughing (W=0.956***) Salivation (W=0.885***) Abortion W=0.881***) Lameness (W=0.900***) Mortality (W=0.871***) Reduced milk production (W=0.543***) Loss of body weight (W=0.814***) Skin lesion (W=0.782***) Teat lesion (W=0.762***) Hair over growth (W=0.902***) Panting (W=0.778***) Seek shade (W=0.768***) Decrease fertility W(=0.840***) Decrease market value (W=0.624***) Disease affect wild life (W=0.569***) Transmitted by tick (W=0.9245***) Transmitted with contact (W=0.840***)

CBPP (Sombesa) 25(17-25)

FMD (Hooyale) 0(0-8)

LSD (Suuki) 0(0-0)

Black (Haarka) 0(0-0)

leg

Mastitis (Nakarsa) 0(0-0)

0(0-6)

25(15-25)

0(0-0)

0(0-7)

25((18-25)

0(0-7)

0(0-2)

0(0-0)

0(0-3)

13.5(6-24)

0(0-3)

7.5(1-13)

0(0-7)

7.5(3-11)

3(0-5)

4(0-6)

10.5(7-22)

0(0-6)

0(0-8)

11(6-18)

3(0-9)

0(0-8)

10(2-13)

1(0-8)

16(8-25)

6(0-12)

0(0-2)

0(0-4)

0(0-0)

7(0-10)

0(0-5)

2(0-8)

0(0-0)

0(0-5)

15.5(1215) 5(0-17)

0(0-0)

25(17-25)

0(0-8)

0(0-0)

19.5(825) 0(0-0)

3(0-15)

22(10-25)

0(0-5)

0(0-2)

0(0-0)

0(0-10)

20.5(14-25)

0(0-5)

0(0-4)

0(0-0)

22(14-25)

2.5(0-11)

0(0-0)

0(0-9)

9(6-12)

7.5(4-13)

0(0-0)

5(0-13)

0(0-0)

19(0-25)

0(0-7)

0(0-14)

0(0-5)

0(0-0)

0(0-12)

0(0-0)

25(13-25)

4.8(2-8)

6(5-10)

7(4-10)

0.5(0-4)

0(0-2)

Number of informants groups =12; W = Kendall’s Coefficient of Concordance (*P<0.05; **P<0.01; ***P<0.001). W values vary from 0 to1; the higher the value the higher the level of agreement between informants groups. The number out of parentheses indicates median scores of the 12 groups and minimum and maximum limits are shown in Parentheses. Age-specific incidence and mortality of acute FMD (hoyaale) is shown in Fig 1. Pearson correlation coefficients for incidence and mortality by age group were −0.12 (p>0.05) and -0.59 (p<0.001) respectively. The Pearson correlation coefficient for incidence caused by chronic FMD (gaandile) by age group was 0.47 (p<0.001).

355

Mean incidance and mortality of acute FMD and 95% CI

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 30

0 I (r= -0.12,p=0.08)

-10

M (r= -0.59,p=0.00) Calves

Weaner

Young

Adult

Figure 1: The mean annual incidence and mortality of acute FMD in different age groups of cattle in Borana pastoral area (Dec. 2004-Nov. 2005) A summarized seasonal calendar for livestock diseases, rainfall, tick infestation, cattle movement, and contact with wildlife is shown in Table 2. Moderate to good agreement (W=0.29 to 0.99) was evident between informant groups for the diseases CBPP (sombessa), acute FMD (hoyaale), LSD (sukii), blackleg (haarka) and mastitis (nakarsa), and also for seasonal rainfall and cattle movement. The incidence of acute FMD (hoyaale) peaked during the two dry seasons called short dry season (adoolessa) and long dry season (bona), with relatively higher incidence in the main dry season. Cattle movement followed a similar seasonal pattern to the incidence of acute FMD. Table 2: Summarized seasonal calendar on the occurrence of different diseases of cattle in Borana pastoral area (Dec. 2004-Nov. 2005) Borana seasons Gana (Long rain) 53.1.1.1.1.1.1.1

Hagayya (Short rain)

Bona (Long dry)

Months by Gregorian calendar M

356

Adoolessa ( Cold dry)

Rainfall (W=0.995)***

12(10-13)

2(0-3)

6(4-8)

0(0-2)

CBPP (Sombesa) (W=0.700)***

2(0-5)

5.5(2-8)

1.5(0-4 )

10(3-13)

FMD (Hooyale) (W=0.938)***

1(0-3)

5(4-7)

1(0-3)

12.5(10-16)

LSD (Suuki) (W=0.870)***

3(1-5)

12.5(9-18)

4(0-6)

0(0-2)

Blackleg (Haarka) (W=0.919)***

13.5(9-18)

2(0-3)

4(0-8)

0(0-1)

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Mastitis (Nakarsa) (W=0.787)*** Tick infestation (W=0.397)**

4.5(2-7)

3(2-4)

11.5(7-16)

0.5(0-6)

9.5(0-13)

3(0-7)

5(3-9)

2(0-10)

Increased cattle movement (W=0.839)***

1.5(0-5)

5(3-6)

2(0-6)

11.5(9-14)

Wildlife Contact

8.5(2-12)

4.5(0-10)

4.5(2-8)

2.5(0-10)

(W=0.287)*

N=10; W, Kendall’s coefficient of concordance (*P <0.05; **P <0.01; ***P< 0.001). The number out side the bracket represents medians and minimum and maximum values are in the bracket. The agreement was termed weak, moderate and good if W-values were less than 0.26, between0.26 and 0.38 (p < 0.05) and greater than 0.38(p< 0.01 to 0.001), respectively (Seigel and Castellan, 1994). From the 116 herds examined for the presence of antibodies to 3ABC non-structural protein of FMD virus, at least one animal tested positive in 64 herds (55.2% of herds) and 21% of individual (n = 920) tested positive. Out of 120 samples randomly selected from 193 positive sera for serotyping using liquid phase blocking ELISA, 99.2%, 95.8%, 80.0%, and 67.5% were positive, respectively for O, A, SAT2 and C serotypes. Of the 50 herds that were sampled, 41 (82 %) of the herders reportedly observed clinical FMD in their herds at various times between Dec. 2004-Nov. 2005. The accuracy of the herder diagnosis was further shown by calculating the positive predictive value (PPV) based on the herd level diagnosis and the 3ABC ELISA test results. Calculations demonstrated that the PPV of herder diagnosis of FMD at herd level is 93.1% (95% CI = 78.0%, 98.1%). This validates the matrix scoring and Proportional piling data on disease characterisation, incidence and mortality. The detection of a non-structural protein, 3ABC is currently regarded as reliable indicator to show FMD virus infection. The antibody to the 3ABC protein persists in affected animals for over one year (up to 395 days) post infection. 4. DISCUSSION The study showed that Borana pastoralists described clinical and epidemiological features of FMD according to typical descriptions of the disease in veterinary textbooks. Pastoralist perceptions of the clinical signs of acute and chronic FMD (Table 1), age specific incidence and mortality due to FMD (Fig 1) and seasonal variations in FMD outbreaks associated with cattle movement (Table 2) were all consistent with modern veterinary thinking (Radostits et al.1994). The clinical signs of chronic FMD (gaandile) reported by Borana herders were similar to those reported in cattle in pastoralist areas of Afar region (Tadesse, 2003) and Somali region (Eshetu 2003) in Ethiopia, and in Maasai areas of Tanzania (Catley et al. 2004). Up to 12.8% of acute FMD cases in Borana cattle later developed signs of chronic FMD, and this was similar to the 12.0% of Maasai cattle developing chronic FMD after acute disease (Catley et al. 2004). Incidence estimates (Fig 1) and herd seroprevalence estimates (Table 3) indicated that FMD was endemic in Borana pastoral system. FMD was also reported as one of the main diseases of Borana cattle more than 15 years ago (Coppock 1994), indicating the long-term presence of the disease. Compared to other cattle diseases, FMD cases in our study were observed by pastoralists more frequently than any other disease. The overall seroprevalence of 21% (55.2% of Borana herds) was similar to the 26.5% seroprevalence reported elsewhere in Ethiopia (Sahle 2004). Although we did not aim to quantify the risk factors associated with FMD outbreaks, the seasonal calendar (Table 2) suggested a link between FMD outbreaks and cattle movement during the two dry seasons. In Borana production systems, the forra herd comprises the majority of the cattle and these herds are moved during the dry season to access grazing and water. Herds from different areas congregate around these resources as the dry season progresses and theoretically, contact between herds would enable transmission of FMD virus. Research on transhumant Fulani cattle in

357

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Cameroon showed that important risk factors in FMD transmission included transhumance and mixing of herds at watering points (Bronsvoort et al. 2004). The detection of FMD serotypes O, A, SAT2 and C in Borana cattle agreed with surveys in the Omo National Park and Bale mountain area of Ethiopia (Sahle 2004), and samples collected during FMD outbreaks in dairy farms in and around Addis Ababa (Dejene 2004). Serotypes O, A, and SAT2 were also isolated from cattle by National Veterinary Institute in Ethiopia during outbreak investigations between 1982 and 2000 (Gelaye et al.2001). This study has shown that participatory epidemiology is one of the options that best complement veterinary disease surveillance. In developing countries where disease surveillance and diagnostic services are not smooth, standardized participatory epidemiological tools can produce quantitative as well as qualitative data that can be used to formulate appropriate global FMD control. 5. ACKNOWLEDGEMENTS The authors thank Dr Keith Sumption for his help and FAO EMPRS for the financial support and Dr Andy Catley and Dr Brihnu Admasu for reviewing the manuscript. 6. REFERENCES [1] Bronsvoort, B.M.de C., Nfon, C., Hamman, S.M., Tanya, V.N., Kitching, R.P., Morgan, K.L., 2004. Risk factors for herdsman-reported foot-and-mouth disease in the Adawa Province of Cameroon. Preventive Veterinary Medicine 66, 127–139 [2] Catley, A. (1999): Method on the move. Review of veterinary uses of participatory approaches and method focusing on experience on dry land Africa. International Institute for Environment and Development .London, UK. 33-42. [3] Catley, A., Okoth, S., Osman, J., Fison, T., Njiru, Z., Mwangi, J., Jones, B, A., Leyland, T, J. (2001): Participatory diagnosis of a chronic wasting disease in cattle in southern Sudan. Prev.Vet. Med. 51 (3/4): 161-181. [4] Catley, A., Chibunda, R. T., Ranga, E., Makungu, S., Magayane, F. T., Magoma, G., Madege, M. J., Vosloo, W. (2004): Participatory diagnosis of heat intolerance syndrome in Cattle in Tanzania and Association with foot-and-mouth disease. Prev.Vet. Med. 65: 17 30. [5] Catley, A. (2005): Participatory Epidemiology: A Guide for trainers. African union/ InterAfrican Bureau for animal resources, Nairobi. 1-42. [6] Catley, A. (2006): The use of participatory epidemiology to compare the clinical and veterinary knowledge of pastoralists and veterinarians in East Africa. Tropical Animal Health and Production 38, 171–184 [7] Coppock, D. L. (1994): The Borana plateau of southern Ethiopia: Synthesis of Pastoral Research Development and change, 1980-1991. ILRI, Addis Ababa, Ethiopia. 15-33 [8] Dejene, A. (2004): Foot-and-mouth outbreak investigation in smallholder and Commercial Dairy Farms in and around Addis Ababa, DVM, Thesis, FVM, Debre Zeit. Pp: 30-39. [9] Eshetu, T. (2003): Participatory studies on heat intolerance syndrome associated with FMD in indigenous cattle of Somali pastoral area in Shinille Zone, Ethiopia. DVM, Thesis, FVM, Debre Ziet. [10] Ferris, N.P. (2004): FMD ELISA kit bench protocol: liquid phase blocking ELISA for Detection of FMD virus serotypes O, A C and SAT2 for Ethiopia. Institute for animal health, Pirbright laboratory, UK. 3-42. [11] Gelaye, E., Beyene, B., and Ayelet, G. (2001): Foot-and-mouth disease virus serotypes [12] Jost, C.C., Mariner J.C., Roeder P.L., Sawitri, E., Macgregor-Skinner G.J. (2007): Participatory epidemiology in disease surveillance and research. Rev. Sci. Tech. off. int. Epiz. 26(3), 537-547. [13] Kitching, R.P. (1999): Foot-and-mouth disease: Current world situation. Vaccine, 17: 17721774. [14] Leforaban, Y. (2005): Report of a mission on Foot and Mouth disease in Ethiopia, Proposals for a Strategic plan for a control program oriented to the export, 10-22 April 2005. 12-42. [15] Radostits, O. M., Blood, D. C., Gay, C. C. (1994): Veterinary Medicine, 8th edition. London: Bailliere Tindall. 345-372. [16] Sahle, M. (2004): An epidemiological study on the genetic relationships of foot-and-mouth disease viruses in east Africa. University of Pretoria, South Africa,. Pretoria, PhD Thesis. 84-107. [17] Siegal, S. and Castellan, N.J., 1988. Nonparametric Statistics for the Behavioural Sciences, second edition, (McGraw-Hill, New York)

358

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [18] Tadesse, G. (2003): Participatory studies on Heat intolerance syndromes Associated with FMD in indigenous cattle in Afar pastoral area of Ethiopia. Faculty of veterinary Medicine, Addis Ababa University, Debre Zeit DVM, Thesis. [19] Thomson, G.R. and Bastos, A.D.S. (2005). Foot-and-mouth disease. In: J.A.W. Coetzer and R.C. Tustin (eds), Infectious Diseases of Livestock, (Oxford University Press, New York), 1324– 1366 [20] World Organisation for Animal Health (OIE) (2007): Final report of the 17th conference of the OIE Reginal commission for Africa: strategy for strengthening epidemiological surveillance in Africa, Asmera, Eritria, 26 February- 1 March. OIE, Paris, 53-54.

359

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

360

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 63

DEVELOPMENT OF AN IMPROVED CAPABILITY IN SUPPORT OF NATIONAL BIO-SECURITY FOR THE SURVEILLANCE AND CONTROL OF FOOT-AND-MOUTH DISEASE IN CATTLE AND PIGS IN VIET NAM C. Morrissy1*, N.T. Long2, L. Wright1, D.M. Hoa2, P.P. Vu2, N.T. Phong2, L.H. Vu2, I. Pritchard1, S. Riddell1, D. Schafer1, D. Eagles1, W. Ha1, W. Goff1, J. Hammond3, S. Juzva1, M. Johnson1 and P. Daniels1 1

CSIRO Livestock Industries, Australian Animal Health Laboratory (AAHL), Geelong, Victoria, Australia. 2 Regional Animal Health Office No. 6, Ho Chi Minh City (RAHO6-HCMC), South Viet Nam. 3 Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey, United Kingdom.

INTRODUCTION Development and implementation of disease control strategies depend upon a good understanding of FMD epidemiology which in turn requires accurate laboratory testing for both virus typing and serosurveillance. The establishment of an integrated National laboratory network is vital, with National Biosecurity being an important issue for the Viet Namese Government. The aims of the project were to determine why FMD vaccination of livestock does not give the expected protection against disease and to fully determine which serotypes of FMDV are circulating in Viet Nam. This will enable better vaccination strategies to be employed. Regional diagnostic laboratories will be established with the capacity to carry out rapid and accurate FMDV serology, virus isolation and detection of antigen and viral genome. MATERIALS AND METHODS Pilot zones were established in provinces near Viet Nam’s borders to gain insight into FMD serotypes circulating in these areas. Molecular sequencing of the FMD isolates from these provinces provided information on the effectiveness of border control and the origin of FMDV circulating in Viet Nam each year. FMD serology was carried out on 15,000 sera using the AAHL 3ABC NS ELISA & LP-ELISA (O, A & Asia 1). RESULTS Nucleotide sequence analysis of 100 FMDV isolates revealed 3 topotypes for serotype O [Cathay, ME-SA (PanAsia) and SEA (Myanmar 98)], only Thailand/Malaysia 97 for serotype A and both Jiangsu-China-2005 and Myanmar 98 for serotype Asia 1. Serosurveillance data showed vaccination efficacy and prevalence of infection. CONCLUSION This project has established an improved diagnostic capacity for FMD in Viet Nam through the development of an integrated network of laboratories. The improved capacity has allowed the early detection and identification of FMD thus enabling better control of disease and helping reduce livestock losses and so improve productivity. Nucleotide sequencing has provided vital new information on circulating viruses. The project is aligned with the National FMD plan and the information generated is aiding the control of FMD in Viet Nam.

361

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

362

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 64

DETECTION OF FMDV SEROTYPES O, A AND ASIA 1 BY REAL-TIME RT-PCR S. M Reid1*, N. J Knowles1, M. H. N.Shirazi2, D. P King1 1

Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey, GU24 0NF, UK 2 Iran Veterinary Organisation, Central Veterinary Laboratory, Tehran, Iran

ABSTRACT In countries like Iran where foot-and-mouth disease (FMD) is endemic, identification of the serotypes of the causative virus strains is important for vaccine selection and for tracing the source of the outbreaks. Real-time reverse transcription polymerase chain reaction (rRT-PCR) assays for the specific detection of FMD virus (FMDV) serotypes O, A and Asia 1 were evaluated using primers/probe sets designed from the VP1 region of the FMDV genome. RNA extracted from suspensions of vesicular epithelia of representative strains of serotypes O, A and Asia 1 was tested by new one-step rRT-PCR assays run in parallel with pan-serotype specific assays (5′ UTR and 3D). These new assays incorporated primers/probe sets intended for specific detection of serotype O, A and Asia 1 virus sequences, respectively. Strains were chosen from subgroups of each serotype recently circulating in the Middle East. Serotype O strains belonged exclusively to the PanAsia-2 lineage, type A strains were from the Iran-05 lineage while the Asia 1 viruses were selected from three relevant lineages. The assays with the universal primers and probes were useful in confirming the presence of FMDV in each sample. All three serotype-specific primers/probe sets were strongly positive against RNA from homotypic viruses and no cross-reactivity was observed with heterotypic viruses except with one serotype O virus which also gave a weaker reaction with the serotype A-specific primers/probe set. The cause of this apparent non-specificity is being investigated but analysis of the sequence did not reveal a significant similarity with those of the type A-specific primers and probe. The results illustrate the potential of using rRT-PCR for the detection and discrimination of FMDVs belonging to geographically distinct sub-groups of serotypes O, A and Asia 1 from the Middle East. A similar approach could be used to develop typing assays for other geographical regions of the world. 2. INTRODUCTION Foot-and-mouth disease (FMD) is a highly contagious vesicular disease of wild and domestic cloven-hoofed animals, particularly cattle, sheep, pigs and goats. It is the most economically important viral disease of domesticated livestock throughout the world today, being endemic in many countries of Africa, Asia and South America. However, recent outbreaks have occurred in countries that are normally free of FMD including Japan and Korea (2000), France and The Netherlands (2001) and the United Kingdom (UK) (2001 and 2007). The causative agent, FMDV virus (FMDV), is a single-stranded positive sense RNA virus of around 8.4 kilobases in length, belonging to the genus Aphthovirus within the family Picornaviridae (Belsham, 1993). There are seven immunologically distinct serotypes: O, A, C, SAT 1, SAT 2, SAT 3 and Asia 1 and a diverse antigenic spectrum of virus strains within each serotype. The seven serotypes are not distributed equally around the world (Knowles and Samuel, 2003). Serotypes O and A are widely disseminated but type C viruses are rare; having recently only been found only in Brazil, Kenya and Ethiopia while Asia 1 is mainly confined to the Asian continent. The SAT (Southern African Territories) serotypes are normally, but not exclusively, restricted to sub-Saharan Africa. In countries normally free of FMD, rapid confirmation of FMDV as the causative virus in material from suspect cases is the primary goal of field and laboratory investigations. A number of panserotype specific real-time reverse transcription polymerase chain reaction (RT-PCR) assays have been developed to target highly conserved regions of the RNA genome of FMDV (Reid et al., 2002;

363

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Callahan et al., 2002; Moniwa et al., 2007) to detect all seven serotypes in clinical samples. These assays have undergone extensive evaluation: parallel testing of samples has shown the sensitivity of these molecular assays to be at least equal to that of the current gold standard method of invitro virus isolation in cell culture (Reid et al., 2003; Shaw et al., 2004; King et al., 2006). Further optimisation including the use of automated robitic equipment to prepare template RNA has simplified and speeded up the assays for routine use (Shaw et al., 2007; Reid et al., 2008, manuscript submitted for publication). A limitation of RT-PCR procedures targeting conserved sequences in non-structural and untranslated regions of the genome is that they cannot determine the serotype identity of the causative FMDV. Therefore, it is not currently possible to serotype samples that are RT-PCR positive but VI/ELISA negative. Serotyping such samples may only be accomplished by use of serotype-specific primers and probes and/or through nucleotide sequencing. In countries such as Iran where the disease is endemic, identification of the serotypes of the causative virus strains is important for vaccine selection, disease containment and for tracing the source of the outbreaks. Previous attempts have shown that it is difficult to achieve maximum sensitivity with serotype-specific primers by conventional RT-PCR to cover the genetic diversity within all FMDV serotypes (Reid et al., 1999; Reid et al., 2001). While these procedures could be used in conjunction with antigen-detection ELISA and VI to provide additional information, they were insufficiently sensitive to replace them for primary diagnosis of FMD (Reid et al., 1999). Alternative RT-PCR assays reported by Suryanarayana et al. (1999) and Alexandersen et al. (2000) for serotype-specific diagnosis were extremely cumbersome and unsuitable for routine use in an epidemic. Giridharan et al. (2005) describe a conventional RT-PCR procedure for differentiation of FMDV serotypes native to India using multiple primers based mostly on nucleotide sequences of viruses circulating in that geographical area. Whilst able to differentiate Indian FMDV serotypes O, A, C and Asia 1; these primers would be too restrictive for detection of viruses belonging to other lineages or sub-groupings within these serotypes. However, this work demonstrated the potential to develop tailored molecular tools for detection of serotypes of FMDV. Real-time RT-PCR (rRT-PCR) assays for the specific detection of FMDV serotypes O, A and Asia 1 recently circulating in the Middle East were evaluated in this study using primers/ probe sets designed from the VP1 region on clinical samples submitted to the FAO World Reference Laboratory for Foot-and-Mouth Disease, Pirbright (WRLFMD) for virological investigation and on samples held at the Central Veterinary Laboratory in Tehran. To our knowledge, this is the first reported use of serotype-specific primers/probe sets for the detection of FMDV in clinical samples by rRT-PCR. 3. MATERIALS AND METHODS Specific primers/probe sets were designed at the WRLFMD using PrimerExpress (Applied Biosystems) from VP1 sequence alignments of FMDV serotype O strains belonging exclusively to the PanAsia-2 lineage, from serotype A viruses of the Iran-05 lineage and from Asia 1 viruses from three relevant lineages. For serotypes O and A, two primer sets with the same probe for each (O1, O2 and A1, A2 respectively) were designed while four forward primers, two reverse primers and four probes were designed for detection of the Asia 1 strains (Table 1). Table 1; Sequences and serotype-specificity of the primers/probes sets Primer/probe

FMDV serotypespecificity O O O O O

Sequence (5’ – 3’)

Forward 1 Reverse 1 Forward 2 Reverse 2 A Iran-05 probe

A A A A A

CACGACCATCCACGAGCTT GCAGAGGCCTGGGACAGTAG ACGACCATCCACGAGCTYC RCAGAGGCCTGGGACAGTAG CGTGCGCATGAAACGTGCCG

Forward 1 Forward 2

Asia 1 Asia 1

GCTGTAAAGGCTGAAACCATCAC GCAGTWAAGGCYGAGAGCATYAC

Forward 1 Reverse 1 Forward 2 Reverse 2 Panacea probe

364

CCGAGACAGCGTTGGATAACA CCATACTTGCAGTTCCCGTTGT CCGAGACAGCGTTGGAYAAYA CCATACTTGCAGTTCCCGTTRT CCGACTTGCACTGCCTTACACGGC

Name of set O1a O2b O1/O2 A1c A2d A1/A2

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Forward 3 Forward 4 Reverse 1 Reverse 2 Probe 1 Probe 2 Probe 3 Probe 4

Asia Asia Asia Asia Asia Asia Asia Asia

1 1 1 1 1 1 1 1

GCAGTWAAGGCYGAGASCATYAC GCAGTWAAGGCCGAGASCATYAC GCAAAGGCCTAGGGCAGTATG GCARAGGCCTAGGGCAGTATG TTTGATTCGCATGAAACGTGCGGAG AGCTTTTGATTCGCATGAAACGTGCG AGCTGTTGATYCGCATGAAACGYGCG AGCTGTTGATCCGCATGAAACGYGCG

Forward primer 1/Reverse primer 1/PanAsia O probe. Forward primer 2/Reverse primer 2/PanAsia O probe. c Forward primer 1/Reverse primer 1/A Iran-05 probe. d Forward primer 2/Reverse primer 2/A Iran-05 probe. b

Total nucleic acid was extracted from suspensions of vesicular epithelia of serotype O PanAsia-2, serotype A Iran-05 and Asia 1 viruses from the three lineages by an automated procedure as previously described (Shaw et al., 2004). The best performing primers/probe sets (O1, A2 and the Asia 1 set: Forward 3/Reverse 2/Probe 4) were then selected following evaluation of the new primers/probes on templates of the serotype O, A and Asia1 viruses with the one-step rRT-PCR protocol described by Shaw et al. (2007). All templates were tested by these selected one-step rRT-PCR assays (incorporating the primers/probe sets intended for specific detection of serotype O, A and Asia 1 virus sequences, respectively) run in parallel with pan-serotype specific assays (5′ UTR and 3D) using the protocol of Shaw et al. (2007). To check the specificity of the primers/probe sets, total nucleic acids were similarly extracted from FMDV serotype C, SAT 1 and SAT 2 and from swine vesicular disease virus (SVDV) isolates and each were tested by the new assays. The O1, O2, A1 and A2 primers/probe sets were also used to test homo- and heterotypic FMDV O and A viruses at the Central Veterinary Laboratory in Tehran using a locally-employed rRT-PCR procedure. 4. RESULTS The assays with the universal primers and probes were useful in confirming the presence of FMDV in each sample. All three serotype-specific primers/probe sets were strongly positive against total nucleic acid from homotypic viruses and no cross-reactivity was observed with heterotypic viruses except with one serotype O virus which also gave a weaker reaction with the serotype A-specific primers/probe set (Table 2). Analysis of the sequence of the type O virus did not reveal a significant similarity with those of the type A-specific primers and probe but this result will be investigated further by repeat testing. No cross-reactivity was observed when the serotype-specific primers/probe sets were tested against the other FMDV serotypes of C, SAT 1 and SAT 2 or with SVDV isolates. Table 2: Summary of the results obtained at the WRLFMD using parallel one-step rRT-PCR assays with pan-serotypic (5’UTR and 3D) or serotype-specific primers/probe sets FMDV serotype or virus tested

Lineage

Ratio of number of samples positive to total number tested by rRT-PCR 5’UTR 3D Oa Ab Asia 1c 19/21 21/21 21/21 1/21 0/21

PanAsia-2

A Asia 1

Iran-05 3 lineages

18/18 12/12

18/18 11/12

0/18 0/12

18/18 0/12

0/18 12/12

C, SAT 1, SAT 2

Various

9/10

10/10

0/10

SVDV

Various

0/3

O1 primers/probe set. A2 primers/probes set. c Asia 1-specific primers/probe set (Forward 3/Reverse 2/Probe 4 – see Table 1). b

365

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Evaluation of serotype-specific primers/probes in Tehran indicated that the sets A1 and A2 performed better than sets O1 and O2. When four serotype O viruses were tested, three were positive with the O1 and O2 sets but weaker positive with the A1 and A2 sets – the other virus being negative with the O1 and O2 sets but positive with sets A1 and A2. Two serotype O/A viruses were positive with the A1 and A2 sets but negative with the O1 and O2 sets and 5 out of 6 serotype A viruses were positive with sets A1 and A2 (the other being positive with set A1 only and all 6 negative with sets O1 and O2). 5. DISCUSSION While pan-serotypic rRT-PCR assays for FMDV have been extensively validated for routine use, several challenges face the design of serotype-specific primers/probes for rRT-PCR. In-silico analysis can be used to attempt to find conserved sequences for primer/probe binding sites that allow the specific recognition of all viruses within a particular serotype, but do not cross-react with viruses of other serotypes. This work is made difficult by the high variability of the FMDV genome and lack of consistent sequences that are conserved within, and restricted to a particular serotype. For these reasons, efforts have concentrated on detecting viruses of particular serotypes from distinct geographic regions or lineages rather than all strains within a serotype. The results from this study have demonstrated the potential of using rRT-PCR for the detection and discrimination of FMDVs belonging to geographically distinct sub-groups of serotypes O, A and Asia 1 currently circulating in the Middle East. This is the first step in developing a suite of molecular tools for different countries and regions. Typing assays could similarly be developed for other geographical regions of the world. For this study, a simple approach was employed whereby specific primers/probe sets were designed from sequence alignments using basic PrimerExpress software. To complete this study, more strains (particularly reference strains) from the desired lineages need to be tested to further evaluate the diagnostic sensitivity of each primers/probe set and the analytical sensitivity of the assays remains to be determined. 6 AUTHOR’S CONCLUSIONS The results have demonstrated the potential of using rRT-PCR for the detection and discrimination of FMDV strains belonging to geographically distinct sub-groups of serotypes O, A and Asia 1 currently circulating in the Middle East. To our knowledge, this is the first reported use of serotype-specific primers/probe sets for the detection and discrimination of FMDV in clinical samples by rRT-PCR. The study is the first step towards the development of a suite of molecular tools for different countries and regions. 7 AUTHOR’S RECOMMENDATIONS Molecular typing assays should similarly be developed by reference laboratories for other geographical regions of the world. 8. ACKNOWLEDGEMENTS The authors thank Dr Nigel Ferris and Geoffrey Hutchings from the WRLFMD, Pirbright for characterisation and supply of viruses and colleagues from the Central Veterinary Laboratory, Tehran for their co-operation. This work was funded by Defra, UK (Project number: SE1124). 9. REFERENCES [1] Belsham, G.J. 1993. Distinctive features of foot-and-mouth disease virus, a member of the picornavirus family; aspects of virus protein synthesis, protein processing and structure. Prog. Biophys. Mol. Biol., 60: 241-260. [2] Knowles, N.J. & Samuel, A.R. 2003. Molecular epidemiology of foot-and-mouth disease virus. Virus Res., 91: 65-80. [3] Reid, S.M., Ferris, N.P., Hutchings, G.H., Zhang, Z., Belsham, G.J. & Alexandersen, S. 2002. Detection of all seven serotypes of foot-and-mouth disease virus by real-time, fluorogenic reverse transcription polymerase chain reaction assay. J. Virol. Methods, 105: 67-80. [4] Callahan, J.D., Brown, F., Osorio, F.A., Sur, J.H., Kramer, E., Long, G.W., Lubroth, J., Ellis, S.J., Shoulars, K.S., Gaffney, K.L., Rock, D.L. & Nelson, W.M. 2002. Use of a portable real-time

366

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 reverse transcriptase-polymerase chain reaction assay for rapid detection of foot-and-mouth disease virus. J. Am. Vet. Med. Assoc., 220: 1636-1642. [5] Moniwa, M., Clavijo, A., Li, M., Collignon, B. & Kitching, P.R. 2007. Performance of a footand-mouth disease virus reverse transcription-polymerase chain reaction with amplification controls between three real-time instruments. J. Vet. Diagn. Invest., 19: 9-20. [6] Reid, S.M., Grierson, S.S., Ferris, N.P., Hutchings, G.H. & Alexandersen, S. 2003. Evaluation of automated RT-PCR to accelerate the laboratory diagnosis of foot-and-mouth disease virus. J. Virol. Methods, 107: 129-139. [7] Shaw, A.E., Reid, S.M., King, D.P., Hutchings, G.H. & Ferris, N.P. 2004. Enhanced laboratory diagnosis of foot-and-mouth disease by real-time polymerase chain reaction. Rev. Sci. Tech., 23: 1003-1009. [8] King, D.P., Ferris, N.P., Shaw, A.E., Reid, S.M., Hutchings, G.H., Giuffre, A.C., Robida, J.M., Callahan, J.D., Nelson, W.M. & Beckham, T.R. 2006. Detection of foot-and-mouth virus: comparative diagnostic sensitivity of two independent real-time reverse transcription-polymerase chain reaction assays. J. Vet. Diagn. Invest. 18: 93-97. [9] Shaw, A.E., Reid, S.M., Ebert, K., Hutchings, G.H., Ferris, N.P. & King, D.P. 2007. Implementation of a one-step real-time RT-PCR protocol for diagnosis of foot-and-mouth disease. J. Virol. Methods, 143: 81-85. [10] Reid, S.M., Ebert, K., Bachanek-Bankowska, K., Batten, C., Sanders, A., Wright, C., Shaw, A.E., Ryan, E.D., Hutchings, G.H., Ferris, N.P., Paton, D.J. & King, D.P. 2008. Performance of diagnostic real-time RT-PCR during the outbreaks of foot-and-mouth disease in the United Kingdom in 2007. Manuscript submitted to Journal of Veterinary Diagnostic Investigation for publication. [11] Reid, S.M., Hutchings, G.H., Ferris, N.P. & De Clercq, K. 1999. Diagnosis of foot-andmouth disease by RT-PCR: evaluation of primers for serotypic characterisation of viral RNA in clinical samples. J. Virol. Methods, 83: 113-123. [12] Reid, S.M., Ferris, N.P., Hutchings, G.H., De Clercq, K., Newman, B.J., Knowles, N.J. & Samuel, A.R. 2001. Diagnosis of foot-and-mouth disease by RT-PCR: use of phylogenetic data to evaluate primers for the typing of viral RNA in clinical samples. Arch. Virol., 146: 2421-2434. [13] Suryanarayana, V., Madanamohan, B., Bist, P., Natarajan, C. & Tratschin, J.D. 1999. Serotyping of foot-and-mouth disease virus by antigen capture reverse transcriptase /polymerase chain reaction. J. Virol. Methods, 80: 45-52. [14] Alexandersen, S., Forsyth, M.A., Reid, S.M. & Belsham, G.J. 2000. Development of reverse transcription-PCR (oligonucleotide probing) enzyme-linked immunosorbent assays for diagnosis and preliminary typing of foot-and-mouth disease: a new system using simple and aqueous-phase hybridization. J. Clin. Microbiol. 38: 4604-4613. [15] Giridharan, P., Hemadri, D., Tosh, C., Sanyal, A. & Bandyopadhyay, S.K. 2005. Development and evaluation of a multiplex PCR for differentiation of foot-and-mouth disease virus strains native to India. J. Virol. Methods, 126: 1-11.

367

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

368

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 65

LINEAR-AFTER-THE-EXPONENTIAL (LATE) PCR: NEW DETECTION TECHNOLOGIES FOR PAN-FMDV AND SEROTYPE-SPECIFIC ASSAYS IN THE FIELD K.E. Pierce1*, R. Mistry2, S. Bharya2, S.M. Reid3, K. Ebert3, D.P. King3, L.J. Wangh1 1

Department of Biology, Brandeis University, Waltham, MA 02454, USA 2 Smiths Detection, Watford, WD23 2BW, United Kingdom. 3 Institute for Animal Health, Pirbright, GU24 0NF, United Kingdom.

INTRODUCTION Detection and serotype identification of Foot-and-mouth Disease virus (FMDV) RNA is complicated by high sequence variation between strains. Linear-After-The-Exponential (LATE)-PCR efficiently generates single-stranded DNA that is freely available to hybridize with fluorescently labelled, mismatch-tolerant probes over a wide temperature range, enabling the detection of numerous sequence variants. Two assays are being developed; a pan-FMDV assay intended for detection of all strain variants and a second to distinguish between FMDV serotypes. MATERIAL AND METHODS Limiting and excess primers were designed for the relatively conserved sequences in the FMDV 3D (RNA polymerase) gene or in the 1D (VP1 capsid protein) to 2B genes using LATE-PCR design criteria. RNA targets are pre-incubated with primers, and then amplified using a one-step RT-PCR reaction. Fluorescent signals from the probes are measured over a wide range of temperatures following RT-PCR. Control RNA sequences are co-amplified and detected with a separate probe to guard against false negatives. RESULTS Initial experiments using synthetic DNA templates demonstrated that the fluorescent intensity of the probe at end point was proportional to the initial concentration of the templates over a wide concentration range, 10 copies to 1 million copies. Subsequent testing showed that viral RNA from each of the 7 FMDV serotypes could be amplified and then detected with a single mismatchtolerant probe in the pan-FMDV assay. All samples with unrelated viruses remained negative for FMDV signals. All non-FMDV samples and other control samples without virus showed amplification of the internal RNA control. Tests with the serotype-specific assay are ongoing. A probe for the Asia 1 serotype generated a strong fluorescent signal with synthetic targets for Asia 1 variants, but no signal with an O serotype target with the most similar sequence. DISCUSSION AND CONCLUSIONS LATE-PCR with a single mismatch-tolerant probe can be used for pan-detection of FMDV. A second assay with specific probes should provide serotype characterization of the virus. These quantitative end-point assays can be used in laboratory thermal cyclers or in field instruments, such as the BioSeeqTM-Vet Portable Diagnostic Laboratory, for rapid detection of FMDV in samples from suspect animals.

369

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

370

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 66

EVALUATION OF A LATERAL FLOW DEVICE FOR THE PEN-SIDE DIAGNOSIS OF FOOTAND-MOUTH DISEASE N. P Ferris*1, A. Nordengrahn2, G. H Hutchings1, S. M Reid1, D. P King1, K. Ebert1, D. J Paton1, T. Kristersson2, E. Brocchi3, S. Grazioli3 and M. Merza2 1

BBSRC Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey GU24 0NF 2 Svanova Biotech AB, Uppsala Science Park, Glunten, S-751 83 Uppsala, Sweden 3 Istituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia Romagna Via Bianchi, 9 – Brescia, Italy

ABSTRACT A lateral flow device (LFD) for the pen-side diagnosis of FMDV has been developed using a monoclonal antibody (Mab 1F10) and evaluated in the laboratory for its sensitivity and specificity for the detection of FMDV. Suspensions of 1288 vesicular epithelia and two vesicular fluids from worldwide suspect cases of FMD were evaluated by the LFD. The collection represented 304 samples of different geographical origin and antigenic and molecular variation within each of the FMDV serotypes and 986 samples in which FMDV had not been detected. Additionally, five samples containing viruses of the other vesicular diseases of SVD and five of VS plus seven from truenegative samples were also tested. Simple sample homogenisers (a disposable pellet pestle with microtube and a plastic rod with disposable bijou) were evaluated for their potential abilities to prepare epithelial suspensions under field conditions for LFD use. The LFD detected antigens of FMDV of wide diversity of all seven serotypes but weaker reactions were often evident with viruses of type SAT 2. The diagnostic sensitivity and specificity of the device at 84% and 99%, respectively, compared favourably to 85% and 99.9% of the antigen ELISA. Both the simple sample homogenisers yielded suspensions which reacted in the LFD to a similar degree to those arising from routine pestle and mortar preparation. The results illustrate that the 1F10 LFD has the potential of providing a specific, simple, cheap and disposable FMD diagnostic test that would be capable of providing results within minutes of taking a clinical sample and which could be used on a suspect premises by a veterinarian in support of his clinical analysis. A prototype FMDV type SAT 2 specific LFD is currently being evaluated to complement the 1F10 device to decrease the likelihood of SAT 2 FMDVs going unrecognised. 1. INTRODUCTION In the event of an outbreak of FMD in a previously FMD-free country, diagnosis is dependent on the early recognition of signs of disease by the farmer and rapid reporting to the relevant veterinary authorities to enable the clinical symptoms to be evaluated. This is followed by the submission of samples to a reference laboratory for confirmatory tests for the presence of FMD virus, antigen and genome. The availability of an objective ‘point-of-care’ or ‘pen-side’ diagnostic test would have the advantage of providing support to veterinary clinical judgment in the first instance and could reduce the time taken for test confirmation in secondary cases of disease. In FMD-endemic areas, the period between reports of disease and collection and dispatch of samples to a laboratory for disease investigation can be protracted, allowing for the possibility of sample degradation, and FMDV is very often not detected in received submissions (A. Naci Bulut, personal communication). Long distances between the sample site and the laboratory can delay the diagnostic result, which can hinder the effectiveness of local actions in the face of outbreaks. On other occasions, episodes of disease might go uninvestigated, creating uncertainty as to the true FMD status of a region. In such situations, and additionally where FMDV of more than one serotype might be circulating, the availability of a test that could be performed at the site of a suspected FMD outbreak would help in the selection of appropriate material that need be sent to the laboratory for diagnostic serotyping

371

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 and further characterization such as virus sequencing and vaccine matching analyses. In other endemic regions, where the animal health status and productivity is low, FMD control is not considered a priority but the availability of a simple and inexpensive field test might increase FMD awareness and improve epidemiological information. The laboratory validation of a LFD using an FMDV pan-reactive Mab for the pen-side diagnosis of FMD is presented. 2. MATERIALS AND METHODS 2.1 Monoclonal antibody (Mab) for use in the lateral flow device (LFD) A Mab, designated 1F10 and produced at IZS, Brescia against the FMD type O virus strain UKG 31/2001 by methods as described previously (Brocchi et al., 1986), was chosen for incorporation into a LFD for the full laboratory validation study. 2.2 1F10 LFD assembly The Mab 1F10 was coupled to 40 nm colloidal gold particles using a proprietary method and the conjugate was stored at 4 oC until use. The Mab 1F10, dissolved in 0.5 M Tris buffer, pH 8.0 to a final concentration of 1.8 mg/ml, was applied to the nitrocellulose membrane (Hi-flow membrane, Millipore, USA) using Bio-Dot air-brush equipment (Bio-Dot, UK). Fifty microliters of the Mab solution were added per 30 cm of membrane. Rabbit anti-mouse antibodies (DAKO, Denmark) were applied (control band) at a concentration of 1.8 mg/ml parallel to the Mab line (test band). The membranes were then dried at 37 oC for 45 min and stored in sealed foil sachets until use. The gold conjugate was applied to the filter (Whatman, UK) using Bio-Dot air-brush equipment (BioDot, UK) at a volume of 1 µl per mm filter. The filters were dried at room temperature for 45 min and then stored in sealed foil sachets until required. The filter was overlaid onto the base of the nitrocellulose membrane, parallel to the control and the antibody bands, stuck to the membrane with adhesive and cut into 0.8 cm wide strips. The fibre/membrane strips were assembled into a device as described previously (Brüning et al., 1999) so that the filter constituted the sample pad situated above a nitrocellulose membrane strip. 2.3 Test samples Twelve hundred and eighty seven vesicular epithelia and a vesicular fluid from suspect cases of FMD from around the world that had been submitted to the FAO WRL for FMD from 1965 to 2008 (plus one other epithelium and a fluid sample examined under field conditions) were evaluated by the LFD. The samples were selected from a collection representing different geographical origins and antigenic and molecular variation within each of the FMDV serotypes (n=304) as well as others in which FMDV had not been detected (n=986; classified as ‘no virus detected’). In addition, samples containing viruses of the other vesicular diseases of SVD (n=5) and VS (n=5) and from samples collected from naïve animals (n=7) were also tested. 2. 4 Virus detection tests The majority of tests for virus detection were undertaken at the time of sample receipt and encompassed the following procedures. Virus isolation was performed using primary calf thyroid cells and a permanent cell line of IB-RS-2 cells; indirect sandwich ELISAs for FMDV, SVDV and VSV were used to characterize the specificity of the virus serotype in original material and cell culture antigens derived from them (Roeder and Le Blanc Smith, 1987; Ferris and Dawson, 1988; Ferris and Donaldson, 1988); in addition, real-time RT-PCR procedures (King et al., 2006; Shaw et al., 2007) were undertaken on samples received post-July 2002 (such assays had yet to be implemented and used routinely for diagnosis prior to this date). 2.5 Lateral flow device sample buffer and test operation The results are based on the use of a proprietary buffer of Svanova Biotech AB. Aliquots of the epithelial suspensions were mixed with an equal volume of the LFD sample buffer and 200 µl of the mixture was applied to the sample pad of the LFD. If present in the sample, FMDV binds to the gold conjugate and forms an immune complex, which migrates by capillary action along the membrane until it reaches the immobilised antibody in the ‘T’ (test) window where it is trapped resulting in the accumulation of colloid gold, which can be visualised as a red line to signify a positive result. Excess (or unbound) Mab-labelled gold particles continue to migrate along the device until being captured by the immobilised rabbit anti-mouse antibody and the formation of a red ‘C’ (control) line, to validate the test. The test (and control) lines were observed for colour development at intervals over a period up to 1h from sample addition and reactions scored subjectively from

372

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 negative to strong. A diagrammatic representation of negative and positive LFD test development is illustrated in Fig. 1 2.6 Simple sample homogenizers Two alternative devices for sample homogenization were evaluated for their potential abilities to prepare epithelial suspensions under field conditions for use in the LFD in comparison with the pestle and mortar that is employed routinely in the laboratory. These were (i) a disposable pellet pestle with microtube (Anachem) and (ii) a plastic rod (The BMC Research Workshop, Uppsala University, Uppsala, Sweden) in combination with a disposable glass bottle. Approximate 0.1 g amounts of similar pieces of each of 20 positive epithelia were ground in 1 ml volumes of the LFD sample buffer by each homogenizer. After leaving the homogenates for 2-3 min to allow the tissue to settle from the suspension, 200 µl of the supernatants were added to the LFD and reactions scored. 2.7 SAT 2 FMDV type-specific lateral flow device Three SAT 2 FMDV type-specific Mabs (designations: 2H6, 3C5 and 4A6) produced at the IZS, Brescia against the virus strain ZIM 5/81 were selected for the production of prototype LFDs for SAT 2 FMDV antigen detection for preliminary evaluation (Grazioli et al., 2006). 3. RESULTS 3.1 Sensitivity and specificity of the 1F10 lateral flow device (LFD) Table 1 shows the results observed with the 1F10 LFD in comparison with the antigen ELISA. Samples are sub-grouped into epithelia suspensions classified as either positive for one of the seven FMDV serotypes, negative (from naïve animals) or “NVD” (i.e. no virus detected by virus isolation/ELISA and/or RT-PCR). It can be seen that the overall specificity and sensitivity of the LFD is comparable to the ELISA with individual variations depending on the virus serotype – higher for FMDVs of serotypes A, C, SAT 1 and Asia 1, slightly lower for types O and SAT 3 but lower still for type SAT 2. Five epithelial suspensions containing high amounts of SVDV and VSV, respectively, (as judged by high OD values from homologous ELISAs) were all negative in the LFD. 3.2 Field sample preparation Suspensions from 20 FMDV positive epithelia were prepared by each of the three sample extraction methods (i.e. [i] conventional pestle and mortar in comparison with [ii] the Anachem pellet/pestle with microtube and [iii] plastic rod with bottle) and tested in the 1F10 LFD. Positive reactions were produced in each case, although there were slight variations in the strength of test reactions: LFD reactions arising from the pellet/pestle and rod preparations were occasionally weaker than those from the pestle and mortar; three samples were slightly stronger by pellet/pestle than rod preparation and vice versa for three others (results not shown). 3.3 Prototype lateral flow devices for FMDV type SAT 2 detection Thirty SAT 2 FMDV samples were tested in each of the three different SAT 2 Mab based prototype LFDs. All 30 reacted positively in the 2H6 and 4A6 devices, while 23 were also positive in the 3C5 LFD but with weaker reactions. For comparison, one of the 30 samples was negative by ELISA and only 19 had reacted positively in previous testing using the 1F10 LFD (and also to a weaker extent than now). Three type O, three A, two C, three SAT 1, two SAT 3, three Asia 1 and two SVDV positive samples, six negative and 18 NVD test samples failed to react in the 2H6 and 3C5 devices; conversely, some very weak non-specific reactions resulted in the 4A6 LFD. 4. DISCUSSION Speed of diagnosis is paramount in maximizing the efficiency of the control measures that are implemented to stop the spread of disease and bring about its eradication when an outbreak of FMD occurs in a previously FMD-free country. Laboratory confirmation of the disease agent and its serotype are an essential component of this process, especially for primary outbreaks of the disease, where a battery of tests is often used to give the highest possible confidence in the results obtained. In secondary outbreaks where there may be less urgent need for characterization of the strain of virus involved, delays in sample shipment to the laboratory for confirmatory diagnosis can place reliance on clinical and epidemiological considerations for urgent decisions, such as whether or not to slaughter stock. For example, in the 2001 UK FMD outbreak, all susceptible livestock had to be slaughtered on infected premises (IPs) within 24 hours of disease confirmation and those on

373

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 dangerous contacts and contiguous premises within 48 hours (Anon, 2001); the consequence being that the majority of outbreaks were ‘confirmed’ on the basis of clinical judgment before laboratory results were known. Unfortunately, clinical diagnosis of FMD can often be fraught and syndromes such as idiopathic mouth ulcers or ovine mouth and gum obscure disease (Watson, 2004), foot rot, trauma, grazing on rough pasture and agents other than FMD can be confused with it and can lead to the overreporting of disease (Ferris et al., 2006). The availability of a specific, simple, cheap and disposable test that would be capable of providing results within minutes of taking a clinical sample and which could be used on a suspect premises by a veterinarian might help mitigate this. Accordingly, the development of rapid tests for FMD diagnosis was recommended by more than one of the committees of enquiry that investigated aspects of the 2001 UK FMD outbreak (Anderson, 2002; Royal Society, 2002) The aim of this study was to develop a lateral flow device (LFD) that would meet this requirement and a pan-reactive FMDV Mab (1F10) was identified as the basis for its construction; thereafter a full laboratory validation study was undertaken. It is evident from the validation that the specificity (99%) and overall sensitivity of the 1F10 LFD (83.6%) are almost identical to those of the slower and more complicated antigen ELISA (85%) for detection of FMDV antigen. Although the LFD reacted with FMD viruses of all serotypes, it performed less well with FMDVs of the SAT 2 serotype and often weak reactions resulted, with several isolates not reacting at all. Steps are now being taken to address this possible shortcoming through the development and further validation of a SAT 2 type-specific LFD using the Mab 2H6; trials of prototype devices have produced encouraging results and suggest that the two devices would decrease the likelihood of SAT 2 FMDVs going unrecognized if used in concert. In common with the antigen ELISA, the LFD is most suited for the detection of FMDV in vesicular fluid or epithelia but it is not recommended for investigating other sample types including blood, throat swab (probang) and milk samples, which would normally be expected to contain insufficient amounts of virus for recognition. For epithelia, the sample must first be homogenized to release virus antigen from the tissue. The evaluation of two simple sample preparation devices suggested that either could be used successfully for preparing such homogenates at the pen-side but that the rod/bijou device might be more practical and easier to manipulate. A pair of scissors and forceps would be additionally useful both for sample collection in the first place and then for finely mincing the specimen to facilitate grinding. It is important to remember that unlike virus isolation or RT-PCR, the device does not have exquisite sensitivity for antigen detection (rather that it is equivalent to the antigen ELISA) and a negative result will not necessarily negate the possibility of an FMD infection. Nevertheless, the correct samples taken from animals in the acute stage of disease or soon after would normally be expected to contain sufficient amounts of virus to be detected by the 1F10 LFD. An additional advantage of its speed and potential ease of use in the field, is that if an unexpected result occurs it is very easy to repeat tests on the same or other affected animals. Experience from using the LFD in the UK 2007 outbreak also demonstrated that the LFD was extremely useful for confirmatory diagnosis in the laboratory because a result could be obtained within an hour of a samples arrival, compared to four hours for ELISA or RT-PCR. The results of this study indicate that the 1F10 LFD has the potential for use in the pen-side next to an animal and for reassuring the veterinarian on the accuracy of his FMD diagnosis from clinical examination. Furthermore, the use of the LFD in FMD-endemic countries may also prove fruitful in, for example, overcoming problems associated with delays in submission of samples to reference laboratories, poor communication and under reporting of disease. The device is specific, sensitive, rapid to use and disposable. In order to incorporate use of FMDV detecting LFDs into procedures for the diagnosis of FMD, consideration will have to be given as to who should receive and use the devices and for the requirement for continued submission of samples to central laboratories for further characterization (e.g. sequencing) and for confirmation of LFD diagnoses. 5. ACKNOWLEDGEMENTS

374

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 This work was supported financially by the Department for the Environment, Food and Rural Affairs (DEFRA; project numbers SE1120 and SE1123) and the EU project Lab-on-site (SSPE-CT-2004-513 645).

6. REFERENCES [1] Anderson, I. 2002. Foot-and-mouth Disease 2001: Lessons to be Learned Inquiry Report. London, The Stationery Office. [2] Anon. 2001. EI 2001/69 Emergency instruction 2001/69/VEXDT, Ministry of Agriculture, Fisheries and Food/Scottish Executive Rural Affairs Department/National Assembly for Wales Agriculture Department. London, UK, March 26, 2001. [3] Brocchi, E., Capucci, L., De Simone, F. & Panina, G.F. 1986. Potential of monoclonal antibodies (Mabs) for FMD diagnosis and characterisation of the isolates. Report of the Session of the Research Group of the Standing Technical Committee for the Control of Foot-and-Mouth Disease, Madrid, Spain, 14-17 October, 1986. Appendix 5: 30-31. [4] Brüning, A, Bellamy, K., Talbot, D. & Anderson, J. 1999. A rapid chromatographic strip test for the pen-side diagnosis of rinderpest virus. J. Virol. Methods 81: 143-154. [5] Ferris, N.P. & Dawson, M. 1988. Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Vet. Microbiol. 16: 201-209. [6] Ferris, N.P. & Donaldson, A.I. 1988. An enzyme-linked immunosorbent assay for the detection of vesicular stomatitis virus antigen. Vet. Microbiol. 18, 243-258. [7] Ferris, N.P., King, D.P., Reid, S.M., Shaw & A.E., Hutchings, G.H. 2006. Comparisons of original laboratory results and retrospective analysis by real-time reverse transcriptase-PCR of virological samples collected from confirmed cases of foot-and-mouth disease in the UK in 2001. Vet. Rec. 159: 373-378. [8] Grazioli, S., Moretti, M., Barbieri, I., Crosatti, M. & Brocchi, E. 2006. Use of monoclonal antibodies to identify and map new antigenic determinants involved in neutralization on FMD viruses type SAT1 and SAT 2. Report of the Session of the Research Group of the Standing Technical Committee for the Control of Foot-and-Mouth Disease, Paphos, Cyprus, 17-20 October, 2006. Appendix 43: 289-300. [9] King, D.P., Ferris, N.P., Shaw, A.E., Reid, S.M., Hutchings, G.H., Giuffre, A.C., Robida, J.M., Callahan, J.D., Nelson, W.M. & Beckham, T.R. 2006. Detection of foot-and-mouth disease virus: comparative diagnostic sensitivity of two independent real-time reverse transcriptionpolymerase chain reaction assays. J. Vet. Diag. Invest. 18: 93-97. [10] Roeder, P.L. & Le Blanc Smith, P.M. 1987. Detection and typing of foot-and-mouth disease virus by enzyme-linked immunosorbent assay: a sensitive, rapid and reliable technique for primary diagnosis. Res. Vet. Sci. 43: 225-232. [11] Royal Society 2002. Infectious Diseases in Livestock. London, Royal Society. [12] Shaw, A.E., Reid, S.M., Ebert, K., Hutchings, G.H., Ferris, N.P. & King, D.P. 2007. Protocol: Implementation of a one-step real-time RT-PCR protocol for diagnosis of foot-and-mouth disease. J. Virol. Methods 143: 81-85. [13] Watson, P. 2004. Differential diagnosis of oral lesions and FMD in sheep. In Practice 26: 182197.

375

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig.1: Diagrammatic depiction of negative and positive results produced in the 1F10 lateral flow device. Table 1: Overall sensitivity and specificity of the 1F10 lateral flow device (LFD) using LFD sample buffers A and B with positive and negative epithelial suspensions in comparison with the FMDV antigen ELISA Positive samples FMDV 1F10 LFD serotype No. tested O 131 A 41 C 24 SAT 1 24 SAT 2 34 SAT 3 10 Asia 1 40 Total 304

No. positive 121 36 15 16 20 7 39 254

Sensitivity (%) 92.4 87.8 62.5 66.7 58.8 70.0 97.5 83.6

0 0 0 1 4 5

Specificity (%) 100 100 100 99.9 98.4 99.5

Negative samples Sample SVDVa VSVb Negativec NVDd NVDe Total a

5 5 7 735 251 1003

ELISA No. tested 128 41 24 24 34 10 40 301

5 5 7 725 251 993

No. positive 122 32 14 13 30 9 36 256

Sensitivity (%) 95.3 78.0 58.3 54.2 88.2 90.0 90.0 85.0

0 0 0 1 0 1

Specificity (%) 100 100 100 99.9 100 99.9

SVDV, swine vesicular disease virus epithelial suspensions positive by SVDV antigen ELISA VSV, vesicular stomatitis virus epithelial suspensions positive by VSV antigen ELISA c negative epithelia from naïve animals d NVD, no virus detected by virus isolation and antigen ELISA e NVD, no virus detected by virus isolation, antigen ELISA or RT-PCR b

376

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 67

DETECTION OF PERSISTENTLY FOOT-AND-MOUTH DISEASE INFECTED CATTLE BY SALIVARY IGA TEST

J. K Biswal*, D. Paton, G. Taylor and S. Parida Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey, GU240NF, UK.

Introduction The new European council Directive 2003/85/EC on FMD has made provision for vaccination and the use of post-vaccination serosurveillance to detect sub-clinical infection. Carrier state can develop following clinical or sub-clinical disease, even in vaccinated ruminants exposed to live virus. Since carriers may be considered a risk for transmitting infection, they must be identified by post-vaccination serosurveillance to substantiate freedom from infection and to regain the FMDfree status for the purpose of international trade. Therefore the main aim of this study was to develop and validate an indirect IgA ELISA for the detection of persistently FMDV infected cattle. Materials and methods 875 naïve saliva and serial saliva samples collected weekly from five vaccine challenge experimental cattle (n=106) were used to develop and validate the assay. The assay was performed as per Parida et al., (2006) with an addition of a heterologous antigen control to increase the specificity.. The individual sample O.D value was normalised with respect to an inhouse positive control as percentage of positivity. Standardisation of cut-off value for IgA ELISA was performed by using two-graph receiver operating characteristic (TG-ROC) analysis. Results of different virological and serological assays were used for analysis of IgA results. Results Using the standardised cut-off value of 40 percentage of positivity, a specificity of 98.74% and a sensitivity of 89.18 % was found for IgA assay. The anti-FMD IgA response was significantly higher (P<0.05) in carrier animals as compared to non-carriers as a result of local replication of virus in the oro-pharynx. Conclusion The detection of salivary mucosal IgA has potential to work as screening or confirmatory DIVA test to identify subclinical infection in vaccinated population. INTRODUCTION The new European council Directive 2003/85/EC on FMD and the OIE Terrestrial Animal Health Code (OIE, 2004) have made the provision for emergency vaccination and the use of post vaccination serosurveillance to detect sub-clinical infection. The European Directive on FMD control specifies that serosurveillance should be carried out at least one month after an outbreak has finished or one month after the last use of vaccine, whichever is the later. Further it states that entire vaccinated population should be sampled and tested, or enough should be sampled and tested to give 95% confidence to detect a within-herd prevalence of infection of 5%. Countries using the above approach can regain their FMD-free status after six months of the last reported infection (OIE, 2004).To demonstrate the absence of infection in vaccinated population, serological surveillance need to be based on the detection of antibodies to the non-structural proteins of FMDV as vaccination elicits antibodies against structural proteins whereas infection elicits antibodies against both structural and non-structural proteins. The sensitivities and specificities of currently available commercial and in-house NSP tests were compared at an international workshop in Brescia, Italy (Brocchi et al., 2006). The specificity of the tests ranged between 97 %-98 % whereas the sensitivity to detect the viral carriers in vaccinated and subsequently infected cattle range from 68-94%, particularly 86.4% in Cedi-NS test (Brocchi et al., 2006). Thus if, NSP serology were to be used in large scale post-outbreak serosurveillence to identify potential carriers, a certain level of inconclusive false positive and false negative results are inevitable. Therefore it is

377

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 a requirement to develop new NSP tests or alternative tests to NSP which can be used as screening or confirmatory tests to the existing NSP tests. An indirect IgA-ELISA, using saliva samples, to detect FMD carrier cattle following vaccination and challenge exposure has been developed earlier (Parida et al., 2006). Although this IgA-ELISA has a fairly good sensitivity for detecting carrier cattle in vaccinated populations, the non-specific reaction to naïve saliva is a major problem. Here we report, on the further development and validation of IgA-ELISA as confirmatory or screening DIVA test. MATERIALS AND METHODS Saliva samples were collected and processed as described before (Parida et al., 2006). In order to determine the cut-off value and specificity of the mucosal IgA detection test, 875 saliva samples had been collected from FMD free cattle, from the Compton dairy farm, UK and from Republic of Ireland. Saliva samples were also collected at different time-points from four vaccine challenge experiments each consisting of 25 Holestein-Friesian cattle (Steers), aged 4-8 month, carried out in biosecurity containment at the world reference laboratory (WRL) for FMD, Pirbright, U.K. Each experiment involved vaccination of cattle with oil adjuvant O Manisa vaccine and subsequent challenge by contact with donor cattle that had been previously inoculated in the tongue with O UKG 2001 (Cox et al., 2007; Parida et al., 2006). Further, to study the effect of multiply vaccination on the salivary anti-FMDV IgA level in cattle, saliva samples were also collected from one multiply vaccinated-challenge experiment in cattle. In this experiment six calves had been inoculated with (1X) O1 Manisa vaccine (18PD50) 3 times at 21 day intervals and then challenged on the 35th day after the 3rd vaccination by a contact challenge. Known virological and serological results that had been obtained prior to this study were used for analysis of current IgA-ELISA findings. In order to increase the specificity of the IgA-ELISA, a heterologous FMDV antigen was taken. This negative antigen control was added during the antigen addition step of indirect IgA-ELISA. The optical density (O.D) values of negative antigen control wells were subtracted from the O.D values of positive antigen wells, in order to calculate the corrected O.D values. Later on individual sample O.D values were normalized with respect to an in-house positive control as percentage of positivity (PP). The test specificities and sensitivities were plotted against different cut-off values (Fig. 1), and a suitable cut-off value was selected as the intersection point of the two graphs based on the principle of “two-graph receiver operator characteristic” (TG-ROC) analysis (Greiner, 1995). The student’s t-test was carried out to compare the anti-FMDV IgA responses between vaccinated carriers and noncarriers.

RESULTS Principal outcomes from the four vaccine challenge experiments in cattle including clinical protection and laboratory findings were described else where (Parida et al., 2005, 2006; Cox et al., 2007). From these four vaccine challenge experiments 37 cattle (vaccinated, n=32 and unvaccinated, n=5) had been detected as FMD virus carrier by both virus isolation (VI) and RT-PCR tests beyond 28 days post-challenge. In the current IgA assay 33 out of 37 virogically detected carriers were found positive. The point of intersection of the two graphs in TG-ROC analysis was found in the range of 30-40 PP (Fig.1), which gave equal weightage to both sensitivity (Se) and specificity (Sp). Keeping the cutoff value as 40 PP, a specificity of 98.74% and sensitivity of 89.18% was obtained in IgA assay.

378

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Sp, Sn

120 100 80 60 40 20 0

Cut-off values Sp

Fig. 1: Two graph curves (specificity and sensitivity) for determination of suitable cut-off values. The point of intersections of the two graphs indicates the suitable cut-off value in the range of 3040 PP. From the four vaccine challenge experiments 48 out of 80 vaccinated cattle were detected as noncarriers by virological tests. The mean anti-FMDV IgA antibody response (Fig. 2.A) in the saliva samples of these 48 vaccinated non-carrier animals was below the cut-off value throughout the sampling period whereas the mean anti-FMD IgA response in vaccinated carrier animals was remained above the cut-off value (Fig. 2.B) and a significant difference (P<0.05) was observed.. (A)

(B)

80 40

161 DPC

140 DPC

112 DPC

Days post vaccination/challenge

Days post vaccintion/challenge

154 DPC

131 DPC

112 DPC

84 DPC

63 DPC

42 DPC

100 80 60 40 20 0 0 DPV

91 DPC

77 DPC

63 DPC

49 DPC

35 DPC

21 DPC

7 DPC

14 DPV

0DPV

PP Values

100 80 60 40 20 0

21 DPC

(D)

0 DPC

(C)

PP Values

84 DPC

0 DPV

Days post vaccination/challenge

63 DPC

20 0

161 DPC

140 DPC

112 DPC

84 DPC

63 DPC

42 DPC

21 DPC

0 DPV

0 DPC

42 DPC

21 DPC

0 DPC

PP Value

100

PP value

100

Days post vaccination/challenge

Fig. 2: Mean anti-FMDV IgA antibody response in the saliva samples: (A) Vaccinated non-carrier animals, (B) vaccinated carrier animals, (C) unvaccinated carrier animals and (D) unvaccinated non-carrier animals. Out of 20 unvaccinated challenged cattle, five were detected as carriers by both virological and by IgA tests. The mean anti-FMDV IgA response in saliva samples of these 5 unvaccinated carriers cattle was found above the cut off value on or after 35 days of post-challenge (Fig.2.C). The mean IgA antibody response against FMDV in the saliva samples of 15 unvaccinated and subsequently challenged non-carrier animals was below the cut-off value (40 PP) throughout the sample collection period (Fig.2.D). All the 20 unvaccinated animals (5 carriers and 15 non-carriers) were scored positive by Cedi NSP test.

379

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Multiply vaccinated animals were clinically protected and none of them scored as carriers after 28 dpc by VI or RT-PCR tests. The IgA antibody responses in saliva samples of five out of six multiply vaccinated animals remained below the threshold value during post-vaccination period (Fig.3.A). However, one animal was seen marginally positive in IgA assay for one day immediately after 2nd vaccination and remained negative even after the 3rd vaccination. Out of six multiply vaccinated cattle, two were scored positive for anti-FMD NSP antibody (Cedi test) after 3rd vaccination.

(Fig.3.B). (B)

70 60 50 40 30 20 10 0

PI Values

PP Values

(A) 70 60 50 40 30 20 10 0

Days post vaccination VC14 VC17

VC15 VC18

Days post vaccintion VC16 VC19

VC14 VC17

VC15 VC18

VC16 VC19

Fig.3. Antibody response in multiply vaccinated animals: (A) Anti-FMDV IgA antibody in saliva; (B) Anti-NSP antibody response in serum. DISCUSSION: In the process of development and validation of the IgA-ELISA, efforts had been made to increase the specificity of the IgA assay. Non-specificity in the old assay format may be due to the high content of detached cells, proteases and tissue particles in the saliva samples, which often contribute to the high back ground reaction by non-specifically binding to the test antigen. The use of tissue cultured inactivated crude antigen or polyclonal FMD specific rabbit hyper immune sera as a trapping antibody may be other reasons for the less specificity obtained in the past assay. Thus, in order to have a good specificity, a heterologous negative antigen control was included in the new test system. Parida et al., (2006), after analysing the saliva samples from 173 naïve cattle, suggested a cut-off optical density (O.D) value of 0.6, resulting in a test specificity of 99%. When more samples were collected from naïve cattle from the Compton farm, IAH, UK and from the Republic of Ireland 10-20% nonspecificity was observed in the old assay format (unpublished results). In the current study, 875 normal saliva samples from Compton farm, IAH, UK and from the Republic of Ireland are analysed in the new format of IgA test and a test specificity of 98.74 % was found at a cut-off value of 40 PP. In the current IgA assay 33 out of 37 virogically detected carriers were found positive, which resulted in a sensitivity of 89.18 %. However, all the 37 carrier animals were not constantly scored positive either by VI or RT-PCR or by combining the results of both the tests as seen some times by IgA ELISA. The achieved level of sensitivity (89.18%) and specificity (98.74%) are fairly good for the detection of infection in a vaccinated herd as seen in NSP tests. However when screening of individual vaccinated animal is to be considered for surveillance the amount of sensitivity found in the IgA assay may not be sufficient to detect all the carrier animals in which case the IgA test may be used as a confirmatory test to the other established screening tests. The mean anti-FMDV IgA response in the vaccinated carrier animals was above the cut-off value after 28 days and a significant difference (P <0.05, t-test) was observed between carriers and noncarriers for anti-FMDV IgA response. This IgA response should be viewed as genuine FMD specific

380

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 antibody response sustained by resident antibody secreting B-cells, rather than plasma transudation (Archetti et al., 1995). Therefore, the persisting virus in carrier animals stimulated the local immune response for the production of anti-FMDV IgA antibody which supports the suggestion that high level of anti-FMDV specific IgA is an indicator of oro-pharyngeal replication (Parida et al., 2006). It has been well established that inactivated FMD vaccine when administered parenterally stimulates very little or no FMDV specific mucosal immune response (Francis et al., 1983; Parida et al., 2006). However, an immediate question arises about the effect of repeated vaccination on the mucosal anti-FMDV IgA response, particularly in the endemic countries where bi-annual prophylactic vaccination is carried out. In order to address this question, the opportunity was taken to test saliva samples for the detection of anti-FMDV specific antibody in multiply vaccinated (3 times emergency vaccination with 21 days intervals) animals. The results indicate that IgA essay should not suffer from non-specificity when multiple vaccinations are in use in endemic countries as seen from our multiple vaccination experiment. Cedi-NSP was found to be the most suitable DIVA test (Brocchi et al., 2006) as it is commercially available and detects 86.4% NS seropositive animals with 98% specificity. In the four vaccine challenge experiments described in this study, Cedi test and IgA ELISA detected 34 and 33 carriers respectively out of 37 total carriers. However, Cedi test also detected all unvaccinated non-carrier animals (n=15). Thus NSP tests can not entirely be used for the detection of carriers, as NSP antibody response in an animal may be due to replication of virus either in acute or persistent phase of infection and animal found sero-positive in NSP tests may or may not be actually virus carriers. Further, the analysis of serum samples originated from multiply vaccinated animals showed an increased humoral immune response to the anti-FMDV NSP antibody following 3rd vaccination in two animals. The detection of NSP antibody in multiply vaccinated animals prior to challenge may be due to the presence of trace amounts of contaminating NSP proteins in the commercial FMD vaccine and dependent on antigen pay load of the vaccine and the frequency of vaccination. CONCLUSION

Levels of FMDV specific IgA become elevated during acute phase of infection and were stronger in FMDV carrier animals, irrespective of vaccination status. The IgA essay should not suffer non-specificity from multiple vaccinations in endemic countries.

RECOMMENDATION

The salivary mucosal IgA detection test may be considered as a potential DIVA test for the detection of persistently infected animals after application of vaccinate-to-live policy as screening or confirmatory test.

ACKNOWLEDGEMENT Authors acknowledge Nigel Ferris for providing rabbit polyclonal anti-sera against FMDV O1 Manisa. Thanks are also due to Dr Sarah Cox and Dr Paul Barnet for allowing SP to collect various samples from their ongoing experiments at Pirbright. JB is a recipient of Commonwealth Scholarship-INCS2007-155. This work has also been funded by the UK Department for Environment Food and Rural Affairs through grant SE 1122 and FMD Improcon project of the EU 6th Framework Programme, SSPE-CT-2003-503603. DP and GT are Jenner fellows and SP is an adjunct professor to Murdoch University, Australia. REFERENCES [1] Archetti, I. L., Amadori, M., Donn, A., Salt, J. & Lodetti, E. (1995). Detection of foot-andmouth disease virus-infected cattle by assessment of antibody response in oropharyngeal fluids. J Clin Microbiol 33: 79-84. [2] Brocchi, E., Bergmann, I. E., Dekker, A., Paton, D. J., Sammin, D. J., Greiner, M., Grazioli, S., De Simone, F., Yadin, H., Haas, B., Bulut, N., Malirat, V., Neitzert, E., Goris, N., Parida, S., Sorensen, K. & De Clercq, K. 2006. Comparative evaluation of six ELISAs for the detection of antibodies to the non-structural proteins of foot-and-mouth disease virus. Vaccine 24: 6966-6979.

381

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [3] Cox, S. J., Parida, S., Voyce, C., Reid, S. M., Hamblin, P. A., Hutchings, G., Paton, D. J. & Barnett, P. V. 2007. Further evaluation of higher potency vaccines for early protection of cattle against FMDV direct contact challenge. Vaccine 25: 7687-7695. [4] Francis, M. J., Ouldridge, E. J. & Black, L. 1983. Antibody response in bovine pharyngeal fluid following foot-and-mouth disease vaccination and, or, exposure to live virus. Res Vet Sci 35: 206-210. [5] Greiner, M. 1995. Two-graph receiver operating characteristic (TG-ROC): a Microsoft-EXCEL template for the selection of cut-off values in diagnostic tests. J Immunol Methods 185: 145-146. [6] OIE 2004. Terrestrial animals health code. 13th ed. 75017 paris, France: 12 rue de Prony; 2004. [7] Parida, S., Cox, S. J., Reid, S. M., Hamblin, P., Barnett, P. V., Inoue, T., Anderson, J. & Paton, D. J. (2005). The application of new techniques to the improved detection of persistently infected cattle after vaccination and contact exposure to foot-and-mouth disease. Vaccine 23: 5186-5195. [8] Parida, S., Anderson, J., Cox, S. J., Barnett, P. V. & Paton, D. J. 2006. Secretory IgA as an indicator of oro-pharyngeal foot-and-mouth disease virus replication and as a tool for post vaccination surveillance. Vaccine 24: 1107-1116.

382

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 68

DEVELOPMENT AND EVALUATION OF IGM ELISA FOR THE DETECTION OF FMDV SPECIFIC IGM ANTIBODIES IN BOVINE AND OVINE SERA U. Waheed1&2, D. Gibson1, D. J Paton1, Q. M Khan2 and S. Parida1* 1 2

Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey,GU24 0NF, UK., National Institute for Biotechnology and Genetic Engineering (NIBGE), Jhang Road, 38000, Faisalabad, Pakistan.

Introduction FMDV infected cattle and pigs usually develop obvious signs of the disease but in sheep and goats diagnosis is more difficult because the manifestation of the disease is often mild. Therefore, an easy laboratory diagnostic assay is necessary to detect disease in early phase of outbreak. It is well known that the first serum neutralisation antibodies are IgM that appears within 7 days following infection or vaccination. Therefore the main aim of this study was to develop and evaluate IgM assay for detection of early FMDV infection and to find out whether this assay can be used as a DIVA test. Materials and methods An indirect ELISA was developed for the detection of FMDV specific IgM antibodies in bovine and ovine sera from FMDV vaccinated and subsequently infected experimental animals (cattle=50 and sheep=36) as well as from field outbreaks animals (n=521) with and without previous vaccination status. Results 234 naïve cattle and 36 naïve sheep serum samples were used to find out the normal frequency distribution and 100% specificity of the assay is obtained at a cut off of 0.3 OD. High level of IgM antibody was evident on 7th day of post-infection in unvaccinated animals. The vaccinated animals revealed high titre of IgM antibody starting from 7 days of post-vaccination up to end of second week and in some case up to end of 3rd week of post-vaccination. Vaccinated and subsequently contact challenged transiently infected animals were seen positive in IgM assay from 7 to 21 days of post-challenge and if the animal acquired a carrier status the IgM antibody level increased further. Sera obtained from early phase of outbreaks were found positive whereas sera from late phase of outbreak were found mostly negative in IgM assay. Conclusions The IgM test has potential to identify early infection in field outbreaks, particularly in unvaccinated populations of sheep where lesions are very difficult to find out. Though conventional FMD vaccine induces IgM antibody in early stage of vaccination and the antibody does not persist long and only reappears if the animal receives FMDV infection, the test has potential to identify infection in vaccinated population as a DIVA screening or confirmatory test. 1. INTRODUCTION: During and following the 2001 FMD outbreak in the UK and the Netherlands, there has been a growing demand for vaccination as an alternative to large-scale slaughter for the control of FMD. To demonstrate the absence of infection in vaccinated population, serological surveillance needs to be based on the detection of antibodies to the non-structural proteins of FMDV. ELISAs that measure antibodies to FMDV NSPs can be used for differentiating infection in vaccinated animals (DIVA) as purified FMDV vaccine elicits antibody against only structural proteins whereas natural infection elicits antibodies against both structural and non-structural proteins. Available NSP antibody tests were validated in Brescia, Italy under the scope of an EU funded International research group, a consortium of European reference laboratories (Brocchi et al., 2006). The

383

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 specificity of the tests ranged between 97 to 98% whereas the sensitivity of the tests to detect viral carriers in vaccinated and subsequently infected cattle ranged from 68 to 94% depending on the test used. The workshop concluded that two tests should be carried out simultaneously to increase the specificity up to 99.99% .Therefore, there is a need to develop further serological test which can work as a confirmatory test to existing NSP antibody test. Further, FMDV infected cattle and pigs usually develop obvious signs of the disease but in sheep and goats diagnosis is more difficult because the manifestation of the disease is often mild. Therefore, an easy laboratory diagnostic assay is necessary to detect disease in early phase of outbreak. It is well known that the first serum neutralisation antibodies are IgM that appears within first week following infection or vaccination (Collen, T. 1994). Therefore the main aim of this study was to develop and evaluate IgM ELISA for detection of early FMDV infection and to find out whether this assay can be used as a DIVA test. 2. MATERIALS AND METHODS IGM ELISA An indirect ELISA for the detection of IgM antibodies to structural proteins was developed. Semipurified, concentrated, inactivated FMDV vaccine antigen in blocking buffer was added to alternate columns of 96 well ELISA plates (Nunc™, Denmark) that had been pre-coated with a polyclonal rabbit anti-FMDV antibody, as in the IgA ELISA [Parida et al., 2006]. After washing, 50 µl of diluted test serum was added in blocking buffer in the plates and then incubated for one hour at 370C. After a further wash step, specific bovine IgM was detected using a polyclonal rabbit anti-bovine IgM HRPO conjugate. After a final wash the test was developed by the addition of substrate. The reaction was stopped after suitable colour development by the addition of 1M sulphuric acid and the plates read on a multi-channel spectrophotometer at 490nm (A490). ELISA results were expressed as optical density values after subtracting the OD of without antigen well from OD of antigen positive well. Test serum samples Serum samples were collected from vaccinated challenged cattle (Cox et al., 2005 & 2006; Parida et al., 2006) and sheep (Parida et al., 2008) experiments conducted at Pirbright high containment isolation facility. Serum samples from 2001 UK outbreak were also used in the assay. 3. RESULTS AND DISCUSSION Serum from 234 naïve cattle and 36 sheep were used to find out the normal frequency distribution (Fig. 1) and at a cut off of 0.3 OD value 100% specificity was obtained.

Frequency Distribution 120

Frequency

100 80 60 40 20 0 0

≥0<.1

≥.1<.2

≥.2<.3

≥.3<.4

≥.4<.5

≥.5<.6

optical density Fig 1: Normal frequency distribution of negative sera originated from cattle and sheep.

384

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 High level of IgM antibody was observed during first week of post-infection which stayed up to 3rd week post-infection in unvaccinated sheep and cattle (Fig 2).

B Control animals exp I

Controls (Sheep Exp.)

UV22 1.2

UV23

UV24

0.8

UV25

0.6

UV26

VR49

VR50 VR51

0.5

VR52 0

OD Values

1.5

0.4 0.2 0

VR53 7 DPI

10 DPI

14 DPI

21 DPI

28 DPI

35 DPI

40 DPI

VR54

-10 DPI

VR55

Days

OD Value

VR 48

P 7 V D PV 2 dp 10 c dp 16 c dp 35 c dp 56 c dp 77 c dp 98 c d 11 p c 9 d 14 pc 0 dp 16 c 1 dp c

Days

Fig 2: IgM antibody titre remains high in unvaccinated infected sheep (A) and cattle (B) postchallenge. The vaccinated animals revealed high titre of IgM antibody starting from 7 days of post-vaccination up to end of second week and after challenge if the cattle acquired a carrier status the IgM antibody level increased further. All the 9 carriers detected by virus isolation and real-time RT-PCR were detected by IgM assay where as IgA and Cedi NSP test could detect only 7 carriers (Parida et al., 2005 and 2006). However, one carrier, UV 2 which was detected only twice by RT-PCR after 28 days post-challenge was undetected by Cedi test, but detected by IgM test at some time points. Therefore IgM test may have potential as a confirmatory test to the NSP test. Carrier Animals Vaccine I 1.4 1.2

UV2 UV14

1 OD Values

UV5 UV10

0.8

UV13 0.6

UV11 UV19

0.4

UV9 UV17

0.2

V dp c 7 dp 12 c dp 16 c dp 28 c dp 42 c dp 56 c dp 70 c dp 84 c dp 98 c dp 11 c 2 d 12 pc 6 dp 14 c 0 d 15 pc 4 dp 16 c 8 dp c 2

Days

Fig. 3: High IgM antibody titre observed in vaccinated cattle during 2nd week of vaccination and after challenge if the animals become FMDV carrier. The serum samples were also analysed from cattle vaccinated 3 times with 21dys intervals. IgM antibody was observed in all the 6 multiply vaccinated animals after 4 days of first vaccination and the level of antibody continued to stay up to end of second week of vaccination. There was no elevated IgM response observed after the second vaccination though 2 out of six cattle showed high IgM antibodies after 3rd vaccination. However 5 out of 6 vaccinated cattle scored positive in IgM assay after challenge though all the 6 animals were free from clinical lesions and persistent infection.

385

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 521 serum samples of ovine origin submitted in the early stage of 2001 outbreak were analysed in the IgM assay. These samples were originated from in and around the farms where clinical lesions were seen. 16 samples were scored positive in IgM test. Another 147 serum samples from sheep from 2001 outbreak those were submitted after the outbreak during serosurveillance period were found negative in the IgM assay. This results correlates well with the results obtained from experimental samples that IgM detect infection in early days of infection. 4. CONCLUSION: Levels of FMDV specific IgM antibodies become elevated after vaccination, during acute phase of infection and continue to be higher in FMDV carrier animals. Therefore IgM assay may be a useful tool to detect disease in early phase of outbreak, especially in sheep. 5. RECOMMENDATION:

The IgM detection test may have potential to work as DIVA test for the detection of infection in vaccinated population.

6. ACKNOWLEDGEMENT Authors acknowledge Nigel Ferris for providing rabbit polyclonal anti-sera against FMDV O1 Manisa. Thanks are also due to Dr Sarah Cox and Dr Paul Barnet for allowing SP to collect various samples from their ongoing experiments at Pirbright. UW is recipient of International Research Support Initiative Program (IRSIP) fellowship awarded by Higher Education Commission, Pakistan to work at Pirbright for 6 months. This work has also been funded by the UK Department for Environment Food and Rural Affairs through grant SE 1122 and FMD Improcon project of the EU 6th Framework Programme, SSPE-CT-2003-503603. DP is a Jenner fellow and SP is an adjunct professor to Murdoch University, Australia. 7. REFERENCES [1] Brocchi, E., Bergmann, I. E., Dekker, A., Paton, D. J., Sammin, D. J., Greiner, M., Grazioli, S., De Simone, F., Yadin, H., Haas, B., Bulut, N., Malirat, V., Neitzert, E., Goris, N., Parida, S., Sorensen, K. & De Clercq, K. 2006. Comparative evaluation of six ELISAs for the detection of antibodies to the non-structural proteins of foot-and-mouth disease virus. Vaccine 24: 6966-6979. [1] Collen, T. (1994) Foot-and-mouth disease Virus; viral T cell epitopes.In cell-mediated immunity in ruminants, PP173-197. Edited by B.M.Goddeeris7 W. I. Morrision.CRC press Inc. [1] Cox S.J, Voyce C, Parida S, Reid S.M, Hamblin P. A, Paton D.J and Barnett P.V.(2005) Protection against direct contact challenge following emergency FMD vaccination of cattle and the effect on virus excretion from the oropharynx. Vaccine, 23: 1106-1113 [1] Cox S J , Voyce C, Parida S, Reid S.M, Hamblin P. A, Paton D.J and Barnett P.V. (2006). Effect of antigen payload on protection, sub-clinical infection and persistence following direct contact challenge, Vaccine 24:3184-3190. [1] Parida, S., Cox, S. J., Reid, S. M., Hamblin, P., Barnett, P. V., Inoue, T., Anderson, J. & Paton, D. J. (2005). The application of new techniques to the improved detection of persistently infected cattle after vaccination and contact exposure to foot-and-mouth disease. Vaccine 23: 5186-5195. [1] Parida, S., Anderson, J., Cox, S. J., Barnett, P. V. & Paton, D. J. 2006. Secretory IgA as an indicator of oro-pharyngeal foot-and-mouth disease virus replication and as a tool for post vaccination surveillance. Vaccine 24: 1107-1116 [1] Parida S , Fleming L, Oh Y, Mahapatra M, Hamblin P, Gloster J and Paton D J. (2008)Emergency vaccination of sheep against foot-and-mouth disease: significance and detection of subsequent sub-clinical infection. Vaccine 26, 3469-3479.

386

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 69

ENHANCED PROSPECTS FOR FMDV ANTI-VIRALS TARGETED TO THE 3C PROTEASE S. Curry1*, P. Zunszain1, T. Sweeney1, N. Roqué-Rosell2, S. Knox2, A. Jaulent2 and R. Leatherbarrow2 1

Biophysics Section, Blackett Laboratory, Imperial College, London SW7 2AZ 2 Department of Chemistry, Imperial College, London SW7 2AZ

ABSTRACT Introduction In infected cells the RNA genome of FMDV is translated as a single polypeptide precursor that must be cleaved into functional proteins by virally-encoded proteases. Most of these cleavages are performed by the highly-conserved 3C protease (3Cpro), making the enzyme an attractive target for antiviral drugs. Such drugs could provide a valuable prophylactic weapon during FMD epidemics in previously disease-free regions by reducing transmission in the period before vaccination takes effect. Materials and methods We have developed a soluble, active, recombinant form of FMDV 3Cpro that may be expressed at high levels in E. coli and used this material to investigate the structure and specificity of the protease with X-ray crystallography and peptide cleavage assays. Results The crystallographic analysis of FMDV 3Cpro (including the first determination of the structure of the complex of 3C with a peptide substrate) provides us with a detailed understanding of the protease structure that will be invaluable for inhibitor design. Peptide cleavage assays showed that the recognition sequence spans at least four residues either side of the scissile bond (P4-P4´), a result that is consistent with our structural analyses. We have also developed a fluorescent peptide substrate that can be used in high-throughput cleavage assays. Collectively these results establish a valuable framework for the development of FMDV 3Cpro inhibitors. I will briefly review our recent results and discuss the challenges ahead for the development of FMDV antivirals. 1. INTRODUCTION The picornavirus foot-and-mouth disease virus (FMDV) causes a serious vesicular disease of a wide range of mammalian hosts, including domesticated livestock such as cattle, pigs, sheep and goats. The disease is rarely fatal but it is extremely contagious: infected animals rapidly produce high viral loads that may be excreted or exhaled and are easily transmitted to uninfected hosts in aerosols. As a result of this pathogenic profile, FMD is endemic in many parts of the world. Although many regions, such as the EU and the USA, are officially disease-free and strive to remain so via strict import controls, the risk of sporadic epidemics is real and ongoing. The EU faces particular difficulties due to the proximity of countries suffering frequent outbreaks; for example Turkey, Israel and Egypt all reported outbreaks in early 2007. Control of FMDV outbreaks places a severe economic burden on the affected country: the last major epidemic in the EU, which occurred in the UK in 2001, inflicted total costs of around £6 billion. Although vaccination remains a powerful weapon in the struggle to control FMD, current vaccines have problems that inhibit their practical benefit in endemic and epidemic situations (see Discussion). It is therefore worthwhile to consider alternative or supplementary control measures (Grubman & Baxt, 2004; Kitching et al, 2006; Sutmoller et al, 2003). The development of such measures is clearly dependent on a fuller understanding of the molecular basis of viral pathogenesis. A potentially powerful approach, as has been adopted for HIV, is to target essential viral enzymes (e.g. proteases, polymerases) with inhibitory drugs. We propose to investigate the potential of FMDV 3C protease (3Cpro) as a viable target for drug design.

387

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 As with other picornaviruses such as poliovirus (PV), human rhinovirus (HRV) and hepatitis A virus (HAV), FMDV comprises a protein capsid containing a single-strand, positive-sense RNA genome that is translated as a long polyprotein precursor shortly after cell entry. The FMDV polyprotein must be cleaved at 13 distinct locations by virally encoded proteases in order to release the proteins needed for capsid assembly and RNA replication. The proteolytic activity of FMDV 3Cpro is absolutely crucial to virus replication because it performs 10 of the 13 cleavages and is therefore an attractive target for therapeutic intervention. Despite the potential benefits of anti-viral therapy for FMD, there has so far been relatively little effort to develop drugs targeted to key viral enzymes. In part this is due to the lack of structural information and the appropriate tools to investigate enzyme inhibition. Recent crystallographic work has uncovered the structures of the viral polymerase (Ferrer-Orta et al, 2006) and the L (Guarne et al, 1998) and 3C (Birtley et al, 2005; Sweeney et al, 2007) proteases. These advances are likely to stimulate the quest for effective antiviral compounds. In our lab we have made substantial progress in investigating the structure and activity of FMDV 3Cpro. We are now in a strong position to advance these investigations and to be able to initiate the search for effective inhibitors. 2. MATERIALS AND METHODS The 3C protease from type A1061 was mutated to enhance solubility, expressed in E. coli, purified and crystallised as described previously (Birtley & Curry, 2005; Birtley et al, 2005; Sweeney et al, 2007). A customised fluorescent peptide substrate [4-(4-dimethylaminophenylazo) benzoic acidAPAKQLLD [5-(2-amino-ethyl) amino-1-naphthalenesulfonic acid (EDANS)] FDLLK (3C-FRET4)] was synthesised and used in cleavage assays as reported (Jaulent et al, 2007b). 3. RESULTS We have engineered a soluble, active recombinant form of the enzyme that can be expressed at high levels in E. coli (Birtley & Curry, 2005; Birtley et al, 2005), yielding around 25 mg per litre of culture. This has allowed us to determine crystal structures of 3Cpro, revealing different conformational states and providing important new insights into the catalytic mechanism and substrate recognition (Birtley et al, 2005; Curry et al, 2007; Sweeney et al, 2007). More recently we have determined the co-crystal structure of a 3Cpro-peptide complex, a first for any picornaviral protease (Fig. 1; unpublished data). The active site of the enzyme involves the participation of a flexible loop (known as the -ribbon) that folds in contact with the bound substrate; the importance of this interaction has only recently been revealed by crystallography and emphasises the need for structural information in the search for new inhibitors (Sweeney et al, 2007). We have also developed a continuous assay of 3Cpro cleavage activity—based on a quenched fluorescent peptide—with which we have begun to examine determinants of cleavage specificity, both on the enzyme and within the peptide substrate, and to probe inhibitor activity (Jaulent et al, 2007a; Sweeney et al, 2007). More recent work, performed in collaboration with Prof. Julie Frearson, (Scottish Hit Discovery Facility, Dundee University), has established that our fluorescent peptide cleavage assay can be adapted to the 384-well format, greatly reducing the enzyme load (Fig. 2; unpublished data). Under these assay conditions the %CV is 5.4 and the Z-factor is 0.83: the assay therefore comfortably meets the criteria for use in HT screening. Using synthetic peptides in our cleavage assays we have identified the peptide corresponding to the VP1-2A junction as the best substrate sequence within the viral polyprotein (Birtley & Curry, 2005). Using the results generated from these experiments we have synthesised peptide-based inhibitors that specifically block FMDV 3Cpro in in vitro assays (Fig. 3; unpublished data). Although such peptide based reagents are unlikely to be feasible drug candidates, they are valuable research tools since they permit exploration of the structural determinants of high affinity binding. With this exciting series of breakthroughs we now have the reagents and methodologies in place to forge ahead with our investigations. In particular, we can now make a systematic search for small molecule inhibitors of FMDV 3Cpro and follow up potential hits with further activity and structural analyses. But the development of effective drugs from initial hits is a long and complex process and a key question for discussion is whether there is a viable market for such drugs in today’s world. 4. DISCUSSION

388

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Current FMDV vaccines, consisting of inactivated virus particles, have significantly reduced the incidence of disease worldwide, but their use as a control measure in the event of new outbreaks, especially in previously disease-free countries, is beset by a number of problems. Initial diagnosis is complicated by the fact that FMDV is symptomatically similar to other diseases (e.g. vesicular stomatitis). Once an outbreak has been confirmed, vaccine doses must be formulated from frozen stocks, a process that takes several days; since FMDV exists in seven distinct serotypes and multiple subtypes, diverse vaccine stocks must be maintained so that doses can be formulated to protect against the outbreak strain. There is a further delay of up to a week post-injection before vaccinated animals become fully protected. Even then, vaccination does not provide sterile immunity, i.e. does not prevent infection and has no impact on the development of the carrier state, the persistence of infection in animals that have apparently recovered from the disease. In countries that strive to remain diseasefree, control by slaughter is sometimes seen as a more effective way to terminate outbreaks and has the attraction of incurring a shorter delay before the resumption of livestock trading with the rest of the world. However, pre-emptive slaughter (before diagnosis) is a difficult control strategy to implement because of farmers’ resistance; moreover, a slaughter policy can threaten the existence or valuable rare breeds. Although it is certainly vital that vaccine development work should continue, it is also important to explore alternative control measures (Follet, 2002). Until very recently the knowledge and tools did not exist to permit such exploration but structural work on key FMDV enzymes, including our studies on FMDV 3Cpro, creates an opportunity to investigate the role that antivirals could play in effective disease control. The major attraction of such compounds is that they could overcome many of the short-comings of FMDV vaccines. They could act rapidly, especially if orally bioavailable. Moreover, since the viral enzymes such as 3Cpro and 3Dpol are highly conserved across all FMDV serotypes (Carrillo et al, 2005), it is likely that inhibitors will have broad spectrum (cross-serotype) activity, a major advantage over current vaccines. In addition, a drug-treated animal that was not infected would be easy to distinguish (using existing tests) from an infected (or vaccinated animal). The prophylactic use of antivirals could provide a powerful weapon for disease control; conceivably such drugs could be used for short periods only in conjunction with vaccines to close the window of opportunity that the slow onset of the immune response offers to the virus. Moreover, it is conceivable that an antiviral might also be able to cure carrier state animals. Previous research on the 3C proteases of HAV, HRV and PV revealed that they belong to a unique class of enzymes that combines a chymotrypsin-like fold with a cysteine protease mechanism; since there are no known cellular homologues, these proteases are attractive targets for drug design. Determination of the crystal structures greatly stimulated the search for inhibitors as potential therapeutics compounds; much of the research has focused on HRV 3Cpro—since effective vaccines exist for HAV and PV infections—and has led to the development of compounds such as rupintrivir (Patick, 2006; Patick et al, 2005), which acts as a potent irreversible inhibitor of HRV 3Cpro. This inhibitor was demonstrated to protect against viral challenge, thus establishing the principle that picornaviral 3C inhibitors may provide an effective prophylactic defence against infection (Hayden et al, 2003). The progress made with HRV 3Cpro has recently led to proposals that anti-HRV 3Cpro compounds may be used to initiate development of antivirals to prevent PV infection (Katz et al, 2006) (to overcome the risks of production and use of the live poliovirus vaccine in a post-eradication world) and to combat potentially fatal SARS CoV infections in humans (Anand et al, 2003). Proof of principle of antivirals targeted to proteases has already been demonstrated for HRV 3Cpro and HIV protease. More recently antivirals to pestiviruses (designed to target the viral polymerase) have exhibited potential as a disease control measure (Paeshuyse et al, 2006; Vrancken et al, 2008). In fact it has been shown that a compounds developed against one pestivirus, bovine viral diarrhea virus (BVDV), can confer protection in pigs from challenge by a different pestivirus, classical swine fever virus (CSFV) - a disease with symptomatic similarities to FMDV. With regard to FMDV, recent reports indicate that the 3D polymerase may be a valuable drug target (Goris et al, 2006; Sakamoto et al, 2006). These data underscore the very real potential of antiviral drugs to help control infectious diseases of livestock. In a very similar vein, it is conceivable that inhibitors of FMDV 3Cpro could be developed as effective antivirals. The key question remaining is whether the international community is willing to provide the resources needed to take these efforts forward. 5. ACKNOWLEDGEMENTS SC and RJL are grateful for grant support from the BBSRC and access to synchrotron facilities at Daresbury SRS (UK), EMBL-DESY (Germany) and ESRF (France). NRR was funded by a Marie Curie Host Fellowship for Early Stage Research Training. We thank Stuart McElroy and Prof Julie Frearson (Dundee University) for optimising HT assays and providing the data for Fig. 2.

389

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 6. REFERENCES [1] Anand K, Ziebuhr J, Wadhwani P, Mesters JR & Hilgenfeld R. 2003. Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs. Science 300(5626): 1763-1767 [2] Birtley JR & Curry S 2005. Crystallization of foot-and-mouth disease virus 3C protease: surface mutagenesis and a novel crystal-optimization strategy. Acta Crystallogr D 61(Pt 5): 646650 [3] Birtley JR, Knox SR, Jaulent AM, Brick P, Leatherbarrow RJ & Curry S 2005. Crystal Structure of Footand- Mouth Disease Virus 3C Protease: New Insights into Catalytic Mechanism and Cleavage Specificity. J Biol Chem 280(12): 11520-11527 [4] Carrillo C, Tulman ER, Delhon G, Lu Z, Carreno A, Vagnozzi A, Kutish GF & Rock DL 2005. Comparative genomics of foot-and-mouth disease virus. J Virol 79(10): 6487-6504 [5] Curry S, Roqué-Rosell N, Zunszain PA & Leatherbarrow RJ 2007. Foot-and-mouth disease virus 3C protease: Recent structural and functional insights into an antiviral target. Int J Biochem Cell Biol 39(1): 1-6 [6] Ferrer-Orta C, Arias A, Agudo R, Perez-Luque R, Escarmis C, Domingo E & Verdaguer N 2006. The structure of a protein primer-polymerase complex in the initiation of genome replication. EMBO J 25(4): 880- 888 [7] Follet B 2002. Infectious diseases in livestock. London: 1-139 [8] Goris N, De Palma A, Toussaint J-F, Musch I, Neyts J & De Clerq K 2006. 2’-CMethylcytidine, a potent and selective inhibitor of the replication of the foot-and-mouth disease virus. Research Group of the Standing Technical Committee of EuFMD: International control of foot-and-mouth disease: tools, trends and perspectives Paphos: 404-413 [9] Grubman MJ & Baxt B 2004. Foot-and-mouth disease. Clin Microbiol Rev 17(2): 465-493 [10] Guarne A, Tormo J, Kirchweger R, Pfistermueller D, Fita I & Skern T 1998. Structure of the foot-and-mouth disease virus leader protease: a papain-like fold adapted for self-processing and eIF4G recognition. EMBO J 17(24): 7469-7479 [11] Hayden F, Turner R, Gwaltney J, Chi-Burris K, Gersten M, Hsyu P, Patick A, Smith G & Zalman L 2003. Phase II, randomized, double-blind, placebo-controlled studies of ruprintrivir nasal spray 2-percent suspension for prevention and treatment of experimentally induced rhinovirus colds in healthy volunteers. Antimicrob Agents Chemother 47(12): 3907-3916 [12] Jaulent AM, Fahy A, Knox SR, Birtley JR, Roqué-Rosell N, Curry S & Leatherbarrow RJ 2007a. A continuous assay for foot-and-mouth virus 3C protease activity. Anal Biochem (in press; available online) [13] Jaulent AM, Fahy AS, Knox SR, Birtley JR, Roque-Rosell N, Curry S & Leatherbarrow RJ 2007b. A continuous assay for foot-and-mouth disease virus 3C protease activity. Anal Biochem 368(2): 130-137 Katz S, Andino R, Joseph-McCarthy D, Modlin J, Nathanson N, Whitley R & Wimmer E (2006) Exploring the Role of Antiviral Drugs in the Eradication of Polio: Workshop Report: National Academies Press. [14] Kitching P, Hammond J, Jeggo M, Charleston B, Paton D, Rodriguez L & Heckert R 2006. Global FMD control-Is it an option? Vaccine (in press; available online) [15] Paeshuyse J, Leyssen P, Mabery E, Boddeker N, Vrancken R, Froeyen M, Ansari IH, Dutartre H, Rozenski J, Gil LH, Letellier C, Lanford R, Canard B, Koenen F, Kerkhofs P, Donis RO, Herdewijn P, Watson J, De Clercq E, Puerstinger G & Neyts J 2006. A novel, highly selective inhibitor of pestivirus replication that targets the viral RNA-dependent RNA polymerase. J Virol 80(1): 149-160 [16] Patick A 2006. Rhinovirus chemotherapy. Antiviral Res 71(2-3): 391-396 [17] Patick A, Brothers M, Maldonado F, Binford S, Maldonado O, Fuhrman S, Petersen A, Smith G, Zalman L, Burns-Naas L & Tran J 2005. In vitro antiviral activity and single-dose pharmacokinetics in humans of a novel, orally bioavailable inhibitor of human rhinovirus 3C protease. Antimicrob Agents Chemother 49(6): 2267- 2275 [18] Sakamoto K, Ohashi S, Yamazoe R, Takahashi K & Furuta Y 2006. The inhibition of FMD virus excretion from the infected pigs by an antiviral agent, T-1105. Research Group of the Standing Technical Committee of EuFMD: International control of foot-and-mouth disease: tools, trends and perspectives Paphos: 414-420 [19] Sutmoller P, Barteling SS, Olascoaga RC & Sumption KJ 2003. Control and eradication of foot-and-mouth disease. Virus Res 91(1): 101-144 [20] Sweeney TR, Roqué-Rosell N, Birtley JR, Leatherbarrow RJ & Curry S 2007. Structural and mutagenic analysis of foot-and-mouth disease virus 3C protease reveals the role of the betaribbon in proteolysis. J Virol 81(1): 115-124 [21] Vrancken R, Paeshuyse J, Haegeman A, Puerstinger G, Froeyen M, Herdewijn P, Kerkhofs P, Neyts J & Koenen F 2008. Imidazo [4, 5-c] pyridines inhibit the in vitro replication of the classical swine fever virus and target the viral polymerase. Antiviral Res 77(2): 114-119

390

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

391

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

392

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 70

AN ANTIVIRAL AGENT, T-1105 PREVENTS FROM VIRUS EXCRETION FROM PIGS INFECTED WITH PORCINOPHILIC FOOT-AND-MOUTH DISEASE VIRUS S. Ohashi1, K. Sakamoto1*, K. Fukai 1, K. Morioka1, R. Yamazoe1, K. Takahashi2 and Y.Furuta2 1

Exotic Diseases Research Station, National Institute of Animal Health, 6-20-1 Jousui-honcho, Kodaira, Tokyo 187-0022, Japan. 2 Toyama Chemical Co., Ltd., 3-2-5 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan

ABSTRACT Introduction It is a great challenge to control the spread of foot-and-mouth disease virus (FMDV) from the infected animal, especially pigs. We evaluated effectiveness of T-1105, one of pyrazinecarboxamide derivatives, in the virus excretion from the pigs infected with O/TAW/97, known as porcinophilic strain. Materials and methods One hour before the virus inoculation of 106.2 TCID50 of FMDV O/TAW/97, 200 mg/kg of T-1105 was orally administered to four pigs. The same dose of T-1105 was administered twice a day for 7 days. Two control pigs were all done without administration. Virus excretion in nasal swab and virus contents in plasma were examined by real-time PCR. Antibody to FMDV was measured by virus neutralization test and liquid phase blocking (LPB) ELISA. Results The control pigs showed the typical clinical signs. In the administered group two pigs showed no clinical sign but other two pigs formed vesicles at the limited site of the injection and the viral RNA was detected from nasal swab samples. Viremia was detected three of the four pigs at early stage of infection. But amounts of viral RNA in plasma were ten times lower than non-administered group. Both antibodies titers of LPB ELISA and the virus neutralization test were lower than those of non-administered group. Discussion By the oral administration of T-1105, some pigs inoculated with porcinophilic FMDV created mild symptoms but the duration of viremia became shorter and the virus excretions from nasal route were nothing or minor than that in non-administered group. The antibody responses to FMDV were so low that it was considered there was no or low virus replication in the pigs. It was suggested that administration of T-1105 also controlled virus excretion from pigs infected with porcinophilic FMDV. 1. INTRODUCTION The foot-and-mouth disease (FMD) is the most contagious disease in cloven-hoofed animals, including cattle, swine, sheep and goats as well as a variety of wild animal species. FMD virus (FMDV) classified in the genus Aphthovirus, family Picornaviridae. The virus is antigenically variable and seven distinct serotypes of the virus O, A, C, Asia1 and the South African Territories types 1, 2 and 3. FMDV can spread rapidly in susceptible animal herds. In 1997, because of pig industries in Taiwan had hit by the pig-adapted strain (porcinophilic strain) of FMDV, more than 4 million pigs were killed and made Taiwan losing FMD free status (Yang et al. 1999). Strain O/TAW/97 has been shown to have a species-specific adaptation to pigs (Dunn and Donaldson, 1997) and only caused severe clinical disease in pigs. FMD has the severe socioeconomic impact. To control the disease there are two policies of “test and slaughter” and/or “vaccination”. The countries where the FMD outbreaks occur will decide to

393

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 use either or both approaches, depending on the epidemiological situation of their outbreaks. Since the 2001 FMD outbreak in the UK, there has been changes concerning control measure in Europe which use vaccination as a means of reducing dependence on culling of animals. Though the best currently available, chemically inactivated vaccines only confer complete clinical protection against homologous challenge 7 days after vaccination and partial protection in 4 days (Golde et al. 2005). Therefore, the use of current FMD vaccines to induce early protection is limited and alternative/supplementary methods to rapidly reduce the spread of FMDV in outbreak situations are needed. We evaluated effectiveness of T-1105, one of pyrazinecarboxamide derivatives, on the virus excretion from the pigs infected with O/TAW/97, known as porcinophilic strain. Vaccination is the one of the gold standards for FMD control measure. Moreover, we would be able to provide new alternative option to control FMD outbreak. 2. MATERIAL AND METHODS 2.1 Compounds T-1105 was provided by Y. Furuta of Toyama Chemical Co., Ltd. (Toyama, Japan). T-1105 was dissolved in minimum essential medium (MEM) for in vitro and vivo studies for animal studies. 2.2 Viruses and cells: Porcinophilic FMDV, strain O/TAW/97, was propagated on baby hamster kidney (BHK-21) cells. The cells were maintained in Eagle's MEM (Nissui Pharmaceutical Co., Tokyo, Japan) supplemented with 0.295% tryptose phosphate broth (Difco Laboratories, Detroit, Mich.), 0.15% sodium bicarbonate, 2 mM L-glutamine, and 5% fetal bovine serum. 2.3 Animal experiment: Animal experimentation was carried out in accordance with National Institute of Animal Health guidelines for Animal experiments. Six female conventional pigs weighing 20 kg were used in this experiment. One hour before virus inoculation, the administered group (n=4) was orally administered T-1105 at a dose of 200 mg/kg/day. Administered group were challenged intradermally into the heel bulb with 106.2 TCID50 of FMDV O/TAW/97. Nonadministrated pigs (n=2) were treated in same manner administration of T-1105. After virus challenge, the oral administration with T-1105 at the same dose was continued twice a day for 7 days. As a control, two pigs were all done in the same manner except administration of the compound. The pigs were monitored daily for clinical signs, including rectal temperature. Nasal swabs and heparinized blood were collected daily for monitoring virus excretion and viremia. Serum samples were collected daily for serological assay. 2.4 RNA extraction and quantitative real-time reverse transcription (RT)-PCR Viral RNA from plasma and saliva were extracted with a High Pure Viral RNA kit in accordance with the manufacturer's instruction (Roche Diagnostics-Boehringer Mannheim, Mannheim, Germany). For the quantification of FMDV RNA, real-time RT-PCR was carried out as described elsewhere with minor modification (Oleksiewicz et al. 2001). The one-step quantitative RT- PCR assay was performed using a pair of primers TaqManFMD-IRES-F 5'-CTGTCTCGTAGCGGAGCATG-3' and TaqManFMD-IRES-R 5'-GCCCCGTGGGTCCTT-3', targeting the internal ribosomal entry site region and a TaqManFMDProbe-IRES 5'-VIC-TGGCCGTGGGAACTCCTCCTTG-TAMRA-3' with TaqMan OneStep RT-PCR Master Mix Reagents Kit (Applied Biosystems). 2.5 Serum neutralization test Serum neutralizing antibodies to FMDV were measured by a micro plate assay using IB-RS-2 cells and FMDV strain O/TAW/97. Briefly, One hundred 50% tissue culture infective doses of virus were added to each serum dilution. The mixtures were incubated at 37°C for 1 h, and then IB-RS-2 cells suspended in MEM were added to each well. After incubation at 37°C for 3 days in a humidified 5% CO2 atmosphere, the neutralizing antibody titers were expressed as the reciprocal of the highest dilution of sera that completely neutralized for 50% of the challenge virus. 2.6 Detection of antibody in sera by ELISA: A LPB ELISA was used to measure antibodies against FMDV, as described previously (Ferris et al. 1990). 3. RESULTS The control pigs showed the typical clinical signs such as fever, severe vesicles on their feet within 1 to 2 days post infection. Vesicles were ruptured until 3 days post infection and pigs exhibited lameness (Table). In control pigs, viral RNA in nasal swabs were detected between 1 and 7 days post infection and viremia lasted for 3 days (Figures 1 & 2). In administered group, two pigs didn’t show any clinical signs but other two pigs formed vesicles only at the limited site of the injection

394

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 and the viral RNA was detected from nasal swab samples at 2 and 3 days post infection. Viremia was detected three of the four pigs at early stage of infection (Figures 1 & 2). But amounts of viral RNA in plasma from them were ten times lower than those of control group. Both antibodies titers of LPB ELISA and the serum neutralization test were also lower than those of non-administered group (Figures 3 & 4). 4. DISCUSSION Pyrazinecarboxamide derivatives are known to have a potent anti-RNA viral activity (Furuta et al. 2002, Gowen et al. 2007., Julander et al. 2007). T-705, one of the Pyrazinecarboxamide derivatives, inhibited influenza virus RNA polymerase (Furuta et al. 2005). T-1105, a substituted pyrazine compound, has been found to exhibit potent anti-FMDV activity in vitro and in vivo (Sakamoto et al. 2006). The inhibitory mechanism of T-1105 is considered to be the inhibition of FMDV RNA-dependent RNA polymerase. Pigs are important amplifiers of FMDV because of the abundance of infectious material excreted out of the body (Sellers et al. 1977). Recent intensification of agricultural systems has led to a massive increase in the size and density of susceptible pig populations provides potential for large scale and rapid spread of FMDV. Our study is aimed to evaluate the efficacy of T-1105, antiviral compound, in giving inhibition to pigs from infection, virus replication, virus excretion and clinical disease. At last EUFMD in Cyprus, we demonstrated that pigs which were administered with T-1105 developed no clinical symptoms by challenging the FMDV strain O/JPN/2000. Virus excretion from these pigs was not detected from their nasal swabs by the real-time RT-PCR assay. Although only slight increase of the both ELISA and serum neutralizing antibodies was observed, it is considered that inoculated FMDV could not or slightly replicate in the pig. These results are suggesting that T1105 is effective antiviral agent against FMD infection in pigs. We confirmed inhibitory effect of T-1105 in case of infection with porcinophilic strain of FMDV. By the oral administration of T-1105, some pigs inoculated with porcinophilic FMDV created mild symptoms but the duration of viremia became shorter and the virus excretions from nasal route were nothing or minor than that in control group. The antibody responses to FMDV were so low that it was considered there was no or low virus replication in the pigs. In infection with O/JPN/2000, T-1105 was completely inhibited viral replication in pigs. It was suggested that administration of T-1105 also controlled virus excretion from pigs infected with porcinophilic FMDV. This anti-FMDV agent has advantages of an immediate effect at the early stage of the virus infection. It is considered that the use of this kind of excellent antiviral agents in FMD outbreaks of FMD free countries without vaccination can be a strong tool to control the disease and to reduce the spread of FMD outbreaks. 5. ACKNOWLEDGEMENTS The authors thank Dr. Ming-Hwa Jong and Dr. Yeon-Liang Lin in the National Institute for Animal Health of Taiwan, for providing FMDV strain O/TAW/97. 6. REFERENCES [1] Dunn, C.S. & Donaldson, .A.I. 1997. Natural adaption to pigs of a Taiwanese isolate of footand-mouth disease virus, Vet. Rec. 141:174–175. [2] Ferris, N.P., Kitching, R.P., Oxtoby, J.M., Philpot, R.M. & Rendle, R. 1990. Use of inactivated foot-and-mouth disease virus antigen in liquid-phase blocking ELISA. J. Virol. Methods 29: 33–41. [3] Furuta, Y., Takahashi, K., Fukuda, Y., Kuno, M., Kamiyam, T., Kozaki, K., Nomura, N., Egawa, H., Mimani, S., Watanabe, Y., Narita, H., & Shiraki, K. 2002. In vitro and in viro activities of anti-Influenza virus compound T-705. Antimicrob. Agents Chemother. 46:977-981. [4] Furuta, Y., Takahashi, K., Kuno-Maekawa, M., Sangawa, H., Uehara, S., Kozaki, K., Nomura, N., Egawa, H. & Shiraki, K. 2005. Mechanism of Action of T-705 agaist influenza virus. Antimicrob. Agents Chemother. 49:981-986. [5] Gowen B.B., Wong M.H., Jung K.H., Sanders A.B., Mendenhall M., Bailey K.W., Furuta, Y. & Sidwell, R. W. 2007. In vitro and in vivo activities of T-705 against arenavirus and bunyavirus infections. Antimicrob Agents Chemother. 51:3168-3176. [6] Golde, W.T., Pacheco, J.M., Duque, H., Doel, T., Penfold, B., Ferman, G.S., Gregg, D.R. & Rodriguez, L.L., 2005. Vaccination against foot-and-mouth disease virus confers complete clinical protection in 7 days and partial protection in 4 days: use in emergency outbreak response. Vaccine. 23: 5775–5782.

395

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [7] Julander JG, Furuta Y, Shafer, K. & Sidwell R.W. 2007. Activity of T-1106 in a hamster model of yellow Fever virus infection. Antimicrob Agents Chemother. 51:1962-1966. [8] Oleksiewicz B.M., Donaldson, A.I. & Alexandersen, S. 2001. Development of a novel realtime RT-PCR assay for quantitation of foot-and-mouth disease virus in diverse porcine tissues. J.Virol. Methods. 92: 23-35. [9] Sakamoto K., Ohashi, S., Yamazoe, R., Takahashi, K., & Furuta, Y. 2006. The inhibition of FMD virus excretion from the infected pigs by an antiviral agent T-1105. Report of the European Commission for the Control of Foot-and-mouth disease, Session of the Research Group of the Standing Technical Committee, Paphos, Cyprus, 17-20 October 2006, Appendix64 , pp 418- 423. [10] Sellers R.F., Herniman, K.A. & Gumm I.D. 1977. The airborne dispersal of foot-and-mouth disease virus from vaccinated and recovered pigs, cattle and sheep after exposure to infection. Res Vet Sci 23:70–75. [11] Yang, P.C., Chu, R.M., Chung, W.B., & Sung, H.T. 1999. Epidemiological characteristics and financial costs of the 1997 foot-and-mouth disease epidemic in Taiwan. Vet Rec. 145:731-734.

Table Effects of T-1105 clinical signs days post infection Pig No. T-1105 Administration group

Unadministration (Control) group

2 3 4 5

+ +

+ ++ + - - - - + ++ +

1 6

+ +

+ ++ ++ ++ ++ ++ ++ ++ + ++ ++ ++ ++ ++ ++ ++

＋ Creation of vesicular legion ++ Rupture of vesicle

106

106 #2

105

103 102 10

10 PFU/0.1ml

PFU/0.1ml

#1 105

103 102 10

1 0dpi

1dpi

2dpi 3dpi 4dpi 5dpi days post infection

6dpi

7dpi

0dpi

1dpi

2dpi 3dpi 4dpi 5dpi days post infection

Figure 1: Viremia on pigs infected with FMDV strain O/TAW/97. A: T-1105 administered pigs (n=4), B: T-1105 non-administered pigs (n=2)

396

6dpi

7dpi

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

106

106 #2

105

PFU/0.1ml

103 102 10

PFU/0.1ml

104

105

104 103 102 10

1 0dpi

1dpi

2dpi 3dpi 4dpi 5dpi days post infection

6dpi

7dpi

0dpi

1dpi

2dpi 3dpi 4dpi 5dpi days post infection

6dpi

7dpi

Figure 2: Virus excretion in saliva from pigs infected with O/TAW/97 A: T-1105 administered pigs (n=4), B: T-1105 non-administered pigs (n=2)

Log2 LPBE titer

10000

1000

100

10 0dpi

2dpi

4dpi

6dpi

8dpi

10dpi

days post infection

Figure 3: LPBE antibody responses. Solid lines and dotted lines indicate T-1105-administrated pigs (n=4) and control pigs (n=2), respectively.

397

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Log2 SN antibody titer

7 6 5 4 3 2 1 0 0dpi

2dpi

4dpi

6dpi

8dpi

10dpi

days after infection

Figure 4: Serum neutralizing antibody responses. Solid lines and dotted lines indicate T-1105administrated pigs (n=4) and control pigs (n=2), respectively.

398

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 71

ELEMENTS OF A GLOBAL STRATEGY J. Domenech1, Juan Lubroth2 and Keith Sumption3, in consultation with OIE 1

CVO-FAO EMPRES, FAO 3 EU FMD

399

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

400

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

401

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

402

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

403

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

404

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

405

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

406

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Stakeholder and national buy-in to FMD control (Session 3) Opportunities: – international standards and policy objectives that encourage investment at farm/containment level – FAO/OIE/partners to promote critical risk control analyses for endemic countries – socio economic analysis to determine sustainable and cost effective national FMD risk reduction strategies that address critical risk control points in the market chain risk

Veterinary service capacity to prevent and control (Session 6) Opportunities: – FMD can only be f ought at the f ront-line • with stakeholder support • OIE PVS assessments • international need to assess and monitor capacity build-up

– utilise PRA to rapidly assess FMD epidemiology, reasons f or control f ailure, int ervention options – invest in epide miosurveillance a nd laborat ory networking and upgradi ng

Coherent and feasible global strategy to achieve change (Session 8) • Keynote:

GF TADs context and Elements of a global FAO/OIE strategy for FMD control • Panel discussion

– OIE, EC, EuFMD – WRL Pirbright (OIE/FAO Ref Lab network) – Global FMD Research Alliance (GFRA) • Opportunity

– to shape major elements to be considered in the international strategy to be discussed at OIE/FAO Paraguay Conference in 2009

407

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

408

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

POSTER SESSION Appendix 72 A FOETAL GOAT TONGUE CELL LINE FOUND HIGHLY SENSITIVE FOR FOOT-AND-MOUTH DISEASE VIRUS

K. Brehm*, M. Lenk, R. Riebe and B. Haas Friedrich-Loeffler-Institute, Südufer 10, 17493 Greifswald-Insel Riems, Germany

INTRODUCTION While infectious FMDV usually is present with high titres in fresh vesicular material, titres found in sera, nasal swabs, saliva and oropharyngeal samples (probang) are much lower, necessitating highly sensitive detection systems. The most sensitive cells for FMD virus isolation are primary bovine thyroid (BTY) cells, but they can´t be passaged or frozen without impairing their sensitivity. Ensuring that there is always a fresh and suitable batch of primary BTY cells available for diagnostic purposes is quite laborious and expensive. Therefore, most diagnostic laboratories use other cells which are more convenient to handle, either cells of bovine, ovine or porcine origin or permanent cell lines as BHK-21 or IBRS-2. These protocols do not facilitate a reliable overnight detection of virus. MATERIALS AND METHODS A foetal goat tongue cell line (ZZ-R 127) was supplied by the CCLV (Collection of Cell Lines in Veterinary Medicine) of the FLI and was inoculated with cell culture virus as well as with virus originating from vesicular material. RESULTS FMDV infection could always be detected visually within 18-24 hours. Strains representing all seven serotypes of FMDV could be isolated on ZZ-R 127 cells with a higher sensitivity than in BHK-21/CT or IBRS-2 cells. Furthermore, the CPE was consistently observed in the first passage in ZZ-R 127 cells and also earlier than in BHK-21 and IBRS-2 cells, in which the detection of low amounts of FMDV often takes several days and may even require several passages. The foetal goat tongue cell line maintains its sensitivity for FMDV at least from the 76th to the 160th passage. DISCUSSION A foetal goat tongue cell line was found to be a sensitive, rapid and convenient tool for the isolation of foot-and-mouth-disease virus (FMDV) with significant advantages over established permanent cell lines. It is recommended that the new cell line is introduced into diagnostic laboratories to improve FMD isolation.

409

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

410

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 73 EPIZOOTIOLOGICAL STUDY OF FOOT-AND-MOUTH DISEASE IN THE SUDAN [THE SITUATION AFTER TWO DECADES] M. Habiela1*, M. A. Gaffar2, Y. A. Raouf1, Y. H. Ali

Unit of Foot-and-mouth disease - Central Veterinary Research Laboratories (CVRL) Animal Resources Researches Corporation (ARRC)-Sudan. 2 Department of Virology - Central Veterinary Research Laboratories-(CVRL) Animal Resources Researches Corporation (ARRC) Sudan. * Corresponding author: -E-mail addresses: mhabiela979@hotmail.com (M. Habiela), Present address: Central veterinary research laboratories (CVRL) - Unit of FMD- Madani HighwayKhartoum, Box: 8067 Amarat- Khartoum- Sudan, Cell: 00249 922 846856

ABSTRACT To update the information on the situation of foot-and-mouth disease (FMD) in the Sudan, a serosurvey and disease survey were conducted. Recently collected data on FMD in the Sudan showed that FMD is a major constrain to cattle production in the country, but causes no threat or mild disease in sheep and goats. Disease with obvious clinical signs was detected in cattle only, which was caused by serotype "O" and "SAT2". Disease seasonal occurrence in the cold dry season was observed and animal movement seems to play a major role in virus dissemination. A total number of 1069 sera were collected from cattle, sheep, goats, and camel; from seven states in the Sudan, for detection of antibodies to FMDV. Application of liquid phase blocking (LPB) ELISA, revealed that antibodies to four serotypes were present in ruminants; namely O, A, SAT1 and SAT2. No antibodies to FMD were detected in camels’ sera. The results differed from early reports regarding the prevalence of serotype specific antibodies. In cattle antibodies to type "A" (78.13%) surpassed that of type "O"(69.39%) and antibodies to type "SAT2"(44%) surpassed that to type "SAT1"(20.2%). This work elucidates the current epidemiology of FMD in some parts of the Sudan. 1. INTRODUCTION Foot-and-mouth disease (FMD) is a highly contagious vesicular disease of cloven-footed animals. Infection of FMD was reported in cattle, sheep, goats, swine, and antelopes, as well as more than 70 wild animal species (Shahan, 1960). FMD virus belongs to the Family Picornaviridae, Genus Aphthovirus. It has seven distinct serotypes O, A, C, South African Territories 1-3 (SAT-1, SAT-2, & SAT-3) and Asia-1 (Murphy et al. 1999). In most of sub-Saharan Africa, serotypes O, A, SAT-1 and SAT-2 are predominant (Rweyemamu et al., 2001). In the Sudan FMD is endemic and FMD outbreaks occur annually, the first record of the disease in the Sudan was in 1903 (Eisa and Rweyemamu, 1977), and four FMD serotypes out of the seven were reported in the country. These are O, A, SAT-1 and SAT-2 (Abu Elzein, 1983). Serotype O was firstly isolated then Serotype SAT1 before1952, Serotype A in 1957, and lastly Serotype SAT-2 in 1977, (Abu Elzein and Crowther 1979). Antibodies to these four FMDV serotypes were detected in cattle, sheep and goats sera, but their prevalence rate was quite different between species (Abu Elzein, 1987). camels' sera has been screened by the agar gel immunodiffusion test (AGID) for the presence of antibodies against FMD virus infection associated (VIA) antigen and proved to be negative (Abu Elzein et al., 1984). Since 1987 no study on FMD in the Sudan was carried out and type and subtype situation needs to be updated. However, recently Efforts were regenerated; a serosurvey was conducted in Khartoum state (Raouf et al., 2008), moreover, samples of suspected FMD outbreaks were sent more regularly to the World Reference Laboratory (WRL) at Pirbright in the UK (Anon, 2007). Detailed survey of FMD in different animal species in the Sudan is insufficient. In this study, an attempt was made to update and evaluate the FMD situation in some states of the Sudan, questionnaire and interviews with herdsmen and veterinarians, beside conduction of a serosurvey and a disease survey were carried out.

411

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 2. MATERIALS AND METHODS 2.1 Questionnaire and serosurveyThe survey was conducted from 2006 to 2008. A simple standardized questionnaire was formed and used for collection of information from the herdsmen emphasizing on the data on host and environment. Data was collected on blood-sampled animal (Species, age, sex, and breed) beside history of FMD in the herd. A total of 1069 sera were randomly collected from Cattle (469 sera), Sheep (319sera), Goats (88sera) and dromedary Camels (193sera) from seven states in the Sudan; namely, Gezira (Wad madani), Northern Kordofan (ElObied), Southern Kordofan (ElDeling), White Nile(Rabak), Gadarif (Gedarif), River Nile (Atbarah) and El-Shemalyah (Dongola) ( Fig. 1). These localities were selected according to the disease history and their density in animals’ population. 2.2 FMDV antibody Detection by LPB ELISA The LPBE Kits were provided by the Arab Organiztion for Agricultural development and the Federal Ministry of Animal Resources and Fisheries. They were obtained from the Institute for Animal Health (IAH), Pirbright Laboratory, UK. Screening assay for detection of FMD antibodies were carried out according to the instructions of the manufacturer after optimization of antigen dose for the test (Raouf et al., unpublished data). 2.3 Collection of FMD virus samples 2.3.1 Collection of Oesopharyngeal (OP) fluid (Probang samples) Samples were collected from suspected FMD- carrier animals using a probang cup, according to the method of Hedger (1968), added to equal volume of transport media ( 0.08 M phosphate buffer pH 7.2-7.6) with antibiotics (Kitching and Donaldson, 1987), Samples were transported in ice box to the Laboratory, and then stored at -70° C. 2.3.2 Collection of epithelial samples The epithelium samples were collected during the course of field outbreaks as described by Kitching and Donaldson (1987). Mouth lesions were taken from infected animals, put in transport media (0.04 M phosphate buffer pH 7.2.-7.6) with 50% glycerol and antibiotics, kept on ice, transported to the laboratory and stored at -30 to -5°C. 2.4 FMD virus isolation and serotyping Nineteen Probang samples and two swabs were collected from previously suspected FMD-infected cows. Eight epithelial samples that were collected from three out breaks, one in Gezira State and two outbreaks in White Nile State ( Table 4) . Some of the probang samples were packaged, as described by Kitching and Donaldson, (1987) and were submitted to the FAO-OIE World Reference Laboratory (WRL) for FMD at Pirbright, UK. Virus isolation was carried out in bovine thyroid (BTY) cell culture (Snowdon, 1966) and primary bovine kidney (PBK) cell culture (Patty et al., 1962) at CVRL. Virus isolates were serotyped using FMD antigen detection ELISA that obtained from the Institute for Animal Health (IAH), Pirbright Laboratory, UK. The ELISA procedures used were similar to those described by Roeder and Smith (1987) and Ferris and Dawson (1988). 3. RESULTS 3.1 Questionnaire and interviews From 50 distributed questionnaires, 23 were returned, eleven questionnaires from Gezira stats, four from White Nile state, three from each North Kordofan and Gedarif states, and one from each South Kordofan and El-Shymaliyah States. The collected data showed that FMD, which is locally known as (Abu Lisan = Tongue disease), is well-known to herdsmen and they are well-awarded of the disease, its clinical signs, seasonality, duration and transmission. The practiced- husbandry systems in the investigated herds, were either extensive or semiintensive with free animal movement. The questionnaire data showed that FMD clinical signs were only observed in cattle; and causes mild or no clinical signs in small ruminants specially that were intermingling with cattle. It was predominantly encountered in the cold dry season (November to March). The morbidity rate may approach 100%; especially in cross-bred cattle; while the mortality rate was low and only occurred in young animals. Losses were largely due to the death of newborn and suckling calves, loss of weight and milk production, decrease of draft power and infertility.

412

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 The data revealed that the current applied control policy is restriction of animal movement and quarantine. To reduce the effects of the disease, herdsmen apply powder of the ground seeds of the Acacia nilotica in the drinking water of infected animals to cure the mouth ulcers. These seeds are known to contain a high concentration of tannic acid. They also use glycerin and antibiotics for protection of infected animals from secondary bacterial infection.

= Cattle.

= Sheep.

= Goats.

= Camel.

= main location of sampling.

Figure 1: Map of the Sudan showing the area of the Study and locations of sampled livestock between 2006- 2008 3.2 Serology The overall percent of positive sera for FMDV in the four tested animal species were 79.24 % in cattle, 22.95 % in sheep, 28.57 % in goats and no positive serum was observed in the tested camels’ sera (Table 1). Antibodies to the four FMDV serotypes used in the study, were observed in the animal sera from all the investigated states; with the highest prevalence in cattle, (Table 2).The results obtained by LPB ELISA showed that serotype A (78.1 %) was the most prevalent in cattle followed by serotype O (69.4 %), SAT2 (44%) and SAT1 (20.2%). In sheep, serotype O (27.5 %) was the most prevalent followed by SAT-2 (9.1%), A (8.7 %), and SAT-1 (5.1%). In goats, serotype O (27.5%) was the most prevalent followed by A (15.9%), SAT-1 (8.5 %) and SAT-2 (2.4 %) (Table 2; Figure 2). Table 1: Overall positive sera for FMDV antibodies per species Animal species

Positive sera a

Cattle Sheep Goat

374 / 472 70 / 305 28 / 98

Camel

0 / 176

Positive % 79.24% 22.95% 28.57% 0

a = positive sera b= tested sample

Table 2: The percentage of positivity of the total sera for each of the four FMDV serotypes using LPBE

413

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Serotype Species Positive sera a 297 *Cattle b /428 *Sheep *Goats *Camel

52 /189 22 /80 0 /193

O Positive % 69.39

A Positive sera a 275 b /352

Positive % 78.13

SAT1 Positive Positive sera % a 60 20.2 /297 b

27.51

8 /92

8.7

5 /98

5.1

27.5 0

11 /69 0 /193

15.94 0

4 /47 0 /193

8.51 0

SAT2 Positive Positive sera % a

176 /400

16 /178 2 /84 0 /193

44 8.99 2.38 0

a = Number of positive sera, b= Number of tested sample.

*= some of the serum samples were tested for more than one serotype.

Figure 2: Overall results of screened sera of different animal species for FMDV antibodies using LPBE 2006-2008 3.3 FMD virus isolation and serotyping From the nineteen probang samples and two swabs that were inoculated in BTY cell culture, seventeen samples caused cytopathogenic effects (CPE) and progressive changes in cell culture within 24 -72 hours following inoculation. Eighteen cell culture harvests were re-inoculated in bovine kidney (BK) cell culture; ten samples produced CPE in BK cell culture (Table 3). A 10% suspension of collected epithelium samples from infected cows during FMD outbreaks were inoculated in BK cell culture and a progressive CPE started within two to three hours following inoculation. After 24 hours the monolayer was completely destroyed. (Table 3) 3.4 Identification of virus isolates by Antigen Detection ELISA All probang samples derived cell culture material that were tested by the antigen detection ELISA showed optical density less than 0.1 and were considered negative (Table 5). Probang samples dispatched to the FAO-OIE World Reference Laboratory (WRL) for FMD at Pirbright, United Kingdom (UK) also proved to be negative (Table 4). Epithelium samples from Gezira state (Al-Kiraiba) were positive for serotype SAT-2, whereas, one epithelium sample from the White Nile state (Jabal Biyout) was positive for serotype O. Epithelium samples from Alkonoz and Omshatain were negative for FMDV (Table 5).

414

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 4a: Gezira state results

Figure 4b: North Kordofan state results

Figure 4c: White Nile State results

Figure 4d: South Kordofan State results

Figure 4e: River Nile State results

Figure 4f: Gedarif State results

Figure 4: Results of screened sera of different animal species for FMDV antibodies using LPBE in some states in the Sudan between 2006- 2008:

415

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 3: Results of inoculation of specimens sampled from cattle in different cell cultures Type specimen Probang Swabs Epithelium a b

Number of samples passaged in cell culture 21a 2a 7b

Number of samples caused CPE in cell culture 19 2 7

= Passaged in BTY cells and twice in BK cells = At least two passages in BK cells only

Table 4: Pirbright Laboratory- UK results of detection and serotyping of FMDV specimens sampled from cattle in Sudan-2007

416

Sample identity

Description sample

SUD 7 -NK 1 SUD 7 -NK 2 SUD 7- NK 3 SUD 7 -NK 4 SUD 7 -NK 5 SUD8 -WN 1 SUD8- WN 2 SUD8- WN 3 SUD8 -WN 4 SUD8- WN 5 SUD8 -WN 6 SUD8- WN 7 SUD 9 -GAD 1 SUD 9 -GAD 2 SUD 9 -GAD 3 SUD10 DONG 1 SUD10DONG 2 SUD 11GEZ-MD 19N SUD 11GEZ-MD 21T SUD12 GEZ-MS 1 SUD12GEZ-MS 2

PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG PROBANG FLUID PROBANG FLUID FLUID PROBANG PROBANG

WRL Reference code

Serotyping results by cell culture and ELISA

PCR results

SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD SUD

NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD NVD

NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG NEG

18/2007 19/2007 20/2007 21/2007 22/2007 8/2007 9/2007 10/2007 11/2007 12/2007 13/2007 14/2007 15/2007 16/2007 17/2007 6/2007 7/2007 4/2007 5/2007 2/2007 3/2007

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 5: Results of epithelium and probang samples tested by Antigen detection ELISA

Sample ID GAD-3 GAD-2 DONG-2 NK-3 NK-4 GZ-MD19N GZ-MD21T WN-3 WN-7 AlKiraiba-2 AlKiraiba -3 AlKiraiba -4 Kuwait 133 AlKonoz calf Omshatain

State Gedarif Gedarif ElShymalyiah North Kordofan North Kordofan Gezira Gezira White Nile White Nile Gezira Gezira Gezira White Nile White Nile White Nile White Nile

Original sample Probang Probang Probang Probang Probang Swab Swan Probang Probang Epithelium Epithelium Epithelium Epithelium Epithelium Epithelium Epithelium

description of tested sample by Ag ELISA Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest epithelium suspension Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest Cell culture harvest

Result Negative Negative Negative Negative Negative Negative Negative Negative Negative SAT2 positive SAT2 positive SAT2 positive Negative Type O positive Negative Negative

4. DISCUSSION In studying the current status of FMD in the Sudan, it is obvious that FMD is still endemic in the country. It occurs mostly, in the cold dry season. The extensive livestock husbandry systems adopted in the Sudan seems to favor the conditions for spread of FMD virus. Cattle reared under nomadic conditions in the Sudan use their feet to wander around for grazing which may extend for many kilometers and use the tongue for prehension of grass while eating, but when these functions are impaired by FMD lesions in feet and mouth they will be recumbent and mostly suffer from starvation (Abu Elzein, personal communication, 2008). These observations add a further dimension to the economic significance of FMD in the Sudan and clearly enfeeble the notion in enzootic areas that FMD is not particularly a serious disease and its relevance is only to international trade. Doel (2003) observed that FMD might have devastating effects on animals and herdsmen regardless of animal population. In the present study beside Gezira, North Kordofan and River Nile States that were previously screened by Abu Elzein et al. (1987), another four states were included in the recent study viz. South Kordofan, White Nile, Gedarif and El-Shymalyah states. After 20 years from the last study on FMD in the Sudan, and without applying any vaccination or eradication programmes, the prevalence of antibodies to FMDV serotypes has changed both in magnitude and order. In cattle antibodies to serotype "A" surpassed that of serotype "O" and antibodies to serotype "SAT2" surpassed that to serotype "SAT1". In this work in cattle, antibodies to serotype "A" and "O" showed a prevalence rate of 78.1% and 69.4% compared to 18% and 75.6% respectively in the previous study (Abu Elzein et al, 1987). Antibodies to serotype"SAT2" and "SAT1" showed a prevalence rate of 44% and 20.2% compared to 0.2% and 6.4% respectively in the previous study (Abu Elzein et al, 1987). These results coincided with the results of the recent serosurveilance in cattle species in Khartoum state (Raouf; 2007 report of FMD UNIT). In sheep and goats, the prevalence rate was also changed, serotype O was found to be the most prevalent serotype in sheep and goats (27.5% each) instead of serotype A in sheep and SAT1 in goats in previous reports (Abu Elzein et al, 1987), moreover, antibodies to SAT2 were detected for the first time in sheep and goat sera from investigated states 8.99% and 2.38 % respectively. Similar result was observed in recently surveyed sheep and goats' sera in Khartoum state (Habiela et al., unpublished data). Similar to previously published data ( Abu Elzein et al, 1987), detected antibodies to FMDV serotypes in sheep and goats sera revealed much lower prevalence rates than that detected in

417

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 cattle sera. This could be due to the fact that some of the screened flocks of sheep and goats graze without intermingling with cattle, as in some parts of North Kordofan, River Nile and Gedarif states.

Figure 3: Prevalence antibodies to FMDV in different animal species sera screened by Abu Elzein et al. in 1987. The high prevalence rates of FMDV serotypes "A", "O" and "SAT2" antibodies detected in cattle species were consistent with the isolation of serotype "O" and "SAT2" in this work and with the recent reports of FMDV isolates of serotype O, A and SAT-2 from the Sudan by Pirbright Laboratory, UK (Anon, 2007 ) . In the Sudan, camels frequently browse in contact with other ruminants under free range conditions and at watering points. Screening of camels’ sera by LPB ELISA, which was used for the first time in the Sudan, revealed that no antibodies to any of the four FMDV serotypes used in the present study were detected. This result is in agreement with the findings of Abu Elzein et al (1984) who reported that Sudanese camels were seronegative to FMDV antibodies. Moreover, these findings are consistent with recent reported showed that dromedary camels are not susceptible to FMDV and do not show detectable serological response against it even under experimental conditions (Wernery and Kaaden 2004; Larska et al., 2008). Theoritically, FMDV carrier animals might harbor infectious virus in their esopharyngeal region (Sutmoller et al, 2003; Alexandersen et al, 2002), then the risk of transmitting the virus to other susceptible animals cannot be ruled out. Collection of OP samples from previously infected animals revealed that no FMDV was recovered or detected. In spite of that 81% of probang samples caused CPE in BTY cell culture and 56% of re-inoculated probang samples in BK caused CPE. This could be due to the known difficulty of isolation FMDV from such samples (Alexandersen et al., 2002; Abu Elzein, personal communication, 2008), or that, the examined animals were actually free from FMDV. SAT2 serotype being the most recently serotype introduced in the Sudan in seventies (Abu Elzein, 1979), this serotype is often associated with outbreaks in sub-Saharan Africa (Bastos et al. 2003). It showed wide spread as it could be deduced from our results of serosurvey and disease survey. Serotype SAT2 antibodies were detected in all the investigated states, and these virus serotype was isolated from one outbreak during the course of this study. In conclusion, Sudan is a vast country with different ecosystems and massive diverse animals' species. Our results indicated that FMD was detected in all the seven investigated states. The prevalence picture of the virus serotypes has changed. For instance, the SAT-2 serotype has spread and was involved in most of the recent FMD outbreaks in the country; this coincided with the results that obtained from Khartoum state survey (Raouf et al., 2008; Habiela et al., unpublished data). Further extensive serosurveillance, disease monitoring and phylogenetical analysis of the recent FMDV isolates are required so as to evaluate the real situation of FMD in the country. 5. ACKNOWLEDGEMENTS: We thank Prof. M. Tibin, Director General of ARRC for his support, Ministry of animal resources and fisheries, and all colleagues who helped us in this study. We indebted to the staff of both FMD unit and Department of Virology at CVRL, to Dr. K. Sumption- EUFMD- Rome and Pirbright staff for their

418

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 assistance and to Prof. E. Abu Elzein and Dr. M. Badawi for reading the manuscript. This paper is published by permission of the Ministry of Animal Resources and Fisheries, Republic of the Sudan. 6. REFERENCES [1] Abu Elzein, E. M. E.; Newman, B. J. ; Omer, E. A. & Haroon, B. (1984). The prevalence of serum antibodies to the foot-and-mouth disease virus infection associated antigen (VIA) in camel, sheep and goats of the Sudan, Sudan J. Vet. Res., Vol. 6. [2] Abu Elzein, E. M. E.; Newman, B. J; Crowther, J. R. ; Barnett, I. T. R. & McGrane, J. J.(1987). The Prevalence of Antibodies against foot-and-mouth disease in various species of Sudanese livestock following natural infection, Rev. Elev. Vet. Pays trop., 40(1): 7-12. [3] Abu Elzein, E.M.E. & Crowther, J. R. (1979), Serological comparison of a type SAT-2 footand-mouth disease virus from Sudan with other type SAT2 strains, Bull. Anim. Hlth. Prod. Afr., 27, 245-248, [4] Abu Elzein, E.M.E. (1983), Foot-and-mouth disease in Sudan, Rev. sci. tech. off. Int. epiz. 2 (1), 177-188), [5] Alexandersen, S. ; Wernery, U. ; Nagy, P. ; Frederiksen, T. & Normann, P., 2006, Dromedaries (Camelus dromedarius) are of very low susceptibility to experimental, high dose inoculation with FMDV Serotype O and do not transmit the infection to direct contact camels or sheep International relevance of current research managing, 2006 Session of the Research Group of the Standing Technical Committee of EuFMD, Paphos, Cyprus 17-20 October 2006, App. 24 [6] Alexandersen, S.; Zhang, Z. & Donaldson, A. I. (2002), Review: Aspects of the persistence of foot-and-mouth disease virus in animals—the carrier problem, Microbes and Infection, 4, 1099– 1110. [7] Anon (2007), Quarterly Report of FAO/OIE FMD Reference Laboratory, April-June2007. [8] Bastos, A. D. S., Haydon, D. T., Sangare, O., Boshoff, C. I., Edrich, J. L. & Thomson, G. R. ( 2003), The implications of virus diversity within the SAT 2serotype for control of foot-andmouth disease in sub-Saharan Africa. J. Gen. Virol., 84, 1595–1606. [9] Doel, T.R. (1999), Optimisation of the immune response to foot-and-mouth disease Vaccines. Vaccine 17 (1999) 1767-1771 [10] Eisa, M. &Rweyemamu, M. M. (1977). A note on the epizootiology of foot-and-mouth disease in the Sudan, Bull. Anim. Hlth. Prod. Africa, 25: 108-115. [11] Ferris, N. & Dawson, M. (1988), Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases, Vet. Microbiol. 16, 201–209. [12] Hedger, R. S. (1968), the isolation and characterization of Foot-and-mouth disease virus from clinical normal herds of cattle in Botswana, J. Hyg., 66: 27-36. [13] Kitching, R.P. & Donaldson, A. I. (1987), Collection and transportation of specimens for vesicular virus investigation, Rev. Sci. tech. off. Int. Epiz., 6 (1), 263-272. [14] Larska, M., Wernery, U., Kinne, J., Schuster, R., Alexandersen, G.and S. Alexandersen, (2008), Differences in the susceptibility of dromedary and Bactrian camels to footand-mouth disease virus. Epidemiology and Infection, Published online by Cambridge University Press 08Aug 2008 doi: 10.1017/S0950268808001088 [15] Murphy,F. A. ; Gibbs, E. ; Horzinek, M. & Studdert, M. (1999), Veterinary Virology Text book, 3rd ed. Academic Press, San Diego, California , , pp. 521-527. [16] Patty, R. E. ; Tozzini, F.; Seibold, H. R. & Callis, J. J. (1962), Growth of Foot-and-Mouth Disease Virus in Dispersed Tissue Cells.1. Methods of Production, Can. J. Comp. Med. Vet. Sci.,Vol. 26. [17] Raouf, Y, Habiela, M. & Yagoub, I. (2008), Control of transboundary and export diseases :Foot-and-mouth disease, The first Scientific conference on animal health, production, fisheries and wildlife of Animal Resources Research Corporation (ARRC), 17-21 August 2008 ,Khartoum- Sudan. [18] Roeder, P.L. & Le Blanc Smith, P.M. (1987), Detection and typing of foot-and-mouth disease virus by enzyme-linked immunosorbent assay: a sensitive, rapid and reliable technique for primary diagnosis. Research in Veterinary Science, 43, 225-232. [19] Rweyemamu, M.; Paskin,R.; Benkirane,A.; Martin,V.; Roeder,P. & Wojcichowski, K. (2000), Emergin diseases of Africa and Middle East, Annals, New York Academy of Sciences, Issue TROPICAL VETERINARY DISEASES: CONTROL AND PREVENTION IN THE CONTEXT OF THE NEW WORLD ORDER, December 2000, Volume 916, Pp. 61-70. [20] Shahan, M. S. (1960), A review of current knowledge of the vesicular diseases, Can. Vet. Jour., 1, (10). [21] Snowdon, W.A. (1966), Growth of foot-and-mouth disease virus in monolayer cultures of calf thyroid cells. Nature 210, 1079-1080

419

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [22] Sutmoller , P. ;Barteling , S.S. ; Olascoaga ,R. C. & Sumption, K.J. (2003), Control and eradication of foot-and-mouth disease, Virus Research, 91, 101-144) [23] Wernery, U. & Kaaden, O.R. (2004), Foot-and-mouth disease in camelids: a review, The Veterinary Journal 168, 134–142.

420

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 74

STATUS OF FOOT-AND-MOUTH DISEASE IN PAKISTAN S.M. Jamal1, 2, *, S. Ahmed2, M. Hussain3 and Q. Ali1 1

National Veterinary Laboratory (NVL), Park Road, 44500, Islamabad, Pakistan. Department of Biological Sciences, Quaid-i-Azam University, Islamabad, Pakistan. 3 FAO project (GTFS/INT/907/ITA) for Central Asian Countries, NVL, Islamabad Pakistan.

ABSTRACT Foot-and-mouth disease is a world-wide known disease of economic significance. The disease is endemic in south Asian region. The present study reports observations and spatial distribution of 153 laboratory confirmed field outbreaks taken place during 2002 to 2007 and serotypes of the virus involved. Distribution of different serotypes from outbreaks in Pakistan and samples tested either in Pakistan or sent to WRL-FMD for virus typing during 1952 to 2007 has been shown. Serotype O was found to be the most prevalent serotypes followed by serotype Asia-1 and A. No sample was found positive for serotype C during the period 1996-2007. Mixed infection of either serotypes O and Asia-1 or A and Asia-1 was also detected in some samples. The disease was found to be more prevalent in cattle than buffaloes. Moreover, higher number of outbreaks of the disease was noted between the months of January to March during the period of 2002-07, which may be attributed to the livestock movement in the country due to religious festival, Eidul Azha, in which the animals are slaughtered. 1. INTRODUCTION Foot-and-mouth disease (FMD) is an infectious and highly contagious viral disease of cloven-hoofed animals that causes heavy economic losses to the livestock industry in term of high morbidity in adult animals and mortality in young stock, reduced milk production, loss of work efficiency in draught animals and sanctions on export of animals and animal products due to sanitary measures. The disease is endemic in south Asian region. Historically, the disease was well-known in the IndoPakistan subcontinent but it is not known as to what part/place was first affected. However, occurrence of the disease in pre-partition Punjab from 1900 onward is mentioned in the literature. Only limited and isolated studies, mainly on farm/district level, have been conducted on FMD in Pakistan (Rauf et al. 1981; Riaz et al. 1992; Ahmad et al. 2002; Klein et al. 207; 2008), which do not reflect the true status of the disease in the country. Country-wide systematic study on FMD and serotype involved in disease outbreaks has not been conducted in Pakistan. Therefore, exact situation of the disease in the country is unknown. Aim of the present study is to identify the virus serotypes from FMD field outbreaks taken place during 1952 to 2007, spatial distribution of serotypes responsible for 153 laboratory confirmed cases of the outbreaks between 2002 to 2007, species of the animals affected and seasonal prevalence of the disease. 2. MATERIALS AND METHODS Epithelial samples received from suspected cases of FMD from various parts of the country during 2002-2007 were tested for diagnosis of FMD and virus typing using indirect sandwich ELISA at National Veterinary Laboratory (NVL), Islamabad, Pakistan following Roeder and La Blanc Smith (1987). Some samples were also sent to the World Reference Laboratory for FMD, Pirbright, UK for virus typing. Historical data on FMD between 1952 to 2002 were collected from different sources. The data were analyzed for distribution of serotypes of the virus among positive samples, species affected and to determine association between serotypes and species of animals. Chi square test was applied to know the comparative incidence of serotypes involved in disease outbreaks. Fisher’s exact test was used to know the association between species and serotypes of the virus involved in disease outbreaks. All the analyses were carried out analyzed using statistical software, R (www.rproject.org).

421

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 3. RESULTS Distribution of virus types among the positive samples collected during 1952 to 2007 are shown in Figure 1. Serotype O was detected in 61% of the total samples tested positive. Mixed infection of serotypes O and Asia1 or A and Asia-1 were also detected in some samples. A significant difference in incidence of disease outbreaks due to different FMD virus serotypes was observed in 153 field outbreaks taken place during 2002-2007. Serotype O was found responsible for significantly higher (p<0.0001) number of outbreaks, followed by serotypes Asia-1 and A. However, no significant difference (p>0.05) in disease outbreaks due to serotypes Asia-1 and A was observed. Significantly higher numbers (p<.0001) of cattle were affected in FMD outbreaks compared to buffaloes. However, serotype of the virus was found to be independent of the species of large ruminants. Exotic breeds of cattle and their crosses were found to be more susceptible to FMD than local cattle. Higher number of outbreaks of the disease was noted between the months of January to March during the period of 2002-2007 (Figure 2). Spatial distribution of different serotypes responsible for FMD outbreaks in the country during 2002-2007 is shown in Figure 3. 4. DISCUSSION The study shows that FMD is endemic and widely spread throughout Pakistan. The study reveals that major FMD outbreaks are caused by serotype O virus compared to serotype A or Asia-1. This finding is in agreement with Rweyemamu et al (2008), who reported that serotype O was the most widely prevalent serotype in the world. Asia-1 was first detected by WRL-FMD in 1954 from an epithelial sample collected from Pakistan. Asia-1 is still responsible for disease outbreaks in the country. Serotype C was for the first time detected in Pakistan in 1954. Last outbreak of FMD in the neighboring country, India, was reported in 1995 (Hemadri et al. 2003). Serotype C is probably no longer present is Asia (Rweyemamu et al. 2008) and appears to have disappeared from the world as a whole with last outbreak in Kenya in 2004. Historically, this is the rarest of the FMD serotypes to have occurred in the world. In the present study, cattle were found to be the most affected species followed by buffaloes with only few reports of FMD in small ruminant. This finding may be due to a high likelihood that cattle are affected compared with buffaloes or clinical signs/severity of the disease are les pronounced in buffaloes than cattle, which would result in differential under-reporting of buffaloes cases of FMD. Higher number of FMD outbreaks noted between the months of January to March during the period of 2002-2007 may be due to livestock movement in the country due to religious festival, Eidul Azha, falling from December to February during the reported period. Progressive control of FMD is the most pragmatic approach for Pakistan. The critical components of such an approach should include a national FMD control initiative, continuous surveillance and monitoring of the FMD situation, facility for FMD virus isolation and sub-typing, and availability of an effective vaccine. Some studies on the antigenic characterization of Pakistani isolates originating from Landhi cattle colony, Karachi have been done (Jamal et al. 2007) and such studies on country-wide basis has already been planned as a collaborative project between Government of Pakistan and Danish Government with financial support of an FAO funded project, GTFS/INT/907/ITA. 5. REFERENCES [1] [2] [3] [4] [5] [6] [7]

422

Ahmad et al (2002). Pak Vet J. 22: 201-203. Hemadri et al. (2003). Vet Microbiol. 92: 25-35. Jamal et al (2007). Report submitted to the EU-FMD, Rome. Klein et al. (2007). Virol J. 4:122. Klein et al. (2008). Virol J. 5:53. Rauf et al. (1981). Pak Vet J. 1: 13-14. Riaz et al (1992). Pak Vet J. 12: 86-88.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

1000 900 800

Serotypes

700 600 500 400 300 200 100 0 O

Asia-1

Number

Figure 1. Frequency distribution of different serotypes from suspected cases of FMD submitted to the laboratory during 1952 to 2007

40 35

Number of outbreaks

30 25 20 15 10 5 0 Jan

Feb

March

April

May

June

July

Aug

Sept

Oct

Nov

Dec

Month

Figure 2. Seasonal distribution of laboratory confirmed FMD outbre aks taken place during 2002-2007 (n=153)

423

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Figure 3: Spatial distribution of serotypes from laboratory confirmed cases of FMD during 20022007

424

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 75

THE RELATION ANTIBODY AND PROTECTION AFTER FOOT-AND-MOUTH DISEASE VACCINATION CANNOT BE STANDARDISED A. Dekker1*, N. Goris2, S.M. Jamal3, Y. Li4 1 2

Central Veterinary Institute of Wageningen UR, P.O. Box 65, 8200 AB Lelystad, The Netherlands Epizootic Diseases Section, Virology Department, Veterinary and Agrochemical Research Centre, Groeselenberg 99, 1180 Brussels, Belgium. 3 National Veterinary Laboratory, Park Road, 45500, Islamabad, Pakistan 4 Institute for Animal Health (IAH), Ash Road, Pirbright, Surrey GU24 0NF, UK

1. INTRODUCTION Many studies have been performed into the relation antibody response and protection after footand-mouth disease (FMD) vaccination (Black et al. 1984; Pay et al. 1987; Pay et al. 1992). In every study a correlation was observed with the antibodies induced by the vaccine used in the study and protection observed after challenge. In the studies the methodology was never the same, e.g. the type of cells used in the virus neutralisation test (VNT), some studies were using ELISA and the statistical analysis. For countries not in the position to perform challenge experiments a standardised relation between antibody response after FMD vaccination and protection would be very useful. In the current study we try to standardise antibody level detection for FMD type O Manisa in two different ways, first by using a standardised commercial type PrioCHECK® FMDV Type O ELISA and secondly by inclusion of a standard 4 week post vaccination serum from a cow vaccinated with Cedivac® O Manisa FMD vaccine in both the ELISA and the VNT. 2. MATERIALS AND METHODS 2.1 Data cattle potency tests Sera were available from 6 O Manisa potency tests performed in Lelystad (The Netherlands). One of these potency tests were performed with Al(OH)3 adjuvanted vaccine in which the cattle were challenged 3 weeks after vaccination as described in the European Pharmacopoeia. In the other 5 potency tests double oil adjuvanted vaccines were tested, and because the antibody response to oil emulsion vaccines is a bit slower it was decided to challenge the cattle 4 weeks after vaccination. Sera were available from 10 O Manisa potency performed by the Belgium national laboratory (Brussels) at the animal facilities of FGI-ARRIAH using the same batch of a double oil emulsion O Manisa vaccine (Goris et al. 2007). One set of sera from a potency test performed in Pirbright. Serological tests: Sera were titrated for this study in the national reference laboratory that had performed the potency tests, both in the PrioCHECK® FMDV Type O ELISA and the VNT. In each test a titration of a standard serum was included. The units of antibody were calculated by subtracting the 10log titre of the positive control from the observed titre. Statistical analysis: Titres of the control serum were compared using ANOVA followed by a pairwise comparison using Tukey honest significant difference. Serological responses were fitted by logistic regression. In each analysis the contribution of the laboratory performing the tests was included if this resulted in a better fitting model. The latter was tested by the likelihood ratio test. All statistical analyses were performed in R (www.r-project.org). 3. RESULTS Significant differences were found in the titre of the control serum obtained in the various laboratories, in both the ELISA and the VNT. In the ELISA significant (p<0.01) lower titres in Belgium (titre = 1.72 10log) were found in comparison with the titre found in the Netherlands (titre

425

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 = 1.82 10log). In the VNT the titre found in Pirbright was significantly (p<0.01) different from the result in both other laboratories. The observed mean titres are given in Table 1. Only a small difference in the mean titre found in protected and non-protected cattle, this is especially true for the ELISA titres found in Brussels. In both the ELISA and the VNT always a very significant (p<0.01) influence of the location was found on the relation between antibodies and protection, but the slope of the relation between antibodies and protection was the same in each laboratory. In the ELISA the slope of the relation between antibody concentration and protection was lower when the titres of the sera collected at the day of challenge were analysed in comparison with the titres found at the day of challenge (for experiments performed by Brussels and Pirbright cattle were challenged at three weeks, so the titre at three weeks and the day of challenge were the same). The slope of the relation between antibodies and protection was steeper when analysing the results obtained in the VNT in comparison with the results obtained in the ELISA. Table 1: Results obtained in various laboratories Brussels

Lelystad

Pirbright

Number of PD50 exp

Number of vaccinated cattle

150

Percentage protected cattle

All cattle Mean ELISA titre 3 weeks post vaccination

1.6

1.9

1.6

Mean Units of antibodies by ELISA 3 weeks post-vaccination

-0.1

0.1

-0.2

Mean ELISA titre at challenge

1.6

2.1

1.6

Mean Units of antibodies by ELISA at challenge

-0.1

0.3

-0.2

Mean VNT titre 3 weeks post vaccination

1.7

1.3

Mean Units of antibodies by VNT 3 weeks post-vaccination

0.4

0.3

-1.2

Mean VNT titre at challenge

1.7

1.9

1.3

Mean Units of antibodies by VNT at challenge

0.4

0.5

-1.2

1.4

1.6

1.2

Non-Protected cattle Mean ELISA titre 3 weeks post vaccination Mean Units of antibodies by ELISA 3 weeks post-vaccination

-0.3

-0.5

Mean ELISA titre at challenge

1.4

1.8

1.2

Mean Units of antibodies by ELISA at challenge

-0.3

-0.5

Mean VNT titre 3 weeks post vaccination

1.4

1.0

Mean Units of antibodies by VNT 3 weeks post-vaccination

0.1

-1.5

Mean VNT titre at challenge

1.4

1.5

1.0

Mean Units of antibodies by VNT at challenge

0.1

-1.5

1.6

2.1

1.6

Protected cattle Mean ELISA titre 3 weeks post vaccination Mean Units of antibodies by ELISA 3 weeks post-vaccination

0.3

-0.1

Mean ELISA titre at challenge

1.6

2.1

1.6

Mean Units of antibodies by ELISA at challenge

0.4

-0.1

Mean VNT titre 3 weeks post vaccination

1.7

1.9

1.5

Mean Units of antibodies by VNT 3 weeks post-vaccination

0.4

0.6

-1.1

Mean VNT titre at challenge

1.7

2.0

1.5

Mean Units of antibodies by VNT at challenge

0.4

0.6

-1.1

Figure 1 shows the relation found between antibody titre in the VNT at three weeks post vaccination and protection. The maximum difference between the curves at the 50% point is 0.55, so close to a 4 fold difference in titre. When correcting the titre by using the titre of the control serum we get the curves shown in Figure 2. The distance between the laboratory in Brussels and Lelystad becomes smaller (0.17), but the difference between Pirbright and the other two laboratories becomes much bigger. A similar result is obtained when using the antibody titres and

426

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 units obtained by ELISA, although the difference between the laboratories is then smaller the maximum difference at the 50% point is still 0.6. 4. DISCUSSION The current study was set up to standardise the serology for FMD type O Manisa and the relation with protection, either by a standard test (PrioCHECK® FMDV Type O ELISA), by using a standard control serum to calculate units of antibody or by using both. The analysis shows a significant difference between laboratories. There are several explanations possible for this phenomenon, differences in challenge dose in the different laboratories, differences in match between challenge virus and vaccine virus, the difference in time of challenge, differences in adjuvant used and the difference in 140/12 S composition of the vaccine. In a recent comparison between pigs vaccinated intramuscularly and intradermally there was also a significant difference between both application methods and the relation between neutralising antibody titre and protection against virus shedding detected by mouth swabbing (Eblé non published data). The relation between antibody response and protection was valid within a laboratory. In each laboratory only vaccines from one producer had been tested, so producers can determine the relation between antibodies and protection for their own vaccine but that it is not possible to produce a more general estimate of this relation. 5. CONCLUSIONS Standardisation of serological tests relating protection to antibody response is not yet possible Producers should set-up their own relation between antibody response and protection 6. RECOMMENDATIONS

0.0

0.2

0.4

Protection

0.6

0.8

1.0

More data from more sources should be analysed to finalise the conclusion Mathematical optimisation studies should be performed to provide guidance on how producers most effectively can relate antibody response to protection

0.0

0.5

1.0

1.5

2.0

2.5

Log VNT antibody titre

Figure 1: Comparison of protection and the VNT titre of sera collected 3 weeks post vaccination. In solid the results of the laboratory in the Netherlands, dashed the results obtained in Brussels and dotted the results obtained in Pirbright.

427

0.6 0.4 0.0

0.2

Protection

0.8

1.0

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

Log VNT units

Figure 2: Comparison of protection and the VNT units of sera collected 3 weeks post vaccination. In solid the results of the laboratory in the Netherlands, dashed the results obtained in Brussels and dotted the results obtained in Pirbright. It would be interesting to analyse more data from more laboratories performing O Manisa vaccination and challenge experiments. 7. REFERENCES [1] Black, L., Francis, M. J., et al. (1984). "The relationship between serum antibody titres and protection from foot-and-mouth disease in pigs after oil emulsion vaccination." J Biol Stand 12(4): 379-89. [2] Goris, N., Merkelbach-Peters, P., et al. (2007). "European Pharmacopoeia foot-and-mouth disease vaccine potency testing in cattle: between test variability and its consequences." Vaccine 25(17): 3373-9. [3] Pay, T. W. F. and Hingley, P. J. (1987). "Correlation of 140S antigen dose with the serum neutralizing antibody response and the level of protection induced in cattle by foot-and-mouth disease vaccines." Vaccine 5(1): 60-4. [4] Pay, T. W. F. and Hingley, P. J. (1992). "Foot-and-mouth disease vaccine potency tests in cattle: the interrelationship of antigen dose, serum neutralizing antibody response and protection from challenge." Vaccine 10 (10): 699-706.

428

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 76

STUDY OF IMMNUNOBIOLOGIC CHARACTERISTICS OF TYPE ASIA-1 NO. 1987/AMURSKY/2005 FMD VIRUS

A. Timina, S.R Kremenchugskaya., V.V. Mikhalishin, V.I. Diev, V.V. Borisov, V.M Zakharov. OIE Regional Reference laboratory for FMD, Federal Centre for Animal Health, Vladimir, Russia

FMD is an acute contagious disease of cloven-hoofed animals caused by the agent with antigenically distinct serotypes and intratype variants. Convalescence from FMD or vaccination with one of FMD virus types provide no protection against the other serotypes. Moreover, the protection can be insufficient against other subtypes within one type. In this view the study of matching between production and field strains is critical for the disease control. In June 2005 an FMD outbreak caused by type Asia-1 FMD virus was reported in unvaccinated cattle in the village of Busse, Svobodnensky Rayon, Amurskaya Oblast, Russian Federation. Hereafter, in 2005-2006 FMD outbreaks caused by this serotype were registered in the Khabarovsky and Primorsky Krais of the Far East Region, in the aimak of Dornod in the east of Mongolia and in the Chitinskaya Oblast of the Sibirsky Federal Okrug. The Asia-1 No. 1987/Amursky/2005 type FMD virus submitted to the OIE Regional Reference Laboratory for FMD (FGI “ARRIAH”, Vladimir) was adapted to the primary and continuous cell cultures (SP, PGSK-30, IB-RS-2, BHK-21) and to guinea-pigs by 4-6 passages. The study of naturally susceptible animal sensitivity to the virus showed that primary aphthae at the site of inoculation appeared in cattle 48 hours after the inoculation and in pigs and sheep 72-96 hours after the inoculation. Considering that the FMD vaccine from Shamir 3/89 strain is recommended for the immunization of animals against type Asia-1 FMD virus in the world, r1 value was studied using microneutralization test to define the antigenic matching between the epidemic virus and production virus strain. The test was conducted using 2 series of sera from cattle vaccinated with monovalent Asia-1 Shamir 3/89 FMD vaccines against 100 TCID50/50 µl of the homologues production strain and correspondent dose of the field isolate. The r1 of the epidemic Asia-1 No.1987/Amursky/2005 FMDV was found to be 0.25 and 0.18 against production strain Asia-1 Shamir 3/89 and 0.25 against the epidemic FMDV strain Asia-1/Mongolia/2005 (using 2 series of sera). The data interpretation according to Paton et al. (2005) suggests that the field isolate differs from the production strain at r1<0.3 and the vaccine produced from this strain cannot protect animals from the circulating virus. The experiment on cross-infection of cattle vaccinated with monovalent sorbated vaccines against FMDV of A, O, Asia-1 types, with FMD virus of type Asia-1 (production and epidemic strains) demonstrated that PD50 of the inoculation dose was 10.5 against the production Asia-1 strain and 3.4 against epidemic Asia-1 No.1987/Amursky/2005 isolate. This fact indicates that the vaccine is 3 times less immunogenic against the circulating epidemic virus. Thus, performed investigations on the examination of immunobiological properties of type Asia-1 No.1987/Amursky/2005 FMD virus strain showed the significant antigenic difference between circulating virus and production strain of this type. Thereby, in May 2006 the Asia-1 No.1987/Amursky/2005 FMD virus strain was deposited to the All-Russian State Collection of Microorganism Strains, used in veterinary medicine and animal production. It is currently used as a production strain for vaccine and diagnostic preparation manufacture. The work is performed under the ISTC Project No.2538 р.

429

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

430

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 77

PIGLETS WITH MATERNALLY DERIVED ANTIBODIES CAN BE VACCINATED AT 2 WEEKS OF AGE G. Chénard1*, P. Selman1, P. Eblé2, N. Stockhofe2 and A. Dekker2 1

Products Division, Animal Sciences Group of Wageningen University and Research Centre P.O. Box 65, 8200 AB Lelystad, The Netherlands 2 Central Veterinary Institute of Wageningen University and Research Centre, P.O. Box 65, 8200 AB Lelystad, The Netherlands

ABSTRACT We investigated to what extent maternally derived antibodies (MDA) interfere with FMD vaccination in order to determine when piglets should be vaccinated in emergency situations. Groups of piglets with MDA were vaccinated at different time points following birth and the antibody titres to FMD virus were measured using virus neutralisation tests (VNT). We used 50 piglets from 5 sows that had been vaccinated three times intramuscularly in the neck during pregnancy using a high potency trivalent Cedivac-FMD vaccine containing strains of FMD virus serotypes O, A and Asia1. Four groups of 10 piglets were vaccinated intramuscularly in the neck at 2, 4, 6 or 8 weeks of age using a 6 PD50 monovalent Cedivac-FMD vaccine (serotype A). One group of 10 piglets with MDA was not vaccinated and another group of 10 piglets without MDA was vaccinated at 2 weeks of age. Sera samples were collected and (serotype A) antibody titres were determined using VNT. The antibody responses of piglets with MDA vaccinated at 6 or 8 weeks of age were similar to the responses of piglets without MDA vaccinated at 2 weeks of age and reached the level corresponding with 6 PD50 at 3-4 weeks post vaccination. The antibody responses of piglets with MDA vaccinated at 2 or 4 weeks of age were lower compared to the responses of piglets without MDA. Nevertheless, a partial immune response is better than a further decline in antibody levels in an emergency situation. 2. INTRODUCTION Vaccination and revaccination are effective intervention tools in combating Foot-and-Mouth Disease (FMD). Revaccination is especially relevant for the pig industry given the relatively short gestation period of the species. Passive immunization by maternally derived antibodies (MDA) interferes to a varying degree with active immunization against FMD. It has been shown that the degree of suppression of the immune response is directly proportional to the MDA titre at the moment of vaccination (3). Factors such as the vaccination regimen of the sow, the antigenic payload of the vaccine, the type of vaccine adjuvant, the amount of colostrum uptake and the general health status and number of the piglets in a litter affect the quantity of MDA that are ultimately transferred to individual animals. Given the high turnover of animals in the porcine population, it has been suggested that the immaturity of the immune response of piglets makes vaccination of the sow during pregnancy a more effective method to protect the litter (4). In the knowledge that MDA interferes with the immune response of young animals, it has been suggested that effective immunization can only be achieved in the absence of MDA (1). In piglets, the MDA levels drop to about 1-3% of their initial value after about 60 days (5). It is important to realize that the observed drop of MDA levels in piglets is largely due to the rapid increase in blood volume. Vaccination of piglets with MDA has generally been recommended to be performed only after levels are very low. Indeed, in order to circumvent interference of the immune response due to MDA, vaccination of piglets born from vaccinated sows can be delayed until approximately 8-10 weeks of age. Such a strategy however is more suited to increasing the herd protection levels in a number of months and is less suited to emergency or other situations with imminent threat of

431

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 exposure to live virus. In an emergency situation, it is of paramount importance to maintain the highest possible levels of herd protection. Any delay in vaccination of piglets with MDA results in a further drop in herd protection levels. However, vaccination in the presence of high levels of MDA may be ineffective. The optimal vaccination moment is likely dependent on a number of factors including the quality and potency of the vaccine being administered. For this reason, it is advisable to generate data to support the decision-making process in an emergency situation. 2. MATERIALS AND METHODS 2.1 Vaccines The vaccines administered were manufactured by the Products Division of the Animal Sciences Group (ASG), Lelystad, The Netherlands and contained inactivated, purified FMD virus antigens using a mineral oil as adjuvant in a double oil emulsion formulation. The pregnant sows were vaccinated with a trivalent Cedivac-FMD vaccine formulated to contain a high antigen payload. This payload corresponded with approximately 15 times the antigen required for one 50% protective dose (PD50) for each of the strains: A Turkey 14/98 (A Iran96 analogue), Asia1 Shamir and O1 Manisa. A monovalent Cedivac-FMD vaccine containing the strain A Turkey 14/98 was used to vaccinate the piglets. The antigen payload of this monovalent vaccine corresponded with a potency of at least 6 PD50. 2.2 Animals and husbandry The 5 pregnant sows in this study were primipari and purchased (at day 21 of gestation) from Van Beek SPF Varkens BV (Putten, The Netherlands). The sows were individually housed in farrowing boxes with tenderfoot floor in the animal experimental facilities of ASG and were fed rationed portions appropriate for their age. Each of the 5 vaccinated sows had between 11 and 13 piglets; 10 piglets of each litter were used in this study. The piglets were allowed colostrum uptake and distributed evenly over 5 groups so that each group had 2 piglets from each sow. 2.3 Vaccination regime The sows were subject of a vaccine safety study in pregnant animals following the European Pharmacopoeia monograph for FMD vaccines (2). The animals were vaccinated intramuscularly in the neck in each trimester of pregnancy using a double dose (first trimester) or regular dose (second and third trimesters) of the trivalent vaccine described earlier. Four groups of 10 piglets were vaccinated intramuscularly in the neck at 2, 4, 6 or 8 weeks of age using a dose volume of 2 ml of monovalent A Turkey Cedivac-FMD vaccine described earlier. In addition, one group of 10 piglets with MDA was not vaccinated. Another group of 10 piglets without MDA was vaccinated at 2 weeks of age. Both groups served as a control group. 2.4 Virus neutralization tests Blood samples were collected from all piglets periodically and antibody titres in the resulting sera were assessed against homologous FMD virus (A Turkey) by virus neutralisation tests (VNT) performed as described in the OIE manual (6) using secondary porcine kidney cells. Blood samples were collected until 6 weeks post vaccination. 2.5 Antibody titre expected for 6 PD50 vaccines Using the virus neutralization test described above, a mean antibody titre of log10 1.50 for FMD virus strain A Turkey has previously been observed 28 days post vaccination in adult pigs vaccinated with a 6 PD50 vaccine. 3. RESULTS All sows had normal pregnancies and normal size litters. Following colostrum uptake, the piglets born from the immunized sows had MDA against A Turkey with a mean antibody titre of log10 1.73, which is above the VNT titre expected for a 6 PD50 vaccine. The rate of decline of the MDA was similar in all the groups (Figure 1). MDA titres declined to below the detection level of the VNT (log10 0.30) at approximately 7 weeks of age. The half-life of antibodies was approximately 16 days when corrected for increase in body weight. The antibody profiles are shown again in Figure 2 but shifted to the same moment of vaccination to facilitate comparison of immune responses. Piglets without MDA vaccinated at two weeks of age showed a normal immune response and antibody levels reached the 6 PD50 level at 4 weeks. The

432

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 neutralising antibody responses of piglets with MDA vaccinated at 6 or 8 weeks of age were similar to the responses of piglets without MDA vaccinated at 2 weeks of age and reached the 6 PD50 level 3-4 weeks post vaccination. The neutralising antibody responses of piglets with MDA vaccinated at 2 or 4 weeks of age were supressed compared to the responses of piglets without MDA. The levels of MDA continued to decline at the same rate for 2 more weeks after vaccination and by 3 weeks post vaccination, the mean VNT titre had stabilized or increased marginally. The neutralising antibody response of piglets vaccinated at 4 weeks of age was reasonable and the mean titre approached the 6 PD50 level at 6 weeks post vaccination. In the case of piglets vaccinated at 2 weeks of age, the immune response resulted in antibody levels that remained below the 6 PD50 level. All vaccinated piglets had mild to moderate injection site reactions, transient temperature increases and histology of vaccination sites revealed no unusual findings. 4. DISCUSSION The objective of this study was to determine when piglets with MDA should be vaccinated in an emergency situation. The main obstacle to effective vaccination is undoubtedly the presence of maternally derived antibodies. The results presented here confirm that MDA levels at the time of vaccination are directly proportional to the degree of interference to active immunization. In this study, the MDA titres in piglets declined to below the detection level of the VNT after approximately 7 weeks, which is in agreement with earlier findings. This explains why groups of piglets (with MDA) that were vaccinated at either six or eight weeks of age had very good immune responses. It is therefore tempting to designate this time point as the “best” moment to vaccinate. However, delaying the vaccination of piglets until protective antibody levels are very low allows the herd protection level to drop to undesirable levels. This is due to the sheer numbers of piglets born and the dilution effect this has on the herd. The optimal vaccination moment it seems, is at the trade-off point where levels of maternally derived antibodies still offer a reasonable level of protection but still allow a reasonable immune response following vaccination. When observing the antibody level profiles of piglets vaccinated at two weeks of age, it is interesting to note that following vaccination, protective antibody levels declined for a further two to three weeks before reversing. This time gap is important as any further delay in vaccination of piglets allows an undesirable window of susceptibility to open. In order to maintain the highest herd immunity levels possible in the farrowing period, it is necessary to prevent drops in levels of protective antibodies in piglets. This means vaccinating as early as possible and brings us back to the problem of vaccination in the presence of antibodies of maternal origin. In fact, it is not the vaccination that is the problem, but rather achieving an effective immune response. There is only limited experimental evidence available of effective vaccination in the presence of maternally derived antibodies. However, it is reasonable to expect that vaccines of good quality can, to a certain limit, overcome maternally derived antibodies to induce a good immune response. FMD vaccines using an oil emulsion adjuvant, such as CedivacFMD, are known to be effective in pigs. This study shows that piglets respond to vaccination at 2 weeks of age and that the decline in levels of protective antibodies in piglets can be stopped by vaccination using Cedivac-FMD at 2 weeks of age. 5. CONCLUSIONS The immune system of piglets is capable of responding to vaccination at 2 weeks of age. The immune response following vaccination of 2 week old piglets was partially suppressed by the presence of MDA. The decline in levels of protective antibodies (and the corresponding decline in herd protection levels) can be stopped by vaccination of piglets at 2 weeks of age. 6. REFERENCES [1] Chappuis, G. 1998. Neonatal immunity and immunisation in early age: lessons from veterinary medicine. Vaccine, 16(14-15): 1468-1472. [2] European Pharmacopoeia. 2006. Foot-and-mouth disease (ruminants) vaccine (inactivated) 04/2005:0063, 5th Edition. [3] Francis, M.J. & Black, L. 1984. Effect of the sow vaccination regimen on the decay rate of maternally derived foot-and-mouth disease antibodies in piglets. Res Vet Sci., 37(1): 72-76.

433

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [4] Francis, M.J. & Black, L. 1986. Response of young pigs to foot-and-mouth disease oil emulsion vaccination in the presence and absence of maternally derived neutralising antibodies. Res Vet Sci., 41(1): 33-39. [5] Kitching, R.P. & Salt, J.S. 1995. The interference by maternally-derived antibody with active immunization of farm animals against foot-and-mouth disease. Bri. Vet. J., 151(4): 379-89. [6] OIE (Office International des Epizooties/World Organisation for Animal Health). (2004).Footand-mouth disease. In: Manual of standards for diagnostic tests and vaccines. 5th edn. Ed. OIE Standards Commission. Paris, France, Office International des Epizooties. Chapter 2.1.1, version adopted May 2006.

2.40

2.10

A Turkey Log 10 VNT Titre

1.80 VNT titre 1.50 in adult pigs 4 weeks postvaccination with a 6 PD50 vaccine

1.50

1.20 Piglets vaccinated at age of:

0.90

2 weeks 4 weeks 6 weeks 8 weeks not vaccinated 2 weeks no MDA

0.60

0.30

0.00 0

Age of piglets in weeks

Figure 1: Antibody profiles (FMD strain A Turkey 14/98) in piglets with MDA vaccinated at 2, 4, 6, 8 weeks of age. Profiles are shown real time.

2.00 1.80

Mean VNT titre 1.73 in piglets after colostrum uptake

1.60 VNT titre 1.50 in adult pigs 4 weeks postvaccination with a 6 PD50 vaccine

A Turkey Log 10 VNT Titre

1.40 1.20 1.00 Piglets vaccinated at age of:

0.80

2 weeks 4 weeks 6 weeks 8 weeks not vaccinated 2 weeks no MDA

0.60 0.40 0.20 0.00

-12

-10

-8

-6

-4

-2

Weeks post vaccination

Figure 2: Antibody profiles (FMD strain A Turkey 14/98) in piglets with MDA vaccinated at 2, 4, 6, 8 weeks of age. Profiles are shifted to allow comparison of responses.

434

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 78

IN VIVO AND IN VITRO TESTS TO DETECT NON-STRUCTURAL PROTEINS IN FOOT AND MOUTH DISEASE VACCINES. M. Trotta1, 2, D. Compaired1, O. Zabal1, M. Pérez-Filgueira1, 2, J. La Torre3 and N. Fondevila1 ¹Instituto de Virología, CICVyA, INTA-Castelar, Buenos Aires, Argentina; ²Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina; ³Instituto de Ciencia y Tecnología “César Milstein”, CONICET

ABSTRACT The detection of antibodies against non-structural proteins (NSP) is the main methodology used for differentiation between vaccinated and infected animals. This strategy assumes that only active forms of the virus may induce antibodies against NSP since they are synthesized only during replication cycle. For this reason, commercial vaccines based on inactivated particles should prove to be free of NSP contaminants that are synthesized during the production process and may persist as residues in the final antigenic preparation. The OIE recommends a protocol to test the presence of NSP contaminants based on the repeated vaccination in cattle and the detection of anti-NSP antibodies up to 120 days post-primo vaccination. Here alternative in vivo and in vitro assays are presented with the aim of avoiding the use of cattle in the assays and reduce the time required for determination of NSP contaminations in FMDV antigenic preparations and vaccines. 1. INTRODUCTION Foot-and-mouth disease virus (FMDV) is considered the most contagious viral pathogen in mammals. Outbreaks of the disease are often responsible for severe economic problems in countries affected due, among other reasons, to serious restrictions imposed to livestock trade and most of the industrial activities derived from it. The Office International des Epizooties (OIE) Terrestrial Animal Health Code stipulates a series of criteria for regaining FMD-free status following an outbreak as well as for status of disease under regular vaccination (OIE, 2006). In such cases, all immunized animals should prove to be free of antibodies (Ab) against FMDV non-structural proteins (NSP), as they reflect replication of the virus in the natural hosts and thus, its circulation in the field. The same criteria applies for differentiation between infected and vaccinated animals in areas under systematic vaccination; in such conditions it is critical to guarantee that vaccines formulated using inactivated FMD virions do not contain NSP which may induce Ab against them. In fact, the OIE currently recommends controlling FMD vaccines through the detection of anti-NSP Ab induced after repeated vaccination. Such immunization protocols are demanding regarding time and expenses: they should be performed in cattle and comprise 3 consecutive inoculations every 30 days with a final testing at 120 days post primo-vaccination (dpv). In this work, in vivo and in vitro procedures are presented for detection of the FMDV NSP, with the aims of (i) assessing presence of anti-NSP Ab in animal models other than cattle and (ii) developing methods to detect FMDV NSP in vaccine antigens at different stages of production by enzyme-immuno analysis. 2. MATERIALS AND METHODS 2.1 Animal models for the detection of immune response against FMDV NSP. Detection of immune response against FMDV-NSP in commercial vaccines: Calves, aged 6-12 months, were divided into 8 groups (n=3 or 4) and received 3 intramuscular doses of the vaccines at one-month intervals. Serum samples were obtained before each vaccination and 1 month after the last vaccination. Dose-response experiments: Groups of 3 bovines and 5 mice were vaccinated with experimental vaccines (water-in-oil single emulsions 50%-50%), containing different concentrations of a recombinant 3ABC (Robiolo et al., 2006) in the aqueous phase. Cattle received 3 vaccinations with 30 day intervals and mice followed the same protocol, including antigen amount per dose, but with 15 days intervals.

435

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

NSPs antibody testing I-ELISA 3ABC Panaftosa (Bergman et al, 2000) and 3AB1 ELISA (Nanni et al, 2005) were used for the detection of anti-NSP antibodies following protocols described by the manufacturer and previous references. Sera from the experiments carried out in cattle were analyzed with the both kits. Mice serum samples were analyzed with 3AB1 ELISA modified to detect mouse antibodies (cut off OD 0.162 at 492 nm). 2.2 Development of immunoassays to detect the presence of NSP in FMDV vaccine antigens. Production of mAb MVT3AB1 Water in oil emulsions containing 30 µg of purified 3AB1 protein (Nanni et al, 2005) were prepared and used to vaccinate BALB/c mice (8 weeks old) by the intraperitoneally route (i.p.). Mice were boosted (30 µg of 3AB1 per dose) 3 times at 2-week intervals and final immunizations were given 4, 3 and 2 days before fusion procedures. After production, cloning and subcloning of hybridomas, ascites fluid was obtained in BALB/c mice following standard procedures and stored at -20ºC until use. The resulting MVT3AB1 mAb (IgG2A isotype) was characterized by indirect ELISA, by measuring the ability of the mAb to detect the recombinant 3AB1 adsorbed to the plates, being able to detect 0,83 ng of recombinant 3AB1 in dilutions up to 1:364,500 (data not shown). Development of Capture NSP ELISA ELISA plates (Immulon 2HB, Dynatech) were coated with the MVT3AB1 mAb diluted 1:4000 in coating buffer (15 mM Na2CO3–35 mM NaHCO3, pH 9.6). After overnight incubation at 4ºC, plates were washed and blocked with PBST-OVA 1%. Recombinant proteins 3ABC or 3AB1 were diluted in PBST-OVA 1% and each dilution added to duplicate wells. Plates were incubated for 1h at 37ºC, washed 3 times with PBST and a polyclonal guinea pig anti-3AB1 serum was added as detector and incubated for 30 min. at 37ºC. After washing with PBST, HRPO-conjugated goat anti-guinea pig (KPL) diluted 1:2000 was added to the plates and incubated for 30 min. at 37ºC. After thoroughly washing, the reaction was developed using a color substrate containing 4 mg/ml OPD-0.08% H2O2 in 0.05 M phosphate-citrate buffer pH 5.0. The reaction was allowed to develop for 2 min at room temperature, and then stopped by the addition of H2SO4 3N and read at 492 nm in an ELISA microplate reader (Multiskan EX, Thermo Electron Corp.). 3. RESULTS Antibody induction against FMDV NSP for some vaccines is only evident after repeated immunizations. A set of commercial vaccines, all of them already assayed for NSP antigens by the Argentinean sanitary authorities (SENASA), were used to vaccinate cattle following the OIE recommended protocol. Antibody responses to FMDV-NSP were detected by an in-house indirect 3AB1 ELISA (Nanni et al, 2005) and the I-ELISA PANAFTOSA (Bergman et al, 2000). Results for both assays were expressed as number of negative (white), positive (light grey) or doubtful (dashed bars) animals for anti-FMDV NSP antibodies, as determined by the corresponding cut-off levels of each ELISA (figure 1). In all cases, results coincided with those obtained by SENASA, showing that the repeated vaccination protocol proposed by the OIE is adequate to detect variable levels of the NSP existing in commercial FMDV vaccines since anti-NSP antibodies were only evident after the second or third vaccination for some of the vaccines tested. Cattle induced antibodies against recombinant 3ABC protein at 90 dpv and after 3 immunizations with doses containing at least 42 ng of the protein. To test the minimum amount of a recombinant FMDV NSP (3ABC) necessary to induce detectable humoral responses in cattle, experimental oil vaccines were prepared containing 670, 170, 42.5, 10.6, and 2.6 ng/dose of recombinant 3ABC and applied following a repeated vaccination scheme. Five groups of 3 animals received 3 vaccinations at 0, 30 and 60 days post-primo vaccination (dpv) with each of the 3ABC doses and serum samples were taken monthly up to 90 dpv, and analyzed by 3AB1 ELISA and I-ELISA PANAFTOSA as described before. Results in figure 2, show that vaccines containing 670 and 170 ng of 3ABC were detected after the first inoculation by both assays. Doses containing 42 ng of 3ABC were also detected but only after the third vaccination; doses below 42 ng were undetectable for both NSP-Ab detection assays.

436

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Mice induced antibodies against recombinant 3ABC protein after 3 immunizations with doses containing at least 42 ng of the protein but after only in 45 days. Following the model described above for cattle, 5 experimental vaccines were prepared using the same amount of 3ABC NSP per dose as described above. Groups of mice (n=5) were inoculated each with one of the 5 doses and received 3 vaccinations over a period of 30 days (at 0, 15 and 30 dpv). Serum samples were taken at 0, 15, 30, 45 and 60 dpv and analyzed by ELISA 3AB1. As described for cattle, the 2 highest doses were detected soon after the first inoculation while that corresponding to 42 ng was only detected after the third inoculation, in this case at 45 dpv (figure 3). Again, doses below 42 ng did not induced antibody levels that were detectable for the ELISA 3AB1 assay up to 60 dpv. A mAb-based indirect capture ELISA is capable of detecting up to 1,56 ng of both 3AB1 and 3ABC recombinant proteins. With the aim of producing an in vitro test capable of detecting trace amounts of FMDV NSP in antigenic preparations, an indirect sandwich ELISA was developed and tested. The assay was designed using a high titer 3AB1-specific mAb developed at our lab, to capture different amounts of 3AB1 or 3ABC recombinant antigens, followed by detection using a polyclonal antibody against 3AB1. Results presented in figure 4, show that the assay was able to detect up to 1.56 ng for any of the recombinant proteins with a coefficient of variation < 10% after ten repetitions. Mockinfected supernatants from BHK-21 cells were always below the cut-off OD (data not shown). 4. CONCLUSIONS 1. Humoral responses to 3ABC/3AB NSP were positive for 5 out of 7 vaccines tested in cattle following OIE recommended procedure and 3 of these vaccines developed anti-NSP antibodies only after the second immunization (detected at 60 dpv.). This supports the utility of the multiple vaccination protocol to detect NSP contaminations in current commercial FMDV vaccines. 2. Dose-response experiments performed in cattle using water in oil emulsions including different amounts of recombinant 3BAC, indicate that ~ 40 ng is lowest protein amount detected at 90 dpv and after 3 immunizations. Mice were able to induce specific antibodies with the same amount of recombinant 3BAC and number of immunizations but in a shorter period of time (45 dpv) and thus, opening the possibility to obtain the same results in a shorter period of time. 3. Preliminary results indicate that an indirect capture ELISA based on a anti-3AB1 mAb developed at our lab, was able to detect ~ 1.5 ng of recombinant versions of 3AB1 and 3ABC proteins. These initial results indicated that the in vitro assay was able to detect ~ 25 times less recombinant 3AB1 or 3ABC proteins than both animal models. This assay would then represent a useful tool to control vaccine manufacturing at different stages of the production process and before the final formulation of large amounts of the vaccine. Experiments are currently in progress to asses the ability of the assay to detect these recombinant proteins and naturally developed FMDV NSP in different antigenic preparations including final vaccine products. 5. BIBLIOGRAPHY [1] Bergmann, I., Malirat, V., Neitzert, E., Beck, E., Panizzutti, N., Sánchez, C., and Falczuk, A. Archives of Virology 145: 473-489 (2000) [2] Nanni M, Alegre M, Compaired D, Taboga O, Fondevila N. J Vet Diagn Invest. 17: 248-51 (2005) [3] Organizacion Internacional de Epizootías. International Animal Health Code – CHAPTER 2.2.10. FOOT AND MOUTH DISEASE (2006). http://www.oie.int/eng/normes/mcode/code2006_back/en_chapitre_2.2.10.htm [4] Robiolo B, Seki C, Fondevilla N, Grigera P, Scodeller E, Periolo O, La Torre J, Mattion N. Vaccine 24: 997–1008 (2006).

437

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Vaccine 2-ELISA 3AB1

Vaccine 1-ELISA 3AB1 2

POS

NEG

No. of animals

POS

NEG

0 0 DPV

30DPV

60DPV

90DPV

0 DPV

120DPV

POS NEG

No. of animals

Doubt 1

POS NEG

30DPV

60DPV

90DPV

0 DPV

120DPV

30DPV

Vaccine 5-ELISA-3AB1

60DPV

90DPV

120DPV

Vaccine 4-ELISA 3AB1

Vaccine 6A-ELISA 3AB1

2 POS 1

NEG

POS NEG

No. of animals

3 No. of animals

0 0 DPV

30DPV

60DPV

90DPV

120DPV

0 DPV

Vaccine 5-ELISA PANAFTOSA

30DPV

60DPV

90DPV

120DPV

0 60DPV

90DPV

POS 1

120DPV

NEG

30DPV

Vaccine 6B-ELISA 3AB1

60DPV

90DPV

NEG

No. of animals

POS

1 0 30DPV

60DPV

90DPV

POS NEG

0 DPV

NEG

No. of animals

POS

1 0

30DPV

60DPV

90DPV

120DPV

POS

NEG

120DPV

0 DPV

30DPV

60DPV

90DPV

120DPV

Control Vaccine-ELISA PANAFTOSA 4 Doubt

POS 1

NEG

0 60DPV

90DPV

Vaccine 3-ELISA PANAFTOSA

Doubt

60DPV

30DPV

Control Vaccine-ELISA 3AB1

Vaccine 6B-ELISA PANAFTOSA

30DPV

NEG

120DPV

0 DPV

POS

0 DPV

0 0 DPV

Doubt

Vaccine 3-ELISA 3AB1

120DPV

90DPV

0 0 DPV

60DPV

3 Doubt

0 30DPV

30DPV

Vaccine 6A-ELISA PANAFTOSA

No. of animals

NEG

No. of animals

POS

0 DPV

NEG

0 DPV

3 Doubt

POS

Vaccine 4-ELISA PANAFTOSA

3 2

No. of animals

120DPV

0 0 DPV

No. of animals

90DPV

No. of animals

60DPV

Vaccine 2-ELISA PANAFTOSA

Vaccine 1-ELISA PANAFTOSA 2 No. of animals

30DPV

No. of animals

PANAFTOSA

3AB1-ELISA

PANAFTOSA

3AB1-ELISA

PANAFTOSA

3AB1-ELISA

Figure 1: A set of commercial vaccines were tested following the OIE recommended method and serum samples were assayed by 3AB1 ELISA and I-ELISA PANAFTOSA. Results for both assays were expressed as number of negative (white), positive (light grey) or doubtful (dashed bars) animals for anti-FMDV NSP antibodies, as determined by the corresponding cut-off levels of each ELISA.

Doubt

POS NEG

1 0

0 DPV

30DPV

60DPV

90DPV

120DPV

0 DPV

30DPV

60DPV

90DPV

120DPV

Figure 2: Experimental oil vaccines were prepared containing 670, 170, 42.5, 10.6, and 2.6 ng/dose of recombinant protein 3ABC. Groups of 3 animals received 3 vaccinations at 0, 30 and 60 dpv. Serum samples were analyzed by 3AB1 ELISA and I-ELISA PANAFTOSA and results are expressed as described for figure 1. (

neg pos 1

670 170 42,5 10,6 2,6 670 170 42,5 10,6 2,6 670 170 42,5 10,6 2,6 670 170 42,5 10,6 2,6 ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng ng 0 dpv

30 dpv

60 dpv

I-ELISA 3ABC PANAFTOSA-BOVINES

438

90 dpv

(

No. of animals

neg pos 1

0 670 170 42,5 10,6 2,6 ng ng ng ng ng

670 170 42,5 10,6 2,6 ng ng ng ng ng

0 dpv

30 dpv

60 dpv

90 dpv

3AB1 ELISA-BOVINES

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Figure 3: Experimental oil vaccines were formulated containing the same amount of recombinant 3ABC protein as described for cattle, and tested for anti-NSP antibodies. Groups of mice (n=5) received 3 vaccinations over a period of 30 days (at 0, 15 and 30 dpv). Serum samples were taken at 0, 15, 30, 45 and 60 dpv and analyzed by the ELISA 3AB1. Results are expressed as described for the previous figure 1. 5

No. of animals

neg pos

0 dpv

15 dpv

45 dpv

2,6 ng

10,6 ng

170 ng

42,5 ng

2,6 ng

670 ng

10,6 ng

170 ng

30 dpv

42,5 ng

2,6 ng

670 ng

10,6 ng

170 ng

42,5 ng

2,6 ng

670 ng

10,6 ng

170 ng

42,5 ng

2,6 ng

670 ng

10,6 ng

170 ng

42,5 ng

670 ng

60 dpv

3AB1 ELISA-MURINE

Figure 4: An indirect ELISA assay was set up to detect FMDV NSP using a 3BA1-specific mAb as capture antibody. Decreasing amounts of 3AB1 or 3ABC recombinant antigens were tested to establish the detection limits of the assay. Bars indicate mean OD values obtained from 10 repetitions of each antigen and vertical lines in each bar are the corresponding standard deviations (CV < 10%). Cut-off value for the assays was established as the mean OD obtained for samples without antigen plus 3 times their corresponding standard deviation (OD= 0.23).

2,0

3ABC 3AB1

1,6 1,2 O.D

0,8 0,4 0,0 25

12,5

6,25

3,12

1,56

0,8

0,4

0,2

ng NSPs/ well

439

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

440

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 79

TARGETING FMDV MINIGENES TO SLA II POSITIVE CELLS ENHANCES THE INDUCTION OF CELLULAR RESPONSES IN SWINE AND CONFERS PROTECTION AGAINST VIRAL CHALLENGE B. Borrego*1, J. M. Argilaguet2, E. Pérez-Martín2, A. Ezquerra3, M. Pérez-Filgueira3, 4, J. M. Escribano3, F. Sobrino1, 5 and F. Rodríguez2 1

CISA-INIA, Valdeolmos, Madrid, Spain; 2 CReSA, UAB, Bellaterra, Barcelona, Spain; Dpto de Biotecnología INIA Madrid, Spain; 4 Instituto Virología-CICVyA INTA Buenos Aires, Argentina; 5 CBMSO, UAM Cantoblanco, Madrid, Spain

ABSTRACT In this report we present some preliminary results obtained with the last generation of DNA vaccines based on FMDV minigenes designed in our laboratory. Using previously characterised FMDV B- and T-cell epitopes, we have tested several targeting signals as carriers for our vaccine antigens. From the different molecules assayed, the signal peptide from the CCL20 porcine chemokine showed to be very effective in vitro, both in terms of antigen expression levels and in its capacity to secrete the peptides to the supernatant of the transfected cells; however when tested in vivo, constructs in which viral antigens were fused to this sp failed in inducing an efficient neutralizing antibody response and pigs were unprotected after viral challenge. In contrast, immunization with a DNA plasmid encoding the same FMDV epitopes fused to a single chain variable fragment (scFv) of an antibody that recognizes the Class II Swine Leukocyte Antigen (SLA II) induced a significant T-cell response that, even in the absence of anti-FMDV antibodies, resulted in total protection of 50% of the animals. Protection correlated with the presence of FMDV specific T-cells prior to challenge and with the development of seroneutralizing activity immediately after FMDV-challenge. Thus, both humoral and cellular responses seemed to play relevant roles in conferring sterile protection against FMDV. 1. INTRODUCTION Foot-and-mouth disease (FMD) is one of the most devastating diseases for animal health, and development of novel, safer marker vaccines is essential to avoid the problems associated to the inactivated virus vaccines currently in use. DNA vaccination has become a promising alternative because of its several advantages over conventional vaccines, i.e.: it avoids the manipulation of infectious virus, which results risky and expensive. Furthermore, DNA is easy to manipulate to generate DIVA-vaccines as well as “a la card” vaccines including conserved epitopes capable to confer protection against more than one virus isolate. In addition, they do not need of maintaining the cold chain until its distribution and costs are low. Although promising, previous DNA vaccines formulations against Foot-and-Mouth Disease Virus (FMDV) induced weak immune responses needing to be improved (Cedillo-Baron 2001, Wong 2002). Besides the advantages mentioned above, the feasibility of manipulating the DNA constructs allows application of different strategies for optimization and/ or modulation of the immune responses induced (Rodríguez and Whitton 2000). We have previously shown the protective capability of DNA vaccines encoding FMDV B- and T-cell antigenic determinants (“minigenes”) fused to the human prion signal peptide, both in mice and in pigs (Borrego 2006, Ganges 2008). In an attempt to enhance the immune responses induced, in this work we have generated a new panel of DNA constructs encoding different combinations of same viral epitopes fused to new targeting molecules: i) the signal peptides from two porcine molecules, expected to improve the introduction of the antigen into the secretory pathway; and ii) a single chain variable fragment (scFv) of an antibody that recognizes the Class II Swine Leukocyte Antigen (SLA II), in order to target antigens to APCs. The vaccine potential of all the constructs generated was first analyzed in vitro and in a mouse model, and based on the results obtained, two of the constructs were finally selected for a further evaluation in an immunization/challenge experiment in swine, natural host for FMDV. In this report we are showing the results of this vaccination experiment.

441

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

2. MATERIALS AND METHODS PLASMIDS The previously described pCMV-spBTT plasmid (Borrego 2006) was used as template for PCR amplification of different combinations of FMDV Cs8c1 epitopes B, T3A and TVP4 fused in tandem. They were cloned within the pCMV plasmid fused in frame to: i) the ORF encoding the signal peptide from the porcine chemokine CCL20, also named mip3γγ (Acc. nr: AJ311716); ii) the ORF encoding the signal peptide from the porcine CD163 molecule, a surface antigen of monocytes (Acc. nr: AJ311716, Sanchez 1999); and iii) the ORF encoding the single chain of the variable fragment (scFv) of an antibody with affinity to porcine MHC class II antigens (Bullido 1997, Gil 2003). Sequences were confirmed by automatic sequencing and plasmids were produced in a large scale (Endofree QIAGEN kits) to be used in vivo. EXPRESSION IN CELL CULTURE. Monolayers of BHK-21 cells were transfected using the lipofectamine-Plus reagent (GIBCO-BRL). At 48 h after transfection, supernatants and cells extracts were collected and analyzed by immuno-dot using monoclonal antibody SD6, recognizing the B FMDV Cs8c1 epitope included in the constructs (Mateu 1987). Expression of the antigen in transfected cells was also analyzed by immunofluorescence and flow cytometry using MAb SD6. IMMUNIZATION OF MICE. Groups of 4 Swiss mice received 3 intramuscular (IM) doses of 100 microg each of the different plasmids every two weeks. Specific FMDV antibodies in serum samples (before inoculation and after the last boost) were detected by a neutralization assay (Mateu 1987). One month after the last DNA boost, mice were euthanized and their spleens collected in order to analyze the specific cellular response by intracellular Cytokine Staining (ICCS) (Borrego 2006). IMMUNIZATION OF PIGS. Pigs received 400 micrograms per dose of the corresponding plasmid every two weeks. Animals 1 to 4 received 3 shots of the pCMV-spCCL20-BTT plasmid; animals 7 and 8 received 3 shots of the pCMV-scFv-BTT plasmid and another two pigs (animals 11 and 12) received only one shot of the same plasmid (pCMV-ScFvBTT) at the time of the second boost for the rest of the pigs. Pig 5 was inoculated with an irrelevant pCMV plasmid. 15 days after receiving the last DNA dose, all animals were needle-challenged with 104 TCID50 of FMDV Cs8c1. Pigs were daily monitored for clinical signs of disease for 10 days, when they were euthanized. Blood and swab samples were taken at different time points along the experiment. Samples collected at day 2 and 3 after infection were analysed for viral detection by isolation on IBRS cells monolayers and/or RT-PCR targeted to the 3D protein (Borrego 2006, Saiz 2003); specific antibodies were determined by neutralization assays (Mateu 1987) and 3ABC-ELISA (Blanco 2002). Peripheral blood mononuclear cells (PBMCs) obtained before FMDV challenge and at day 10 after viral infection were used to measured cellular responses by IFN-gamma ELISPOT (Díaz and Mateu 2005). 3. RESULTS Construction of plasmids and in vitro expression. Aiming to improve our previous results with the human prion signal peptide (Borrego 2006, Ganges 2008), we generated a new panel of DNA constructs fusing the FMDV BTT epitopes to new targeting signals. We selected on one hand, two signal peptides (sp) from two different porcine molecules: the chemokine CCL20, and CD163, a surface antigen of monocytes. On the other hand, the FMDV peptides were fused to the scFv of an antibody that recognizes the Class II Swine Leukocyte Antigen (SLA II), aiming to target the vaccine encoded antigens to the APCs . Antigen expression driven by these plasmids was first analyzed by immunodot, both in supernatants and cell extracts from transiently transfected cells. As shown in fig.1a., clear differences in the expression levels of the viral antigen were observed, being the CCL20 sp the most efficient. Interestingly, the expression levels obtained with this new construct resulted 50-fold higher than those obtained with the pCMV-spBTT construct, encoding the FMDV BTT antigens fused to the human PrP signal peptide, previously showed to confer partial protection in vivo (Borrego 2006, Ganges 2008). Surprisingly, no specific signal was detected in cells transfected with the

442

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 pCMV-scFv-BTT construct (not shown). However, low, albeit detectable expression levels of the antigen in these cells was observed by immunofluorescence (data not shown). We next generated a new set of plasmids containing different combinations of the viral minigenes fused to the CCL20 sp and evaluated their in vitro expression (fig.1b). Surprisingly, the presence or absence of the VP4 FMDV peptide clearly affected not only the expression levels but overall the antigen localization, since its elimination allowed a much more efficient secretion of the antigen. Immunogenicity in adult mice. The higher levels of expression achieved with the new constructs encouraged us to test them in mice as a first evaluation of their vaccine potential. Therefore, groups of Swiss mice received 3 IM injections of each one of the plasmids. As negative controls one group was inoculated with the empty plasmid pCMV, and a second group received plasmid pCMV-CCL20neg, encoding the CCL20 sp fused to the complementary strain of the BTT ORF. After the last DNA dose, sera were analyzed for detection of FMDV neutralizing antibodies (fig.2a). The only construct found to generate high levels of neutralizing antibodies was pCMV-CCL20sp-BTT, with one only animal showing neutralizing titers (PRN50= 1.7), that well might assure protection against FMDV (Borrego 2006). Surprisingly, all the other constructs failed in inducing good antibody responses, in spite of their high expression levels observed in vitro. Surprisingly, mice immunized with pCMV-CCL20sp-BT3A did not improve the humoral responses induced in spite of the efficient secretion of the encoded antigen (fig. 2a). Cellular responses were evaluated by Intracellular IFN Staining of splenocytes after in vitro stimulation with a mixture of the 3 FMDV peptides included in the vaccine (Borrego 2006). Interestingly, a strong positive response was exclusively observed in animals inoculated with pCMV-scFv-BTT (fig 2b). In two of the 4 mice of this group it was possible to detect both CD4+ and also CD8+ T-cells that specifically secreted IFN in response to ex vivo stimulation. The percentages of specific CD8+ T-cells in these animals were surprisingly high (23.2% and 8.7%). As mentioned, no responses were detected for the rest of the animals. In summary, antigenicity and in vivo immunogenicity did not seem to correlate, at least upon DNA vaccination with FMDV minigenes. DNA immunization and viral challenge in pigs. Since immune responses had been only detectable in mice inoculated with two of the five plasmids tested: pCMV-CCL20sp-BTT (neutralizing antibodies) and pCMV-scFv-BTT (IFN-secretion by CD4+ and CD8+ T-cells after specific stimulation), we decided to further analyse the protective capacity of these two plasmids in swine, a natural host for FMDV. Thus, groups of four pigs were immunized and subjected to viral challenge as described under Materials and Methods. Progression of disease in the pigs was evaluated daily using a clinical score based on a semi-quantitative rating of clinical signs (rectal temperature, lameness, vesicle formation on each one of the four feet, on the tongue, mouth and snout, and vesicle size) (Fig.3). All parameters regarding development of disease in control animal, pig 5, were as expected for the virus and dose used for challenge (Borrego, unpublished results). All the animals immunized with the pCMV-CCL20sp-BTT showed clinical signs of disease undistinguishable from those found in the control pig. In clear contrast, two animals inoculated with the pCMV-scFv-BTT construct (pigs 8 and 12) showed no signs of disease during the 10 monitored days. Interestingly, animal 12 had received a single DNA dose before viral challenge. The other two animals within this group (pigs 7 and 11), showed a delay on the disease onset, with a shorter period of acute disease, and milder clinical signs (small-size vesicles appearing only in the feet and tongue) when compared to the control animal (pig 5). Viremia for all the animals immunized with the pCMV-CCL20sp-BTT peaked at day 2, one day earlier than for pig 5 (Table 1), correlating with clinical signs of disease. In contrast, pigs 8 and 12, inoculated with the pCMV-scFv-BTT construct, showed no detectable virus at any time and sample tested, in agreement with the lack of signs of disease. Curiously enough, the other two animals within this group (pigs 7 and 11) had similar viremia titres than control animal (table 1, first row) in spite showing milder symptoms of disease (fig. 3). Remarkably, viral detection by RT-PCR in swabs from this pCMV-scFv-BTT group resulted mainly negative, while the rest of the animals scored positive both at day 2 and 3 after FMDV infection (Table 1, second row). Immune response after dna vaccination. Disease outcome and viral loads suggested that immunization with the pCMV-scFv-BTT construct had induced a protective immune response in the pigs. In an attempt to identify the mechanism(s) responsible for protection, we analysed both humoral and cellular responses in samples collected at day 43 (prior to challenge). No neutralizing antibodies at the time of challenge were detected in

443

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 any of the DNA-vaccinated pigs, independently of the plasmid used. Interestingly, a clear positive response was detected by IFN ELISPOT, in which animals inoculated with the scFv-construct showed a significant number of IFN producing cells that specifically responded to BEI-inactivated virus (fig 4), thus indicating that the protection afforded was mediated by cellular mechanisms. No significant responses (neither humoral nor cellular) were found in the pigs immunized with pCMVCCL20sp-BTT or the control pig. Immune response after challenge. Prior to viral challenge (day 43) and at days 3, 6, 8 and 10 post challenge, serum samples were collected and their neutralizing activity was analyzed. The kinetics of the development of seroneutralizing activity (fig 5) was very similar between animals within group CCL20sp and control animal, pig 5, with high titers of neutralization (PRN50 between 3.0-4.0) detected at first at day 6. In group scFv, in contrast, first detection of neutralization occurred earlier, with significant titers (PRN50 between 1.0-2.0) already detected at day 3 in 75% animals (pigs 8, 11 and 12). Curiously, in pigs 8 and 12 (the two animals protected against disease) this neutralization was not detected at day 6 but reappeared again at later times post-challenge. Analysis of antibodies to the 3ABC protein, indicative of viral replication, revealed that at day 10 post-infection no seroconversion had occurred in pigs 8 and 12, while the rest of the pigs showed high antibody titers against these non structural FMDV antigens (Table 1, third row), thus supporting the total protection achieved by vaccination with pCMV-scFv-BTT . 4. DISCUSSION Vaccination against FMDV is still an unsolved question and development of novel and safer effective vaccines circumventing the problems associated to conventional vaccination is a must. In addition to other strategies such as subunit vaccines or recombinant viruses expressing the main antigenic proteins of FMDV (SanzParra 1999, Sobrino 1999, Pacheco 2005, Wang 2002), DNA vaccination has become a promising alternative, as evidenced by results from several works (Beard 1999, Cedillo-Baron 2001, Wong 2002). Besides safety and economical advantages related to their production, the feasibility of manipulating the DNA constructs allows application of different strategies for optimization and/ or modulation of the immune responses induced (Rodriguez and Whitton 2000). In an attempt to design effective DNA vaccines we decided to target FMDV epitopes to pathways expected to play a key role in protection against FMDV. In previous works we have shown the vaccine potential of a DNA vaccine encoding the FMDV BTT minigenes fused to a strong signal peptide from the human Prion protein (pCMV-spBTT) (Borrego 2006, Ganges 2008). We decided to further explore this strategy, and selected some new signal peptides aiming to improve the introduction of the vaccine encoded antigens into the porcine secretory pathway, increasing so the chances to induce FMDV neutralizing antibodies. It is known for long that neutralizing antibodies are one of the main arms capable to confer protection against FMDV (van Bekkum 1969, McCullough & Sobrino 2004). The signal peptides (sp) selected belong to two different porcine molecules: the chemokine CCL20, and the CD163, a surface antigen of monocytes (Sanchez 1999). When fused to these sp, the levels of expression in vitro of viral antigens were greatly enhanced, and the antigen was efficiently exposed on the cell membrane or even secreted to the milieu, depending on the minigene combination. Unfortunately, our plasmids based on the CCL20sp, even those with most of the antigen being secreted, failed in inducing detectable immune responses in adult mice. Neutralizing antibodies were only detected in one of the animals receiving the pCMVCCL20sp-BTT construct, and no cellular response was either detected. When tested in swine, this construct failed again in inducing antibodies and animals were indeed unprotected. For this same FMDV Cs8c1-B epitope, this lack of correlation between antigenicity and antigen secretion with immunogenicity has been already observed (Sobrino et al, unpublished results), probably indicating an extreme sensitivity of this epitope to structural conformation, so that changes from its original conformation within the FMDV capsid might strongly affect its immunogenicity . On the other hand, some works have demonstrated the important role that other mechanisms different from neutralizing antibodies play in protection against FMDV (SanzParra 1999, Takamatsu 2006), although the precise contribution of cellular responses against FMDV is still to be defined. Because of their minimalism, our DNA constructs may be a good tool to get an insight into it. The fusion of antigens to targeting molecules expected to drive them to the sites of the immune induction has been successfully assayed, both in animal models and in livestock species (Leifert 2004). Targeting the antigen to dendritic cells by single-chain Fv antibodies (scFv) specific for APC surface molecules has resulted in an enhancement of immunogenicity (Demangel 2005). Following this strategy, we fused our FMDV BTT antigens to an scFv that specifically recognizes porcine SLAII

444

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 molecules (Bullido 1997), previously used as recombinant fusion protein to improve the immune responses induced against the fused antigens (Gil 2003). In spite of the low expression levels obtained in vitro, the construct pCMV-scFv-BTT showed to be highly immunogenic in mice, and also in swine. Immunization with pCMV-scFv-BTT, even after one single shot, resulted in full protection for 50% of the animals, in which no signs of disease or viral replication were observed, and in partial protection for the rest of the pigs receiving the same construct. Total protection seemed to correlate with the strong induction of specific T-cells prior to FMDV-challenge that secrete IFN in response to BEI-FMDV stimulation. The potent antiviral activity of IFN on FMDV, both in vitro and in vivo, has already been reported (Moraes 2007, Zhang 2002) as well as a positive correlation with protection (Parida 2006). On the other hand, the capability of pigs 8 and 12 to produce a very early seroneutralizing response seems to play an important role in the protection achieved. Further work is currently going on in our lab in order to better characterize the responses induced both before and after viral challenge. In summary, vaccination with pCMV-scFv-BTT, expected to drive FMDV antigens to SLA II positive cells, can protect pigs from FMDV challenge. Both induction of specific T-cells before FMDV challenge and a fast neutralizing activity after viral challenge seemed necessary to confer protection. Despite much work remains to be done in order to elucidate immune mechanisms mediating protection, our results open the possibility of using these vaccines to confer protection, even perhaps against heterologous FMDV strains. 5. CONCLUSIONS 1. Both humoral and cellular responses seemed to play relevant roles in conferring sterile protection against FMDV 2. Targeting of antigens to APCs seems to be a very effective approach to induce protective cellular responses 3. Immune responses elicited against highly conserved FMDV T-cell epitopes could overcome the variability of FMDV strains 6. ACKNOWLEDGEMENTS We thank M. G. Esguevillas, N. delaLosa and people at the animal facilities at CISA for technical assistance, and D. Andreu for the generous supply of synthetic peptides. Work supported by Ramón y Cajal, BIO2005-07592-C02-01, AGL2004-07857-C03-01, TRT2006-00035-C02-00, CONSOLIDER-INGENIO 2010 and EPIZONE. 7. BIBLIOGRAPHY [1] Beard, C., G. Ward, E. Rieder, J. Chinsangaram, M. J. Grubman, & P. W. Mason. 1999. Development of DNA vaccines for foot-and-mouth disease, evaluation of vaccines encoding replicating and non-replicating nucleic acids in swine. J Biotechnol 73:243-9. [2] Blanco, E., L. J. Romero, M. El Harrach, & J. M. Sanchez-Vizcaino. 2002. Serological evidence of FMD subclinical infection in sheep population during the 1999 epidemic in Morocco. Vet Microbiol 85:13-21. [3] Borrego, B., P. Fernandez-Pacheco, L. Ganges, N. Domenech, N. Fernandez-Borges, F. Sobrino, & F. Rodriguez. 2006. DNA vaccines expressing B and T cell epitopes can protect mice from FMDV infection in the absence of specific humoral responses. Vaccine 24:3889-99. [4] Bullido R, N. Domenech, B. Alvarez, F. Alonso, M. Babin, A. Ezquerra, E. Ortuno, & J. Dominguez. 1997. Characterization of five monoclonal antibodies specific for swine class II major histocompatibility antigens and crossreactivity studies with leukocytes of domestic animals. Dev Comp Immunol;21(3):311-22) [5] Cedillo-Barron, L., M. Foster-Cuevas, G. J. Belsham, F. Lefevre, & R. M. Parkhouse. 2001. Induction of a protective response in swine vaccinated with DNA encoding foot-and-mouth disease virus empty capsid proteins and the 3D RNA polymerase. J Gen Virol 82:1713-24. [6] Demangel C, Zhou J, Choo AB, Shoebridge G, Halliday GM, & Britton WJ. 2005. Single chain antibody fragments for the selective targeting of antigens to dendritic cells. Mol Immunol. May; 42(8):979-85. [7] Díaz I. & E. Mateu 2005. Use of ELISPOT and ELISA to evaluate IFN-gamma, IL-10 and IL-4 responses in conventional pigs.Vet Immunol Immunopathol. Jun 15; 106 (1-2):107-12. [8] Ganges, L, B Borrego, P Fernández-Pacheco, C Revilla, N Fernández-Borges, J Domínguez, F Sobrino & F Rodríguez. 2008. DNA immunization with FMDV minigenes in pigs: Immunological consequences of antigen targeting. Manuscript in preparation.

445

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [9] Gil, F. 2003. Desarrollo de estrategias de sobreexpresión de antígenos vacunales en plantas transgénicas. Universidad Complutense, Madrid. Ph.D. Thesis. [10] Leifert, JA, M P. Rodriguez-Carreno, F Rodríguez, J. L Whitton. 2004. Targeting plasmid-encoded proteins to the antigen presentation pathways Immunol Rev Vol. 199: 40–53 [11] Mateu, M. G., E. Rocha, O. Vicente, F. Vayreda, C. Navalpotro, D. Andreu, E. Pedroso, E. Giralt, L. Enjuanes, & E. Domingo. 1987. Reactivity with monoclonal antibodies of viruses from an episode of foot-and-mouth disease. Virus Res 8:261-74. [12] McCullough, K. C., & Sobrino, F. 2004. Immunology of foot-and-mouth disease, p. 173222. In F. Sobrino, Domingo, E. (ed.), Foot-and-mouth Disease: Current Perspectives. Horizon Bioscience, Norfolk, UK [13] Pacheco, J. M., M. C. Brum, M. P. Moraes, W. T. Golde, & M. J. Grubman. 2005. Rapid protection of cattle from direct challenge with foot-and-mouth disease virus (FMDV) by a single inoculation with an adenovirus-vectored FMDV subunit vaccine. Virology 337:205-9. [14] Parida, S., Y. Oh, S. M. Reid, S. J. Cox, R. J. Statham, M. Mahapatra, J. Anderson, P. V. Barnett, B. Charleston, & D. J. Paton. 2006. Interferon-gamma production in vitro from whole blood of foot-and-mouth disease virus (FMDV) vaccinated and infected cattle after incubation with inactivated FMDV. Vaccine 24:964-9. [15] Rodriguez, F., & J. L. Whitton. 2000. Enhancing DNA immunization. Virology 268:233-8. [16] Saiz, M., D. B. De La Morena, E. Blanco, J. I. Nunez, R. Fernandez, & J. M. SanchezVizcaino. 2003. Detection of foot-and-mouth disease virus from culture and clinical samples by reverse transcription-PCR coupled to restriction enzyme and sequence analysis. Vet Res 34:10517. [17] Sánchez C, N. Doménech, J. Vázquez, F. Alonso, A. Ezquerra & J.Domínguez. The porcine 2A10 antigen is homologous to human CD163 and related to macrophage differentiation. J. Immunol. 1999, 162:5230-7. [18] Sanz-Parra, A., M. A. Jimenez-Clavero, M. M. Garcia-Briones, E. Blanco, F. Sobrino, & V. Ley. 1999. Recombinant viruses expressing the foot-and-mouth disease virus capsid precursor polypeptide (P1) induce cellular but not humoral antiviral immunity and partial protection in pigs. Virology 259:129-34. [19] Sobrino, F., E. Blanco, M. Garcia-Briones, & V. Ley. 1999. Synthetic peptide vaccines: foot-and-mouth disease virus as a model. Dev Biol Stand 101:39-43. [20] Takamatsu, H. H., M. S. Denyer, C. Stirling, S. Cox, N. Aggarwal, P. Dash, T. E. Wileman, & P. V. Barnett. 2006. Porcine gammadelta T cells: possible roles on the innate and adaptive immune responses following virus infection. Vet Immunol Immunopathol 112:49-61. [21] van Bekkum, J. G. 1969. Presented at the Session of the Research group of the standing Technological Commitee of the European Commission for the control of foot-and-mouth disease. Brescia, Italy. [22] Moraes MP, T de los Santos, M Koster,T Turecek, H Wang,V G. Andreyev, & M J. Grubman. 2007. Enhanced Antiviral Activity against Foot-and-Mouth Disease Virus by a Combination of Type I and II Porcine Interferons. J. Virol.Vol. 81, No. 13 p. 7124–7135 [23] Wang, C. Y., T. Y. Chang, A. M. Walfield, J. Ye, M. Shen, S. P. Chen, M. C. Li, Y. L. Lin, M. H. Jong, P. C. Yang, N. Chyr, E. Kramer, & F. Brown. 2002. Effective synthetic peptide vaccine for foot-and-mouth disease in swine. Vaccine 20:2603-10. [24] Wong, H. T., S. C. Cheng, F. W. Sin, E. W. Chan, Z. T. Sheng, & Y. Xie. 2002. A DNA vaccine against foot-and-mouth disease elicits an immune response in swine which is enhanced by co-administration with interleukin-2. Vaccine 20:2641-7. [25] Zhang, Z.D., G. Hutching, P. Kitching, & S. Alexandersen. 2002. The effects of gamma interferon on replication of foot-and-mouth disease virus in persistently infected bovine cells. Arch Virol 147: 2157–2167.

446

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig 1 a: Detection of the FMDV B epitope in serial dilutions of protein extracts obtained from cells transfected with plasmids encoding the FMDV BTT minigenes fused to different signal peptides.

Fig 1 b: Detection of the FMDV B epitope in serial dilutions of a) cell-protein extracts and b) cellsupernatants from cells transfected with plasmids encoding combinations of the FMDV BTT minigenes fused to the CCL20 signal peptide

Fig. 2a: Detection of neutralizing antibodies in mouse sera. Each rhomboid represents a single animal.

447

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Fig. 2b: Intracellular IFNg detection in splenocytes from mice (#1 to 4) immunized with pCMVscFv-BTT after in vitro stimulation with a mix of the three specific FMDV synthetic peptides encoded in the vaccine.

Fig. 3: Clinical score after viral challenge in the DNA-immunized pigs.

Fig 4: IFNg ELISPOT in DNA-immunized pigs. Average number (of three replicates) of IFN-gamma producing cells per 106 PBMCs upon in vitro stimulation with: Infectious FMDV Cs8c1 (gray bars); BEI-inactivated FMDV Cs8c1 (black bars) or media alone (white bars).

448

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Table 1: Viral load after FMDV-challenge in DNA-immunized pigs. (a) For those samples in which cytopathic effect was detected, results are shown as TCID50/10 microl (log10). Some of the negative samples were also assayed by RT-PCR. (b) Results expressed as P (positive) or N (negative) for each one of the two swab (nasal / pharyngeal) samples. (c) Results expressed as OD450nm at day 10 - OD450nm at day 0 in a 3ABC-ELISA.

Fig 5: Seroneutralization after viral challenge in DNA-immunized pigs. Results are represented as PRN50, i.e., dilution of serum (log10) causing a reduction of 50% in the number of pfu in a plaquereduction assay (Mateu). A: pigs immunized with the pCMV-spCCL20-BTT construct (:nr.1, O: nr.2, :nr.3, :nr. 4); B: pigs immunized with the pCMV-scFv-BTT construct (:nr.7, O: nr.8, :nr.11, :nr. 12). Animal 5, negative control, discontinuous line and black squares, is included in both figures for comparison. Results are the mean of at least two independent experiments.

449

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

450

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 80

CONTROL OF FOOT AND MOUTH DISEASE UNDER PUBLIC-PRIVATE PARTNERSHIP (PPP)

S.N. Singh M.D. BIOVET PVT LTD, #560; 1st Main; ‘C’ Block; AECS Layout; Kundalahalli; Bangalore – 560037; E-mail: snsingh_2002@yahoo.com

The Indian dairy farming and agriculture business has grown to Himalayan heights to put the country as number one milk producer in the world today with above 92 million tons record production. The country has a huge livestock population of cattle, buffaloes, sheep and goats 170 million mainly in desert and hilly track of Indian continent e.g HP, J & K, Rajasthan, Tamil Nadu are the major habitant for small ruminants for rural economy with a marginal and landless community. The production of livestock products like milk, meat, wool, hides and other byproducts have also increased tremendously in recent years. Goattery is becoming more popular as compare to poultry in recent times the problems are mainly breading, the good quality of mutton breed and the milk breed so that the vegetarian and non vegetarian dietery system in the Indian family has been popularized, the problem has also due over grazing in the forest area at grass root villages since stall feeding system of farming is on way as viable alternative for economic production but the disease is like haemorrhagic septicaemia and other viral disease are need to be control through vaccination program Improvement in the genetic potential of the livestock by means of cross breeding have made the cross bred animals, apart from increasing stressful productivity management practices have made animals more susceptible to diseases like Foot and Mouth disease, infectious bovine rhinotracheitis, haemorrhagic septicaemia, black quarter, brucellosis. Most of these diseases "an be controlled by systematic vaccination and monitoring The export market of livestock products is ever increasing. The increase is particularly significant to other Asian countries, countries of the Middle East and Europe. The world trade organization (WTO) plays a significant role in determining trade policies, it would be of utmost importance that export products be free of important infectious diseases especially those listed by the OIE. A stringent monitoring and control policy should be implemented to prevent spread of these diseases (FAO / OIE / WTO). India has followed the OIE proposal for the eradication of Rinderpest disease. Sero surveillance and disease diagnosis was vigorously followed along with compulsory vaccination of all animals. It should be noted that the OIE would recognize India to be Rinderpest free soon. An immune belt has been created along the borders of the endemic areas. Many of the developed nations are free from most of the diseases listed in OIE particularly FMD and hence their livestock products are better accepted worldwide. The prevalence of FOOT AND MOUTH DISESE in India is a major trade barrier. Foot and Mouth disease is an acute infection caused by a virus. The virus belonging to the picornaviridae family, is the smallest virus known so far. The disease is characterized by formation of blisters, followed by ulcers on the mucosa of the. mouth cavity and on the skin of the feet, hence the name: "Foot and Mouth Disease" . Recent event in Foot and Mouth Disease control 2006 FAO Meet, Cyprus clearly defines the following important considerations. 1. Lack of early warning of emergence of new antigenic types of Foot and Mouth Disease type A has contributed to the scale of the subsequent regional epidemic in the I.R. of Iran, and Turkey 2. Incursion of an African type A virus into the Mediterranean region has occurred for first time in Egypt in 2006, leading to a widespread and severe epidemic in the naïve animal production. 3. Regional or national vaccine banks do not currently exist in the countries of the middle east and that there is often a prolonged lead time before delivery of vaccine from commercial supplier. 4. That delay in diagnosis of the new virus incursions has resulted from the use of diagnostic methods and reagents that did not sensitively detect emergent viruses of different type or antigencity. 5. There is a need to identify the extent of biosecurity measures to prevent new farm infections given the cost and impact of culling and vaccination programmes, given the widespread

451

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 dissemination of the type A epidemics in Turkey and Egypt, and the type O epidemic in the northern Europe in 2001 which occurred during periods of cool and humid conditions which favored virus transmission. 6. Significant quantities of animal products are brought by air traffic into European countries every day by passengers from Foot and Mouth Disease endemic countries in Africa and elsewhere. 7. There is an increasing trade-driven movement of livestock commodities from Foot and Mouth Disease-endemic areas in Asia Animals Susceptible for Foot and Mouth Disease All cloven-footed animals including cattle, buffaloes, sheep, goat, pigs, elephants and other ruminants are susceptible to the virus. For obvious reasons, the disease is more important in cattle, buffaloes, bulls and bullocks in India. Foot and Mouth Disease outbreaks in UK sheep has been the source of spreading the infection through trade channel all over the region this clearly proves that unnoticed clinical leasons keeps the disease symptoms unobserved during incubation for a longer period. Sheep and Goat always shows the clinical leasons at the later days even they are disseminating the virus in the population. Thus there is a scientific reason to believe that the virus may be spread in the susceptible. Disease Spread The infected animal is the main source of infection. Infection may spread either through direct contact or by indirect means, the infected feed and fodder, infected utensils, and infected means of transportation or, through carrier cattle attenders. Infections has also been reported to travel through air. However, at most times, the wide spread of infection results from congregation of animals in cattle fairs, cattle markets or large-scale transportation of agricultural produce in bullock carts. The incubation period for Foot and Mouth disease is as short as 48 to 72 hours and as long as 10 to 14 days. On an average, it varies from 3 to 7 days. Seven immunologically distinct serotypes of Foot and Mouth disease viruses have been reported worldwide. There is no cross-immunity between serotypes, immunity to one does not confer immunity against any of the other types. Four serotypes 0, A, Asia-! and C are the reported serotypes in India. Disease Economics Besides the acute stage of the disease, characterized by the formation of ulcers in the mouth, feet and udder, the virus of foot-and-mouth disease exhibits its pathology in some of the vital hormonal glands, which control the metabolic processes of the body. Disordered functioning of heat regulating centres leading to panting is one example. The disturbance in physiological process of lactation leads to a significant reduction in milk yield. In milch animals lesions on teat and udder can lead to mastitis, which may damage the teats and thereby affect the milk yield on permanently. The economic losses to the livestock industry attributed to Foot and Mouth disease are large. The OIE / FAO / APHCA place a massive significance in their attempts to eradicate Foot and Mouth disease worldwide. Direct losses to livestock sector are due to 1. 2. 3. 4. 5.

Abortion in 25 % pregnantanimals. Reduction in meat production by 25 % in endemic area. A drop in milk production by 50%. A reduction in wool production by 25 % in affected sheep. Mortality rate of up to 5.5 % of the affected cases.

Indirect losses to livestock sector are due to a. b. c. d. e. f.

452

Loss of production function during the acute phase of disease. Loss of milk yield on a permanent basis. Loss of breeding capacity including abortions. Loss due to reduced draught capacity in working bullocks. Interference with food production programme. Loss in cattle trade both national and international.

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 g. h. i. j.

Loss resulting from temporary cessation of A.I. programmes. Loss in flesh in meat-animals. Mortality in young calves due to heart failure. Flare up ofinter-current infections like Theileriosis and Anaplasmosis.

‘Prevention is better than cure' This adage is very relevant in the case of Foot and Mouth Disease than other diseases. In countries where Foot and Mouth Disease is wide spread regular programmes of large-scale vaccination using Foot and Mouth Disease vaccines are being followed. Prevalence Foot and Mouth disease is enzootic in Africa, Asia. South America and parts of Europe. The disease has been reported from various parts of the world except Japan, New Zealand, Australia, Canada and the United State of America. Foot and Mouth disease is a reportable disease in most countries and attempts are made by the FAO/OIE to collect data on the prevalence of the disease in various countries. The identification of the various virus serotypes is based upon complement fixation test, liquid phase blocking ELISA and recently by nucleic acid recognition method acid recognition method (Polymerase chain reaction). Overall it has been found that the serotype '0' and the serotype 'A' occur more frequently than the other serotypes. The disease is endemic in India. Strategy to control Sero-type spredominantly occuring in the country are mainly type '0' (70%) followed by Asia-l and type, A '. There has been no report or minor occurrence of type 'C' outbreak(s). In Punjab, Uttar pradesh and Maharastra in 1998 were predominantly by Type '0'. The various serotypes of footand-mouth disease virus are antigenically distinct and do not cross react. Depending upon the prevalence of the type of the virus causing disease, the vaccine used in the area are determined. If a single type of virus is seen prevalent, a monovalent vaccine (with only one type of the virus antigen) is used. If two types seen, bi-valent vaccine with two types of antigens are used. In India, a tetravalent vaccine with antigen from type A, O, C and C and Asia-1 are used. To prevent antigenic drift, vaccines usually with more than one strain are used inmanufacture. Hoechst Roussel Vet maintains a reference collection of the vaccine strains of FMDV. By comparing outbreak strains with the vaccine strains could be identified. A repertoire, an appropriate vaccine strains could be identified. A repertoire of antibodies is also developed to determine shared neutralizing epitopes, thereby giving an indication of the potential value of vaccine strains in helping control the outbreak. Linkages are being established within India, with IVRI, the United Kingdom Institute for Animal Health, at Pirbright and other institutes globally via the Internet. In 1951-52 over 900,000 outbreaks were reported in Europe, the European countries finally eradicated the disease and from 1992 the member countries of European Economic Community (EEC) are adopting, a non-vaccination and stamping-out policy. This had largely come about by the maintenance of solid vaccination coverage because the European FMD commission in 1957 accepted the systematic vaccination would be necessary for number of years to reduce the incidence of the disease so as to make other measures like slaughter policy an economically feasible. FMD Vaccines International standards for FMD vaccines can be found in the British Veterinary Pharmacopoeia, British Veterinary pharmacopoeia Codex, European Pharmacopoeia (Veterinary) (1993) and the OIE Manual of Standards for Diagnostic Tests and Vaccines. National Veterinary Authorities usually exercise control of the use of Foot and Mouth Disease vaccines. Indian Veterinary Pharmacopoeia is being planned to release shortly. The FAO/OIE is formulating International standards for safety, potency and antigenic mass requirements for the various vaccine strains. The dosage of the vaccine mass depending upon the epidemiological situation of the area is also being worked out. History of FMD vaccine development The three critical elements of FMD vaccine production are antigen production, virus inactivation and the addition of suitable adjuvants. Historically, the original source of FMDV for vaccine production was clinically derived material, such as infected cattle tongues in 1926.

453

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

In 1951, Franked described a new technique for the production of FMDV on an industrial scale in tongue explants. It was the FMD vaccine made in this system that was used in the Netherlands in the first of the highly successful mass annual prophylactic vaccination campaigns to be carried out in Europe. The advantages of this production system were its simplicity, low / lack of cellular protein contamination of the virus harvest and the fact that adaptations of the virus to the culture system was not required. A significant development in FMDV antigen production was the transition to tissue culture methods of virus growth. Initially, small-scale production in roller bottles using primary calf kidney cells was instigated in Italy in 1963. However, following the introduction in 1964 of a continuous cell line derived from baby hamster kidney fibroblasts (BHK 21) that supported the growth of FMDV, this system gained wide acceptance in FMDV vaccine production. Later a variety of monolayer systems were devised to increase culture vessel surface area. and thus productivity. However, the greatest scale-up capacity for FMDV production was the advent of technology, which exploited the ability of BHK-21 cells to grow in deep suspensions culture in fermentors (bioreactors) that are used widely now days. Vaccine Development, production and selection application needs further development alternative potency methods, early detection evaluating different DIVA tests for susceptible spaces mainly ruminants. Both ‘carrier’ to be seen as a major challenge for production and control. Therefore, Foot and Mouth Disease vaccine development quality control quality assurance including NSP antigen and 3D & A analysis is important to understand Antigen production Commercially available FMD vaccines are still based upon inactivated whole virus particles, mostly grown in BHK-21 in a battery of fermentors 100 1200 liters capacity located in strict containment area under controlled air conditions. Virus growth in cell culture system is followed by a series of treatments to clarify, inactivate, purify and concentrate the viral harvest. During FMD antigen production, temperature and pH have to be closely controlled because of thermal instability and the low tolerance of the virus to pH conditions outside the ranges 7.0 - 8.0. Whole virus particle (146S) content is critical to the potency of the final product and measurement of 146S is used for vaccine formulation calculations. Inactivation Inactivation is one of the most critical steps in the production of FMD vaccine. Initially formaldehyde was used to inactivate alum adsorbed virus. Inprocess evaluation of this system proved to be difficult and it has mostly been superseded by the use of first order kinetics inactivants of the aziridine group of chemicals, most recently binary-ethylene-imine (BEl). Ideally this procedure is performed twice in separate inactivation vessels. Good Manufacturing Practices (GMP) In recent years, lhere has been a move away from end product testing towards the philosophy of in-process control. This policy has been encouraged at HR Vet. Through the promotion of Good Manufacturing Practice (GMP). In process inactivation controls are performed upon the virus harvest by tissue culture titration in sensitive cells, spectrophotometeric analysis or serological assays. Further most inactivation concentration and purification by ultrafiltration or precipitation with polyethylene glycol (PEG) could yield a final virus product with a concentration factor of upto 1000 fold. Post inactivation purification reduces the non viral protein component of the antigen harvest, which is important in the reduction of potential hypersensitivity reaction in vaccinated animals. Adjuvants

454

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Inactivated whole virus vaccines against FMD are formulated as mono or polyvalent products with suitable stabilizers, buffers and adjuvants to enhance their potency. In aqueous formulations, the inactivated viral antigen is adsorbed to aluminium hydroxide [AI (OH) 3] and further adjuvanted with saponin. Such vaccines are used successfully world wide for the immunization of ruminants. However, commercial aqueous vaccines have not been successful in immunizing pigs (reactions at the site of injection were observed), and concentrated, inactivated antigens formulated as oil adjuvanted vaccines have been used widely in this species. Oil adjuvanted FMD vaccines are also used in cattle, particularly in South America. Improved formulations have reduced the local reactions initially seen in this species. Advantages are claimed for the use of oil-adjuvanted FMD vaccines in cattle in the areas of duration of immunity and the ability to immunize calves. Simple water-in-oil preparations can be made by the emulsification of the antigen in aqueous solution with light mineral oil and an emulsifying agent. Silverson and Ystral on-line pumps are used for the emulsification process and to ensure stability of the emulsion under field conditions. Alternatively, a more easily injectable formulation can be made by further emulsification in a second aqueous phase to produce a stable water emulsion [double oil emulsified (DOE)]. There are several reports of the successful experiment use of these DOE FMD vaccines in cattle and pigs. Following the completion of the blending process and addition of suitable preservatives, the vaccine bottled should be subjected to prescribed in vitro sterility test, safety I innocuity and potency tests in cattle, as described in the European Pharmacopoeia (Veterinary). Safety tests are performed in vivo using the whole vaccine inoculated into susceptible animals and in vitro using eluted antigen inoculated onto sensitive cell culture. Minimum potency assurance required is assessed by a variety of serological and I or animal challenge procedures. FMD vaccines have a shelf life of one year if stored at 40C. Production capacity: The production capacities have been increased tremendously to meet the demands of country as well as India current production capacity is 80 million doses in totality and need to increase the capacity to 200 million doses by 2008. Use of FMD Vaccines In order to establish satisfactory immunity, it is usual to give a primary course of two inoculations with an interval of 2-4 weeks. Re-vaccination may be given at 4-12 month intervals depending upon local epidemiological conditions and the quality of the vaccine. Therefore, the primary vaccination course may be delayed until four months of age in the offspring of regularly vaccinated mothers, although there is some evidence that calves can respond at one month old or younger. THE ROLE OF VACCINATION IN FMD CONTROL STRATEGY Prophylactic The successful control of FMD in countries with endemic or epizootic disease has often been based upon the regular use of inactivated whole virus vaccines as part of a regional FMD control policy. The short lived nature of protective immunity in cattle following vaccination compared to FMD infection has led to the need to vaccinate annually or bi-annually, and even thrice a year in areas with a high risk of exposure to the virus. Antigenic variation within a serotype has made it common practice to include more than one strain of a particular serotype in FMD vaccines.

455

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008 Mass prophylatic vaccination against FMD, usually practiced only in the cattle population, is the first step towards controlling FMD in endemic areas. The aim of this policy is that, over a period of years the load of FMDv in the environment will be reduced as the number of outbreaks, and therefore animals, with clinical disease will fall. Obviously, good veterinary services are essential to maintain the vaccination campaign and monitor disease status in the country. If the level of immunity to FMD in the target population in excess of 75% is achieved, the disease, should be adequately under control so that extra measures, such as importation control, quarantine and stamping out foci of infection, can be effective. An example of the successful implementation of these policies was the reduction in outbreaks of FMD in Europe from 30000/year in 1965 to less than lOOO/year by 1975. It is extremely important that an antigenically appropriate vaccine should be used. It is essential that the antigenic relationship between field isolates and the vaccine strains in use should be ascertained regularly. The next stage in FMD control is to stop mass prophylactic vaccination and, by means of stringent surveillance, rapid diagnosis and importation control, a state of freedom from infection could be achieved. This is the current situation for Uruguay and the European Union countries. Emergency General vaccination is recommended for countries where the disease is enzootic, or where the threat of an outbreak is very great. If an outbreak occurs, a booster vaccination with the relevant serotype will increase the resistance of the population. The process of stamping out of infection is difficult under Indian conditions because of social reasons. Mass vaccination coverage to 80% of the animal population will reduce the incidence of foot-andmouth disease in endemic areas. A generation of a vaccine should be advocated to contain the disease. Committed people, a proven vaccine, a good delivery system and effective vaccination coverage along with active support from the farmers, Governmental decision makers, government research institutes, nongovernmental agencies, and manufacturers of vaccine would effectively control footand-mouth disease in India. REFERENCES [1] Anon (1996) Report from the OlE World Reference Laboratory for Foot and Mouth Disease Bulletin office International des Epizooties, 108:804. [2] Anon (1996) Report from the OlE World Reference Laboratory for Foot and Mouth Disease Bulletin office International des Epizooties, 108:867 [3] Carpenter. WC, Rai D.V., Samuel A.R. & Hofner M.C, (1996) Comparison of a radioactive and non radioactive method for sequencing foot-and-mouth disease virus isolates Revue Scientifique et Technique, office International des Epizooties, 15: 875-882. [4] Donaldson, A. & Kihm, U, (1996) Research and technological developments required for more rapid control and eradication of foot-and-mouth disease Revue Scientific Technique, Office International des Epizooties, 15:863-873. [5] Donaldson A.I., Kitchinc R.P. & Barnett, P.V. (1996) Foot-and-mouth disease. In. Manual of Standards for Diagnostic Tests and Vaccines, Lists A and B Diseases of Mammals, Birds and Bees, PP 47-56, Paris, Office international des Epizooties. [6] Donaldson, A.I. (1997) Foot-and-mouth disease in Taiwan Veterinary Recorded, 140:407. [7] Donaldson, A.I. (1997) Risks of spreading foot-and-mouth disease through milk and dairy products Revue Scientifiique et Technique, Office International des Epizooties, 16: 117 -124. [8] Donaldson, A.I. (1997) Global eradication of foot-and mouth disease considerations and proposals. In: Empress Livestock Programme Rome, FAO. . [9] Donaldson, A.I. & Have, P. (1997) Criteria to consider and operational requirements for implementing emergency (ring) vaccination Report of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-mouth diseases, Kibbuttz Ma'ale Hachanisha, Appendix 29: 201-207.

456

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008 [10] Ferris, N.P. (1997) Development and use of the ELISA in the control of foot-and-mouth disease. Proceedings, International Symposium on Diagnosis and Control of Livestock Diseases Using Nuclear and Related Techniques Towards Disease Control in the 21st Century, Vienna. Salf JS & Iloot MC (1997) Virus titres in the development of foot-and-mouth virus persistence in cattle. Report of the Session of Research Group of the Standing Technical Committee of the European Commission for the Control of Foot and Mouth Disease, Kibbutz Ma'ale Hachamisha, Appendix 4:45-51. PUBLIC-PRIVATE PARTNERSHIP; For foot-and-mouth disease which severely constrains the welfare of millions of livestock-owning small scale farmers and their animals in the develouping world, currently available vaccines do not meet many of the basic requirements necessary for sustainable control of this most infectious disease. This is in part due to the qualities of the available technologies and in part due to the strategies of their use. FMD free countries are much less interested in the thermostability of vaccines or even the price.In contrast, developing countries require vaccines that protect for longer (so that herd immunity can be established and maintained in the face of less developed veterinary services), are less reliant on cold chain facilities (given the tropical and subtropical environments) and are affordable in a develouping country context. International standards of vaccine banks; Ref; low appendix-oie chapter 1.1.11, a vaccine bank is a strategic reserves of antigens or ready to use vaccine.Therefore,country concern could decide type of vaccine bank, quantity, period of storage, acquisition, regular standards, safety and efficacy and quality of stored antigen or vaccine in bulk. National authority has to monitor under the guidelines of FAO/OIE. Like FMD, under public private partnership, Biovet is in hands with CSIR and ICAR for john’s disease vaccine and diagnostics development. And Biovet under PPP, is doing lot many of clinical and field trials with Veterinary universities, Veterinary colleges, Government institutions. Under PPP, Biovet is the bridge between the farmers or the endusers and the academic institutions and NGO’s. Presented in the international conference on emerging and reemerging viral diseases in the tropics and subtropics at NEWDELHI on 11-14th December 2007. (Satellite Symposia; Virus vaccinology; strategies and blue prints)

457

The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008

458

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

CLOSED SESSION Appendix 81

LIST OF PARTICIPANTS

Aldo Dekker, Chairman Keith Sumption Secretary. Nine members of the group: (Aldo Dekker (AD), Naci Bulut (NB), Georgi Georgiev (GG), Bernd Haas (BH), Hagai Yadin (HY), Kris de Clercq (KdC), Emiliana Brocchi (EB), David Paton (DP), Soren Alexandersen (SA) Apologies were received from Mark Bronsvoort, Stefan Zientara and Donal Sammin; The work of the latter for the group was represented by Eoin Ryan (ER). One member -Hakan Vigre- had resigned from the group as his professional position had changed. Additional participants for technical items were: Dr Jef Hammond (JH), Head of the WRL Pirbright, Dr Graham Belsham (GB) Lindholm, Dr Gavin Thomson (GT), SADC FMD project. Observers included: Christianne Bruschke (representing the Chairman, EuFMD), Dr Alf Fuessel (AF; DG-SANCO, EC), Susanne Munstermann (SM), FAO Gaborone, Botswana.

459

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

460

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 82

PROVISIONAL AGENDA - STANDING TECHNICAL COMMITTEE OF THE EUFMD COMMISSION

Friday 17th October: Location-Palazzo Sales, “Board Room”-1st Floor. 08.30-13.00 Provisional Agenda Items

•

Item 1.

Agenda of the Session

•

Item 2.

FMD lab minimum containment standards

•

Item 3.

Minimum Diagnostic capacity in EUFMD Member States for the laboratory confirmation of FMD

•

Item 4.

Sampling instructions for field veterinarians to collect samples with the primary biocontainment at the point of sampling

•

Item 5.

Position paper on the options of decentralized testing (DS 1 December 2007)

•

Item 5.1 Optimising procedures for recovery of FMDV from viral RNA as alternative to live virus shipments

•

Item 6.

Sero-surveillance in Turkey: the question of harmonising the performance/interpretation of SP antibody data with/between RG member laboratories

•

Item 7.

Vaccine selection and potency of antigens in the EuVB for use against the current SAT2 FMDV circulating in Botswana/Namibia

•

Item 8.

Remaining business; workplans and tasks

461

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

462

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 83

ITEM2: MINIMUM CONTAINMENT STANDARDS FOR FMD LABORATORIES B. Haas

FOREWORD This document relates to facilities that handle or intend to handle materials containing foot-and – mouth disease virus (FMDV) in a form that could give rise to animal infection (“live FMD virus”). This Standard supersedes the previous Standards adopted by the European Commission for the control of Foot-and-Mouth Disease (EUFMD Commission), in 1993 and 1985. The terminology utilised, and the biorisk management principles were adapted from the latest currently (April 2008) available draft of the CEN/CWA “Laboratory Biorisk Management Standard” The updating of the MINIMUM CONTAINMENT STANDARDS FOR FMD LABORATORIES reflects the change in the terminology of biorisk management practise, the change in the performance characteristics and standards relating to air filtration, and also concerns relating to security of facilities. INTRODUCTION Foot-and-mouth disease is one of the most contagious diseases known and manipulating the virus in the laboratory without adequate precautions is a hazard. It has been shown that as few as 10 TCID can be infective to cattle by the airborne route. However, this is under experimental conditions and the low infective dose may relate to the relatively large size of aerosol droplets, which can be efficiently contained by HEPA filtration of air exhaust from facilities handling infective FMDV. As a consequence of the low infective dose, laboratories handling FMDV must work under high containment conditions, in which the principle objective of the containment measures is to prevent release of virus that would give rise to animal infection outside of the laboratory (veterinary containment). The principles on which the containment measures are based are as follows:

That FMD virus is an animal health but not a human health hazard That inter alia, the containment measures for FMDV laboratories will differ in certain respects from those required of high containment facilities handling pathogens which present a significant human health hazard That the effective implementation and maintenance of the measures will reduce the risk of a consequential release of virus to a level that is acceptable to stakeholders, whose considerations include the importance of the services provided by FMD laboratories.

The containment measures were prepared on the basis of the documented evidence on the physico-chemical properties of FMD virus, its inactivation kinetics, and the form and quantity of virus required to infect susceptible species. Key factors in establishing and implementing a successful containment system include:

Physical and operational barriers to the release of FMD virus that involve multiple containment layers and fail-safe mechanisms primary containment; contain the live virus at source, within closed containers or a class I,II or III safety cabinet, or for animals, by physical containment in specially constructed rooms with treatment of all waste including the HEPA filtration of air;

463

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

secondary containment of infected materials and closed and highly controlled physical environment, and air by validated procedures that will remove or Tertiary containment such as restrictions placed outside containment. Commitment by senior management

staff working with such materials within a and involving the treatment of solids, fluids, inactivate FMDV; on access of staff to susceptible livestock

1. To the provision of the resources required to attain and maintain the containment measures, including the physical and human environment; 2. To recognise management of the risks associated with facilities handling live FMD virus as a top priority; 3. To establish and maintain a management system and a working culture in the facility that facilitates continual improvement in preventing possible release of virus , the effectiveness of containment processes and root cause analysis of possible release incidents so as to prevent their recurrence; 4. To recognise and promote continual improvement. GENERAL REQUIREMENTS FMD risk management system Each facility should establish, implement and maintain a FMD risk management system, appropriate to the level of risk associated with each of the mechanisms and routes by which FMDV could escape or be released. Policy The management of the facility should have in place a policy that clearly states the FMD risk management objectives and the commitment to improving the FMD risk management performance. Risk assessment In such a system, there should be a risk assessment system in place

To identify and address the risks (likelihood and extent of impact) of release or escape of FMDV by each facility (plant) To define the circumstances which would trigger a new or revised assessment, for example the planned construction/modification to facilities, changes to programme, changes to volume of activities or following incidents or elevated levels of biosecurity threat to the facility.

Hazard identification Should identify the situations, and other hazards, associated with the work of the facility that may impact on the risk of FMD release, including emergencies (electrical failure, fire, flood, medical etc). The requirements in this standard do not necessarily identify all hazards that may occur, but are written to reduce the risk associated with the hazards in facilities handling live FMD virus. The main sources of FMD virus or infectious RNA are: 1. 2. 3. 4. 5.

diagnostic specimens Infected tissue cultures Infected baby mice, guinea pigs, rabbits, etc. Laboratory based physical and chemical processing of large quantities of virus Infected pigs, cattle, sheep, goats and other susceptible animals.

The principal routes by which the FMD virus or infectious RNA may escape or be carried out from laboratories include:

Personnel Air Liquid Effluent Solid waste Equipment Samples and reagents.

Risk control

464

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Under the direct responsibility of the management of each facility (plant), the hazards which could lead to a risk of FMD escape should be identified, quantified, prioritised and control options identified. The requirements indicated in this Standard should be considered a minimum, and do not release the management of each facility from the responsibility to undertake a formal risk assessment process. Special attention should be given to:

Replacement and reduction in use of live virus where possible; Security and recording of access to the facility; Security check of personnel handling live FMD virus The responsible behaviour of personnel within and when they leave the laboratory, including the use of changing and showering facilities; The application of rules for primary containment; The maintenance of the physical containment including the air handling systems to ensure a negative air pressure where virus is manipulated and the effective particulate filtration of exhaust air; The decontamination of effluent; The disposal of carcasses in a safe manner; The decontamination of equipment and materials before removal from the restricted area.

Use of alternative procedures The use of alternative procedures for inactivation of FMD virus to those specified in this Standard is permissible provided that the information from the validation of the process has been examined and found equal or superior in performance to those currently specified. Decisions on equivalence of the proposed procedures can be made by national competent authorities. However, national authorities have to inform the EUFMD Standing Technical Committee of such decisions and their scientific basis, which will be reviewed and findings published in the “Report of the Sessions of the EUFMD Standing Technical Committee." Residual Risk The residual risk is the risk of a consequential release of FMDV, after application of the control measures. The biorisk officer, management and ultimately the national regulatory body should consider the overall biorisk management system together with the hazard identification and risk control procedures, and identify if there are residual risks requiring either more effective controls to be put into place, or work to be suspended.

Authorization of laboratories in respect to FMD Laboratories may be authorized 1) To receive samples for application of diagnostic tests for detection of antibody responses to FMD virus, by methods that do not involve live FMDV manipulation. 2) To receive samples for application of diagnostic tests for confirmation of virus infection by methods that do not involve live FMDV manipulation (e.g. RT-PCR). 3) To manipulate live FMD virus for: a) Activities involving in vitro studies that are limited in scale [definition: less than 10 litres of cell culture, and in autoclavable/disinfectable containers, for RESEARCH and DIAGNOSTIC PURPOSES b) activities which produce higher virus outputs as a result of LARGE SCALE VIRUS PRODUCTION [>10 litres] c) INFECTION OF LARGE ANIMALS. Laboratories handling live FMDV (FMD laboratories) must comply with these minimum standards. The FMD-associated risk of diagnostic laboratories that do not handle live FMV is usually much lower than for FMD laboratories. However, authorisation of such diagnostic facilities to receive samples from areas not free of FMD or to perform certain kind of FMD-investigations, e.g. testing for antibodies, antigen or genome in emergency situations or as part of differential diagnostic investigations for other diseases requires a proper risk assessment. Furthermore, such laboratories should comply with the respective guidelines in the Reports of the Standing Technical Committee of the EUFMD Commission. SPECIFIC REQUIREMENTS

465

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The writing of the requirements below is intended to assist self-assessment, bio-risk audit and inspection of facilities. I. MANAGEMENT Specific management requirements 1. Bio-risk policy, delegation of responsibilities and communication The management of a facility is ultimately responsible for biorisk (biosafety and biosecurity) of its premises. The management should therefore define and document roles, responsibilities and authorities related to biosafety and biosecurity management in a formal policy statement and communicate this to all staff members. 2. Formal process of Risk assessment / threat assessment: The management should ensure that a formal process is in place to conduct, review and update a risk assessment. The need for a structured security threat assessment should be considered for each facility. 3. System for continual improvement: The management should put a system in place to guarantee that biosafety and biosecurity procedures and elements are thoroughly reviewed and audited on a regular basis. Records should be maintained of findings of audits, including actions taken to comply with the containment policy. 4. Standard operating procedure: A system should be in place to maintain a complete set of SOPs for all operational processes that are considered critical to the containment of FMD. 5. Biorisk Officer: It is the duty of the management to properly monitor the biosafety and biosecurity by appointing a Biorisk Officer (Biosafety/Biosecurity Officer), arranging for a deputy or replacement, and creating the necessary framework conditions in the facility. 6. To ensure that biosafety and biosecurity is given full consideration in its activities the management should carefully define the status, duties and responsibilities of a biorisk officer: a. The biorisk officer should report directly to the top management (director-general, site director or similar) representative and should have authority to stop work in the event that it is considered necessary to do so. b. The role should be independent and avoid any potential conflict of interest. c. Adequate financial and personnel resources should be allocated to the biorisk officer to carry out his or her duties. d. The biorisk officer should have the possibility of a direct link to the competent authorities responsible for the enforcement of biosafety/biosecurity regulations within the country or geographical area. e. The biorisk officer should have appropriate training in virology, containment techniques and procedures to fulfil his duties. It is to be expected that he/she would also have a broad based knowledge of the FMD virus with particular respect to its physico-chemical properties, mode of transmission and other topics of relevance to his/her role. f. The biorisk officer should review regularly both technical reports concerning the various containment facilities as well as data relating to their day to day monitoring. On the basis of such information, the officer should inform senior management of any concerns he or she may have and as they arrive as well prepare an annual report on all relevant containment elements of the facilities. 7. Record keeping – accessibility to live FMD Access to live FMDV should be limited to key personnel authorised by the management. Detailed records of handling of live FMDV (e.g. virus strains and dates used) should be kept and stored at least 5 years. Inventory information including the location, the virus strain should be maintained and periodically inspected and crosschecked. Laboratory books or other daily records of procedures by staff working with FMDV should be in place to enable retrospective analysis of activities for at least 12 months. 8. Accident/incident reporting system Each facility should have an accident/incident reporting system in place, with a procedure for rating of the risk of the event and a decision making process for recording, reporting and remedial actions. An example of a risk rating system and associated decision tool is given in Annex 1. 9. Accident / Incident review system There should be a system in place to ensure each incident/accident is reviewed to ensure that the lessons learned have been identified, the type of failing in control measures is recognised

466

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

and adequate and proportionate remedial measures set in place. A statistic concerning accidents / incidents should be made available to the management at least annually. 10. Systems to review biorisk changes Changes to the design, operation and maintenance of a facility including biosafety / biosecurity procedures and risk assessment should be reviewed, verified, approved and documented through a formal change control process before implementation. Trigger points for review or drafting of new risk assessments should be identified. 11. Emergency management plans (contingency plans) Types of emergency should be identified, including fire, flooding, loss of essential services, security breaches and major events affecting integrity of buildings, and standard management procedures for each event developed, documented and made permanently available to staff. 12. Access to site Management should implement and document a system for controlling access to areas of the site where the activities of the area pose a potential hazard. There should be physical security measures to restrict access Management should define the different zones on the site, taking into consideration the hierarchy of risk of activities in each zone. A suggested typology is: RED:

[=restricted area = where FMDV is manipulated and/or which contain infected animals]

ORANGE: [= support services and access to the restricted area] GREEN: [general access and administration]. RED, ORANGE and GREEN are within the controlled area = area within the outer security barrier or fence of the facility. The minimum requirements are to clearly define and document the zones under control of the biorisk officer, including definition of the outer perimeter of the site, lower risk areas for personnel and plant access, the location and barriers of the laboratories in which FMDV is handled, and the location and access points to waste treatment (including ventilation systems). II TRAINING 13. The organisation should ensure that personnel are competent for their designated roles and receive appropriate training on a regular basis. In particular, training requirements and procedures for biosafety and biosecurity related training of personnel should be identified (training programme) and established (training manual) and training records should be maintained. 14. Training content and training tools should be defined taking into account the different target audiences and the individual learning differences within a facility. Training efficacy assessment should be considered wherever possible and appropriate. Training should be reviewed on a regular basis. The biorisk officer should be in charge of providing information and advice on biosafety and biosecurity to laboratory staff, cleaning personnel, visitors, contractors as well as to other persons working either in locations in which FMD is handled or adjacent facilities such as service areas. Personnel should be made aware of the responsibilities, the specific containment features and the risks associated with such activities. 15. Training should be provided on the specific properties of FMD, the primary and secondary containment features and the biosafety / biosecurity procedures pertinent to each facility. 16. All staff members must be appropriately informed and regularly trained in emergency evacuation procedures with special attention being given to security requirements in cases of fire. III. LABORATORY BIOSECURITY Note: additional considerations and notes on good practices are given in Annex 1. The objective is to protect biological materials containing FMD virus against deliberate removal from the facility.

467

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

17. It is part of the duty of care of every facility handling FMDV to ensure that it minimizes the risk of virus misappropriation by intruders and people with access rights to the facility, through measures taken following a formal threat assessment process. In a threat assessment the critical assets of a facility should be identified and the facilities’ vulnerability to threats should be assessed. Based on the threat assessment structural (e.g. building design, IT etc.), physical (cameras, fences, access etc.) and organisational (security policy, accessibility etc.) measures should be taken. Decision not to undertake such an assessment requires documentation and justification. 18. The minimum requirements are: a. A Security system that is appropriate to detect and alert security to the presence of intruders, with a security plan in place for rapid response to intrusion. b. Recorded entry: access to the facility should be recorded to provide an audit trail of who was in the facility at any given time. 19. Threat reduction/control measures: due to the unpredictability of the actual threat, controls are required to reduce the risk to an acceptable level. These controls should be consider structural, physical and organisational measures and must address the following: Intruder attempting to remove FMDV from the facility by forced or fraudulent entry. Staff member removing FMDV from the facility Shipment of virus containing materials IV. PERSONNEL 20. Control of Entry into - and exit from - the restricted zones (“RED zones”) must take place only through changing and showering facilities. This means a complete change from private or controlled area working clothes to dedicated restricted area working clothes on entry - and the reverse process on exit but with a shower before leaving the restricted area. 21. A code of FMD containment practice, including instructions for entry into - and exit from control zones/restricted areas, must be available for all employees and visitors on site. 22. The FMD containment rules and other relevant documents provided by the management must have been read and signed by each employee at the beginning of their employment. At this time, it should also be made clear to new staff that any violation of such and similar regulations may result in disciplinary actions by the management and the terms of employment should indicate this. 23. Control of access to controlled zones/critical areas: a level of security checks is recommended for all individuals with access to FMDV laboratories or critical plant/service areas of these laboratories. The performance of such checks will depend on the legislation of the member state and procedures should have been developed in consultation with the police and relevant government agencies of the member state. Access to FMDV virus containing materials in the lab should be restricted to trained and dedicated staff on the basis of a legitimate need. The number of individuals with access to virus storage areas should be kept as small as reasonably possible. 24. Visitors: there must be rules in place governing the access to controlled zones by visitors, covering at least the record keeping and the use, or not, of background checks . The security system should verify the identity of visitors through use of unique identifiers including passport or ID card details. The reasons for each visit and the responsible person must be recorded. 25. Visitors have to be instructed in the specific containment procedures (e.g. decontamination) of each facility before entering the controlled / restricted zones. There must be a system in place that guarantees that these procedures are properly followed. 26. Oversight (mentoring): a system for oversight of new personnel should be established, such that all new staff have someone assigned for oversight who has sufficient understanding of the biosafety rules. 27. A human resources support system should be in place, with appropriate protocols to support staff who may be under pressures that affect their participation in the biosafety practices of the facility. 28. Quarantine: each facility must define and apply quarantine periods for persons authorised to work in each category of controlled zone/restricted area, to reduce the risk that personnel will cause a release of FMD virus as a result of virus carriage on their body. A range of quarantine periods depending on the level of exposure to virus. Depending on the risk assessment application of quarantine rules may be applied to other areas of a facility as well. Persons, including visitors, authorised to enter the FMDV restricted area must agree not to keep any animals which are susceptible to FMD, nor reside on premises where such animals are kept and to abide by minimum standards of quarantine, i.e. no contact with animals susceptible to foot-and-mouth disease for at least three days.

468

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

29. Personal protective equipment; regular supply of appropriate laboratory clothing for use within the restricted area [“Red zone”] V. FACILITY DESIGN 30.

31. 32.

General construction of buildings and their surfaces, including ducting of the air conditioning system: Maintain inward flow of air through doorways and other openings at all times Properly maintained condition with a high standard of airtightness Insect, rodent and bird proof. Windows: Sealed, toughened and preferably double glazed, and able to withstand operating pressures and all but major impact. Equivalent standard in animal rooms and at a height where animals are not able to break. Doors: warning signs at entrances: ACCESS FOR AUTHORISED PERSONNEL ONLY BIOLOGICAL HAZARD

33. a.

b. 34.

35.

Access restricted by locked doors where locks are operated from the outside [by a non-key System centrally controlled by the biosafety department: UNDER STUDY]. This prevents, Cutting of spare keys and allows biosafety department to reset access rights as necessary Airlocks provided with airtight doors which are interlocked to prevent opening of both doors Simultaneously with particular respect following a gaseous decontamination cycle Doors should be fitted with windows to allow staff outside of a room to see actions inside and provide assistance if necessary. Walls, floors, ceilings: In many respects, the surfaces and material appropriate to Pharmaceutical facilities respecting GMP standards are also relevant to laboratories handling FMD virus. Notably, surfaces should be impervious, smooth, crevice free and easily cleaned and disinfected. Cavities within the fabric of the facility should be avoided (e.g. cavity walls) unless all penetrations of the walls, floors and ceilings are thoroughly sealed with suitable materials such as silicone mastic. Crevices and joins between surfaces should also be sealed with similar materials. Continuity of seal should be maintained between floors and walls. A continuous cove floor finish up the wall is recommended in particular for areas where major spillages will occur, e.g. animal and post mortem rooms. Sealed (airtight) entry of service lines Communication: all areas equipped with telephones and, in some areas, cameras, to ensure additional security outside of normal operations and allow staff to report issues including accidents and incidents without leaving work area. Emergency back-up power: the laboratory facility should be equipped with a back-up source of electricity (an emergency generator) which starts with a delay of no more than a few minutes in the event of power failure. Alternatively, it is acceptable if the commercial power supplier is able to guarantee a supply from an alternative source within a few minutes of the main power failure. The delay period that is permissible will depend on the airtightness of the key buildings in the facility where virus in aerosol form may be present. In the design of a restricted area facility, special attention should be paid to the critical electrical supply circuits such as air handling systems, cold stores, safety cabinets, and other equipment and installations relating to the security and safety of the facility. There should be no possibility of the emergency supply being diverted from critical circuits by less important demand from non-critical equipment. Thus, the critical supply circuits would include air handling systems, cold stores, safety cabinets and other equipment and installations relating to security and safety of the facility.

VI. HANDLING OF FMD VIRUS 36. Recording receipt of virus containing materials: a system should be in place for recording receipt of specimens or samples known or reasonably be suspected (to contain FMDV. The accompanying type and strain identification, or such information generated by the laboratory, respectively, should be recorded.

469

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

37. Except in cases when this not technically feasible (e.g. during large animal experiments and post-mortem examinations), materials known or expected to contain FMD virus must either be kept within closed vessels or in devices that in combination with suitable operating procedures will function as primary containment. Such devices should be equipped with suitable filters, (HEPA filters (requirements defined in the Glossary) or equivalent off-gas or vent filters) (primary containment). A suitable disinfectant should be kept close to the work areas such that a spillage can be rapidly dealt with. 38. In areas where only small quantities of virus are handled (10 litres or less), liquids and suspensions containing FMD virus should be inactivated by a validated procedure (for example, dilution into disinfectants) before disposal into the liquid waste system of the facility. 39. When large quantities of virus are processed (e.g. for vaccine production), it is necessary to transfer virus with a contained system of vessels, pipes and other equipment. To permit fluid transfers, air needs to enter and exit equipment and infectivity must be efficiently removed by a suitably validated procedure. Usually, this is done by filtration and a number of manufacturers supply filters capable of removing FMD virus with very high levels of efficiency. Procedures are also required for decontamination of vessels, pipes and other equipment after the process has finished and before the process is either repeated or items are opened or stripped down for cleaning or maintenance. Usually this will require a chemical decontamination stage followed by steam sterilization. 40. Inoculation of animals, maintenance of infected animals and post-mortem examinations must take place within the restricted area in rooms (normally dedicated animal or post-mortem rooms, respectively) that in combination with suitable operating procedures function as a primary containment. [See glossary] Personnel must wear appropriate and comprehensive protective clothing to minimise exposure of body surfaces to virus splashes and aerosols when handling virus suspensions and when inoculating or handling infected animals. [On exit from an animal and post-mortem rooms, protective clothes and footwear must be left inside these rooms or in ante-rooms to these rooms. Showering and complete change of clothes is required before the operator can move to an area not operating under a negative pressure/airfiltration system. 41. Movement of materials known or expected to contain FMD virus out of one zone (e.g. laboratory), to another zone (e.g. animal rooms) on the same site must be governed and made by a set of procedures that prevent possible loss or spillage of virus in a non-restricted area of the facility. As a minimum requirement, such materials are transported between the zones within a leak proof and break-proof container. Staff making such transfers should be fully authorised to do so and be familiar with the emergency response procedures in the event of accident or incident. 42. Laboratory facilities and equipment must be cleaned and appropriately disinfected at regular intervals. In particular, benches and other flat surfaces exposed to virus should be wiped down with a suitable disinfectant as soon as open work has finished. VII. AIR HANDLING – LIVE VIRUS FACILITIES Ventilation systems 43. Negative pressure ventilation system: All facilities used for the handling of FMD virus must operate under a negative pressure ventilation system with HEPA filtration of exhaust air and system to prevent air escape on the inlet supply. In areas where only small quantities of virus are handled (10 litres or less), the minimum negative pressure should be 35 pascals (3.5 mm water) but due consideration needs to be given to ensure a gradient from the periphery of the restricted area to the area where virus is handled. From a practical perspective, it is difficult to achieve gradient steps of less than 10 pascals and this will tend to dictate the choice of pressure in the most negative part of the restricted area. For areas where larger quantities of virus are handled such as large scale virus production rooms and large animal rooms, the minimum negative pressure should be 50 pascals (5.0 mm water). A system should be in place to prevent a positive pressure occurring within the building due failures or faults within the restricted area ventilation system. 44. Exhaust air filtration system:Laboratories: double HEPA filtration of exhaust air. Use of a single HEPA filter may be acceptable, provided that it is demonstrated that open work with live virus is at all times restricted to within biological safety cabinets (BSC) which have H filtration of exhaust air, thereby maintaining an effective double H filtration following open work. 45. Animal rooms and production laboratories: double HEPA filtration of exhaust air.

470

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

46. Inlet air supply: a system must be in place to prevent escape of air via the inlet in case of a ventilation shut-down. This may be achieved by a single HEPA filter or automatic dampers in the air inlet system. 47. The air pressures within the different rooms of a restricted area should be continuously monitored by manometers and a system must be in place so that staff working in these areas are informed if significant loss of air pressure occurs and the actions to be taken. Manometers should be labelled to indicate the working pressure and the minimum and maximum limits within which open virus work is permitted. Under any of these alarm conditions, the primary action is to cease all open virus work and secure the workplace by sealing virus containers and disinfection of surfaces and protective clothing. The opening of doors leading to the contained area or to rooms containing infected animals or carcasses should be avoided as far as possible until the pressure difference has been restored. 48. All critical filters (HEPA) should be incorporated into a preventative maintenance programme. In particular, the efficiency of HEPA filters should be checked at least once per year, and in line with requirements of EN 14644. 49. When HEPA filters are installed or replaced, an in-situ efficiency test must be carried out by trained personnel with validated equipment. Replacement of HEPA filters must be performed in accordance with an authorised procedure. Strict precautions must be taken to prevent the spread of virus with used filters or contaminated air. Replacement of filters from outside the restricted area must take place after decontamination “in situ”or in "safe change" air-handling units. Filter specifications and test results supplied by the manufacturer should be incorporated into the maintenance records but cannot replace in-situ testing because filters may have been damaged during transportation or may not have been fitted into the gaskets properly during installation. 50. Filters must be changed when the pressure difference exceeds certain limits in accordance with the instructions given by the manufacturer, or sooner if the filter fails one of the prescribed efficiency tests. Additionally, it may be necessary to change some filters more frequently if they are subject to high humidity or high particle challenge. 51. Animal rooms – prefilters should be designed in a way that they can be changed without shutdown of the ventilation system 52. HEPA filters in safety cabinets should also be checked at least once per year. Movement of safety cabinets should be accompanied by re-validation of the filter integrity due to possible flexing and movement on the filter cartridge or filter housing. 53. Off-gas or vent filters require testing on installation and at least once per year. VIII. WASTE MANAGEMENT 54. Effluent from restricted area laboratories and from facilities holding FMD infected or potentially infected animals must be treated in a manner which ensures that there is no residual infectivity in the effluent using a suitable validated procedure. Both heat or chemical treatment may be used to process the effluent provided all of the material in the effluent is exposed to the specific treatment. 55. The treatment must be validated for the highest virus load and the most difficult matrix that can reasonably be expected. The possibility that virus particles may be protected from inactivation by proteins or lipids, and/or by aggregation or precipitation, must be taken into account in the validation process. 56. The entire effluent treatment system must comply with high containment conditions. In every case it must be ensured that no leakage from the primary containment system into the environment can occur. 57. There must be sufficient storage capacity (tanks) for the storage of untreated effluent. 58. The equipment must have automatic monitoring systems to ensure proper function. These systems must ensure that the required conditions for inactivation of FMDV have been reached before the effluent is discharged. The systems should be continuously monitored and all critical data recorded. The system should be designed in a way that in case of any failure, the likelihood of a release of potentially infectious material is minimised. 59. Treatment options: Heat treatment FMD virus is quite sensitive to heat and 100° C for 1 hour or an equivalent heat effect has been shown to be sufficient to inactivate FMDV in effluent to extent that no residual infectivity can be detected. The treatment process should be monitored by multiple, automatic and continuous time and temperature measurements, combined with automatic measurement of flow rates or volumes. Any system must ensure homogeneity of the effluent during the inactivation process. All data

471

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

relevant to the inactivation process and the release of effluent must be recorded. Critical data measuring and logging equipment must be validated by qualified personnel at least annually. Chemical treatment FMD virus is quite sensitive to acid and alkaline pH conditions. NaOH or Na2CO3 or other alkaline treatment at pH 12 for at least 10 hours has been shown to be sufficient to inactivate FMDV in effluent and are particularly effective because of their action on concentrated biological effluents. As with heat, thorough mixing of the materials must be ensured. The treatment process should be monitored by multiple, automatic and continuous time and pH measurements. After treatment, the materials must be neutralized and the pH checked before the effluent is released. All data relevant to the inactivation process and the release of effluent must be recorded. Critical data measuring and logging equipment must be validated by qualified personnel at least annually Solid waste (Animal carcasses, feedstuffs, laboratory waste etc.) 60. The principle requirement is on-site inactivation of FMDV in waste using a validated method. 61. These methods include:

Sterilisation by steam using an autoclave (at least 115 for 30 minutes: or equivalent heat effect). It is essential that the different autoclave load types (e.g. plastic waste, paper waste, waste liquids) are each validated for the maximum load size with thermocouples at different locations within the load including the centre of the load. Typically, autoclave periods are 30 min or more. Autoclaves should be double-ended so that treated waste does not need to re-enter the restricted area. Autoclaves should be revalidated at least annually by experienced personnel. Depending on the national requirements, it may be necessary to dispose of the autoclaved waste by incineration on or off of the site.

Rendering of carcasses, in compliance with the requirements of Chapter III of Annex V of regulation EC 1774/2002.

Incineration on site. The incinerators must comply with current safety standards and be fitted with afterburners.

62. Emergency procedures– a similar level of safety must be demonstrated for procedures used when normal waste treatment procedures can not be followed, e.g. because of a breakdown of equipment. Emergency procedures must be documented in the laboratory emergency plans, and include procedures for storage until treatment and final disposal. IX. EQUIPMENT AND MATERIALS Laboratory fittings 63. Benches: -Smooth, impervious and resistant for any chemicals used in the facility. Junction between horizontal and vertical surfaces should be radiused 64. Centrifuges, sonicators, homogenizers etc.: -equipment must be designed so as to contain aerosols or be used within safety cabinets where any aerosols generated will not escape to the atmosphere of the restricted laboratory Removal of equipment and other material 65. Before removal from restricted areas, equipment must be decontaminated according to the size and use of the equipment: Either by: -Steam sterilization within an autoclave, at 115°C for 30 minutes, or an equivalent heat effect. Or after surface disinfection, fumigation with formaldehyde (10 g/m3 at 70 % RH) for at least 10 minutes or (3 g/m3 for 24 hours or equivalent with other aldehydes, e.g. glutaraldehyde, or ethylene oxide (0.8 g/litre for 1.5 hrs. at 50°C). Equipment which is fumigated out of a restricted area should be cleaned and be opened as much as reasonably possible to allow penetration of the gaseous fumigant. For example, contractors tool boxes, laptops. Or thorough wash in an appropriate chemical disinfectant such as: 4 % SODIUM CARBONATE OR 10% washing soda (Na2CO3 DECAHYDRATE) 0.5 % caustic soda (NaOH)

472

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

0.2 % citric acid 4 % formaldehyde or equivalent with other aldehydes. e.g. glutaraldehyde Or other disinfectant officially approved for the purpose Note: The efficiency of these chemical disinfectants is considerably improved by the addition of a non-ionic detergent (concentration 0.005 %). 66. Decontamination of clothing before removal from the restricted area for laundry must include a wet autoclavation step (at least 115°C for 30 min. or equivalent heat effect). A laundry process without autclavation is permitted if performed on-site in a double-ended pass-through laundry device. Such a laundry process must include a validated alternative inactivation step. 67. Documents should be sent out of the restricted zone by fax, scanning or electronic processes. In case papers have to be taken out of the restricted zone, they must be treated by a validated procedure e.g. autoclaving, irradiation or ethylene oxide treatment. In cases when only low levels of contamination can reasonably be expected and following risk assessment, paper can be sealed and kept at > 20 0C for two years before being taken out of the restricted zone. Removal of biological material from the restricted area 68. Before sending non-FMD biological material to another laboratory which lacks the required level of containment, the necessary precautions must be taken to ensure that the material does not contain FMD virus. Thus if the source of the biological material is a restricted laboratory area, it is essential that it is subject to an innocuity test to demonstrate freedom from FMD virus or a validated treatment that destroys FMDV infectivity. The recipient laboratory must be informed about the potential risk of material coming from a laboratory manipulating FMD virus. The recipient laboratory must further sign a statement that it is prepared to receive the material and that it will take the necessary precautions. 69. Shipment of FMDV containing materials to other laboratories; an innocuity test is not required if the material is sent to a high containment laboratory licensed to handle live FMDV. The laboratory which provides FMDV to another laboratory has a duty of care to ensure that the recipient laboratory is authorised to handle FMDV. Before shipment, it has to ask for a statement from the recipient laboratory that it is requesting the virus only for legitimate purposes and will not redistribute the virus to other laboratories without written consent. The sending of materials containing FMD is subject to international requirements governing transportation. X. DECOMMISSIONING CONTAINMENT COMPARTMENTS FOR MAINTENANCE OR RENOVATION PURPOSES. Additional considerations and notes are given in Annex 1 70. Maintenance or renovation work that may compromise the integrity of the containment barrier [i.e. that may allow air or liquids to escape] must be preceded by an assessment of the risk and a safety plan. 71. Decontamination of rooms/compartments, to reduce the risks to an acceptable level, are required before these can be decommissioned permanently of temporarily, for example during renovation. Standard Treatment procedures include: fumigation with formaldehyde or vaporised hydrogen peroxide, after making the room effectively air-tight. 72. Waste building materials generated by demolition and redevelopment; potentially contaminated materials must be treated in a way that any residual infectivity is inactivated. If autoclavation is not feasible, it should be sprayed or fumigated to disinfect surfaces, and then stored on site for 6 months before removal. GLOSSARY Terms are in line with the proposed “Laboratory Biorisk Management Standard” (CEN draft document for public comment, 2007-07-25) Biorisk (adapted from OHSAS 18001:2007): combination of the likelihood of the occurrence of an adverse event involving exposure to biological agents and toxins and the consequence (in terms of accidental infection, toxicity or allergy or unauthorised access, loss, theft, misuse, diversion or release of biological agents or VBMs) of such an exposure Biorisk officer (BSO) or biorisk advisor (Biosafety / Biosecurity Officer): a staff member of an institution who has expertise in the biohazards encountered in the organisation and is competent to advise top management and staff on biorisk management issues.

473

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Biosafety (adapted from: WHO/CDS/EPR/2006.6): Laboratory biosafety describes the containment principles, technologies and practices that are implemented to prevent the unintentional exposure to biological agents and toxins, or their accidental release Biosecurity (adapted from: WHO/CDS/EPR/2006.6): Laboratory biosecurity describes the protection, control and accountability for valuable biological materials within laboratories, in order to prevent their loss, theft, misuse, diversion of, unauthorised access, or intentional release Restricted Area: area of the facility where FMDV is manipulated and/or which contain infected animals, bounded by physical barriers to prevent air and fluid escape except through air filtration and waste treatment systems. Controlled area: area within the outer security barrier or fence of the facility, containing the restricted area, the services for the restricted area, and zones for access and administration. Open virus work, or open work: describes the handling of materials containing FMDV (usually liquids) in which exposure to room air occurs, for example during the pipetting of liquids into containers, and the subsequent exposure of the liquid handling object (pipettes etc) to air. Primary containment: measures that contain the live virus at source, within closed containers or within a class I, II or III safety cabinet, or for animals, by physical containment in specially constructed rooms with treatment of all waste including the HEPA filtration of air. HEPA filter: High Efficiency Particulate Air filter: the classification of HEPA filters is on the basis of efficiency of removal of the most penetrating particle size, and set by international standards (EN1822). In the context of this minimum standard, all HEPA filters must at least meet H13 requirements. However in order to increase the margin of safety, H14 filters are recommended. HEPA filter performance requirements are defined by EN1822; to classify as H13, the filter must remove > 99.95% of particles of the most penetrating particle size (0.3 um). A leak is defined as penetration > 5 times the required integral efficiency, i.e. 5 times 0.05% = 0.25%. To classify as H14, the filter must remove > 99.995% of particles of the most penetrating particle size (0.3 um). A leak is defined as penetration > 5 times the required integral efficiency, i.e. 5 times 0.005% = 0.025%. ANNEX 1 AdditIonal Considerations and Examples Section I: Establishing an FMD incident risk rating system Each facility should establish a risk rating system and an associated set of incident management procedures, including reporting and responsibilities in the event that a high risk incident occurs. Risk is the product of consequence and likelihood. The consequence of an FMD escape into susceptible livestock (resulting in an outbreak) is huge. In establishing a risk rating system, the following factors should be considered:

Where does the incident occur? (for example in an animal room) What type of event? (for example a visitor leaving without showering) How much potential virus exposure or loss? (for example number of persons, time or volume) To where was the virus release? (for example outside of the high containment area, to ruminants, to areas within the perimeter of the facility)

Each facility should establish their own risk rating system, taking into consideration e.g. the history of incidents, estimations of likelihood, objective data, and computer simulations. The risk rating system and reporting requirements should be agreed at the level of the top management of the facility, and reviewed on a regular basis. Once established, the risk rating system can be used in training of staff on their reporting requirements, setting out the types of event or that should be reported to the line manager and/or biorisk officer.

474

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Example of a risk rating system Where 5 Animal room infected pigs.

containing

4 Animal room containing infected animals (not pigs).

What How much * To where FMD 5 Potentially 5 Unknown or very high 5 Outside contaminated or long time: containment, person, without > 1 L or Kg fluid or probable showering material/day. exposure of FMD >10 days air. susceptible > 50 persons. animals. FMD 4 Potentially 4 High: 4 Outside contaminated 10 – 100 ml or gram containment, to waste. fluid of material / day. Yard or farm with FMD susceptible 1 – 10 days leakage animals. of air. In contact with 5 – 50 persons. other (not FMD) Vet.Bios.Level 3 and 4 susceptible animals.

3 Moderate: 3 Outside 3 Potentially contaminated air. 1 – 10 ml or gram containment, to Or Or fluid or material / day. NON FMD 1 – 24 hour leakage of Potentially susceptible air. contaminated During the first half of the FMDV animals 2 – 5 persons. person, after disinfection process of showering formaldehyde or steam autoclaves or EthyleneOxide sterilizers.

3 Lab undertaking FMD virus work

2 Lab not handling FMD virus but 2 Potentially within common contaminated building/containment to labs fluid. handling FMDV

2 Little: 2 Outside high < 1 ml or gram fluid containment or material / day. suite but on <1 hour leakage of terrain of the air. institute 1 person.

Or During the second half of the FMDV disinfection process of formaldehyde or steam autoclaves or Ethylene Oxide sterilizer. 1 In engineering maintenance areas 1 Other Potentially 1 Very little 1 In engineering – HEPA filter replacement, etc contaminated << 1 ml or gram fluid maintenance items or material / day. areas – HEPA <<1 hour leakage of filter air. replacement, etc

*Temperature, humidity, expired time will also have influence on this issue Relative risk = where x what x how much x to where Example: A person who was working in the laboratory where live FMD is handled was observed to pass to the area outside of high containment, without taking a shower, but did not leave the perimeter of the facility. Risk rating: 3 x 5 x 2 x 2 = 60 relative risk decisions

<20 is ‘Acceptable’ Report Biorisk Officer.

21 – 60 is ‘Low’ Report Biorisk officer. Report Biorisk Committee. Report General Manager.

61 – 250 is ‘Substantial’ Report Biorisk Officer. Report Biorisk Committee. Report General Manager. Call together Crisis Team. Decision about the necessity to inform

>250 is ‘Catastrophic’ Report Biorisk Officer. Report Biorisk Committee. Report General Manager. Call together Crisis Team. Report to Regulatory autority/Chief Vet. Officer

475

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

authorities.

relative risk decisions

<20 is ‘Acceptable’ Report Biorisk Officer.

21 – 60 is ‘Low’ Report Biorisk officer. Report Biorisk Committee. Report General Manager.

61 – 250 is ‘Substantial’ Report Biorisk Officer. Report Biorisk Committee. Report General Manager. Call together Crisis Team. Decision about the necessity to inform authorities.

>250 is ‘Catastrophic’ Report Biorisk Officer. Report Biorisk Committee. Report General Manager. Call together Crisis Team. Report to Regulatory autority/Chief Vet. Officer

SECTION II Improvement of biorisk management through analysis of incidents Management should take a high interest in learning from reported incidents. Each may be considered a form of failure or non-conformity to the expected performance of the risk control measures, and occur as a result of failure in the engineering controls and/or personnel related control measures. The cause of each event may be categorised as: Related to engineering: Hardware (as facilities and equipment) Design (as irrational lay-out and ergonomics) Maintenance (as planning and availability) Procedures (as standard operations and relevance) Defences (as protective equipment and signals) Related to personnel management: Error-enforcing conditions (as occupational health and attitude) Housekeeping (as tidiness and discipline) Incompatible goals (as costs and safety) Communication ( as interpretation and point of time) Organization (as responsibilities and authority) Training (as knowledge and experience). SECTION III Threat assessment In deciding upon undertaking a threat assessment, the following should be considered: a. The threat of criminal use of FMDV for any malicious purpose has to be carefully assessed to determine the additional risk that arises from operating FMDV facilities. FMDV laboratories have exclusively peaceful objectives concerned with development and implementation of control measures. They are critical for the technical cooperation with veterinary services around the world in order to minimize the economic impact of FMD on livestock and economies. The threat of criminal use of FMDV is subject to major change as the political agenda of terrorist group changes. b. The threat and consequences of a terrorist attack will vary by country. Because of the transboundary nature of FMD, there is also the possibility that a deliberate release may occur in another, possibly neighbouring, country. For this reason, effective control measures need to be consistently applied throughout all EU member states that operate FMD laboratories. As the motivation for a deliberate release may change unpredictably over a very short period, effective control measures need to be sustained at all times and be sufficiently flexible to allow an enhanced response if required. Facilities permitted to handle FMDV are obliged to prevent illegal access and removal of the virus. As a consequence, such access to laboratory-held virus must be substantially more difficult than acquiring the virus in the field.

476

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Threat reduction/control measures: Due to the unpredictability of the actual threat, controls are required to reduce the risk to an acceptable level. These controls should be consider structural, physical and organisational measures and must address the following: c. Intruder attempting to remove FMDV from the facility by forced or fraudulent entry. Appropriate controls include 1) physical security measures restricting access to authorised staff and contingency plans in the event of intrusion, 2) secure storage of virus containing materials including maintenance of inventories of stocks. d. Staff member removing FMDV from the facility Appropriate controls include 1) vetting of persons before authorisation of access, and escorts for persons allowed temporary access when security clearance is not available; 2) restricted access to FMDV virus material in the lab to trusted staff on the basis of a legitimate need, 3) access to the facility is logged [and records maintained for at least two years] to provide an audit trail of who was in the facility at any given time. 4) Design of the laboratory or facility such that the number of staff needing to enter the secure areas is limited. E.g. some engineering aspects of the design of the facility can be arranged so that certain services can be maintained from outside of the security envelope e. Shipment of virus containing materials Appropriate controls include standard procedures before authorisation, including receipt of adequate information from the intended recipient of its authority to handle FMDV, and written agreement that the recipient laboratory will not redistribute the virus to other laboratories without applying the same risk assessment and will adhere to relevant national or international legislation relating to shipment and supply of dangerous animal pathogens SECTION VII AIR-HANDLING 1. Depending on the small animal species, route and nature of infection and method of animal containment and handling, quite high titres of virus in relatively uncontrolled conditions might be produced. Consideration should be given to the appropriate negative air pressure requirements, with 35 pascal negative pressure as the minimum. 2. Provisions must be in place to ensure that in the restricted area no overpressure is generated. One approach is to interlock the inlet and extract fans so that the most that can occur is that the air supply and extract fails and the negative envelope decays solely depending on the airtightness of the building. An emergency back-up extract fan is recommended so that the negative envelope can be restored in the event of the main extract fan failing and this also should be interlocked to the supply fan to avoid very high negative pressures which may cause damage to the fabric of the building. As an alternative, the air extraction plant can be divided into several parallel sections so that the negative pressure can be maintained if one section fails or is shut down. 3. It is advisable to have and maintain other filters within the air handling system, notably, prefilters upstream of the HEPA filters. These other filters will conserve the life of the HEPA filters and reduce the need to change at the annual maintenance interval. In properly maintained systems, it is relatively rare to change the terminal extract filter due to the efficiency of particulate removal by all of the filters upstream. However, high levels of humidity will shorten the life expectancy of filters and large amounts of dust generated by nearby building works or other activities will soon blind filters even with efficient prefilters up-stream. 4. Off-gas or vent filters: This type of filter is often steam sterilised and filter efficiency testing involves different approaches such as the water intrusion test. At the smaller scale, disposal cartridge filters may be appropriate as vent filters to allow gas exchange while preventing virus escape from the container to the laboratory environment 5. Although not widely used, sterilisation of extract air may be done by heating the air as it passes through an in-line furnace. 6. To save energy, air extracted from a restricted area may be partially recirculated into the same restricted area provided it passed through a HEPA filter before it re-enters the laboratory. However, the advisability of recirculation and the proportion of air recirculated will need to be considered against the quality of the air leaving and re-entering the work place and the activities within the workplace 7. In the event that HEPA filters become blocked prematurely (i.e. prior to annual testing), this does not normally represent a problem in terms of the integrity of the affected filter(s), but it probable that the increased resistance to airflow and consequent problems of balancing the pressures in the different rooms of the restricted area will necessitate changing the affected filters.

477

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

SECTION X: DECONTAMINATION OF COMPARTMENTS The compartment must be made airtight to make fumigating possible, if necessary by means of temporary panels. Formaldehyde procedure: Check the compartment and accompanying drawings for connections with containment facilities that must be closed. Close down utilities as gas, water, electricity, sewerage, steam and if possible ventilation Empty the compartment, for example by moving objects to other containment facilities. Remove porous material. Discard material via validated procedures like autoclaves and formaldehyde airlocks. Open non removable installation parts to make them accessible to vapour. Clean the compartment and disinfect critical points which are possibly contaminated. Prepare the fumigating equipment and shut the compartment airtight. Disinfect (air) ducts and HEPA filters for example separately by injecting formalin. Use a fumigating method in conformance with a validated procedure used for formaldehyde/VHP airlocks. Use bioindicators, (preferably a rapid bioindicator system) to prove the efficacy of the fumigating process and if found in order neutralize the compartment with ammonia. Set restrictions for access such as clothing, quarantine for people and demolition material, in order to be able to make corrections in case of accidents. Mark the current status of the compartment [where, how?]. Inspect the maintenance and renovation activities to be performed in the compartment.

478

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 84

ITEM2.1 MINIMUM STANDARDS OF BIORISK MANAGEMENT FOR LABORATORIES UNDERTAKING DIAGNOSTIC INVESTIGATIONS OF LOW-RISK SAMPLES DURING AN OUTBREAK OF FMD [Emergency FMDV labs for serology and for testing inactivated samples] B. Haas

The following is a supplement to the “minimum containment standards for fmd laboratories” INTRODUCTION The following Minimum standards for laboratories undertaking diagnostic investigations during an FMD emergency or during FMD surveillance after an outbreak only apply to The testing of blood samples from holdings without clinical signs The testing of samples from holdings with or without clinical signs that have been treated in a way that FMDV infectivity is inactivated by laboratory tests which do not contain or require live FMD virus. Serology by commercially produced FMDV - ELISA kits can be performed in many laboratories, e.g. regional veterinary laboratories, which can process samples with a high throughput. In case of an outbreak, this allows to increase the throughput of diagnostic samples significantly, which will often be a crucial factor for successful disease control. Blood sampling is often combined with surveillance and staff taking samples may also examine the mouth of possibly infected animals, which may increase the risk of a surface contamination of packing material. This risk, as well as the risk of leakage during transport has to be mitigated by appropriate provisions. The risk of FMD occurring as a result of sero-diagnostic activities within laboratories is associated with escape of virus following receipt of blood samples from viraemic animals. While the likelihood of virus being present in samples originating from holdings without clinical signs during an FMD epidemic generally is moderate to low, it is almost impossible to predict due to the dynamic nature of an epidemic. However, the maximum virus titres in blood of viraemic animals is about 10 000 to 100 000 fold lower then in vesicular material. Real-time PCR has been introduced in many laboratories, e.g. regional veterinary laboratories, which can process samples with a high throughput. Testing samples that have been treated at the site under suspicion in a way that infectivity is inactivated for FMDV genome in regional laboratories may facilitate a significantly increased sample throughput and a shortened time between sampling and reception of results. Requirements for testing samples during an FMD outbreak (for any disease or purpose) Classification of samples and overview on required laboratory Sample FMD Risk Level A

Origin of samples

Sample Risk

Holding with signs indicative of FMD

High risk, material has to be considered to contain infectious virus Low risk, to be mitigated by appropriate bio-

Holding without clinical signs indicative of FMD

Laboratory Requirements “Infectious FMDV Lab”

“Serology/inactivated sample FMDV Lab”

479

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Holding with or without signs indicative of FMD, samples treated in a way that FMDV infectivity is inactivated FMD free area

risk management measures

No obvious risk

No specific requirements

-Packaging of samples classified as FMD sample risk level B Samples must be put into watertight primary containers (e.g. plastic tubes) and the primary containers must be packed in watertight secondary packaging, which should be a strong crushproof and leak-proof container, with absorbent material that can absorb the entire contents of all the primary containers. The packaging process must include a disinfection of the secondary packaging. The packaging should comply with the European agreement concerning the international carriage of dangerous goods by road (ADR). Samples should be labelled as biological substance, category B (UN3373). -Treatment of samples to inactive FMDV infectivity Samples from holdings with signs indicative of FMD must be treated in a way that FMDV infectivity is inactivated before shipment by a procedure authorised by the competent national authority in order to be permitted to be investigated at a laboratory not meeting the minimum containment standard for FMD laboratories. Minimum Requirements for laboratory biorisk management Personnel 1. A biorisk officer (BRO) and deputy (DBRO) must be designated, and one or both present onsite at all periods in which samples are being received, and contactable at all periods when diagnostic activities are ongoing. 2. The BRO/DBRO must have sufficient experience and technical training to enable assessment of FMD risk and risk management procedures. 3. There must be a designated restricted area or areas with controls in place to limit human access 4. Personnel must be authorised to enter the restricted area by the BRO/DBRO 5. Authorised personnel working in the restricted area must be trained in biorisk management and evidence of the training recorded. Where facilities for the inactivation of waste from the restricted area are located outside of this area, also staff working with such waste must be trained in biorisk management and evidence of the training recorded. 6. Authorised personnel must change clothing before entering the restricted area and take a shower when leaving the restricted area. Authorised personnel must not have any contact to animals of susceptible species, must not enter buildings or enclosed fields where animals of susceptible species are kept, and must not handle items used in the care of susceptible species, for at least 3 days after leaving the restricted area. The agreement of the authorised personnel to these conditions must be recorded and a reminder notice of these conditions placed in a visible location at the exit point of the restricted area. 7. Entry and exit of personnel to the restricted area should be recorded. 8. Entry and exit points to the restricted area will be kept to the minimum– preferably a single point of entry/exit. 9. A step-over line, or other clearly demarcated boundary shall indicate the exit point. 10. In case the shower facilities are not placed at the border of the restricted area, outer protective garments, including shoes or shoes coverings, shall be removed before exit from the restricted area. All clothing worn in the restricted area must be stored in a secure way, e.g. in designated lockers, until treatment. 11. An incident recording system, SOPs for risk identification and notification procedures and target response time, must be in place to ensure early notification of the authorities responsible for FMD surveillance in the event that -Samples have been received which are considered FMD sample risk level A -Samples have been received in unsatisfactory state of packaging Buildings

480

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

12. Access doors to the restricted area should display a warning sign that access is restricted to authorised personnel only 13. Changing facilities and lockers are required to enable staff to deposit unessential items outside the restricted area 14. Entering of the laboratory premise by farmers or staff working on farms should be avoided. If possible, it should be attempted to separate vehicles bringing samples from vehicles entering the premise for other purposes. 15. Shower facilities must be available onsite, preferably at the border of the restricted area. 16. Sample reception area a. The restricted area must contain a specified area for reception of packages. b. This area must: i. Be easily disinfectable in the event that leakage of samples occurs into packing materials or following opening of the packages; ii. Be equipped to enable packages considered to potentially contain samples of FMD sample risk level A to be re-packaged into appropriate transport containers for dispatch to laboratories licensed for handling FMD virus, iii. have suitable facilities for waste disposal and have hand-washing facilities at exit points 17. Sample preparation area a. The restricted area must contain a specified area for serum separation and/or RNA extraction b. This area must have suitable facilities for surface disinfection and waste disposal and have hand-washing facilities at exit points 18. Testing area 19. a The restricted area must contain a specified area for testing 20. b This area must have suitable facilities for surface disinfection and waste disposal and have hand-washing facilities at exit points 21. Sample storage area a. The restricted area must contain a specified area for the storage of samples b. This area must have suitable facilities for surface disinfection 22. Communications and reporting office space 23. The laboratory must have an adequate provision of office space, computing and communications facilities (e.g. electronic communications, facsimile) to reduce the need to a minimum for staff, papers and physical records to exit the restricted area. 24. Rest rooms 25. The restricted area should have sufficient rest rooms and lavatory facilities in relation to the staff number expected at peak periods of activity, sufficient to reduce the need to a minimum for staff to exit the restricted area. 26. Location of autoclave 27. Facilities for wet heat treatment must be present on the site, preferably with sufficient capacity for throughput at the maximum operating capacity of the laboratory) Waste 23. Liquid waste a. Heat or chemical treatment of all waste water is the PREFERRED treatment, in compliance with the prescribed standards specified for FMD laboratories. b. Alternatively, or additionally, the laboratory may demonstrate that it has put in place a system for inactivation of virus if present in liquid waste that has contacted risk materials. If treatment of all liquid waste from the restricted area (including waste water from the showers) is not possible, at least the ELISA buffers and washing fluids must be collected and treated. 24. Solid waste a. For biological, solid waste, and all solid , disposable materials that have been in contact with specimens, treatment by wet-heat in an autoclave within or at an entrance point to the restricted area is the preferred option. b. If such a treatment of all solid waste is not possible, it may be packed into suitable watertight containers and, after spraying of the containers with disinfectant, removed for treatment at a different site. 26. Removal of equipment, materials and clothing from the restricted area a. Removal of any material and equipment from the restricted area shall be subject to authorisation by the BRO. b. The reason for removal, date, and destination will be recorded.

481

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

The BRO will ensure that materials and equipment which has been in contact with risk materials (specimens) will not be removed from the restricted area without a validated treatment to inactivate FMDV. 27. Declassification of the restricted area a. A decontamination plan must be agreed with the competent authorities, before restrictions can be lifted. b. If heat treatment or scanning of all paper from the restricted area is not possible, it should be packed into suitable containers, which should be disinfected and kept under lock for at least two years. If the containers have to be opened before, this has to be done in a restricted area meeting the standards described above.

482

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 85

ITEM 3 PROPOSED MINIMUM REQUIREMENTS FOR ADOPTION AT THE 38TH EUFMD GENERAL SESSION (2009) AS A MINIMUM FOR MEMBER STATES MINIMUM REQUIREMENTS IN EUFMD MEMBER STATES FOR THE LABORATORY CONFIRMATION OF FOOT-AND-MOUTH DISEASE (FMD)

1. Member States shall ensure that: (a) The veterinary services of the Member State have permanent access to the services of at least one Reference Laboratory (RL), either within their territory (National Reference Laboratory -NRL) or in the territory of another state, that is capable of undertaking i) Laboratory tests for the confirmation of FMD virus, and ii) Of undertaking tests for confirming presence of specific antibodies, to a level that is compliant with the Council Directive 2003/85/EC. Different RL maybe utilised to undertake tests for virus and antibody. (b) The RL / NRL for confirmation of FMDV must be capable of achieving a rapid initial diagnosis, and must be suitably staffed and equipped to initiate laboratory procedures within 12 hours of receipt of samples. (c) Provision is made for emergency situations where the FMD diagnostic services of the Reference Laboratory are unavailable through fire, flood or other contingency. (d) The access to services of a Reference Laboratory on their territory is guaranteed by contracts that adequately equip and staff the reference laboratory with the appropriate numbers of trained personnel to carry out the laboratory investigations required, or through contracts established with laboratories in another state to provide the RL services 2. The functions and duties of Reference Laboratories shall be as follows: 1. All Reference Laboratories handling live Foot-and-Mouth Disease Virus must operate under high security conditions laid down by the most recent decision of the Sessions of the European Commission for the Control of Foot-and-Mouth Disease (EUFMD)2. Reference Laboratories undertaking test procedures that do not involve live virus need not comply all requirements of the minimum bio-security standards for laboratories handling live virus, but must operate procedures which ensure that the risks associated with the possible entry of foot-and-mouth disease virus in samples are recognised and effectively contained. 2. Reference Laboratories shall participate AT LEAST ONCE IN EVERY TWO YEARS in the external quality assurance and standardisation exercises organised by the FAO World Reference Laboratory or the European Community Reference Laboratory (if different). 3. Reference Laboratories shall use tests and standards that meet or exceed the criteria laid down in Annex XIII of the Council Directive 2003/85/EC. 4. Reference Laboratories shall provide the EUFMD Commission on request with data proving that the tests in use meet or exceed the requirements.

2 ‘Minimum Standards for Laboratories working with foot-and-mouth disease virus in vitro and in vivo’, European Commission for the Control of Foot-and-Mouth Disease — 26th Session, Rome, 1985, as amended by Appendix 6 (ii) of the Report of the 30th Session, Rome, 1993.

483

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

REFERENCE TEXTS [1] COUNCIL DIRECTIVE 2003/85/EC of 29 September 2003 on Community measures for the control of Foot-and-Mouth Disease repealing Directive 85/511/ EEC and Decisions 89/531/EEC and 91/665/EEC and amending Directive 92/46/EEC (Text with EEA relevance) Article 68 National Laboratories 1. Member States shall ensure that: (a) Laboratory testing for foot-and-mouth disease is carried out in laboratories authorised for such testing by the competent authorities; (b) Laboratory testing to confirm the presence of Foot-and Mouth Disease virus or other vesicular disease viruses is carried out in accordance with Article 71 by one of the laboratories listed in Part A of Annex XI; (c) One of the laboratories listed in Part A of Annex XI shallbe designated as the national reference laboratory for the Member State on whose territory it is situated, and it shall be responsible for coordinating standards and methods of diagnosis in that Member State; (d) The national reference laboratory carries out at least the functions and duties set out in Annex XV; (e) The national reference laboratory referred to in point (c)liaises with the Community Reference Laboratory provided for in Article 69 and in particular ensures the sending of appropriate samples to the Community Reference Laboratory. 2. The national reference laboratory referred to in paragraph 1(c) of one Member State may provide the services of a national reference laboratory to one or more other Member States. Member States which have no national reference laboratory situated on their territory may use the services of The national reference laboratory in one or more other Member States. That cooperation shall be formalised in a mutual agreement between the competent authorities of the Member States concerned, which shall be notified to the Commission. Such cooperation shall be listed in the special column in the table in Part A of Annex XI. 3. Member States shall ensure that laboratory investigations provided for in this Directive are first of all carried out to confirm or rule out foot-and-mouth disease and to exclude other vesicular diseases. Where an outbreak of foot-and-mouth disease has been confirmed and the serotype of the virus was identified, that virus shall be antigenically characterised in relation to the reference vaccine strains, where necessary with the assistance of the Community Reference Laboratory. Samples from domestic livestock showing signs of vesicular disease which are negative for footand-mouth disease virus and, where relevant, Swine Vesicular Disease virus shall be sent to the Community Reference Laboratory for further investigation. 4. Member States shall ensure that the national reference laboratory on their territory is adequately equipped and staffed with the appropriate numbers of trained personnel to carry out the laboratory investigations required in accordance with this Directive.

484

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[2] COUNCIL DIRECTIVE 2003/85/EC ANNEX XV FUNCTIONS AND DUTIES OF NATIONAL LABORATORIES The functions and duties of National Laboratories referred to in Article 68 for foot-and-mouth and other vesicular diseases shall be as follows: 1. All National Laboratories handling live foot-and-mouth disease virus must operate under high security conditions laid down in ‘Minimum Standards for Laboratories working with foot-and-mouth disease virus in vitro and in vivo’, European Commission for the Control of Foot-and-Mouth Disease — 26th Session, Rome, 1985, as amended by Appendix 6 (ii) of the Report of the 30th Session, Rome, 1993. 2. National Laboratories must provide an uninterrupted service for diagnosing vesicular viral diseases and must be equipped and skilled for providing a rapid initial diagnosis. 3. National Laboratories must keep inactivated reference strains of all serotypes of foot-and-mouth disease virus, and immune sera against the viruses, as well as all other reagents necessary for a rapid diagnosis. Appropriate cell cultures should be in constant readiness for confirming a negative diagnosis. 4. National Laboratories must be equipped and skilled for large-scale serological surveillance. 5. In all suspected primary outbreaks appropriate samples must be collected and quickly transported, according to a set protocol, to a National Laboratory. In anticipation of a suspicion of foot-and-mouth disease, the National Authority shall ensure that the necessary equipment and materials for sample collection and transportation to a National Laboratory are stored in readiness at local sites. 6. Antigenic typing and genomic characterisation must be carried out on all viruses responsible for new incursions into the Community. This can be performed by the National Laboratory, if facilities exist. Otherwise, at the earliest possible occasion, the National Laboratory must send a sample of virus from the primary case to the Community Reference Laboratory for confirmation and further characterisation, including advice on the antigenic relationship of the field strain to vaccine strains in the Community antigen and vaccine banks. The same procedure should be followed for viruses received by National Laboratories from third countries in situations where characterisation of the virus is likely to be of benefit to the Community. 7. National Laboratories should provide disease data to their State Veterinary Service, which shall provide these data to the Community Reference Laboratory. 8. National Laboratories should collaborate with the Community Reference Laboratory in ensuring that members of the field section of State Veterinary Services have the opportunity of seeing clinical cases of foot-and-mouth disease in National Laboratories as part of their training. 9. National Laboratories shall collaborate with the Community Reference Laboratory and other National Laboratories to develop improved diagnostic methods and exchange relevant materials and information. 10. National Laboratories shall participate in external quality assurance and standardisation exercises organised by the Community Reference Laboratory. 11. National Laboratories shall use tests and standards that meet or exceed the criteria laid down in Annex XIII. National Laboratories shall provide the Commission on request with data proving that the tests in use meet or exceed the requirements. 12. National Laboratories should have the competence to identify all vesicular disease viruses and encephalomyocarditis virus in order to avoid delays in diagnosis and consequently in implementing control measures by the competent authorities. 13. National Laboratories shall cooperate with other laboratories designated by the competent authorities for performing tests, for example serological tests, that do not involve handling of live foot-and-mouth disease virus. These laboratories shall not carry out virus detection in samples taken from suspect cases of vesicular diseases. Such laboratories need not comply with the biosecurity standards referred to in Annex XII, point 1, but must have established procedures which ensure that the possible spread of foot-and-mouth disease virus is effectively prevented. Samples giving inconclusive results in tests must be transmitted to the National Reference Laboratory for carrying out confirmatory tests.

485

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

486

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 86

OPTIONS FOR DECENTRALISED DIAGNOSIS OF SECONDARY CASES OF FOOT-AND-MOUTH DISEASE IN ANY FUTURE OUTBREAK. D. Sammin1, N. Ferris2, D. King2, E. Ryan1, S. Zientara3, B. Haas4, H. Yadin5, and D. Paton2 1

CVRL, DAFF Laboratories, Backweston, Celbridge, Co. Kildare, Ireland Institute for Animal Health, Ash Rd., Pirbright, Surrey GU24 0NF, UK. 3 AFSSA, 27-31 Avenue du General Leclerc BP19, Maisons Alfort, 94703, France 4 FLI, Boddenblick 5a, 17493 Greifswald, Insel Reims, Germany 5 Kimron Veterinary Institute, PO Box 12, Beit-Dagan 50250, Israel 2

When foot-and-mouth disease (FMD) occurs in a country previously recognised as free from the disease, that country must report this to international agencies and trading partners with inevitable consequences – loss of disease-free status and disruption to export trade in livestock and animal products. Consequently, huge emphasis is placed on the accuracy of diagnosis of the first case of the disease in any outbreak and this diagnosis will always have to be confirmed by a national (or international) reference laboratory (NRL). Thereafter, the priority for the national veterinary authorities is to bring the disease under control as quickly as possible, to regain FMD-free status and to resume trade. Critical to FMD control is that as soon as the primary case has been diagnosed, all other secondary cases are rapidly identified so that appropriate measures can be taken to prevent further spread of the disease. However, confirming suspect cases at an NRL often involves an unavoidable delay in transporting specimens whilst the reference test methods employed by NRLs can be relatively slow and time-consuming. One approach to this problem is to establish rapid, high throughput methods at NRLs and to speed up the time taken to deliver samples. Alternatively, rapid test systems could be deployed outside of the NRL, making it feasible to move the confirmatory diagnostic process nearer to the location of affected animals in any future outbreak. This paper attempts to summarise options that are currently available for the direct diagnosis of secondary cases of FMD in an outbreak situation and to address some of the issues that arise if testing for FMD is to be performed outside of specialised high security laboratory facilities. A confusing multiplicity of terminologies have been used to describe the concept of testing for FMD outside of an NRL; the terms “rapid testing”, “penside testing”, “on-farm testing”, “on-site testing”, “field testing”, “devolved testing” and “decentralised testing” have all been used synonymously. Assuming that NRLs are the centres designated for testing within their respective national territories, the term “decentralised testing” will be used throughout this paper to refer to all testing outside these laboratories, including that performed in regional laboratories. Direct diagnosis of FMD – basic principles of currently available test methods Direct diagnosis of FMD implies detection of the virus in specimens collected from suspect animals as distinct from indirect diagnosis, which is detection of an FMD specific antibody response. Formerly, direct diagnosis was based on examination of tissue or vesicular fluid by a serotyping ELISA and/or isolation of FMD virus in tissue culture followed by use of a serotyping ELISA. The requirement to use virus isolation in cases where the ELISA on its own was negative or where the sample type was not appropriate for direct use of ELISA meant that testing could only be safely attempted in high security, specialised testing facilities (NRLs). Maximising the sensitivity of virus isolation, requires cultivation on calf thyroid cells (Snowdon, 1966). This is a primary cell line that cannot be maintained indefinitely in culture (and must be continuously harvested de novo), such that for practical reasons, many NRLs must rely on attempted isolation in less sensitive continuous cell-lines such as IB-RS-2, lamb kidney and BHK cells. Although tissues containing large quantities of virus may give a positive test result within a few hours (if positive by ELISA without virus isolation), it takes 4 days (two passages of virus isolation of 48 hours duration each) to rule out the presence of infectivity in those specimens that test negative on virus isolation and antigen ELISA. For these various reasons alternative test methods, which detect either viral antigen or viral

487

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

genome and which give a positive or negative test result within a few hours, have been adopted by most NRLs. Detection of viral antigen (by antigen-antibody immunoassay) An antigen detection ELISA, developed at Pirbright (Roeder and LeBlanc Smith, 1987), has been used routinely for diagnosis of FMD (and swine vesicular disease; Ferris and Dawson, 1988) and for serotyping the virus for the past 20 years. This method, which does not involve an amplification step, requires relatively large amounts of the virus to be present in the tissue under test to yield a positive result. A more recent development in FMD diagnostics has been modification of a technology (initially commercialised in human pregnancy diagnostic kits), which is based on the diffusion of coloured, antibody-coated latex beads or colloidal gold particles through a membrane towards an immobilising band of trapping antibody. These tests are referred to as either “rapid (immuno)chromatographic strip tests” or “lateral flow devices” (LFDs). A prototype LFD has been shown to detect FMD virus with approximately equal sensitivity to the antigen capture ELISA (Reid et al., 2001) and performed satisfactorily with specimens from two suspected premises during the UK outbreak in 2001 (Ferris et al., 2001). An LFD produced by Svanova has been validated recently (Ferris et al, in press) and was used in the laboratory as an initial rapid test during the 2007 UK FMD outbreak besides being used successfully on one farm by a veterinarian to diagnose FMD during that epidemic (Ryan et al, 2008). If FMD viral antigen is present in sufficient concentration in the test specimen, a visible line will appear in the “test window” of the device within minutes. The test is thus rapid and easy to perform. However, current formats of this test do not serotype the virus present, although this may be considered of lesser importance in the confirmation of secondary cases. Reading the device result by eye involves a degree of subjectivity; LFD readers are now available, but a requirement for their use could limit availability of the test and add to disease containment concerns. As the sensitivity of these devices is relatively low (as for the ELISA), they are only suitable for testing epithelial lesion material and vesicular fluid (which is expected to have the highest concentration of virus) and more than one affected animal within any particular group or herd of animals should be sampled and tested (on separate devices) to minimise the risk of a false negative test result. An advantage of their ease of use in the field is that where an unexpected result is obtained a repeat test could be done immediately, on the same or additional animals. The initial step in testing epithelial lesion material by any antigen detection method requires disruption of the tissue to release viral antigen, a process performed in the laboratory by grinding the tissue with sterile sand, pestle and mortar. To facilitate field use of lateral flow devices, a simple dropper bottle containing sand and a buffer has been developed into which an epithelial specimen can be placed; after grinding with a pestle the supernatant can be added to the test device. Vesicular fluid, or a swab which has absorbed vesicular fluid, can also be added to the buffer prior to testing. The buffer fluid is then applied to the device. Detection of viral genome (by RT-PCR) FMDV-specific RNA (viral genome) may be amplified by the twin processes of reverse transcription or “RT”, which produces a DNA copy of an RNA template, and the polymerase chain reaction or “PCR”, which is capable of generating millions of copies of a target sequence of DNA. The amplified product may be detected by electrophoresis in a gel once the reaction is complete (in “conventional RT-PCR”) or now more commonly and with greater sensitivity, specificity and speed in “real time” (i.e. as the amplification process is occurring) by using fluorescent probe technology. There are four steps involved in the testing process: (i) extracting RNA from tissue specimens; (ii) the RT reaction; (iii) the PCR reaction and (iv) the detection of the amplified product. The first of these steps is required to remove inhibitory substances that might be present in the sample. It is both time consuming and labour-intensive if done manually (imposing limits on both throughput and turnaround time) but has been greatly facilitated in recent years by the application of automated (robotic) extraction processes. The PCR reaction itself requires very exacting conditions: a “cocktail” of enzymes, substrates and co-factors in very precise concentrations and repeated “thermal cycling” between two specific incubation temperatures (a higher temperature that allows double stranded DNA to separate and a

488

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

lower temperature that allows the resulting single strands to be replicated). Precise thermal cycling can only be achieved with precision instrumentation. The final step in the process (detection of amplified product), if performed “in real-time”, requires sophisticated equipment and computer software. Amplification processes such as RT-PCR are super-sensitive, such that cross-contamination is a particular concern and stringency is needed to generate reliable test results; in the testing laboratory this is usually achieved by complete separation of the various stages of the process, which are performed in separate rooms, thus avoiding amplified product from one test run contaminating test specimens prior to amplification in subsequent test runs. Centralised testing justifies the acquisition of sophisticated robotic equipment that enables large numbers of samples to be processed at once with a high degree of fidelity. Several different companies have developed portable “real-time” PCR equipment in recent years. These platforms are robust and easy to use, having been designed for use by the military (to detect infectious agents such as Bacillus anthracis, spurred on by fears of bio-terrorism). Combined with mobile robotic RNA extraction equipment (or simple mechanical shaking devices), this machinery is currently being evaluated for decentralised diagnosis of FMD (King et al., 2008). Prepackaged PCR reagents, validated for FMD virus detection, are also commercially available and it will be possible to perform the entire RT-PCR process within disposable sealed tubes considerably reducing the risk of cross-contamination, although most systems currently on the market do not the perform nucleic acid extraction step. With this type of equipment, only small numbers of samples (1-8) can be processed concomitantly. Alternative methods of nucleic acid amplification have been developed which are isothermal (i.e. do not require thermal cycling and such precise instruments) and some of these processes can be formatted to produce a visible colour change if positive, allowing the test to be read without sophisticated equipment (Dukes et al., 2006; Lau et al., 2008). This could form the basis of a disposable decentralised testing kit; but none of these methods are yet routinely applied in FMD diagnosis. Current genome detection methods are not typically designed to serotype FMD viruses. Use of a combined testing strategy involving both rapid antigen detection and rapid genome detection provides separate lines of evidence for the presence of infection and hence greater confidence in the diagnosis. However, in practice, this may not always be feasible (e.g. samples such as blood cannot be tested directly by antigen ELISA) and judgement must be used as to the level of certainty required from laboratory confirmation, according to the strength of the available field evidence (i.e. the clinical basis for suspecting FMD) and the consequences of the diagnosis. Currently available rapid testing options for diagnosis of secondary cases OPTION 1 Retain all testing for FMD at the NRL; take steps to expedite transfer of samples to the laboratory and use a combination of LFD, antigen detection ELISA and RT-PCR testing. The advantages are obvious in that the testing is performed within specialised (QA-accredited) bio-secure facilities by skilled staff experienced in performance characteristics of the test methods and in the interpretation of the test results. A high throughput of samples is possible (such as that required for active surveillance programmes). The principal disadvantages are the delay in getting specimens to the laboratory and in completing the testing process; with the current methodology, it would take about five hours from receipt of the specimen in the NRL until testing is completed by all three procedures and this presumes that there are no discrepant or inconclusive results requiring a retest (but LFDs can be used in NRLs to give very rapid preliminary results as done in the UK in 2007) OPTION 2 Perform real-time RT-PCR in a local, regional or mobile laboratory using RNA extraction and PCR equipment or using a modularised portable PCR unit (such as the Bioseeq Vet that is in development, Smiths), testing samples that have been inactivated by agents such as TRIzol. Local laboratory staff could perform the test and a test result could be available within two hours from receipt of the specimen in the laboratory. It should be noted that the EU Community Reference Laboratory (CRL) see few advantages in using mobile laboratories over local/regional laboratories,

489

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

where these are available; for example in the UK, where the proximity of most farms to a regional laboratory means less than 1.5 hours driving distance. An important caveat is that whilst it would be feasible to set up local laboratories in advance or at short notice with simple automated equipment to enable rapid testing of a low throughput of samples, this would not give a high throughput service (as available at the NRL) for testing large numbers of samples at the same time, in the event of a major local outbreak. However, some regional laboratories may already have capacity for this type of testing for endemic diseases. OPTION 3A Use LFDs on the suspect premises. The official veterinarian that will have examined and sampled the suspect animals could perform this method and a test result could be available within 10-15 minutes providing that a moderate concentration of antigen is present in the test sample. The low cost of the individual LFDs would facilitate their local distribution in “peacetime”, thus ensuring that devices would be quickly available for suspect cases. They could, for instance, be routinely included in the FMD field investigation kit. The official veterinarian would therefore have them immediately available while on the suspect farm. OPTION 3B Use a modularised portable PCR unit (e.g. Bioseeq Vet, Smiths) on the suspect premises. Following notification of a suspect case, this would require transport of the device to the farm, where it could be used to detect viral genome in blood, epithelium, OP fluid or other samples. A result would be available 60-90 minutes later. The advantage of this approach is that it would be more sensitive than the LFD, while still enabling on-farm testing. A disadvantage is that, due to the cost of these devices, they are likely to be stored in some regional depot rather than one being issued to each official vet. Therefore the device would have to be transported to the suspect farm following a request by the investigating vet. This time delay may be insignificant in some circumstances, but in other instances it may be just as quick to simply transport the samples to a regional or local laboratory (as discussed in option 2 above). There is a trade-off or compromise between proximity of the testing process to the affected animal (allowing for speed of diagnosis) and performing the test in ideal laboratory conditions (ensuring reliability or accuracy of diagnosis). This compromise is summarised in Table 1. Of course, there is nothing to stop the national veterinary authority from applying these various options in series until satisfied that there is sufficient evidence to deem a herd to be infected or free from infection. Such a hierarchy of testing options is illustrated diagrammatically in Figure 1. One crucial point in support of this scheme, which may not be readily apparent from Figure 1, is that the animals which pose greatest risk (because they are likely to be shedding the most virus) would most probably be rapidly identified as positive using on-site tests (Options 3a or 3b) whilst the infected animals that pose least risk (as they are shedding the least amount of virus) should eventually be identified as positive by the combination of more sensitive test methods which are employed at NRLs (Option 1). In other words the infected animals that pose the greatest risk should be identified fastest. An issue that is sometimes overlooked is that if samples are not submitted to a NRL they may not become available for subsequent analysis; for example for genome sequencing to trace routes of transmission. A possible way of overcoming this problem would be to ensure that either epithelial suspensions prepared in the field for use in LFDs (or other detection systems) or duplicate vesicular epithelium are sent on to the NRL immediately afterwards. Alternatively, options 1, 2 and 3a might be conducted in parallel and would probably provide the optimum balance of speed and reliability.

490

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Option

Location

Method

NRL

AgELISA and/or rtRT-PCR (Delivery + 5 hrs)

Relative ranking of testing process Speed Reliability Throughput 3

Regional or mobile lab.

rtRT-PCR (Delivery + 1.5 5 hrs)*

2/3

Suspect premises

LFD (15 minutes)

“

Portable rtRT-PCR (1.5 hrs)

1/2

2/3

Assumes conventional testing methods but LFDs and/or a bioseeq can also be used at the NRL for occasional (i.e. low throughput) high speed testing of urgent samples; *Depends on the method used

Table 1: Relative ranking of different options for diagnosis of secondary cases of FMD during an outbreak (1 = best, adapted from Reid et al., 2006) Issues that will arise with decentralised testing for FMD The veterinary authority in each country must decide which of the currently-available options would be most appropriate to their needs in the event of any future outbreak. Some national authorities have already given considerable thought to how they might decentralise testing for FMD (and other OIE-listed diseases) in the event of a future disease outbreak; some examples are summarised at the end of this paper (Annex 1). If a decision is taken to opt for decentralised testing, a number of predictable issues arise; these issues are listed and briefly outlined below. Each of these issues could and should be resolved in “peacetime”. Action points are suggested after each item. 1. Comparative evaluation and field validation of rapid test systems. Several studies have been conducted in which one or other of the rapid test systems described above have been compared with conventional laboratory-based testing methods for FMD (Reid et al., 2001; Hearps et al., 2001; Callahan et al., 2001, Ferris et al, in press), demonstrating the efficacy of these methods. However, to the authors’ knowledge, none of the currently available test systems have been fully validated for field use. The authors recommend that the three testing options described above should be applied in parallel and their performance compared in a real “outbreak” situation. A potential opportunity for such field evaluation and validation exists in Anatolia, where FMD is endemic and where such a study could compliment other work already in progress in that region to better understand and control the spread of FMD. An LFD supplied by Princeton Biomedical Corporation was employed with limited success during a pilot study on FMD outbreak investigation conducted in Erzurum in 2004 (Bulut et al., 2004). Planning is currently underway to carry out the initial phase of a field trial using both the 1F10 LFD (Svanova) and the Bioseeq Vet (Smiths) in Turkey, most likely in Erzurum. Action: Recommend continuing support to representatives of the Şap institute (Ankara), the Turkish veterinary authorities, the EUFMD Research Group, and the CRL in their efforts to carry out this study in late 2008 and early 2009. It is important to note that following a primary diagnosis of FMD in any future outbreak, the NRL must ensure that whatever test system will be applied for diagnosis of secondary cases is capable of recognising the causative strain of the virus. 2. Availability of test devices, reagents and equipment Options 3a and 3b require that sufficient LFDs or Bioseeq devices are purchased and stored or that a contract for supply is agreed with a commercial source, in advance of an outbreak. Both the shelf-life and cost of devices should be considered when making these arrangements. Option 2

491

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

requires that regional or mobile laboratories are equipped with the necessary precision instrumentation; the machinery could be provided in advance (and possibly put to other uses in peacetime) or instruments could be stored at the NRL and delivered to where they are most urgently required during an outbreak (the machines could be relocated during the outbreak as events dictate). With respect to either testing option, commercial companies need incentives to develop, validate, produce and market FMD testing systems during peacetime if such tests are to be commercially available when required during an outbreak. Action: representatives of the national veterinary authority should decide in consultation with colleagues in the NRL; EU member states should consult with DG-SANCO on the possibility of LFDs and/or portable PCR units being included in an EU diagnostic reagents bank; the concept of a reagents bank has been the subject of a previous EUFMD position paper by Haas (2003). 3. Determine the levels of proof required to cull a herd or derestrict a herd Who will decide to cull or derestrict a herd? On what basis will this decision be made? Thought should be given as to how decisions will be made at a herd level, and how test results and other evidence (epidemiological and clinical information) will be considered in the decision-making process. It is critical that this is discussed and agreed in advance of an outbreak whilst there is time available for debate. Action: representatives of the national veterinary authority should discuss with colleagues in the NRL and should attempt to formulate a “decision tree”. 4. Training and/or instruction in the use of rapid test systems Options 3a and 3b require that clearly-written, step-by-step instructions are provided with the test devices, such as those developed by FAO for the use of LFDs in the diagnosis of rinderpest (Roeder, 2002). Option 2 will require staff training. Action: the NRL (and perhaps the CRL) could provide instructions and training as necessary. 5. Biosecurity risks and their mitigation Option 3 above does not pose a biosecurity risk (over and above that associated with any field investigation of suspect cases) as the testing is done on the suspect premises and the LFD devices may be discarded or sent to the NRL once the test has been performed. The Bioseeq Vet device may be completely submerged in disinfectant, and the reaction cartridges (containing inactivated samples) may be discarded or sent to the NRL once the test has been performed. Option 2: to avoid the prospect of live FMD virus being handled in non-biosecure laboratory facilities, samples could be treated on collection so as to destroy infectivity whilst preserving RNA. However, this may limit the scope for subsequent characterisation of virus in samples scored as positive. As regards option 1, all NRLs for FMD should operate to the prescribed standards (Anon., 1993). Action: for option 2, protocols for treatment of samples should be developed and agreed that will ensure loss of infectivity without affecting viral RNA recovery. Rapid testing systems (for FMD diagnosis) could potentially be used in peacetime to give an early indication of infection when investigating suspected cases of FMD. They could also be used in active surveillance for FMD in animals that have been sampled and are being tested for the presence of other infectious agents. However rapid test systems (or “penside tests”) are not the panacea for FMD diagnosis and control that some critics of UK government policy in 2001 would contend. Early recognition of the disease will still largely depend on the awareness, vigilance and goodwill of the farming community and practising veterinarians. A rapid and proportionate response to suspected cases will require that national veterinary services are staffed by competent experienced veterinarians who can recognise the clinical disease and are capable of rational decision-making. Furthermore, regardless of what test systems are deployed in any future outbreak, the services of a national reference laboratory will be indispensable, not only to confirm the primary case but to oversee all subsequent testing for the presence of the disease and to co-ordinate if not prosecute the large-scale serological testing required to substantiate freedom from FMD after an outbreak. REFERENCES

492

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

[1] Anon (1993). Security standards for FMD laboratories. Report of the 30th General Session of EUFMD, Rome, Italy; Appendix 6, pages 67-78. http://www.fao.org/ag/againfo/commissions/docs/SecurityStandards.pdf [2] Bulut A, Sammin D, McDonagh O, Dinler U and Unal N (2004). Field evaluation of rapid “penside” tests for FMD antigen and FMDV-NSP antibody; FAO-EUFMD pilot study in Erzurum, Turkey; 12-25 September 2004. Report of the EUFMD Research Group Closed Session, Chania, Crete, 12-15 October 2004; Appendix 13, pages 98-102. http://www.fao.org/ag/againfo/commissions/docs/greece04/App13.pdf [3] Callahan JD, Brown F, Osorio FA, Sur JH, Kramer E, Long GW, Lubroth J, Ellis SJ, Shoulars KS, Gaffney KL, Rock DL and Nelson WM. (2002). Use of a portable real-time reverse transcriptase-polymerase chain reaction assay for rapid detection of foot-and-mouth disease virus. Journal of the American Veterinary Medical Association 220: 1636-1642. [4] Dukes JP, King DP and Alexandersen S. (2006). Novel reverse transcription loop-mediated isothermal amplification for rapid detection of foot-and-mouth disease virus. Archives of Virology 151: 1093-1106. [5] Ferris NP and Dawson, M (1988). Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Veterinary Microbiology 16: 201-209. [6] Ferris N, Reid S, Hutchings G, Kitching P, Danks C, Barker I and Preston S. (2001). Penside test for investigating FMD. Veterinary Record 148: 823-824. [7] Ferris NP, Nordengrahn A, Hutchings GH, Reid SM, King DP, Ebert K, Paton DJ, Kristersson T, Brocchi E, Grazioli S and Merza M (2008). Development and laboratory validation of a lateral flow device for the detection of foot-and-mouth disease virus in clinical samples. Accepted for publication in Journal of Virological Methods. [8] Haas, B (2003) A European Diagnostic reagent bank – what should it contain? Report of the EUFMD Research Group Closed Session, Gersensee, Switzerland 16-19 September 2003; Appendix 18, pages 132-137. [9] Hearps A, Zhang Z and Alexandersen S. (2002). Evaluation of the portable Cepheid SmartCycler real-time PCR machine for the rapid diagnosis of foot-and-mouth disease. Veterinary Record 150: 625-628 [10] King DP, Dukes JP, Reid SM, Ebert K, Shaw AE, Mills CE, Boswell L, Ferris NP (2008). Prospects for rapid diagnosis of foot-and-mouth disease in the field using reverse transcriptasePCR. Veterinary Record 162:315-6. [11] Lau L.-T., Reid S. M., King D. P., Lau A. M. –F., Shaw A. E., Ferris N. P., and Yu A. C. – H. (2008) Detection of foot-and-mouth disease virus by nucleic acid sequence-based amplification (NASBA). Veterinary Microbiology 126 (1-3): 101-110. [12] Reid SM, Ferris NP, Bruning A, Hutchings GH, Kowalska Z and Akerblom L. (2001). Development of a rapid chromatographic strip test for the pen-side detection of foot-and-mouth disease virus antigen. Journal of Virological Methods 96: 189-202. [13] Reid SM, Dukes J, Ebert K, Ferris N and King D. (2006). Diagnosis of FMD by RT-PCR: prospects for mobile and portable assays. Report of the EUFMD Research Group Open Session, Cyprus, October 2006 (in preparation). [14] Roeder PL, Le Blanc Smith PM. (1987). Detection and typing of foot-and-mouth disease virus by enzyme-linked immunosorbent assay: a sensitive, rapid and reliable technique for primary diagnosis. Research in Veterinary Science 43: 225-232. [15] Roeder PL (2002). Pen-side test for the field diagnosis of rinderpest; instructions for use. FAO, Rome, Italy, 2002. [16] Ryan E, Gloster J, Reid SM, Li Y, Ferris NP, Waters R, Juleff N, Charleston B, Bankowski B, Gubbins S, Wilesmith JW, King DP, Paton DJ (2008). Clinical and laboratory investigations of the outbreaks of foot-and-mouth disease in southern England in 2007. Veterinary Record 163:139-47. [17] Snowdon WA (1966). Growth of foot-and-mouth disease virus in monolayer cultures of calf thyroid cells. Nature 210: 1079-1080. ANNEX 1 DECENTRALISED FMD TESTING IN FRANCE A network of five regional laboratories has been put in place in order to perform FMD serology in the event of a future outbreak. These five laboratories have to participate in an annual ring test organized by the national reference laboratory (NRL), AFSSA, Maisons-Alfort. Moreover, they must test 800 sera per year in order to maintain competence. A similar network of five regional laboratories (each with high containment facilities) is to be put in place for rtRT-PCR testing to

493

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

detect FMD virus. As for serological testing, these laboratories would have to participate in ringtests organized by the NRL; in addition regional laboratory staff would receive training from the NRL in biosafety measures that have to be followed. Suspected cases of FMD would continue to be sent to the NRL. The regional laboratories would only be authorised to perform rtRT-PCR in the event of FMD being confirmed by the NRL and only in the case of suspected secondary cases. Some of the laboratories have systems in place for collection of samples on farm. Alternatively, in the case of remote farm locations, a transport system will be put in place to ensure biosecure delivery of specimens to the nearest laboratory. Testing would be performed under the supervision of the NRL and all results would be sent electronically to the NRL for interpretation; no result would be released without NRL authorisation. The extent to which such a system would be used in any future outbreak is difficult to gauge and will depend on such factors as the number of outbreaks and the outbreak locations. Decentralised rtRT-PCR testing is already in place for both bluetongue and avian influenza and has been extensively used for both; a network of ISO-accredited laboratories has been operating under control of the relevant NRL. The requirement for NRL oversight to ensure confidence in test results, imposes an upper limit on the number of laboratories that could be included in such a network. DECENTRALISED FMD TESTING IN GERMANY There are three strands to either current contingency plans or applied research for decentralised FMD testing in Germany: (i) in case of an FMD epidemic - serological testing (by NSP ELISA) in laboratories of the German States (“Länder”) using reagents from a national test kit bank and testing sera derived from low risk premises in which no clinical signs of FMD have been seen; (ii) in “normal times” - screening of samples (by qRT-PCR at German State laboratories) to rule out FMD as a potential, but unlikely differential diagnosis, e.g. in suspected cases of bluetongue. (Note: Where there is a clinical suspicion of FMD, specimens would be delivered with minimum delay by special courier, e.g. police helicopter, to the national reference laboratory FLI, Insel Riems) and (iii) a research project involving the validation of portable PCR machines that might eventually be deployed for use in the field by special task forces with clinical specimens derived from experimentally-infected animals. DECENTRALISED FMD TESTING IN ISRAEL In Israel FMD control policy consists of annual vaccination (with 150% coverage of the national herd) and quarantine in the event of an outbreak. The main problem in dealing with new outbreaks of FMD is the time interval that elapses between first appearance of clinical signs and first recognition of the disease. Rapid test methods will not necessarily ensure rapid recognition of FMD. Of much greater importance in this regard is that all persons involved with the livestock industry are aware of the disease and remain vigilant (regular training to remind key persons about FMD may be very useful in this regard). An LFD (Svanova), will be distributed this year to all regional veterinary services stations (six throughout the country) and this device will be used to support the local veterinary officer when faced with a clinical suspicion of FMD, in deciding whether or not to place the affected herd under quarantine. This new approach will be used in addition to sending samples to NRL for confirmation by conventional testing methods; the time taken to reach the NRL with samples from anywhere in the country would not exceed four hours.

494

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 87

TRANSPORT OF FMDV RNA RATHER THAN LIVE VIRUS; OPTIMISATION OF VIRAL RECOVERY BY TRANSFECTION OF INFECTIOUS RNA G. Belsham

495

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

496

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

Appendix 88

ITEM 6. SERO-SURVEILLANCE IN TURKEY: THE QUESTION OF HARMONISING THE PERFORMANCE/INTERPRETATION OF SP ANTIBODY DATA WITH/BETWEEN RG MEMBER LABORATORIES N. Bulut and C. Potzsch

497

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008

498