wenz iD - Proefschrift Geert van Hout by wenz iD

Experimental Myocardial Infarction The quest for novel therapeutics © G.P.J. van Hout, 2015

ISBN/EAN 978-90-9029256-4 Design Wendy Schoneveld | wenz iD.nl Printed by Drukkerij Damen B.V., Werkendam

Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowledged Financial support by the Stichting Cardiovasculaire Biologie, Heart and Lung Foundation Utrecht, Biocompatibles UK, Transonic, Servier and ChipSoft B.V. is also greatly acknowledged.

Experimental Myocardial Infarction The quest for novel therapeutics

Experimenteel hartinfarct De queeste naar nieuwe therapieën (met een samenvatting in het Nederlands)

Proefschrift

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op donderdag 22 oktober 2015 des middags te 4.15 uur door Gerardus Petrus Johannes van Hout geboren op 15 november 1986 te Geldrop

Promotoren:

Prof. Dr. G. Pasterkamp Prof. Dr. W.W. van Solinge

Co-promotor:

Dr. I.E. Hoefer

â&#x20AC;&#x153;If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find no such case.â&#x20AC;? Charles Darwin

Aan mijn ouders

CONTENT

Chapter 1

Introduction and thesis outline

PART ONE Optimization of porcine study protocols of myocardial infarction

Chapter 2

Cardiac function in a long-term follow-up study of a moderate and severe porcine model of chronic myocardial infarction Biomed Research International 2015 February ePub ahead of print

Chapter 3

Invasive surgery reduces infarct size and preserves cardiac function in a porcine model of myocardial infarction Journal of Cellular and Molecular Medicine. Accepted June 2015

Chapter 4

Study design: Anti-inflammatory compounds to reduce infarct size in large-animal models of myocardial infarction: A meta-analysis Evidence based preclinical medicine. 2014 December 1(1): 4-10

Chapter 5

Anti-inflammatory compounds reduce infarct size after myocardial infarction: A meta-analysis of large-animal models In revision - Cardiovascular Research

PART TWO Validation of measurements and markers of cardiac function

Chapter 6

Admittance-based pressure-volume loop measurements in a porcine model of chronic myocardial infarction Experimental Physiology November 2013;98(11):1565-75

Chapter 7

Admittance-based pressure-volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction Physiological Reports 2014 April 22;2(4):e00287

115

Chapter 8

Elevated mean neutrophil volume represents altered neutrophil composition and reflects damage after myocardial infarction In revision - Basic Research in Cardiology

129

PART THREE The search for novel therapeutic strategies

Chapter 9

Targeting danger associated molecular patterns after myocardial infarction Expert Opinion on Therapeutic Targets - Accepted August 2015

149

Chapter 10

Selective inhibition of the NLRP3-inflammasome dose dependently reduces infarcts size and preserves cardiac function in a porcine model of myocardial infarction In revision - European Heart Journal

177

Chapter 11

Intracoronary infusion of encapsulated glucagon-like peptide-1eluting mesenchymal stem cells preserves left ventricular function in a porcine model of acute myocardial infarction Circulation Cardiovascular Interventions 2014 Oct;7(5):673-83

201

PART FOUR Summary and discussion

Chapter 12

Summary & General Discussion

Chapter 13

Dutch Summary

235 247

PART FIVE Addendum Review committee Dankwoord (in Dutch) List of publications Curriculum Vitae (in Dutch)

260 261 266 268

LIST OF ABBREVIATIONS 3D-echocardiography area at risk alanine transaminase acute myocardial infarction admittance system aspartate transaminase adenosine triphosphate arbitrary units coronary artery disease closed chest creatine kinase cardiac magnetic resonance imaging c-reactive protein conductance system danger associated molecular patterns end diastolic volume ejection fraction encapsulated mesenchymal stem cells end systolic pressure volume relationship end systolic volume fractional area shortening fractional shortening follow-up glucagon-like peptide-1 heart failure high mobility group box-1 heat shock proteins interleukin ischemia-reperfusion injury infarct size left anterior descending artery left circumflex artery left ventricle left ventricular end diastolic volume left ventricular ejection fraction left ventricular end systolic volume mean arterial pressure myocardial infarction

3DE AAR ALT AMI AS AST ATP AU CAD CC CK CMRI CRP CS DAMPs EDV EF eMSC ESPVR ESV FAS FS FU GLP-1 HF HMGB1 HSPs IL IR-injury IS LAD LCx LV LVEDV LVEF LVESV MAP MI

matrix metallo proteinases mean neutrophil volume magnetic resonance imaging mesenchymal stem cells non-steroid anti-inflammatory drugs open chest pathogen associated molecular patterns peripheral blood mononuclear cells phosphate buffered saline percutenous coronary intervention polymorphonuclear leukocytes pattern recognition receptors pressure volume receptor for advanced glycation and products reactive oxygen species standard deviation standard error of the mean systolic wall thickening trans-esophageal echocardiography toll-like recepors troponin I triphenyl tetrazolium chloride utrecht patient oriented database ventricular fibrilation wall thickness

MMPs MNV MRI MSC NSAIDs OC PAMPs PBMCs PBS PCI PMNs PRRs PV RAGE ROS SD SEM SWT TEE TLRs TnI TTC UPOD VF WT

chapter 1

Introduction and thesis outline

CHAPTER 1

CLINICAL RELEVANCE Coronary artery disease (CAD) is the deathliest disease of the western world1–3. Each year 3.8 million men and 3.4 million women die from CAD worldwide1. Myocardial infarction (MI) comprises an important part of CAD and occurs due to rupture of an instable atherosclerotic plaque, enabling clot formation and occlusion of a coronary artery. Since MI is partly attributable to risk factors such as age, gender, smoking, obesity, hypertension and diabetes, the incidence of MI, especially in the developing countries, is still increasing2. The treatment for MI has evolved rapidly the past decades from conservative approaches to minimally invasive low-risk procedures3,4. The main purpose in patients suffering from MI is to restore coronary blood flow through percutaneous coronary intervention (PCI) and stenting. Due to the development of these treatment strategies, the mortality in the western world after MI has dropped considerably during the last decades5. However, patients that survive MI often have a severely deteriorated cardiac function and are therefore at increased risk of heart failure (HF)6. HF is associated with a poor prognosis and has a 5-year mortality rate of more than 65% after initial hospitalization7. Since HF is a chronic disease, it is responsible for considerable health care costs, decreases the quality of life and is a true burden for society8,9. To prevent a further increase in the incidence of HF and improve survival after MI, novel therapeutics are needed that protect the heart, thereby preventing the progression from MI to HF. Ischemia-reperfusion injury To design novel therapies that target molecular processes involved in MI, extensive knowledge on the pathophysiology of this disease is mandatory. The majority of available data has been gathered by studying artificially induced MI in animal studies (experimental MI), while some of these processes have been confirmed in patients10–12. Coronary occlusion leads to impaired cardiac blood flow. Since the myocardium becomes deprived of oxygen and nutrients, sustained or permanent ischemia results in irreparable damage of the area at risk (AAR). This ‘’wave front’’ of cell death begins in the subendocardium and will extend transmurally over time to the epicardium13. The acute drop in oxygen levels disables oxidative phosphorylation resulting in ATP depletion. Due to the absence of ATP, cardiomyocytes are unable to contract. This leads to the switch from aerobic to anaerobic cellular metabolism, culminating in lactate production and a decreased pH. The increase in hydrogen ions activates the Na+/H+ exchanger, pumping proton ions out from the cell in exchange for sodium ions. In turn, the intracellular accumulation of Na+ enables reverse activation of the 2Na+/Ca2+ exchanger, leading to intracellular calcium overload causing cardiomyocyte hypercontraction and eventual cell death. To reverse this detrimental mechanism that permanently damages the myocardium, the timely restoration of cardiac perfusion (reperfusion) is essential. It has been unequivocally proven that reperfusion salvages cardiac tissue from ischemic cell death, thereby preserving cardiac function and preventing HF. Paradoxically, reperfusion itself can also independently induce damage to cardiomyocytes (reperfusion injury)14,15.

INTRODUCTION AND THESIS OUTLINE

During reperfusion the cardiac tissue is abruptly reoxygenized. Cardiomyocytes switch back to aerobic metabolism enabling the optimal exchange of H+/Na+ for fast pH normalization. However, this rapid rise in intracellular sodium levels causes cellular swelling and may culminate in cell death. To prevent cells from bursting, Na+ is, much faster than during ischemia, exchanged for Ca2+. These elevated intracellular calcium levels lead to hypercontraction of the cardiomyocyte with subsequent membrane rupture and cell death. Additionally, reoxygenation induces an oxygen free radical burst, causing lipid membrane oxidation and cellular destruction. A combination of rapid pH normalization, raised levels of reactive oxygen species and high intracellular calcium levels leads to mitochondrial dysfunction, thereby directly interfering with ATP production. Experimental and clinical studies have shown that, when administering compounds just before reperfusion, infarct size can be reduced with more than a third14. This phenomenon of infarct size reduction through administration of pharmacological compounds prior to reperfusion is termed cardioprotection (figure 1)16 and forms the largest body of evidence that ischemia-reperfusion injury exists. For several decades, investigators have very successfully inhibited mechanisms responsible for myocardial ischemia-reperfusion injury and reduced infarct size after experimental MI13. However, translation of these positive findings to daily clinical practice has been proven to be difficult17.

Infarct Size

2 Revascularization

Coronary artery occlusion

Time

Figure 1. Schematic overview of the rationale behind cardioprotection Area 1 represents the theoretical infarct size due to ischemia alone. Area 3 represents the amount of ischemic tissue that is salvaged by timely PCI. Area 2 shows that the presence of reperfusion results in additional injury and attenuates the positive effect of PCI. The damage depicted in area 2 could be abolished by administration of a cardioprotective pharmacological compound just before reperfusion.

CHAPTER 1

Inflammation and adverse remodeling The inflammatory response plays a prominent role in ischemia-reperfusion injury and eventually contributes to additional damage on top of ischemia. The release of intracellular substances from the damaged myocardium directly after reperfusion leads to platelet activation, endothelial dysfunction and neutrophil plugging, resulting in the no-reflow phenomenon and intramyocardial hemorrhage, causing additional damage18,19. At the same time, these intracellular molecules can serve as danger-associated-molecularpatterns (DAMPs) that are released by the damaged myocardium. These DAMPS can be recognized by surface membrane receptors present on both resident and circulating cells. These so-called pattern recognition receptors (PRRs) can bind DAMPs20. Upon activation, PRRs serve as signal transduction molecules that induce a pro-inflammatory state, which culminates in the synthesis and release of cytokines. Upon reperfusion, a cardiac cytokine burst is initiated that induces leucocyte activation and infiltration21. These circulating cells, of which neutrophils are the first responders, cause injury to endothelial cells due to the excretion of reactive oxygen species, lipids, cytokines and proteases21. As a result, endothelial cells start to express cell adhesion molecules, enabling the transmigration of neutrophils, followed by other inflammatory cells, into the damaged myocardium. Here these inflammatory cells generate more reactive oxygen species and proteases resulting into tissue breakdown and cellular clearance, thereby directly contributing to myocardial ischemia-reperfusion injury22. Although inflammatory cells are required to clear out necrotic tissue and initiate cardiac wound healing, accumulative evidence exists that the inflammatory response in the (sub) acute phase after MI also is detrimental to cardiac function and overall prognosis in the long run, since the severity of this response is directly related to the prognosis of patients23â&#x20AC;&#x201C;26. Once residing in the damaged cardiac tissue, circulating cells exert detrimental effects on the healing myocardium (figure 2). Additionally, neutrophils produce matrix metalloproteinases (MMPs) that degrade the intermyocyte collagen struts thereby destabilizing the ventricular wall. This leads to infarct expansion, a process that is known to occur within hours from initial myocyte injury27. Depending on the size and location of the infarct, the loss of contractile myocardium leads to an alteration of intraventricular pressure and increased wall stress. The increased wall stress gradually induces myocyte slippage in the border and remote area of the left ventricle. Together with infarct expansion, this leads to further left ventricular wall thinning and progressive dilatation within the first weeks after MI. This process, referred to as adverse cardiac remodeling leads to systemic neurohormonal adaptations and eventually culminates in HF27,28. The exaggerated inflammatory response post-MI enhances acute reperfusion injury and is associated with an increased risk of adverse remodeling and a worse prognosis post-MI. Hence, inhibition of inflammation post-MI may serve as a clinically relevant target to preserve cardiac function and reduce infarct size. However, to this date no anti-inflammatory compounds have been implemented in daily clinical practice, since inhibition of various inflammation-related pathways has proven to be unsuccessful in clinical studies 29â&#x20AC;&#x201C;31.

INTRODUCTION AND THESIS OUTLINE

Nevertheless, newly discovered pathways have recently been shown to play a crucial role post-MI and may serve as interesting therapeutic targets for future clinical testing. Large animal models Before clinical testing of therapeutics to preserve cardiac function and reduce infarct size is considered, thorough testing in clinically relevant animal models should be performed32. In vitro experiments and small animal studies are absolutely essential to map molecular pathways and assess the role of different molecules in these signaling pathways. To translate these findings into clinical applications, large animal models of MI are required to confirm findings from studies performed in small animals, since the development of tissue damage post-MI differs in small animals compared to larger mammals17,33. Certain inflammationrelated pathways are also substantially different in small rodents compared to humans34. Additionally, large animal models show great similarity with humans regarding hemodynamics, cardiac anatomy, (coronary) pharmacokinetics and 窶電ynamics35,36 and allow the use of clinical treatment regimens and administration routes along with identical function-related read-outs37,38. Large animal models also allow for a thorough assessment of cardiac function in a controlled experimental setting. The development of new functional read-outs and markers for cardiac function could therefore be of additional value in these studies since these methods could provide essential information on the efficacy of promising therapeutics. Despite these advantages, many compounds that were successful in large animal models eventually failed in clinical trials, suggesting optimization of experimental protocols is necessary to increase the translational value of these large animal MI models10,14. Protocol optimization will not only contribute to therapy development, but also reduce the number of animals required for translational purposes.

Figure 2. Infarct expansion caused by inflammation MI leads to a release of danger molecules into the systemic circulation. As a result, circulating leukocytes migrate to the previously ischemic area (AAR) and attack damaged but viable cardiomyocytes, causing expansion of the infarct area.

CHAPTER 1

Thesis outline Since MI and its consequences are great societal burdens, novel therapeutics that protect the heart from reperfusion injury and adverse remodeling are required. To assess safety and efficacy before exposing patients to experimental compounds, thorough preclinical testing in representative translational animal models is required. However, successful clinical implementation of cardioprotective compounds has not yet been established, while many animal studies have been found positive. This implies that experimental study design is not representative for clinical practice. The quest for novel therapeutics for patients suffering from MI should therefore not only focus on a continued search for novel compounds to treat MI patients but also focus on improving experimental MI study protocols and functional read-outs to increase the translational value of large animal experimental MI studies. Accordingly, the aim of the current thesis is to (1) optimize large animal model study protocols, (2) to validate new measurements that reflect cardiac function and (3) continue the search for therapeutic strategies that attenuate ischemia-reperfusion injury or prevent adverse remodeling in patients post-MI. This thesis is divided into three parts based on these aims, followed by a general discussion and Dutch summary (part four). PART ONE Part one deals with the optimization of porcine models of MI and tries to identify possible confounding factors that hamper clinical translation of promising therapeutics. To increase the translational value of large animal MI models, selection of the appropriate model is essential. Accordingly, extensive knowledge on the disease progression after MI induction in these models is needed. In chapter 2 we therefore studied cardiac function of two distinctive chronic porcine models of MI. In this project we aimed to assess the effect of different coronary occlusion sites on cardiac function and left ventricular geometrics after long-term follow-up. This information will not only contribute to better selection of models for certain therapies, but also increase the knowledge on natural disease progression in patients with different localizations of their MI. Although many different coronary occlusion methods have been described in large animal MI research, two surgical approaches are generally used in this field. The first method requires invasive surgery, while the second method follows a transluminal approach. Since the first method has less clinical similarity, this artificial induction of MI might influence the size of the infarction. For this reason, we assessed the effect of invasive surgery on infarct size in a porcine MI model in chapter 3. Apart from coronary occlusion site and surgical approach, many other different study characteristics can influence the effect of certain therapies. Since anti-inflammatory therapies have been under investigation for the treatment of MI for several decades, while none of these therapies have made it into daily clinical practice, in chapter 4 and 5 we studied the overall effect of these therapies on infarct size in large animal MI studies. Chapter 4 contains the study design of this meta-analysis of anti-inflammatory compounds in large animals of MI and in chapter 5 we provide the results of this study.

INTRODUCTION AND THESIS OUTLINE

PART TWO Part two of this thesis contains studies investigating novel tools to determine cardiac function. Large animal MI studies allow a detailed assessment of cardiac function after therapy administration. Hence, the development of novel cardiac function measurements could contribute to the understanding of the effect of cardioprotective drugs on left ventricular function. Real-time pressure volume (PV) measurements provide extensive information on left ventricular function39. In chapter 6 and 7 we validated cardiac function measurements with a newly developed admittance-based PV system. MI induces an intense inflammatory response. The severity of the inflammatory response is associated with the prognosis of patients after MI. Different parameters of this immune reaction could therefore serve as prognostic markers after MI. Especially neutrophils have previously been shown to predict prognosis in acute MI patients40. In chapter 8 we aimed to investigate the effect of MI on the size of the average circulating neutrophil, referred to as mean neutrophil volume (MNV). PART THREE Since MI and HF are associated with considerable morbidity and mortality, new therapies that protect the heart after ischemia-reperfusion are necessary. Part three consists of the evaluation of novel therapeutics to reduce infarct size and preserve cardiac function, particularly aimed at inflammation. In chapter 9 we discuss the process by which the upstream inflammatory response enhances myocardial reperfusion injury in detail. In this chapter, we provide a concise overview of the discovered DAMPS and PRRs that play a role in MI and postulate that attenuation of some of these interactions form candidates for cardioprotection in patients. An essential upstream mediator of inflammation after MI is the NLRP3-inflammasome41. This intracellular protein complex is activated upon cardiac injury and controls the conversion of inactive interleukin (IL)-1β and IL-1842 into their active state. These cytokines are potent stimulators of the inflammatory reaction and directly influence cardiac performance43. In chapter 10 we therefore investigated the effect of NLRP3-inflammasome inhibition on infarct size reduction and cardiac function preservation by administration of the selective small-molecule inhibitor MCC950 in a pig model of 75-minute ischemiareperfusion injury. Mesenchymal stem cells (MSCs) have been proven to exert beneficial paracrine effects after experimental MI by secreting different cardioprotective substances44,45. CellBeads™ are alginate beads that contain MSCs46. The alginate prevents the host’s immune system from attacking the encapsulated MSCs (eMSCs), while it allows for diffusion from nutrients and paracrine factors to and from the eMSCs. In chapter 11 we evaluated the efficacy of infusion of eMSCs, which have been genetically programmed to produce the cardioprotective glucagon-like peptide-1 (GLP-1), in a porcine model of 90 minutes ischemia-reperfusion injury.

CHAPTER 1

REFERENCES 1. Who. WHO | The Atlas of Heart Disease and Stroke. 2. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit J a., Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart Disease and Stroke Statistics - 2014 Update: A report from the American Heart Association. Circulation. 2014;129:e28–e292. 3. Steg PG, James SK, Atar D, Badano LP, Lundqvist CB, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, Van’T Hof A, Widimsky P, Zahger D, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, FunckBrentano C, Hasdai D, Hoes A, Kirchhof P, Kolh P, McDonagh T, Moulin C, Popescu B a., Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Astin F, Åström-Olsson K, Budaj A, Clemmensen P, Collet JP, Fox K a., Fuat A, Gustiene O, Hamm CW, Kala P, Lancellotti P, Maggioni A Pietro, Merkely B, Neumann FJ, Piepoli MF, Van De Werf F, Verheugt F, Wallentin L, Blömstrom-Lundqvist C, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, van ’t Hof A, Widimsky P, Zahger D. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 2012;33:2569–2619. 4. Kristensen SD, Laut KG, Fajadet J, Kaifoszova Z, Kala P, Di Mario C, Wijns W, Clemmensen P, Agladze V, Antoniades L, Alhabib KF, De Boer MJ, Claeys MJ, Deleanu D, Dudek D, Erglis A, Gilard M, Goktekin O, Guagliumi G, Gudnason T, Hansen KW, Huber K, James S, Janota T, Jennings S, Kajander O, Kanakakis J, Karamfiloff KK, Kedev S, Kornowski R, Ludman PF, Merkely B, Milicic D, Najafov R, Nicolini F a., Noč M, Ostojic M, Pereira H, Radovanovic D, Sabaté M, Sobhy M, Sokolov M, Studencan M, Terzic I, Wahler S, Widimsky P. Reperfusion therapy for ST elevation acute myocardial infarction 2010/2011: Current status in 37 ESC countries. Eur Heart J. 2014;35:1957–1970. 5. Hartstichting. Feiten en Cijfers Hartinfarct. 2008; 6. Hartstichting. Feiten en Cijfers Hartfalen. 2008; 7. Vaartjes I, Hoes a W, Reitsma JB, de Bruin a, Grobbee DE, Mosterd a, Bots MI. Age- and gender-specific risk of death after first hospitalization for heart failure. BMC Public Health. 2010;10:637.

8. Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SMJM, Hofman A, Deckers JW, Witteman JCM, Stricker BHC. Quantifying the heart failure epidemic: Prevalence, incidence rate, lifetime risk and prognosis of heart failure - The Rotterdam Study. Eur Heart J. 2004;25:1614–1619. 9. Mcmurray JJ V, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, Falk V, Filippatos G, Fonseca C, Gomez-Sanchez MA, Jaarsma T, Kober L, Lip GYH, Maggioni A Pietro, Parkhomenko A, Pieske BM, Popescu B a., Ronnevik PK, Rutten FH, Schwitter J, Seferovic P, Stepinska J, Trindade PT, Voors A a., Zannad F, Zeiher A, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Kirchhof P, Knuuti J, Kolh P, Mcdonagh T, Moulin C, Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Bonet LA, Avraamides P, Ben Lamin H a., Brignole M, Coca A, Cowburn P, Dargie H, Elliott P, Flachskampf FA, Guida GF, Hardman S, Iung B, Merkely B, Mueller C, Nanas JN, Nielsen OW, Orn S, Parissis JT, Ponikowski P. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012. Eur J Heart Fail. 2012;14:803–869. 10. Heide RS Vander, Steenbergen C. Cardioprotection and myocardial reperfusion: Pitfalls to clinical application. Circ Res. 2013;113:464–477. 11. Lønborg J, Vejlstrup N, Kelbæk H, Bøtker HE, Kim WY, Mathiasen AB, Jørgensen E, Helqvist S, Saunamki K, Clemmensen P, Holmvang L, Thuesen L, Krusell LR, Jensen JS, Køber L, Treiman M, Holst JJ, Engstrøm T. Exenatide reduces reperfusion injury in patients with ST-segment elevation myocardial infarction. Eur Heart J. 2012;33:1491– 1499. 12. Turer AT, Hill J a. Pathogenesis of myocardial ischemia-reperfusion injury and rationale for therapy. Am J Cardiol. 2010;106:360–368. 13. Hausenloy DJ, Yellon DM. Myocardial ischemiareperfusion injury: a neglected therapeutic target. J Clin Invest. 2013;123:92–100. 14. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 15. Ibáñez B, Heusch G, Ovize M, Van de Werf F. Evolving Therapies for Myocardial Ischemia/Reperfusion Injury. J Am Coll Cardiol. 2015;65:1454–1471. 16. Sluijter JPG, Condorelli G, Davidson SM, Engel FB, Ferdinandy P, Hausenloy DJ, Lecour S, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Van Laake LW. Novel therapeutic strategies for cardioprotection. Pharmacol Ther. 2014;144:60–70. 17. Hausenloy DJ, Baxter G, Bell R, Bøtker HE, Davidson SM, Downey J, Heusch G, Kitakaze M, Lecour S, Mentzer R, Mocanu MM, Ovize M, Schulz R, Shannon R, Walker M, Walkinshaw G, Yellon DM. Translating novel strategies for cardioprotection: The Hatter Workshop Recommendations. Basic Res Cardiol. 2010;105:677–686.

INTRODUCTION AND THESIS OUTLINE

18. Rezkalla SH, Kloner R a. Coronary no-reflow phenomenon: From the experimental laboratory to the cardiac catheterization laboratory. Catheter Cardiovasc Interv. 2008;72:950–957. 19. Betgem RP, de Waard G a., Nijveldt R, Beek AM, Escaned J, van Royen N. Intramyocardial haemorrhage after acute myocardial infarction. Nat Rev Cardiol. 2014;12:156–167. 20. De Haan JJ, Smeets MB, Pasterkamp G, Arslan F. Danger signals in the initiation of the inflammatory response after myocardial infarction. Mediators Inflamm. 2013;2013:206039. 21. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 22. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 23. Van der Laan a. M, Nahrendorf M, Piek JJ. Healing and adverse remodelling after acute myocardial infarction: role of the cellular immune response. Heart. 2012;98:1384–1390. 24. Maekawa Y, Anzai T, Yoshikawa T, Asakura Y, Takahashi T, Ishikawa S, Mitamura H, Ogawa S. Prognostic significance of peripheral monocytosis after reperfused acute myocardial infarction:a possible role for left ventricular remodeling. J Am Coll Cardiol. 2002;39:241–246. 25. Takahashi T, Hiasa Y, Ohara Y, Miyazaki S-I, Ogura R, Suzuki N, Hosokawa S, Kishi K, Ohtani R. Relationship of admission neutrophil count to microvascular injury, left ventricular dilation, and long-term outcome in patients treated with primary angioplasty for acute myocardial infarction. Circ J. 2008;72:867–872. 26. Van Der Laan AM, Hirsch A, Robbers LFHJ, Nijveldt R, Lommerse I, Delewi R, Van Der Vleuten P a., Biemond BJ, Zwaginga JJ, Van Der Giessen WJ, Zijlstra F, Van Rossum AC, Voermans C, Van Der Schoot CE, Piek JJ. A proinflammatory monocyte response is associated with myocardial injury and impaired functional outcome in patients with STsegment elevation myocardial infarction: Monocytes and myocardial infarction. Am Heart J. 2012;163:57–65.e2. 27. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988. 28. Heusch G, Libby P, Gersh B, Yellon D, Böhm M, Lopaschuk G, Opie L. Cardiovascular remodelling in coronary artery disease and heart failure. Lancet. 2014;383:1933–43. 29. Armstrong PW, Granger CB, Adams PX, Hamm C, Holmes D, O’Neill WW, Todaro TG, Vahanian A, Van de Werf F. Pexelizumab for acute STelevation myocardial infarction in patients undergoing primary percutaneous coronary intervention: a randomized controlled trial. JAMA. 2007;297:43–51.

30. Baran KW, Nguyen M, McKendall GR, Lambrew CT, Dykstra G, Palmeri ST, Gibbons RJ, Borzak S, Sobel BE, Gourlay SG, Rundle a C, Gibson CM, Barron H V. Double-blind, randomized trial of an anti-CD18 antibody in conjunction with recombinant tissue plasminogen activator for acute myocardial infarction: limitation of myocardial infarction following thrombolysis in acute myocardial infarction (LIMIT AMI) stu. Circulation. 2001;104:2778–2783. 31. Faxon DP, Gibbons RJ, Chronos N a F, Gurbel P a., Sheehan F. The effect of blockade of the CD11/ CD18 integrin receptor on infarct size in patients with acute myocardial infarction treated with direct angioplasty: The results of the HALT-MI study. J Am Coll Cardiol. 2002;40:1199–1204. 32. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 33. Skyschally A, Van Caster P, Boengler K, Gres P, Musiolik J, Schilawa D, Schulz R, Heusch G. Ischemic postconditioning in pigs: No causal role for risk activation. Circ Res. 2009;104:15–18. 34. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker H V, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, López CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West M a, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier R V, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A. 2013;110:3507– 12. 35. Munz MR, Faria MA, Monteiro JR, Águas AP, Amorim MJ. Surgical porcine myocardial infarction model through permanent coronary occlusion. Comp Med. 2011;61:445–452. 36. Lukács E, Magyari B, Tóth L, Petrási Z, Repa I, Koller A, Horváth I. Overview of large animal myocardial infarction models (review). Acta Physiol Hung. 2012;99:365–81. 37. De Jong R, van Hout GPJ, Houtgraaf JH, Kazemi K, Wallrapp C, Lewis A, Pasterkamp G, Hoefer IE, Duckers HJ. Intracoronary Infusion of Encapsulated Glucagon-Like Peptide-1-Eluting Mesenchymal Stem Cells Preserves Left Ventricular Function in a Porcine Model of Acute Myocardial Infarction. Circ Cardiovasc Interv. 2014;7:673–683. 38. McCall FC, Telukuntla KS, Karantalis V, Suncion VY, Heldman AW, Mushtaq M, Williams AR, Hare JM. Myocardial infarction and intramyocardial injection models in swine. Nat Protoc. 2012;7:1479–1496.

CHAPTER 1

39. Krenz M. Conductance, admittance, and hypertonic saline: should we take ventricular volume measurements with a grain of salt? J Appl Physiol. 2009;107:1683–1684. 40. Horne BD, Anderson JL, John JM, Weaver A, Bair TL, Jensen KR, Renlund DG, Muhlestein JB. Which white blood cell subtypes predict increased cardiovascular risk? J Am Coll Cardiol. 2005;45:1638–1643. 41. Takahashi M. NLRP3 Inflammasome as a Novel Player in Myocardial Infarction. Int Heart J. 2014;55:101–105. 42. Brydges SD, Broderick L, McGeough MD, Pena CA, Mueller JL, Hoffman HM. Divergence of IL-1, IL-18, and cell death in NLRP3 inflammasomopathies. J Clin Invest. 2013;123:4695–705. 43. Toldo S, Mezzaroma E, O’Brien L, Marchetti C, Seropian IM, Voelkel NF, Van Tassell BW, Dinarello C a., Abbate a. Interleukin-18 mediates interleukin1-induced cardiac dysfunction. AJP Hear Circ Physiol. 2014;306:H1025–H1031. 44. Timmers L, Lim SK, Hoefer IE, Arslan F, Lai RC, van Oorschot A a M, Goumans MJ, Strijder C, Sze SK, Choo A, Piek JJ, Doevendans P a., Pasterkamp G, de Kleijn DP V. Human mesenchymal stem cellconditioned medium improves cardiac function following myocardial infarction. Stem Cell Res. 2011;6:206–214. 45. Timmers L, Lim SK, Arslan F, Armstrong JS, Hoefer IE, Doevendans P a., Piek JJ, El Oakley RM, Choo A, Lee CN, Pasterkamp G, de Kleijn DP V. Reduction of myocardial infarct size by human mesenchymal stem cell conditioned medium. Stem Cell Res. 2008;1:129–137. 46. Wallrapp C, Thoenes E, Thürmer F, Jork A, Kassem M, Geigle P. Cell-based delivery of glucagon-like peptide-1 using encapsulated mesenchymal stem cells. J Microencapsul. 2012;30:1–10.

part one

OPTIMIZATION OF PORCINE STUDY PROTOCOLS OF MYOCARDIAL INFARCTION

chapter 2

Cardiac function in a long-term follow-up study of moderate and severe porcine models of chronic myocardial infarction

Biomed Research International 2015 February ePub ahead of print

R. de Jong1, G.P.J. van Hout2, J.H. Houtgraaf1, S. Takashima1, G. Pasterkamp2, I.E. Hoefer2, H.J. Duckers2 1

Molecular Cardiology Laboratory Erasmus University Medical Center, Rotterdam, The Netherlands

Experimental Cardiology Laboratory University Medical Center Utrecht, Utrecht, The Netherlands

PART ONE CHAPTER 2

ABSTRACT Background Novel therapies need to be evaluated in a relevant large animal model that mimics the clinical course and treatment in a reasonable time frame. To reliably assess therapeutic efficacy, knowledge regarding the translational model and the course of disease is needed. Methods Landrace pigs were subjected to a transient occlusion of the proximal left circumflex artery (LCx), (n=6) or mid left anterior descending artery (LAD), (n=6) for 150 min. Cardiac function was evaluated before by 2D-echocardiography or 3D-echocardiography and pressurevolume loop analysis. At 12 weeks follow-up the heart was excised for histological analysis and infarct size calculations. Results Directly following AMI, LVEF was severely reduced compared to baseline in the LAD group (-17.1Âą 1.6%, p=0.009) compared to only a moderate reduction in the LCx group (-5.9Âą 1.5 %, p=0.02) and this effect remained unchanged during 12 week follow-up. Conclusion Two models of chronic MI, representative for different patient groups, can reproducibly be created through clinically relevant ischemia-reperfusion of the mid-LAD and proximal LCx.

MODERATE AND SEVERE MODEL OF CHRONIC MI

INTRODUCTION The treatment of patients suffering from myocardial infarction (MI) is aimed at the preservation of cardiac function. Most treatment regimens directly target pathways that limit infarct size or reduce adverse remodeling, thereby preventing progression into heart failure (HF)1–3. To fully determine the possible effect of such therapies, thorough testing in clinically relevant animal models is needed4,5. Since large animal models enable clinical treatment regimens, delivery route and identical function-related measurements, they are considered to withhold greater translational value than small animal models and are therefore superior for efficacy testing6–10. Importantly, due to optimized logistical and diagnostic health care, the complaint-toneedle-time has decreased considerably in the western world11. This has resulted in a large proportion of patients with a relatively preserved left ventricular function after MI12. These patients could, however, still benefit from therapy that further confines the amount of damage directly post-MI. To test the efficacy of such therapies, an animal model that closely resembles the clinical course of disease in a mildly damaged heart is mandatory. At the same time, therapy optimization has also resulted in an increased incidence of HF, since patients survive with severely deteriorated cardiac function13. This patient group would greatly benefit from therapy that improves cardiac function. For this purpose, an animal model that closely resembles the progression into HF after MI in patients with a severely damaged heart is needed. To create both a moderate and a severe model of chronic myocardial infarction that resemble these different patient-groups, ischemia can be induced in several ways. This includes permanent ligation, progressing occlusion by placement of ameroid constrictors, a bottleneck stent model, coiling or infusion of ethanol in the target coronary artery14–18. However, most of these models do not mimic current clinical reality, where most patients are revascularized within relatively short time after occlusion. Since revascularization of the target vessel remains the cornerstone treatment in MI patients, animal models have to simulate this situation. Even more so since myocardial wound healing and other molecular mechanisms are substantially different after revascularization compared to those during persistent ischemia19,20. Moreover, the chronic coronary occlusion precludes intracoronary infusions, the easiest and fastest technique for myocardial delivery of therapeutics. Hence, models of ischemia/reperfusion, in which coronary flow is restored after an ischemic period, seem most suited to validate novel therapies for the treatment of MI. In most ischemia/reperfusion models, the left anterior descending artery (LAD) is occluded, which is associated with a high mortality rate during infarct induction and reperfusion due to ventricular fibrillation (VF)8,21. As an alternative for LAD occlusion, the left circumflex artery (LCx) can be occluded. The LCx is responsible for approximately 20% of the blood supply to the left ventricle and an occlusion-reperfusion model of this artery may result in a lower mortality rate21. However, it is not fully elucidated to what extent animals that are subjected to LCx ischemia-reperfusion develop cardiac dysfunction during long term follow-up. Moreover, no thorough sequential investigation of cardiac (dys)function and

PART ONE CHAPTER 2

ventricular dilatation has been performed in either model to the best of our knowledge. Therefore, this study was designed to investigate the long-term effect of myocardial infarction on cardiac function in two clinically relevant large animal models of MI: a severe mid-LAD ischemia-reperfusion and a moderate proximal LCx ischemia-reperfusion model. In the present study we investigated the course of both global and regional cardiac function decay in the two models combined with mortality, infarct size and histological analysis. Objectives To determine the long-term course of cardiac function in a moderate LCx and a severe LAD ischemia-reperfusion model for future validation of novel therapeutic strategies. We hypothesized that LAD ischemia-reperfusion will result in fast development of cardiac dysfunction representing a good model for severely affected patients with cardiac dilatation and signs of heart failure. Secondly, we hypothesized that the LCx ischemia-reperfusion model leads to a moderately affected heart representing a large patient group that has been adequately revascularized after an initial ischemic event but could still benefit from additional therapy.

MATERIAL AND METHODS All procedures were approved by the local animal welfare committee of the University of Utrecht (Utrecht, The Netherlands; Permit number 2011.II.04.068). A total of 12 female specific pathogen free (SPF) landrace pigs were included (65.1 ± 1.0 kg; Van Beek, Lelystad, The Netherlands). The pigs were housed in the experimental animal facility at the University of Utrecht (Utrecht, The Netherlands) in a group prior to procedure and individually after the AMI (to prevent hostile behavior after infarct procedure). The animals were checked by a veterinarian upon arrival at the facility and daily scored by a bio-technician and once a week by a veterinarian during follow-up. Six animals were randomized to undergo ischemia-reperfusion of the LCx and 6 animals were randomized to undergo LAD ischemia reperfusion. Medical treatment before infarct induction All animals received dual anti-platelet therapy (Acetyl Salicylic Acid 80 mg/d: Ratiopharm, Haarlem, The Netherlands; clopidogrel 75 mg/d: Apothecon, Barneveld, The Netherlands) and anti-arrhythmic drugs (Amiodarone, 1200 mg loading dose, 800 mg qd; Sanofi-Aventis, Paris, France) starting 10 days prior to infarct induction up until sacrifice at 12 weeks FU. Anesthesia protocol General anesthesia was induced with an intramuscular injection of 0.5 mg/ kg midazolam (Actavis, Zug, Switzerland), 10 mg/kg ketamine (Narketan,Vétoquinol, Lure Cedex, France) and 1 mg of Atropine (Pharmachemie BV, The Netherlands) and maintained with intravenous infusion of midazolam 0.5 mg/kg/h, sufentanil 2.5 µg/kg/h (Janssen-Cilag B.V, Tilburg, The

MODERATE AND SEVERE MODEL OF CHRONIC MI

Netherlands) and pancuronium 0.1 mg/kg/h (Inresa, Battenheim, Germany). Upon infarct induction, all animals were therapeutically heparinized with 2 doses of 5000 IE (Leo pharma, Ballerup, Denmark). All pigs received a Fentanyl patch (25mg, Janssen-Cilag,Tilburg Netherlands) and meloxacam (Boehringer-Ingelheim, Alkmaar, The Netherlands) 0.5 mg/ kg/d, as post-surgery analgesia. Infarct procedure Animals were randomized before the start of the procedure via sealed envelopes. Cardiac function at baseline was quantified by 2D-echocardiography and pressure-volume (PV)-loop analysis. Blood was sampled before AMI and 6 hours after for Troponin I analysis. All animals received a RevealTM event recorder (Medtronic, Tilburg, The Netherlands). Balloon occlusion of the LCx An 8F sheath was inserted in the carotid artery and an 8F guiding catheter (JL 3.5-4.0, Boston Scientific Nederland B.V., Nieuwegein, The Netherlands) was positioned at the ostium of the left main coronary artery. An angioplasty balloon (Trek 3,5-4,0 x12, Abbott) was inflated (8-14 bar) for 150 min in the proximal LCx (figure 1). After balloon inflation, the guiding catheter was carefully retracted to enable normal blood flow through the nonoccluded part of the left coronary system. Position was verified every 15 minutes to ensure appropriate occlusion of the LCx. After the procedure catheters were removed and the wound was closed. Open chest ligation of the LAD An antero-septal myocardial infarct was induced during an open chest procedure in order to reduce peri-procedural mortality due to VF. The thorax was opened via sternotomy. The infarct was induced by a transient ligation of the mid LAD after the first diagonal for 150 min. All pigs subjected to LAD ischemia-reperfusion underwent a 3D epicardial echocardiography before and after infarct induction. Follow-up During the 12 week FU, a 2D-echocardiography was performed at 4 and 8 weeks FU under induction medication as described above. Twelve weeks after infarct induction the animals were anesthetized, and 2D and 3D epicardial echocardiography were performed followed by invasive PV-loop measurements. The animals were sacrificed and the hearts were excised for infarct size determination and histological analyses. Echocardiography Echocardiography was performed using a Phillips iE33 echocardiography machine (Phillips, Eindhoven, The Netherlands). 2D images were obtained of the parasternal long axis and parasternal short axis at basal, mid-ventricular and apical level. The 2D echo was repeated after infarct induction in the animals with LCx infarction. This was not possible in the LAD group because of fluid and air in the chest after open chest procedure. Additionally, a 3D

PART ONE CHAPTER 2

epicardial echocardiogram was performed in the LAD ischemia reperfusion before and after myocardial infarction as previously described 22. In both groups, animals underwent additional 2D-echocardiography at 4 and 8 weeks following infarct induction, using a mild sedation of Midazolam, ketamine and atropine as described earlier. At sacrifice, both animals in the LCx and LAD group underwent transthoracic 2D and an epicardial 3D echocardiography.

Figure 1. Study design A: Flowchart. B ; 1. left anterior descending artery (LAD); 2: Ligation of the LAD was located after the first diagonal branch; 3: Location and example of balloon occlusion of the left circumflex artery (LCx). BL: Baseline before infarct induction; PMI: post myocardial infarct, w: week

MODERATE AND SEVERE MODEL OF CHRONIC MI

Images were analyzed using Velocity Vector Imaging (VVI, Siemens Medical solutions, USA). The end-diastolic and end-systolic volumes were calculated by the modified Simpson rule (LV end diastolic volume (LVEDV)= (AbED) * L/3 + (AmED+ApED)/2) * L/3 + 1/3(ApED) * L/3; LV end systolic volume (LVESV)= (AbES) * L/3 + (AmES+ApES)/2) * L/3 + 1/3(ApES) * L/3, in which Ab is the area at basal level, whereas Am and Ap are the areas at mid and apical level respectively and L is the length of the ventricle)23. LVEF was calculated by: ((LVEDV-LVESV)/ LVEDV)*100. 3D-echocardiographs were analyzed offline using Qlab 10.1 software (Phillips, Eindhoven, The Netherlands) as described before 24. One full volume analysis was used and the end diastolic frame was selected. Markers were placed at the base of anterior, posterior, inferior, lateral side and at the apex. The left ventricle was automatically traced by the software. All frames were checked for correct tracing and manually corrected if needed. The same was repeated for the end systolic phase. Strain analysis Strain is defined as the total deformation of the myocardium during 1 cardiac cycle 25,26. Radial and circumferential strain were analyzed on the 2D-echocardopgraphy short axis views via speckle tracking (VVI, Siemens Medical solutions, USA) as previously described 25 . Strain was analyzed according to the 17-segment echocardiography model. Figure 2 provides a schematic overview of the radial strain and circumferential strain and the 17-segment echocardiography model. In the LAD model anterior and antero-septal segments were analyzed, whereas the inferior and infero-lateral segments were used as reference segments in this group. In the LCx group, inferior and inferio-lateral segments were mostly affected and anterior segments were used as reference segments. Moreover, global strain was calculated by the software. Strain is represented as percentage of left ventricular deformation. Pressure volume loop analysis A 7F conductance catheter (CD Leycom, Zoetermeer, The Netherlands) was placed under fluoroscopic guidance in the apex of the left ventricle as previously described 27. Pressure Volume loops (PV loops) were assessed during apnea to avoid pressure/volume changes due to mechanical ventilation. Calibration was performed as previously described 28. PVloops were performed to obtain data of systolic indices (End-systolic pressure, stroke volume, end systolic pressure volume relations, stroke work, dP/dt max, prerecruitable stroke work), and diastolic indices (end diastolic pressure, tau, dP/dt min and end diastolic pressure volume relations). Analyses were performed using Conduct NT analyses software version 16.1 (CD Leycom, Zoetermeer, The Netherlands). Troponin I levels Troponin I levels were quantified using a standardized ELISA protocol of the Clinical chemistry laboratory of the University of Utrecht.

PART ONE CHAPTER 2

Infarct size calculations and histological analysis After the heart was excised, the atria were removed and the ventricles were cut into 5 slices of approximately 1 cm thickness. The heart was stained using 5% Tetrazolium chloride solution for 10-15 minutes at 37º. The slices were photographed and infarct size was calculated using automatic computer assisted image analysis software (Clemex, Quebec, Canada) as described before29. Infarct size was calculated as percentage of the total LV area. Biopsies were taken from the infarct area, the infarct border zone and the remote myocardial segments followed by embedding into paraffin for further light microscopical analysis. Collagen content in infarct, border and remote myocardial segments was assessed by trichrome stain. Briefly, all sections were deparaffinised, fixed in Buoin’s fixative (SigmaAldrich, St. Louis, USA) at 56º for 15 minutes, Nuclei were stained with haematoxylin for 3 minutes. The slides were submerged in Trichrome-AB solution for 5 minutes after which

Figure 2. Schematic overview of Strain and 17-segment echocardiography model a; schematic overview of strain analysis. Radial strain represents thickening (systole, arrows outwards) and thinning (diastole, arrows facing each other) of the myocardium during 1 cardiac cycle. Circumferential strain: elongation and shortening (arrows facing each other) of myocardial muscle fibers. b; schematic overview of myocardial segments on short axis view at basal level (mitral valve level). A indicates anterior wall; AL: anterolateral wall; IL: infero-lateral; I: inferior; IS: infero-septal; AS; antero-septal. C: Schematic overview of short axis view of myocardial segments at mid-ventricular level (papillary muscle). D: Schematic view of the apex. The apex only consists of 4 segments. L indicates lateral wall.

MODERATE AND SEVERE MODEL OF CHRONIC MI

they were treated with 0.5% acetic acid for 1 minute. Slides were mounted with Entallan (Merck, Darmstadt, Germany). Three random pictures were made at a 10X magnification and collagen content was calculated as percentage collagen of total surface area using automated analysis software (Clemex, Quebec, Canada). Arteriole density was quantified in infarct, border and remote myocardial segments, using alpha smooth muscle actin immunohistochemistry analysis (SMA, clone 1a4, Sigma-Aldrich, St Louis, USA). Endogenous peroxidase activity was blocked by 3% methanol/H2O2 solution for 30% and incubated with SMA 1:1000 overnight. Subsequently, the slides were incubated with secondary HRP-conjugated goat anti-mouse antibody dilution 1:200 (DAKO, Glostrup, Denmark) for 90%. All slides were immersed in DAB solution for 2 minutes (DAKO, Glostrup, Denmark) and mounted with Entallan. A technician that was blinded for group allocation took 3 random pictures at 10 times magnification. Arterioles per field of view were counted and expressed as number per view. Capillary density was only assessed in border en remote areas, because almost all capillaries are destroyed after AMI. All sections were deparaffinised and pre-treated with trypsin EDTA (Lonza, Verviers, Belgium). Endogenous peroxidase activity was blocked as described above. All slides were incubated with Isolectin B4 (Bandeiraea simplicifolia Isolectin B4, Dako, Glostrup, Denmark diluted 1/50) overnight. Subsequently, the slides were immersed in DAB solution and mounted with Entallan. Photographs were taken at 20X magnification and number of capillaries per mm2 was calculated. Statistical analysis All data were analyzed using SPSS Statistics 20 (IBM statistics, Chicago, USA). All data are presented as mean ± SEM. Comparisons between groups at a single time point (histology, infarct size, comparison of 12 week FU data) were analyzed using Student’s t-tests or Mann Whitney U tests. For sequential data, a two-way repeated measures ANOVA was applied. Correlations were analyzed by the Pearson’s correlation coefficient. P<0.05 was considered statistically significant.

RESULTS Mortality and safety All animals that underwent LCx occlusion survived initial infarct induction (100%) compared to 5/6 animals in the LAD group (83 %; p=NS). This animal died of ventricular fibrillation during infarct induction, resistant to defibrillation. Fifty percent of the animals in the LCx group needed defibrillation during infarct induction with an average of 7.8 times, as compared to 5/6 animals in the LAD group with an average of 12.2 times defibrillation. During the 12 week FU period, none of the animals died or were treated for heart failure. No ventricular arrhythmias were recorded on the RevealTM detector during the 12 weeks FU.

PART ONE CHAPTER 2

Cardiac function by echocardiography Both LVEF and LV volumes before infarct induction were comparable between the LCx and LAD group. Post-AMI LVEF decreased with 5.9% to 50.8±1.6 % (p=0.03) in the LCx group and with 17.1% to an average of 39.4±1.6% in the LAD group (p=0.009). Following AMI, LVESV increased with 5.6± 2.0 mL (p=NS) in the LCx group and with 17.8±6.7 mL (p=0.01) in the LAD group (figure 3). There was no significant change in LVEDV following infarct

Figure 3. Echocardiography a-c: LVEF and LV volumes on echocardiography over time. a: Left ventricular ejection fraction (LVEF) over time. b: Left ventricular end-diastolic volume (LVEDV) over time. c: Left ventricular end-systolic volume (LVESV) over time. d-f: 3D-echocardiography at 12 week FU. d: LVEF; e: LVEDV and; f: LVESV at 12 weeks FU as measured by 3D-echocardiography. LCx indicates left circumflex artery; LAD: left anterior descending artery; g: example of echocardiogram at end of systole at mid-ventricular level at 12 week FU. 1 indicates a formation of scar tissue at. Moreover, the end systolic diameter has been increased; h: example of an echocardiogram of an animal in the LCx group at 12 week FU at mid-ventricular level at the end of systole. 2 indicates the location of the scar. No clear myocardial thinning could be observed in the Lcx model. *P<0.05

MODERATE AND SEVERE MODEL OF CHRONIC MI

induction in either of the groups. Table 1 depicts all cardiac dimensions in both groups. At 12 weeks FU, 2D-echocardiography revealed a decreased LVEF of 7.9% to 49.3±0.8% (p=0.06) in the LCx group whereas LVEF of the LAD group declined with 14.7% to 41.7±1.5 % (p <0.001). LVESV increased non-significantly in the LCx group with 10.7mL to 46.0±2.4 mL (p=0.06), where the LAD group did show a significant increase of 30.6 mL to 60.7±4.1 mL (p<0.00,1). No significant dilatation as measured by LVEDV was seen in the LCx group (+10.4mL to 91.4±4.9 mL, p=0.09). LVEDV in pigs subjected to LAD occlusion, however, was significantly increased during follow-up (+35.8mL to 104.1±5.7 mL, p= 0.02). At 12 weeks FU, 3D-echocardiography confirmed the 2D-echocardiography measurements (figure 3). In the LCx group, LVEF was 50.0±2.4% as opposed to 39.2±1.9% in the LAD group. LVEDV in the LCx group at 12 weeks FU was 98.5±2.9 mL as compared to 126.1±3.5 mL in the LAD group.

Table 1. Echocardiography data Echocardigraphy

LCx

p-value to BL *

LAD

p-value to BL*

p-value between groups#

LVEF Baseline

56.7 ± 2.0

56.4 ± 1.5

LVEDV Baseline

80.8 ± 6.6

68.3 ± 3.7

LVESV Baseline

35.3 ± 4.3

30.1 ± 4.0

LVEF PMI

50.8 ± 1.6

0.03

39.4 ± 1.6

LVEDV PMI

83.7 ± 4.3

79.2 ± 5.8

LVESV PMI

40.9 ± 1.1

47.9 ± 3.7

0.01

0.09

LVEF 4 weeks

48.6 ± 1.0

0.02

40.3 ± 0.7

<0.001

0.04

LVEDV 4 weeks

88.6 ± 4.4

96.5 ± 4.3

0.01

0.06

LVESV 4 weeks

45.5 ± 2.0

0.07

57.7 ± 2.8

0.01

0.03

LVEF 8 weeks

51.4 ± 1.0

39.9 ± 1.0

<0.001

0.04

LVEDV 8 weeks

95.8 ± 5.9

97.4 ± 7.8

0.06

LVESV 8 weeks

46.5 ± 3.4

0.06

58.5 ± 5.0

0.01

0.07

LVEF 12 weeks

49.3 ± 0.8

0.06

41.7 ± 1.5

<0.001

0.004

LVEDV 12 weeks

91.4 ± 4.9

104.1 ± 5.7

0.02

0.03

LVESV 12 weeks

46.0 ± 2.4

0.06

60.7 ± 4.1

<0.001

0.04

0.009

0.02

Table 1 represents all data and p-values for echocardiography. The post myocardial infarct values of the LAD study are measured with 3D echocardiography. LCx indicates left circumflex artery; LAD: left anterior descending artery. LVEF: left ventricular ejection fraction; LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume. BL: baseline before infarct, PMI: post myocardial infarct. * repeated measures Anova (values compared to baseline), # student’s t-test.

PART ONE CHAPTER 2

MODERATE AND SEVERE MODEL OF CHRONIC MI

Radial and circumferential strain Decreased strain indicates deteriorated contractile function of the myocardium. In the LCx group, no changes in radial strain were observed in the infarcted inferior segment (figure 4). For the LAD-group, radial strain was significantly attenuated in the anterior segment of the apex during FU and kept progressively declining (42.8±4.8% at baseline and 5.6±7.7% at 12 week FU; p=0.008; figure 4). In this group, total apical radial strain at 12 week FU also decreased to 20.0±4.5% opposed to 45.1±4.1% at baseline (p=0.04). The total ventricular strain at 12 week FU was significantly lower in this group at 12 week FU (48.6±2.0% at baseline and 30.1±3.8 % at follow-up; p=0.03. These findings corroborate with the LVEF echocardiography data. Circumferential strain was significantly reduced in the LAD group in the apex and at mid-ventricular level of the heart throughout the 12 week FU compared to baseline (figure 4). PV-loop analysis, Infarct size, Troponin and Histology Invasive real-time PV-loop analysis was also performed at baseline and 12 weeks FU. dP/ dt – (relaxation of the ventricle) and ESPVR (derivative of contractility) were significantly decreased in the LAD group but not in the LCx group (figure 5). Indices of diastolic dysfunction, Tau and tPFR (time to peak filling rate) and systolic indices tPER (time to peak ejection rate) increased in both groups. Troponin I levels 6 hours after infarct induction in the LCx group were lower than in the LAD group (358.6 ± 79.9 ug/L vs. 560.0 ± 79.9 ug/L, P=0.02; figure 5). Post AMI decrease in cardiac function and Troponin levels 6 hours after infarct correlated significantly (R= -.763; P=0.028). The average infarct size in the LAD group was significantly higher than in the LCx group (23.4 ± 2.1% in the LAD group versus 9.5 ± 1.7% in the LCx group, figure 5). There were no differences in collagen density, capillary density and arteriole density in all myocardial segments (table 2). A trend towards an increase in arteriole density was observed in the LAD group (p=0.09)

Figure 4. Radial strain and circumferential strain A-f: Radial strain anterior and inferior segments. a-b: Radial strain apex, inferior and anterior segments per group. Strain in the anterior section of the LAD study is significantly lower during follow-up than at baseline. C-d: Radial strain mid ventricular level, inferior and anterior segments per group. E-f: Radial strain basal level, inferior and anterior segments per group. G-j: Circumferential strain. g: Apical circumferential strain over time; h: Circular strain at mid ventricular level over time. i; Circumferential strain at basal level over time. j; Total ventricular circumferential strain over time. LCx indicates left circumflex artery; LAD: left anterior descending artery; BL: baseline. *p<0.05 within the LAD group.

PART ONE CHAPTER 2

Table 2. Histological analysis Parameter

measure

LCx

LAD

P-value

Collagen density infarct area

89.2 ± 4.0

89.2 ± 4.2

Collagen density border area

5.1 ± 1.5

5.5 ± 1.3

Collagen density remote area

4.4 ± 1.1

3.0 ± 0.7

Arteriole density infarct area

arterioles/view

11.2 ± 1.5

10.1 ± 1.9

Arteriole density border area

arterioles/view

5.1 ± 1.1

9.2 ± 1.4

Arteriole density remote area

arterioles/view

3.5 ± 0.5

5.4 ± 0.9

Capillary density border area

capillaries/mm²

713.9 ± 74.3

714.4 ± 85.5

Capillary density remote area

capillaries/mm²

470.5 ± 48.0

376.4 ± 37.9

Overview of histological parameters. NS indicates not significantly different between de groups.

DISCUSSION The potential effects of novel therapeutics that confine the damage after MI depend on the severity of the ischemic event.3 To determine which patient group (e.g. moderately or severely affected patients) benefits most from such therapeutics, their efficacy has to be tested in large animal models that resemble the clinical course and severity as closely as possible4,6,21. Therefore, the current study investigated cardiac function over time in two clinically relevant porcine models of ischemia-reperfusion: a severe LAD and a moderate LCx ischemia-reperfusion model. Here, we demonstrated that occlusion of 150 minutes of the LAD resulted in overt deterioration of cardiac function and progressive cardiac dilatation. Therefore, this model is most suitable for testing therapeutics that focus on the prevention of adverse remodeling post-MI in severely affected patients. On the other hand, occlusion of the LCx resulted in limited effects on cardiac contractility and dilatation, resembling the clinical course of adequately revascularized MI patients with only limited cardiac damage. Thus, this model can be used to test compounds that reduce infarct size directly post-MI but is less suitable for prevention of adverse remodeling. As hypothesized, the two models showed a marked difference in response to ischemiareperfusion. First, in the LCx model none of the animals died which was, most likely, based on fewer episodes of VF. Presumably this is due to the involvement of the septum in the infarct area in the LAD model combined with a smaller area at risk after occlusion of the LCx. This difference in the area at risk also resulted in a large difference in infarct size, which was approximately 23% of the LV in the LAD model compared to 10% in the LCx model. In turn, the larger infarct size in the LAD model resulted in a profound decrease in LVEF culminating in adverse remodeling and cardiac dilatation. The decrease in LVEF occurred to a lesser extent in the LCx group and this did not result in any cardiac dilatation post-MI. In concordance with the functional echocardiography data, myocardial strain was

MODERATE AND SEVERE MODEL OF CHRONIC MI

Figure 5. Pressure-Volume loop analysis, troponin I and infarct size a-e: Pressure-volume loop analysis. a; End-systolic pressure (ESP) is significantly lower at follow-up in the LAD group; b: Time to peak ejection rate (tPER); C: Time to peak filling rate (tPFR); d: Tau: Relaxation constant; e: Increase in pressure over time (dP/dt +); f: Troponin I levels 6 hours after myocardial infarction are higher in the LAD group. g-I; Infarct size. g: example of infarct by left circumflex artery infarct (LCx) ischemiareperfusion model. Viable myocardium is red, infarct is depicted in white(arrow); h: Infarct size in left anterior descending artery (LAD) group is significantly higher; i: example of an LAD infarct. LCx indicates left circumflex artery; LAD: left descending anterior artery; BL: baseline; FU: follow-up. *P<0.05

decreased in the affected myocardial segments in the LAD group, whereas strain was preserved in the affected segments in the LCx group. To the best of our knowledge, this is the first study that has studied myocardial strain when comparing different MI models and these findings show that the two models are not only different regarding the global cardiac function but also differ on a regional contractile level. Despite the significant differences in systolic function in the LAD group, no differences between the groups were observed regarding diastolic dysfunction. In both studies, diastolic dysfunction worsened relative to baseline measures, which suggests that these measurements are very robust and are not directly influenced by infarct size or dilatation in the current study.

PART ONE CHAPTER 2

In a previous study of Suzuki in which the mid-LAD was occluded for 60 minutes followed by reperfusion, a comparable mortality (16%) and decrease in cardiac function was observed following infarct induction18. However, in that study LVEF decreased transiently and recovered significantly at 14 days and 28 days follow-up to 47%. This is most likely due to the occlusion time. A 60-minute occlusion period presumably leads to reversible cell damage, with a large fraction of myocardial stunning/hibernation that is known to resolve spontaneously within days to weeks30. In our study, LVEF did not recover during FU after the initial decrease post-MI in either model. In the LAD model this simultaneously occurred with increased LVESV and LVEDV, suggesting progression into heart failure. This provides a larger therapeutic window for experimental therapies that target adverse remodeling and prevent cardiac dilatation in this model compared to spontaneously recovering models. The same holds true for the LCx model. Again, therapeutic efficacy is easier to test in a model in which outcome is not affected by confounding factors such as spontaneous recovery that may exceed and hence mask the therapeutic effect. Importantly, our findings in both the LAD model and the LCx model on cardiac function and infarct size correspond with chronic myocardial ischemia models15,16. This indicates that 150 minute occlusion followed by reperfusion results in the same amount of myocardial damage but more closely resembles the clinical course and treatment of MI. Moreover, in our models, intracoronary therapy to limit myocardial damage is still possible. Despite our best efforts, the study has some limitations. First, we assessed cardiac function by 2D-echocardiography during follow-up and obtained 3D echocardiogram directly postMI (LAD-model only) and at sacrifice (both models). Due to the anatomical position of porcine ribs, it is not possible to obtain a transthoracic 4 chamber view that would be needed for 3D- echocardiography. Although cardiac MRI is considered the gold standard for cardiac function and volumes, due to logistic reasons, it was not feasible to obtain sequential cardiac MRI data in our study. We are confident that our echocardiography data are reproducible and depict the true cardiac function following both ischemia-reperfusion models. Second, in one group we performed a closed chest balloon occlusion model and in the other group an open chest model was applied. The latter was chosen in order to be able to perform epicardial defibrillation in the LAD model to improve survival in the LAD model. Since the study was not designed to directly compare these two models, a similar infarct induction was, however, not essential. To conclude, the current study showed that it is feasible to create two very distinctive models of chronic myocardial infarction to test novel therapeutics post-MI for the possible treatment of different patient groups. Reperfusion after 150 min ischemia of the LAD leads to severe cardiac dysfunction and the development of heart failure over a limited period of time, whereas ischemia-reperfusion of the LCx culminates in stable, moderate cardiac dysfunction. Since these models closely resemble the clinical course and treatment of MI and enable intracoronary therapy administration, they are preferable to models with persistent coronary occlusion. This study adds to refinement of preclinical studies which hopefully will result in improved translational medicine and a reduction of animals needed in preclinical research.

MODERATE AND SEVERE MODEL OF CHRONIC MI

Sources and funding This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs, Agriculture and Innovation. Acknowledgments The authors would like to thank M. Jansen, J. Visser, M. Schurink, E. Velema and C. Verlaan for their excellent technical assistance during the experiments.

PART ONE CHAPTER 2

REFERENCES 1. Mudd JO, Kass D a. Tackling heart failure in the twenty-first century. Nature. 2008;451:919–928. 2. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, Carnethon MR, Dai S, de Simone G, Ford ES, Fox CS, Fullerton HJ, Gillespie C, Greenlund KJ, Hailpern SM, Heit JA, Ho PM, Howard VJ, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A, Matchar DB, McDermott MM, Meigs JB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G, Paynter NP, Rosamond WD, Sorlie PD, Stafford RS, Turan TN, Turner MB, Wong ND, Wylie-Rosett J. Heart disease and stroke statistics--2011 update: a report from the American Heart Association. Circulation. 2011;123:e18–e209. 3. Sluijter JPG, Condorelli G, Davidson SM, Engel FB, Ferdinandy P, Hausenloy DJ, Lecour S, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Van Laake LW. Novel therapeutic strategies for cardioprotection. Pharmacol Ther. 2014;144:60–70. 4. Van Der Spoel TIG, Jansen Of Lorkeers SJ, Agostoni P, Van Belle E, Gyngysi M, Sluijter JPG, Cramer MJ, Doevendans P a., Chamuleau S a J. Human relevance of pre-clinical studies in stem cell therapy: Systematic review and meta-analysis of large animal models of ischaemic heart disease. Cardiovasc Res. 2011;91:649–658. 5. De Jong R, Houtgraaf JH, Samiei S, Boersma E, Duckers HJ. Intracoronary Stem Cell Infusion After Acute Myocardial Infarction: A Meta-Analysis and Update on Clinical Trials. Circ Cardiovasc Interv. 2014;7:156–167. 6. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker H V, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, López CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West M a, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier R V, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A. 2013;110:3507– 12. 7. Hughes HC. Swine in cardiovascular research. Lab Anim Sci. 1986;36:348–350. 8. Suzuki Y, Yeung AC, Ikeno F. The representative porcine model for human cardiovascular disease. J Biomed Biotechnol. 2011;2011:195483. 9. Schaper W, Jageneau a, Xhonneux R. The development of collateral circulation in the pig and dog heart. Cardiologia. 1967;51:321–335. 10. Matsunaga T, Warltier DC, Weihrauch DW, Moniz M, Tessmer J, Chilian WM. Ischemia-induced coronary collateral growth is dependent on vascular endothelial growth factor and nitric oxide. Circulation. 2000;102:3098–3103.

11. Peterson MC, Syndergaard T, Bowler J, Doxey R. A systematic review of factors predicting door to balloon time in ST-segment elevation myocardial infarction treated with percutaneous intervention. Int J Cardiol. 2012;157:8–23. 12. Zhou C, Yao Y, Zheng Z, Gong J, Wang W, Hu S, Li L. Stenting technique, gender, and age are associated with cardioprotection by ischaemic postconditioning in primary coronary intervention: A systematic review of 10 randomized trials. Eur Heart J. 2012;33:3070–3077. 13. Lichtman JH, Froelicher ES, Blumenthal J a., Carney RM, Doering L V., Frasure-Smith N, Freedland KE, Jaffe AS, Leifheit-Limson EC, Sheps DS, Vaccarino V, Wulsin L. Depression as a risk factor for poor prognosis among patients with acute coronary syndrome: Systematic review and recommendations: A scientific statement from the american heart association. Circulation. 2014;129:1350–1369. 14. Crisóstomo V, Maestre J, Maynar M, Sun F, BáezDíaz C, Usón J, Sánchez-Margallo FM. Development of a closed chest model of chronic myocardial infarction in Swine: magnetic resonance imaging and pathological evaluation. ISRN Cardiol. 2013;2013:781762. 15. Rissanen TT, Nurro J, Halonen PJ, Tarkia M, Saraste A, Rannankari M, Honkonen K, Pietilä M, Leppänen OP, Kuivanen A, Knuuti J, Yla-Herttuala S. The bottleneck stent model for chronic myocardial ischemia and heart failure in pigs. Am J Physiol Heart Circ Physiol. 2013; 16. Biondi-Zoccai G, De Falco E, Peruzzi M, Cavarretta E, Mancone M, Leoni O, Caristo ME, Lotrionte M, Marullo AGM, Amodeo A, Pacini L, Calogero A, Petrozza V, Chimenti I, D’Ascenzo F, Frati G. A novel closed-chest porcine model of chronic ischemic heart failure suitable for experimental research in cardiovascular disease. Biomed Res Int. 2013;2013:410631. 17. Timmers L, Henriques JPSPS, de Kleijn DPVP V, Devries JHH, Kemperman H, Steendijk P, Verlaan CWJWJ, Kerver M, Piek JJJ, Doevendans P a a., Pasterkamp G, Hoefer IEE. Exenatide Reduces Infarct Size and Improves Cardiac Function in a Porcine Model of Ischemia and Reperfusion Injury. J Am Coll Cardiol. 2009;53:501–510. 18. Suzuki Y, Lyons JK, Yeung AC, Ikeno F. In vivo porcine model of reperfused myocardial infarction: In situ double staining to measure precise infarct area/area at risk. Catheter Cardiovasc Interv. 2008;71:100–107. 19. Hausenloy DJ, Yellon DM. Review series Myocardial ischemia-reperfusion injury : a neglected therapeutic target. 2013;123. 20. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 21. Halapas a, Papalois A, Stauropoulou A, Philippou A, Pissimissis N, Chatzigeorgiou A, Kamper E, Koutsilieris M. In vivo models for heart failure research. In Vivo (Brooklyn). 2008;22:767–780.

MODERATE AND SEVERE MODEL OF CHRONIC MI

22. Van Hout GPJ, Jansen SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE. Admittance-based pressure-volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014;2:e00287. 23. Folland ED, Parisi a F, Moynihan PF, Jones DR, Feldman CL, Tow DE. Assessment of left ventricular ejection fraction and volumes by real-time, twodimensional echocardiography. A comparison of cineangiographic and radionuclide techniques. Circulation. 1979;60:760–766. 24. Soliman OIIII, Krenning BJJ, Geleijnse MLL, Nemes A, van Geuns R-JJ, Baks T, Anwar AMM, Galema TWW, Vletter WBB, ten Cate FJ, Cate FJT. A comparison between QLAB and tomtec full volume reconstruction for real time three-dimensional echocardiographic quantification of left ventricular volumes. Echocardiography. 2007;24:967–974. 25. Nesbitt GC, Mankad S, Oh JK. Strain imaging in echocardiography: methods and clinical applications. Int J Cardiovasc Imaging. 2009;25 Suppl 1:9–22. 26. Tee M, Noble JA, Bluemke DA. Imaging techniques for cardiac strain and deformation: comparison of echocardiography, cardiac magnetic resonance and cardiac computed tomography. Expert Rev Cardiovasc Ther. 2013;11:221–31. 27. Van Hout GPJ, de Jong R, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittance-based pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75. 28. Steendijk P, Staal E, Jukema JW, Baan J. Hypertonic saline method accurately determines parallel conductance for dual-field conductance catheter. Am J Physiol Heart Circ Physiol. 2001;281:H755– 63. 29. Houtgraaf JH, De Jong R, Kazemi K, De Groot D, Van Der Spoel TIG, Arslan F, Hoefer I, Pasterkamp G, Itescu S, Zijlstra F, Geleijnse ML, Serruys PW, Duckers HJ. Intracoronary infusion of allogeneic mesenchymal precursor cells directly after experimental acute myocardial infarction reduces infarct size, abrogates adverse remodeling, and improves cardiac function. Circ Res. 2013;113:153– 166. 30. Canty JM, Suzuki G. Myocardial perfusion and contraction in acute ischemia and chronic ischemic heart disease. J. Mol. Cell. Cardiol. 2012;52:822– 831.

chapter 3

Invasive surgery reduces infarct size and preserves cardiac function in a porcine model of myocardial infarction

Journal of Cellular and Molecular Medicine - Accepted June 2015

G.P.J. van Hout1*, M.P.J. Teuben2, M. Heeres2, S. de Maat3, R. de Jong4, C. Maas3, L. H.J.A. Kouwenberg1, L. Koenderman2, W.W. van Solinge3, S.C.A. de Jager1, G. Pasterkamp1,3, I.E. Hoefer1,3 1

Experimental Cardiology Laboratory, University Medical Center Utrecht, Utrecht, The Netherlands

Department of Respiratory Medicine, University Medical Center Utrecht, Utrecht, The Netherlands

Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, Utrecht, The Netherlands

Department of Anesthesiology, Erasmus Medical Center, Rotterdam, The Netherlands

PART ONE CHAPTER 3

ABSTRACT Aims Reperfusion injury following myocardial infarction (MI) increases infarct size and deteriorates cardiac function. Cardioprotective strategies in large animal MI models often failed in clinical trials, suggesting translational failure. Experimentally, MI is induced artificially and the effect of the experimental procedures may influence outcome and thus clinical applicability. The aim of the current study is to investigate if invasive surgery, as in the common open chest MI model affects infarct size and cardiac function. Methods Twenty female landrace pigs were subjected to MI by transluminal balloon occlusion. In 10 out of 20 pigs, balloon occlusion was preceded by invasive surgery (medial sternotomy). After 72 hours, pigs were subjected to echocardiography and Evans blue/triphenyl tetrazoliumchloride double staining to determine infarct size and area at risk. Results Quantification of infarct size showed a significant infarct size reduction in the open chest group compared to the closed chest group (infarct size vs. area at risk: 50.9±5.4% vs. 69.9±3.4%, p=0.007). End systolic left ventricular volume and left ventricular ejection fraction measured by echocardiography at follow-up differed significantly between both groups (51±5mL vs. 65±3mL, p=0.033), (47.5±2.6% vs. 38.8±1.2%, p=0.005). The inflammatory response in the damaged myocardium did not differ between groups. Conclusion The current study indicates that invasive surgery reduces infarct size and preserves cardiac function in a porcine MI model. Future studies need to elucidate the effect of infarct induction technique on the efficacy of pharmacological therapies in large animal cardioprotection studies.

INVASIVE SURGERY REDUCES INFARCT SIZE

INTRODUCTION Myocardial infarction (MI) and heart failure (HF) remain the most important cardiovascular causes of death worldwide1,2. Due to improved medical care and revascularization therapy, survival after MI has increased considerably during the past decades3,4. This improved survival increases the risk to develop HF as patients survive with a severely deteriorated cardiac function1. In order to prevent progression into HF and preserve cardiac function post-MI, investigators aim to modulate the predominant mechanisms involved in MI and post-MI healing (reperfusion injury, adverse cardiac remodeling)5,6. Before clinical testing of therapeutics is considered, thorough testing in clinically relevant animal models should be performed7. In this perspective, large animal MI models are required to confirm findings from studies performed in small animals, since the development of tissue damage post-MI differs in small animals compared to larger mammals8,9. Additionally, large animal models show great similarity with humans regarding hemodynamics, cardiac anatomy, (coronary) pharmacokinetics and â&#x20AC;&#x201C;dynamics10,11. Large animal models also allow the use of clinical treatment regimens and administration routes along with identical function-related read-outs12,13. Despite these advantages, many compounds that were successful in large animal models eventually failed in clinical trials, suggesting optimization of experimental protocols remains necessary14. Numerous methods of inducing myocardial ischemia in large animals have been proposed. Despite abundant variations of ischemia induction, there are two main approaches11. The first requires invasive surgery, including lateral or medial sternotomy, to reach the coronary arteries and induce ischemia from an external approach. The second method mimics percutaneous coronary intervention and follows a transluminal approach. The invasive method induces bone trauma and alters intra-thoracic pressure, possibly influencing cardiac physiology whereas trauma induced by the minimal-invasive method is much less extensive11â&#x20AC;&#x201C;13,15. Previous reports from small and large animal studies indicate that abdominal skin injury prior to MI limits infarct size after cardiac ischemia16,17. It has also been established that tissue damage, as can be observed during myocardial infarction and traumatic injury, attracts circulating leukocytes to the site of injury18â&#x20AC;&#x201C;20. Furthermore, surgery can cause a decreased ex vivo responsiveness of leucocytes21,22. Since additional injury is known to influence myocardial infarct size and temporary immune suppression occurs in patients undergoing surgery, we hypothesized that medial sternotomy required for external MI induction decreases myocardial infarct size and preserves cardiac function post-MI. Additionally, we aimed to determine if sternotomy influences the local inflammatory response in the myocardium. This may have significant implications for the interpretation of experimental studies on therapeutic compounds tested in either model.

PART ONE CHAPTER 3

MATERIAL AND METHODS All animal experiments were approved by the institutional animal welfare committee of the University Medical Center Utrecht and were executed conforming to the ‘Guide for the Care and Use of Laboratory Animals’. Twenty female landrace pigs were evaluated in this study (Van Beek, Lelystad, the Netherlands). Pigs (69.8±1.0 kg) were subjected to MI by transluminal balloon occlusion followed by invasive pressure-volume measurements, 3D-Echocardiography and Evans blue / triphenyl tetrazoliumchloride (TTC) double staining to determine infarct size and area at risk at 72 hours follow-up. In 10 out of 20 pigs, balloonocclusion was preceded by medial sternotomy. Surgical procedure Pre-treatment and anesthesia protocols have been described in detail elsewhere12,23. In short, all animals were pre-treated with amiodaron for 10 days (1200mg loading dose, 800mg/day maintenance), clopidogrel for 3 days (75mg/day) and acetylsalicylic acid for 1 day (320mg loading dose, 80mg/day maintenance). All medication was continued until the end of the study. To prevent unnecessary stress and discomfort, animals were anesthetized with an intramuscular injection of 10mg/kg ketamine, 0.4mg/kg midazolam and 0.014mg/kg atropine. Venous access was obtained by insertion of an 18G cannula in the ear vein for intravenous administration of 5mg/kg sodiumthiopental. Depth of anesthesia was then determined by checking eyelid reflex, response to skin stimulus and laryngeal reflex followed by intubation with an endotracheal tube and balloon-ventilation. Pigs were transported to the operating room where anesthesia was maintained by intravenous infusion of 0.5mg/kg/h midazolam, 2.5µg/kg/h sufentanyl and 0.1mg/kg/h pancuronium. Pigs were mechanically ventilated and arterial blood pressure and heart rate were checked when performing surgical actions to determine the depth of anesthesia. Pre-operatively, animals received a fentanyl patch (25µg/h). Arterial access was obtained by introduction of an 8F sheath into the carotid artery after surgical exposure. Pigs were then randomly allocated to either an open or closed chest procedure. Closed chest group After arterial and venous access were obtained, the closed chest animals were left untouched for 20 minutes. This corresponds to the time required to perform the sternotomy in the open chest group. A coronary angiogram of the left coronary tree was acquired using an 8F JL4 guiding catheter (Boston Scientific, Natick, USA). An adequately sized balloon was placed distal to the second diagonal branch and inflated for 75 minutes. After reperfusion, a catheter was temporarily placed in the coronary sinus via an introducer sheath in the jugular vein to draw blood 60 minutes after reperfusion. Animals were observed for 3 hours post-reperfusion and blood samples were drawn at various time points. A dwelling catheter was placed in the jugular vein in 3 pigs of each experimental group to allow additional venous blood sampling at 4 and 8 hours reperfusion. The surgical wound was closed and animals were weaned from anesthesia. Animals were defibrillated

INVASIVE SURGERY REDUCES INFARCT SIZE

in case of ventricular fibrillation (VF). Body temperature was kept constant between 37.0째C and 38.5째C throughout the experiment. Heart rate and arterial blood pressure were measured continuously and documented every 30 minutes. Open chest group The open chest group was treated identical to the closed chest group with the exception that in the 20min period after the insertion of the arterial and venous sheaths, a medial sternotomy was performed. After performing the myocardial ischemia/reperfusion protocol as described above, the sternum and the surgical wound were closed and animals were weaned from anesthesia. Pressure-volume measurements Pressure-volume measurements were performed as recently described23,24. In short, animals were again anesthetized 72 hours after infarct induction. Arterial access was obtained by introducing an 8F sheath into the carotid artery. A 7F tetra-polar admittance catheter (7.0 VSL Pigtail/no lumen, Transonic Scisense, London, Canada) was inserted into the left ventricle (LV) through the sheath in the carotid artery under fluoroscopic guidance. Echocardiography Three-dimensional echocardiography was performed as described before12,24,25. In short, an X3-1 transducer on an iE33 ultrasound device (Philips, Eindhoven, The Netherlands) was used 72 hours after reperfusion. Immediately after PV measurements, a medial (re) sternotomy was performed and a gel-filled flexible sleeve was placed directly on the apex of the heart. The depth and sector size were adjusted to fit the complete ventricle. The images were analyzed offline using QLab 10.1 (3DQ advanced) analysis software. Area at risk & infarct size Before exsanguination, the LAD was externally ligated at the exact site of balloon occlusion. The LV was punctured with a sterile needle attached to a 50mL syringe filled with 2% Evans blue dissolved in 50mL 0.9%NaCl. The aortic root was clamped distal to the origin of the coronary arteries. Evans blue was infused at a rate of 10mL/s. Animals were then sacrificed by exsanguination under anesthesia. The heart was excised and the LV was cut into 5 equal slices from apex to base. Slices were incubated in 1% TTC (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37째C 0.9%NaCl for 10 minutes to discriminate between infarct tissue and viable myocardium. After incubation, photographs of the slices were taken and the remote area, area at risk (AAR) and infarcted tissue were quantified using ImageJ software (NIH, Bethesda, MD, USA). Following quantification, infarcted tissue, tissue from the border zone and tissue from the remote area were collected and either conserved in 4% formaldehyde for histological quantification of neutrophils, or snap frozen in liquid nitrogen for cytokine measurements.

PART ONE CHAPTER 3

Neutrophil numbers, macrophage numbers and inflammatory parameters Circulating leucocyte numbers from blood drawn at multiple time points after reperfusion were measured by whole-blood analysis using an automated hematological cell-counter (Cell-Dyn Sapphire, Abbott, Santa Clara, CA, USA). The Cell-Dyn Sapphire is a routine hematology analyzer, which uses spectophotometry, electrical impedance and laser light scattering to classify blood cells (platelets, erythrocytes and leukocytes). Neutrophil numbers, macrophage numbers and active caspase-3 positive cells in myocardial tissue were measured in paraffin-embedded histological biopsies that were conserved in 4% formaldehyde for at least 7 days. Histological samples were cut into 5μm sections using a microtome and sections were deparaffinized. To assess neutrophil numbers, sections were incubated with a porcine-specific monoclonal mouse anti-pig antibody against porcine neutrophils (Clone PM1, BMA Biomedicals, Augst, Switzerland) for 60 minutes followed by incubation with Brightvision Poly-AP-anti-mouse (ImmunoLogic, Duiven, the Netherlands) for 30 minutes. To determine macrophage numbers, sections were incubated with a monoclonal mouse anti-pig CD107a antibody (Serotec, Raleigh, USA) followed by incubation with the same secondary antibody. Caspase-3 positive cells were assessed using a purified rabbit anti-active caspase-3 antibody (BD Pharmigen, San Diego, USA) followed by incubation with Brightvision Poly-AP-anti-rabbit (Immunologic, Duiven, the Netherlands) for 30 minutes. For microscopic visualization, incubation with liquid permanent red (DAKO, Heverlee, Belgium) for 10 minutes was performed. For every animal, 10 random pictures were made at 200x magnification and quantified using CellSens software (Center Valley, PA, USA). Malondialdehyde (MDA) is a marker of lipid peroxidation and oxidative stress and was measured to determine if sternotomy induced less oxidative stress in the myocardium during reperfusion. MDA was measured from plasma obtained by centrifugation at 1850G for 10 minutes at 4°C. To quantify MDA we used a lipid peroxidation – MDA – kit (Abcam, Cambridge, MA, USA) according to the manufacturer’s instructions.

Figure 1

A Closed Chest Open Chest

150

MAP (mmHg)

Heart rate (bpm)

150

100

150

Time from onset ischemia (min)

200

100

150

Time from onset ischemia (min)

200

Figure 1. Hemodynamic parameters were recorded every 30 minutes during the experiments A. Heart rate did not differ significantly from the onset of ischemia to 180 minutes post reperfusion. B. A significant decrease in MAP was observed at multiple time points during ischemia and reperfusion, peaking at 60 minutes of ischemia. Bpm = beats per minute, MAP = mean arterial pressure, *p<0.05.

INVASIVE SURGERY REDUCES INFARCT SIZE

To assess the inflammatory response in the myocardium, the expression of 9 different cytokines (interleukin (IL)-1β, IL-4, IL-6, IL-8, IL-10, IL-12p40, TNF-α, IFN-α and IFN-γ) was measured using a luminex immunoassay (Procarta™ Multiplex, eBioscience, San Diego, CA, USA) according to the manufacturer’s instructions. Western blotting High molecular weight kininogen (HMWK) was measured as a proxy of bradykinin release. Plasma samples drawn at baseline and 15 minutes reperfusion were diluted 20 times in reducing sample buffer (15.5% Glycerol, 96.8 mM Tris-HCL, 3.1% SDS, 0.003% bromophenol blue, 25 mM DTT) and incubated for 10 minutes at 95 ˚C. Samples were separated on a 4-12% Bis-Tris gel at 165 Volt for 65 minutes in MOPS buffer. Proteins were transferred onto Immobilon-FL membranes at 125 Volt for 60 minutes in blotting buffer (14.4 g/L glycine, 3.03 g/L Tris-HCL, 20% Ethanol) and blocked with blocking buffer (0.5x Odyssey blocking reagent in TBS) for 1 hour at RT. HMWK was detected by overnight incubation at 4˚C with polyclonal affinity purified sheep anti-human HMWK antibody (Affinity Biologicals Inc. Ancaster, ON, Canada). Blots were washed with 0.005% TBS-Tween (TBST), and primary antibodies were detected with Alexa Fluor® 680 Donkey Anti-Sheep IgG (1:7,500 in blocking buffer, Life Technologies Carlsbad, CA, USA). Blots were extensively washed with TBST, followed by washing with water and analyzed on a near infra-red Odyssey scanner (Licor). Statistical Analysis All data are expressed as mean ± standard error unless stated otherwise. Differences in mortality were tested using a Fisher’s exact test. Myocardial infarct size, left ventricular ejection fraction (LVEF) and all other outcomes were compared using a Student’s t-test for unrelated measurements. All statistical analyses were performed in SPSS statistics version 20.0. A two-sided P-value of <0.05 was regarded statistically significant in all analyses.

RESULTS Mortality and hemodynamics Twenty pigs were subjected to 75 minutes myocardial ischemia-reperfusion injury. Four pigs died before the follow-up period (1 in the closed chest and 3 in the open chest group, all due to persistent ventricular fibrillation during cardiac ischemia). Fisher’s exact testing showed that this difference was statistically not significant (p=0.292). This allowed a comparison of 9 animals in the closed chest and 7 animals in the open chest group. During the 3-hour observation period post-MI, heart rate did not significantly differ between both groups (figure 1a). However, from the onset of ischemia onwards, mean arterial pressure (MAP) in the open chest group was significantly lower than in the closed chest group (figure 1b). This difference in MAP reached its peak at 60 minutes post-

PART ONE CHAPTER 3

occlusion (71mmHg vs. 107mmHg, p=0.006) and gradually returned to equal levels by the end of the observational period of 180 minutes (85 mmHg vs. 84 mmHg, p=ns). Myocardial infarct size and Cardiac Function Cardiac function and myocardial infarct size (IS) were determined after a follow-up period of 72 hours. Figure 2a shows representative pictures of the remote area (blue), AAR (red) and infarcted myocardium (white). Quantification of the AAR showed a similar AAR as a percentage of the LV in the open chest and closed chest group (19.1±3.0% vs. 19.4±1.3%, p=ns)(figure 2b). Quantification of IS showed a significant reduction in the open chest group compared to the closed chest group when measured as percentage of the AAR (50.9±5.4% vs. 69.9±3.4%, p=0.007) (figure 2c) and using the total LV as reference (9.2±1.3% vs. 13.6±1.2%, p=0.024) (figure 2d). Cardiac function also differed between the two experimental groups. While end diastolic volume did not differ between the open chest and the closed chest group (99±7mL vs.

Figure 2. After 72 hours of reperfusion, the AAR and IS were determined A Representative pictures of an apical slice of a pig that was subjected to MI without (CC) and with sternotomy (OC) B The AAR as a percentage of the total LV was equal between the closed chest and open chest group C The IS as a percentage of the AAR differed significantly between the closed chest and open chest group D The IS as a percentage of the total LV differed significantly between the closed chest and open chest group. CC = closed chest, OC = open chest, AAR = area at risk, IS = infarct size, LV = left ventricle, *p<0.05

INVASIVE SURGERY REDUCES INFARCT SIZE

Figure 3. Left ventricular function was measured 72 hours after reperfusion A End diastolic volume was not significantly lower in pigs subjected to medial sternotomy B End systolic volume was significantly lower in pigs subjected to medial sternotomy C Ejection fraction was significantly higher in the open chest compared to the closed chest group D dPdt-max was lower in de open chest group E dPdt-min was less negative in the open chest compared to the closed chest group. EDV = end diastolic volume, EDV = end systolic volume, EF = left ventricular ejection fraction, mL = milliliter, dPdt-max = rate in pressure change over time during systole, dPdt-min = rate in pressure change over time during diastole, CC = closed chest, OC = open chest, *p<0.05

106±4mL, p=ns), end systolic volume (51±5mL vs. 65±3mL, p=0.033) and LVEF (47.5±2.6% vs. 38.8±1.2%, p=0.005) differed significantly between both groups (figure 3a-c). Interestingly, the load-dependent rate of pressure increase (dPdt-Max) during systole was lower in the open chest group than in the closed chest group despite a smaller infarct size and better global cardiac function (886±80mmHg/s vs. 1373±87mmHg/s, p=0.002). The load-dependent rate of pressure decrease (dPdt-Min) during diastole was less negative in the open chest group compared to the closed chest group (-964±78mmHg/s vs. -1273±77mmHg/s, p=0.018) (Figure 3d&3e). Troponin was measured 2 hours after reperfusion and was significantly lower in the open chest compared to the closed chest group (251±44ng/mL vs. 1256±281ng/mL, p=0.008) (figure 4a). Inflammatory markers, oxidative stress and bradykinin release To determine if the two models differed in the systemic inflammatory response post-MI, we measured systemic leucocyte numbers up to 2 hours reperfusion. Since no differences in leucocyte numbers were detected during ischemia and the first 2 hours of reperfusion (figure 4b & 4c), we also measured systemic leucocyte numbers 4 and 8 hours postreperfusion in a subset of pigs (n=3 in each group). In the open chest group, circulating leucocyte numbers were significantly higher at 8 hours reperfusion in the open chest compared to the closed chest group (31.6±0.2*106 vs. 23.1*2.0±*106 leucocytes/mL,

PART ONE CHAPTER 3

p=0.014) (figure 4b). Neutrophil numbers were also significantly increased in this group (25.6±0.1*106 neutrophils/mL vs. 18.1±1.2*106 neutrophils/mL, p=0.040) (figure 4c) 8 hours post-reperfusion. Other leucocyte subtypes did not differ significantly between both groups at any time point (data not shown). Since inflammatory cells are an important source of free radical scavengers at the site of injury, MDA, a marker of oxidative stress, was measured in blood directly drawn from the coronary sinus 1 hour after reperfusion in both groups. However, there was no significant difference in MDA levels between both groups (MDA concentration: 483±55 nmol/mL vs. 465±45 nmol/mL, p=ns) (figure 4d). To investigate potential differences in the severity of the local immune response, we measured a variety of cytokines known to play a role in the inflammatory phase post-MI. We did not observe any significant differences in any of the measured cytokines in the myocardium. As examples, IL-6 (infarct area, p=0.15; border area p=0.40) and IL-1β (infarct area, p= 0.9; border area p=0.25) concentrations are shown in figure 5a & 5b.

Figure 4. Different measurements were performed in the systemic circulation A Troponin was significantly higher in the closed chest group compared to open chest group B Systemic leucocyte numbers were measured at baseline and different time points during ischemia and reperfusion C Systemic neutrophil numbers were measured at baseline and different time points during ischemia and reperfusion D Lipid peroxidation measured by malondialdehyde concentration in coronary sinus blood was not different between the open chest and closed chest group. TnI = troponin I, ng = nanogram, L = liter, mL = milliliter, nmol = nanomol, CC = closed chest, OC = open chest, *p<0.05

INVASIVE SURGERY REDUCES INFARCT SIZE

Similarly, we did not observe any significant difference between open and closed chest procedure animals regarding neutrophil and macrophage numbers in the myocardium 72h after reperfusion (figure 5c â&#x20AC;&#x201C; 5f). Finally, we determined cardiomyocyte apoptosis 72 hours after reperfusion. There were no significant differences between the open chest and closed chest group in caspase-3 positive cells (figure 5g & 5h). As bradykinin has previously been reported to mediate the cardioprotective effects of abdominal skin incision on reperfusion injury, we measured the bradykinin precursor HMWK. HMWK fragmentation serves as a proxy for bradykinin, with low HMWK levels reflecting fragmentation and high bradykinin release. Compared to baseline, absolute HMWK concentrations were higher in the closed chest compared to the open chest group (p=0.043) and a trend was observed for relative expression levels between groups (p=0.095), indicating that bradykinin release was higher in the open chest group (figure 5i & 5j).

Figure 5. Local markers of the inflammatory response were measured in the myocardium A Myocardial IL-6 content corrected for protein concentration in both the border and infarct zone differed non-significantly between the open chest and closed chest group B Myocardial IL-1Î˛ content corrected for protein concentration in both the border and infarct zone differed non-significantly between the open chest and closed chest group C Representative picture of histological section of infarcted myocardial tissue containing neutrophils (red cells) 72 hours post-reperfusion D Neutrophil numbers in both the border and infarcted zone of the myocardium are similar between the open chest and closed chest group E Representative picture of macrophages (red) resided in the infarcted myocardium. F Macrophage numbers in both the border and infarcted zone of the myocardium are similar between the open chest and closed group G Representative picture of an active caspase-3 positive cell (red) H Apoptosis of myocardial cells was similar between open chest and closed chest pigs I&J Absolute and relative kininogen levels are higher in the closed chest compared to the open chest group. Pg = picogram, mg = milligram, AU = arbitrary units, CC = closed chest, OC = open chest, *p<0.05, **0.05<p<0.1.

PART ONE CHAPTER 3

DISCUSSION For the development and validation of novel cardioprotective strategies aiming at reduced ischemia reperfusion injury, new compounds need to be tested in models that resemble the clinical situation as closely as possible7,26. The current study indicates that medial sternotomy reduces infarct size and preserves cardiac function in a porcine MI model, possibly confounding the effect of pharmacological therapies in this model. Our observation may partly be explained by the effect of open chest surgery on cardiac pressure, since MAP decreased during the first 3 hours after cardiac ischemia. Cardiac pressure analysis over time revealed a significant decrease in load dependent dPdt-max and dPdt-min 72 hours after surgery despite smaller infarct size and better cardiac function. This indicates that sternotomy unloads the LV and this effect was still measurable after three days. However, a reduction in myocardial workload is unlikely to be the only responsible mechanism for the observed infarct size reduction as previous research reported similar effects of additional trauma on infarct size, which persisted after blood pressure normalization by saline infusion27. Various small and large animal studies have shown that abdominal skin injury preceding experimental MI results in reduced infarct size27,28. This observed reduction seems to be mediated through nociceptive bradykinin-mediated pathways and can be abolished by pharmacologically inhibiting these pathways16,17,29,30. Due to its very short half-life, we were unable to detect circulating bradykinin levels in the current study. However, levels of the bradykinin precursor HMWK were significantly lower in the open chest group compared to the closed chest group, possibly reflecting a higher bradykinin release. This suggests that medial sternotomy exerts its cardioprotective effects through increased bradykinin release. A third, potentially relevant, mechanism is the inflammatory response after MI. Traumatic injury triggers a very complex immune reaction31,32and could therefore alter the inflammatory response that is involved in myocardial reperfusion injury and post-MI repair. However, the number of circulating leukocytes during reperfusion was not decreased in the open chest group, but actually higher at 8 hours reperfusion. Moreover, after 72 hours, we neither observed any significant difference in neutrophil and macrophage accumulation in the myocardium nor myocardial cytokine content, two major determinants of post-MI infarct expansion33â&#x20AC;&#x201C;35. Furthermore, no differences in cardiomyocyte apoptosis were found between the groups. The free radical burst that occurs within the first hours after reperfusion depends on the activation of inflammatory cells during reperfusion in the damaged myocardium 20. MDA levels in blood drawn from the coronary sinus 1 hour after reperfusion did not differ between the open chest and closed chest group. Although this suggests sternotomy does not significantly influence the inflammatory response post-MI, decreased ex vivo cellular responsiveness that has been observed in patients after different surgical procedures could also play a role21,22,36.

INVASIVE SURGERY REDUCES INFARCT SIZE

Regardless of the mechanisms at hand, possible confounders that may obscure the effect of cardioprotective agents in studies of experimental MI should be avoided as much as possible. Better clinical resemblance of animal models will increase translational value and improve reproducibility and translatability towards clinical application. Mortality between the 2 groups was not significantly different. However, this lack of statistical significance might be attributable to insufficient power. Mortality is an essential parameter in translational research and survival bias could be a confounder in cardioprotection studies that have high mortality rates. Unfortunately, the cause of the malignant arrhythmias in the non-surviving animals remains unknown, but something that would be worth investigating. However, in the current study we were unable to clarify this issue, as we did not store any plasma or tissue of the animals that died before the followup time was completed. Despite our best efforts, we could not evade introducing limiting factors ourselves. First of all, experiments were performed under general anesthesia, which cannot be avoided in animal models. The anesthetics used may have a protective effect on the myocardium, although the infarct size compared to the AAR in the minimally invasive group was reasonably high (approximately 70%). Theoretically, differences in blood loss due to the type of intervention may affect blood pressure and thus infarct size. However, blood loss in our experiments was minimal (<200ml) for animals with a circulating blood volume of approximately 5 liters. Furthermore, we did not find any evidence for a modulation of myocardial inflammation by medial sternotomy, indicating that the reduction of infarct size cannot be ascribed to this mechanism. However, to fully conclude this, follow-up studies need to be performed that specifically focus on the inflammatory response (e.g. other leukocyte subsets, cellular responsiveness). Finally, we did not administer any pharmacological agent to block bradykinin-mediated pathways or pro-inflammatory pathways to conclusively determine whether any of the two mechanisms could (partly) be held responsible for the infarct size reduction after medial sternotomy. Determining whether these or other key mechanisms play a role in the reduction of infarct size after medial sternotomy will allow for specifically targeting these pathways. Pharmacological inhibitors could be developed to inhibit these newly discovered cellular signals and determine whether the phenomenon that we have observed in the current study, can directly be translated into clinical applications to salvage myocardium in post-MI patients. Since this was not the primary purpose of this study and therefore remains to be elucidated in future studies. In conclusion, the current study shows that medial sternotomy preceding MI in a porcine model reduces infarct size and preserves cardiac function. Since the mechanisms responsible for this observation are not fully elucidated, it remains unclear how our findings affect the outcome of cardioprotection studies with open chest surgery. The ever-evolving techniques of percutaneous coronary interventions enable animal models to increasingly resemble the clinical situation and in combination with our results call for minimally invasive cardiac ischemia models. This may reduce the possibility of false-positive and â&#x20AC;&#x201C;negative studies and improve translational value of preclinical research.

PART ONE CHAPTER 3

Acknowledgements & Sources of Funding This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs. The authors gratefully acknowledge Marlijn Jansen, Joyce Visser, Martijn van Nieuwburg, Grace Croft, Evelyn Velema, Petra van der Kraak and Lena Bosch for their excellent technical support. All institutional and national guidelines for the care and use of laboratory animals were followed and approved by the appropriate institutional committees. No human studies were carried out by the authors of this article.

INVASIVE SURGERY REDUCES INFARCT SIZE

REFERENCES 1. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit J a., Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart Disease and Stroke Statistics - 2014 Update: A report from the American Heart Association. Circulation. 2014;129:e28–e292. 2. Mendis S, Abegunde D, Yusuf S, Ebrahim S, Shaper G, Ghannem H, Shengelia B. WHO study on Prevention of REcurrences of Myocardial Infarction And StrokE (WHO-PREMISE). Bull World Health Organ. 2005;83:820–828. 3. Peterson MC, Syndergaard T, Bowler J, Doxey R. A systematic review of factors predicting door to balloon time in ST-segment elevation myocardial infarction treated with percutaneous intervention. Int J Cardiol. 2012;157:8–23. 4. Lawesson SS, Alfredsson J, Fredrikson M, Swahn E. Time trends in STEMI--improved treatment and outcome but still a gender gap: a prospective observational cohort study from the SWEDEHEART register. BMJ Open. 2012;2:e000726. 5. Mudd JO, Kass DA. Tackling heart failure in the twenty-first century. Nature. 2008;451:919–928. 6. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988. 7. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 8. Skyschally A, Van Caster P, Boengler K, Gres P, Musiolik J, Schilawa D, Schulz R, Heusch G. Ischemic postconditioning in pigs: No causal role for risk activation. Circ Res. 2009;104:15–18. 9. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker H V, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, López CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West M a, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier R V, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A. 2013;110:3507–12.

10. Munz MR, Faria MA, Monteiro JR, Águas AP, Amorim MJ. Surgical porcine myocardial infarction model through permanent coronary occlusion. Comp Med. 2011;61:445–452. 11. Lukács E, Magyari B, Tóth L, Petrási Z, Repa I, Koller A, Horváth I. Overview of large animal myocardial infarction models (review). Acta Physiol Hung. 2012;99:365–81. 12. De Jong R, van Hout GPJ, Houtgraaf JH, Kazemi K, Wallrapp C, Lewis A, Pasterkamp G, Hoefer IE, Duckers HJ. Intracoronary Infusion of Encapsulated Glucagon-Like Peptide-1-Eluting Mesenchymal Stem Cells Preserves Left Ventricular Function in a Porcine Model of Acute Myocardial Infarction. Circ Cardiovasc Interv. 2014;7:673–683. 13. McCall FC, Telukuntla KS, Karantalis V, Suncion VY, Heldman AW, Mushtaq M, Williams AR, Hare JM. Myocardial infarction and intramyocardial injection models in swine. Nat Protoc. 2012;7:1479–1496. 14. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 15. Ishikawa K, Aguero J, Tilemann L, Ladage D, Hammoudi N, Kawase Y, Santos-Gallego CG, Fish K, Levine R a., Hajjar RJ. Characterizing preclinical models of ischemic heart failure: differences between LAD and LCx infarctions. AJP Hear Circ Physiol. 2014;307:H1478–H1486. 16. Gross GJ, Baker JE, Moore J, Falck JR, Nithipatikom K. Abdominal surgical incision induces remote preconditioning of trauma (RPCT) via activation of bradykinin receptors (BK2R) and the cytochrome P450 epoxygenase pathway in canine hearts. Cardiovasc Drugs Ther. 2011;25:517–522. 17. Gross GJ, Hsu A, Gross ER, Falck JR, Nithipatikom K. Factors Mediating Remote Preconditioning of Trauma in the Rat Heart: Central Role of the Cytochrome P450 Epoxygenase Pathway in Mediating Infarct Size Reduction. J Cardiovasc Pharmacol Ther. 2012;18:38–45. 18. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 19. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 20. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 21. Versteeg D, Hoefer IE, Schoneveld AH, de Kleijn DP V, Busser E, Strijder C, Emons M, Stella PR, Doevendans P a, Pasterkamp G. Monocyte toll-like receptor 2 and 4 responses and expression following percutaneous coronary intervention: association with lesion stenosis and fractional flow reserve. Heart. 2008;94:770–776. 22. Versteeg D, Dol E, Hoefer IE, Flier S, Buhre WF, de Kleijn D, van Dongen EP, Pasterkamp G, de Vries J-P. Toll-like receptor 2 and 4 response and expression on monocytes decrease rapidly in

PART ONE CHAPTER 3

patients undergoing arterial surgery and are related to preoperative smoking. Shock. 2009;31:21–27. 23. Van Hout GPJ, Jansen Of Lorkeer SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE, Jansen SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE. Admittance-based pressure-volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014;2:e00287. 24. Van Hout GPJ, de Jong R, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittance-based pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75. 25. Koudstaal S, Jansen of Lorkeers SJ, Gho JMIH, van Hout GPJ, Jansen MS, Gründeman PF, Pasterkamp G, Doevendans P a, Hoefer IE, Chamuleau S a J. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014;1–10. 26. Sluijter JPG, Condorelli G, Davidson SM, Engel FB, Ferdinandy P, Hausenloy DJ, Lecour S, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Van Laake LW. Novel therapeutic strategies for cardioprotection. Pharmacol Ther. 2014;144:60–70. 27. Ren X, Wang Y, Jones WK. TNF-alpha is required for late ischemic preconditioning but not for remote preconditioning of trauma. J Surg Res. 2004;121:120–129. 28. Jones WK, Fan GC, Liao S, Zhang JM, Wang Y, Weintraub NL, Kranias EG, Schultz J El, Lorenz J, Ren X. Peripheral nociception associated with surgical incision elicits remote nonischemic cardioprotection via neurogenic activation of protein kinase C signaling. Circulation. 2009;120:S1–9. 29. Heusch G, Bøtker HE, Przyklenk K, Redington A, Yellon D. Remote Ischemic Conditioning. J Am Coll Cardiol. 2015;65:177–195. 30. Hausenloy DJ, Yellon DM. “Conditional Conditioning” in cardiac bypass surgery. Basic Res Cardiol. 2012;107:258. 31. Valparaiso AP, Vicente D, Bograd B, Elster E, Davis T. Modeling acute traumatic injury. J Surg Res. 2015;194:220–232. 32. Robertson CM, Coopersmith CM. The systemic inflammatory response syndrome. Microbes Infect. 2006;8:1382–1389. 33. Carbone F, Nencioni A, Mach F, Vuilleumier N, Montecucco F. Pathophysiological role of neutrophils in acute myocardial infarction. Thromb Haemost. 2013;110:501–14. 34. Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res. 2002;53:31–47. 35. Marx N, Neumann FJ, Ott I, Gawaz M, Koch W, Pinkau T, Schömig A. Induction of cytokine expression in leukocytes in acute myocardial

infarction. J Am Coll Cardiol. 1997;30:165–170. 36. Scholtes VPW, Versteeg D, de Vries J-PPM, Hoefer IE, Schoneveld AH, Stella PR, Doevendans P a FM, van Keulen KJK, de Kleijn DP V, Moll FL, Pasterkamp G. Toll-like receptor 2 and 4 stimulation elicits an enhanced inflammatory response in human obese patients with atherosclerosis. Clin Sci (Lond). 2011;121:205–214.

chapter 4

Study design: Anti-inflammatory compounds to reduce infarct size in large animal models of myocardial infarction: A meta-analysis

Evidence Based Preclinical Medicine 2014 December 1(1): 4-10

G.P.J. van Hout1*, S.J. Jansen of Lorkeers2*, K.E. Wever3, E.S. Sena4, W.W. van Solinge5, P.A. Doevendans1,2, G. Pasterkamp1, S.A.J. Chamuleau2, I.E. Hoefer1 *Contributed equally to this work 1

Experimental Cardiology Laboratory, University Medical Center Utrecht, Utrecht, The Netherlands

Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands

Systematic Review Centre for Laboratory animal Experimentation, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands

4 5

Centre for clinical brain sciences, University of Edinburgh, Edinburgh, United Kingdom Department of Clinical Chemistry and Haematology, University Medical Center Utrecht, Utrecht, The Netherlands

PART ONE CHAPTER 4

ABSTRACT Targeting the inflammatory response after myocardial infarction (MI) could potentially prevent infarct expansion, resulting in a preservation of cardiac function. Despite extensive testing in large animal models of MI, anti-inflammatory therapeutics are not incorporated in daily clinical practice. Methodological review of the literature describing the effects of anti-inflammatory compounds in large animal models of MI may provide useful insights into the reasons for the translational failure from preclinical to clinical studies. Moreover, systematic review of these preclinical studies may allow us to determine which antiinflammatory agents have the greatest potential to successfully treat MI in the clinic and guide which pre-clinical setting seems most appropriate to test these future treatment strategies in. The current systematic review protocol provides a detailed description of the design of this systematic review of studies investigating the effects of anti-inflammatory compounds in large animal models of MI.

GENERAL Title of the review Anti-inflammatory compounds to reduce infarct size in large animal models of myocardial infarction: a meta-analysis Authors G.P.J. van Hout and S.J. Jansen of Lorkeers are shared first author and are responsible for the study design, title/abstract screening, full text screening, data extraction and manuscript preparation. G.P.J. van Hout is responsible for the data-analysis. Co-authors of this manuscript are K.E. Wever (study design, data-analysis), E.S. Sena (study design, data-analysis), P.A. Doevendans (manuscript preparation and optimization), W.W. van Solinge (manuscript preparation and optimization), G. Pasterkamp (manuscript preparation and optimization), S.A.J. Chamuleau (study design, manuscript preparation) and I.E. Hoefer (study design, manuscript preparation). Contact person + E-mail address G.P.J. van Hout â&#x20AC;&#x201C; g.p.j.vanhout@umcutrecht.nl Background Cardiac damage after myocardial infarction (MI) induces a sterile immune response that leads to infarct expansion after the initial ischemic event1â&#x20AC;&#x201C;3. Due to the release of proinflammatory mediators from the damaged myocardium post-MI, circulating cells are drawn to the infarcted tissue to clear out dead cardiac resident cells and promote infarct healing. Paradoxically, these circulating cells are known to also target viable tissue and greatly amplify the initial inflammatory response, thereby inducing infarct enlargement regardless of the application of revascularization therapies in patients suffering from MI4,5. Therefore,

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

targeting the immune response could potentially prevent infarct expansion, resulting in a preservation of cardiac function and a prevention of heart failure in patients suffering from ischemic heart disease6. Before effectiveness of new anti-inflammatory therapeutics can be tested in clinical trials, novel compounds are commonly tested in large animal models of MI. These large animal models have shown to possess additive translational value due to comparable hemodynamics, similar heart size and corresponding coronary physiology7–9. Moreover, large animal models enable clinical treatment regimens, delivery route and identical function-related measurements and therefore could provide an evidence base for clinical trial design10–14. Over the last 5 decades, numerous pharmacological therapies that target the inflammatory response have been tested in large animal models of MI. Unfortunately, none of these treatments have made it past clinical trials into daily practice. Methodological review of the preclinical literature describing the effects of antiinflammatory compounds in large animal models of MI may provide useful insights into the reasons for the translational failure from preclinical to clinical studies. Systematic review of these preclinical studies may also allow us to determine which anti-inflammatory agents have the greatest potential to successfully treat myocardial infarction in the clinic and guide which pre-clinical setting seems most appropriate to test these future treatment strategies in. Objectives of this SR Specify the disease/health problem of interest The mortality due to myocardial infarction (MI) has decreased over the past 30 years. This can be mainly attributed to optimized revascularization therapy15,16. However, MI still accounts for a large amount of cardiovascular deaths worldwide and is expected to increase again due to an increased prevalence of obesity and diabetes in the western world17. Moreover, as more patients survive, the prevalence of heart failure – a direct consequence of MI – dramatically increases18. These high mortality numbers, combined with the societal and economic burden of heart failure, call for improved treatment after acute MI. There is considerable heterogeneity regarding the way myocardial infarction is applied in the different studies performed during the past few decades (e.g. open/closed chest, permanent/temporary ligation). In the current systematic review we have included studies that satisfy the following definition: ‘’An intervention that leads to permanent or temporary total occlusion of a coronary artery, disabling blood flow for a period long enough (>30min) to induce permanent damage to the myocardium. ‘’ Specify the population/species studied In this systematic review we will focus on large animal models. We specifically want to study this group of animals because the cardiac physiology and anatomy (e.g. hemodynamics, coronary anatomy) combined with the immune response post-MI is considered relatively similar to humans. Also, route of drug administration and treatment regimens in large animal models allow protocols that resemble clinical treatment. In this

PART ONE CHAPTER 4

review, we focus on four species: pigs, dogs, sheep and goats because these species are mostly used and widely applied for the validation of novel anti-inflammatory treatments for MI9,12,19. Specify the intervention/exposure The intervention applied for the treatment of MI should in this case be a pharmacological treatment with an anti-inflammatory drug. Any route of drug delivery or treatment regimen of anti-inflammatory compounds (IM/IV/SC/PO) is possible. Studies were included in the analyses if the interventions applied met the following criteria: 1. A compound that is FDA-approved for its anti-inflammatory mechanisms of action. AND/OR 2. A compound that directly targets a (recently discovered) mechanisms, that plays a proven role in inflammation. In this perspective it is possible that certain compounds may have multiple mechanisms of action and pleiotropic effects besides being anti-inflammatory. To exclusively investigate anti-inflammatory compounds, we chose to exclude interventions that have relevant and evident pleiotropic effects. Also, if the mechanism of action was unclear regardless of the effect on inflammatory parameters, these interventions were excluded. Finally, the treatment had to be solely pharmacological. According to these criteria, the following interventions were excluded from analysis: Stem-cells, biomaterials, pro-inflammatory compounds, statins, ACE-inhibitors, Î˛-blockers, calcium-channel antagonists, adenosine analogues, prostacyclin analogues, L-arginine analogues, endothelin-1 analogues, thromboxane A2-antagonists, omega-3 fatty acids, gene therapy, anesthetics, extracorporal treatments, aspirin in dosages <6.25mg/kg/day (anti-platelet therapy), flavonoids, flavonols, map-kinase inhibitors. Specify the control population The control population is a group of animals that receive the above-defined MI without any additional (anti-inflammatory) treatment and is preferably placebo controlled. If studies use multiple control groups, the control group that resembles to the interventional group the most in terms of vehicle use and administration route was chosen. Specify the outcome measures Primary outcome measures: Myocardial infarct size (IS) measured as a percentage of the area at risk (AAR) Secondary outcome measures: 1. Myocardial IS measured as a percentage of the total left ventricle (LV) 2. Left ventricular ejection fraction (LVEF) 3. Myocardial scar thinning given in millimetres or as a ratio, divided by the opposite, non-infarcted wall

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

4. Left ventricular end diastolic volume (LVEDV) and left ventricular end systolic volume (LVESV) in mL 5. Mortality after therapy administration State your research question What is the effect of anti-inflammatory compounds on mortality, infarct size, cardiac function and myocardial scar thinning in large animal models of myocardial infarction, when compared to placebo-treated large animal models of MI?

METHODS Search and study identification Identify literature databases to search Pubmed and Embase Define electronic search strategy Pubmed Myocardial ischemia[Mesh] OR Myocardial infarction[tiab] OR Myocardial infarctions[tiab] OR Myocardial infarct[tiab] OR Myocardial infarcts[tiab] OR Myocardial ischemia[tiab] OR Myocardial ischemias[tiab] OR Myocardial ischaemia[tiab] OR Myocardial ischaemias[tiab] OR Myocardial reperfusion injury[Mesh] OR Myocardial reperfusion[tiab] OR (Myocardial[tiab] AND Reperfusion injury[tiab]) OR Coronary occlusion[tiab] OR Coronary occlusions[tiab] AND Large model[tiab] OR Large animal[tiab] OR Swine[Mesh] OR Swine[tiab] OR Pigs[tiab] OR Pig[tiab] OR Porcine[tiab] OR Suidae[tiab] OR Hog[tiab] OR Minipig[tiab] OR Minipigs[tiab] OR Dogs[Mesh] OR Dogs[tiab] OR Dog[tiab] OR Canis[tiab] OR Canine[tiab] OR Hound[tiab] OR Sheep[Mesh] OR Sheep[tiab] OR Ovis[tiab] OR Ovine[tiab] OR Goats[Mesh] OR Goat[tiab] OR Goats[tiab] OR Capra[tiab] OR Capras[tiab] AND Anti-inflammatory agents[Mesh] OR Anti-inflammatory agents[Pharmacological Action] OR Anti-inflammatory[tiab] OR Antiinflammatory[tiab] OR Anti inflammatory[tiab] OR Inflammation[tiab] OR Inflammatory[tiab] OR Pro-inflammatory[tiab] OR Immunosuppressive agents[Mesh] OR Immunosuppressive agents[Pharmacological Action] OR Immunosuppression [Mesh] OR Immunosupressive [tiab] OR Immuno-supressive [tiab] OR Immuno suppressive [tiab] OR Immunosuppressant [tiab] OR Immune suppressant [tiab] OR Immunosupression[tiab] OR Antioxidants[Mesh] OR Antioxidants[tiab] OR

PART ONE CHAPTER 4

Antioxidant[tiab] OR Anti-oxidant[tiab] OR Free radical scavenger[tiab] OR Radicals[tiab] OR Radical[tiab] OR Metalloproteinase[tiab] OR Complement system proteins[Mesh] OR Complement[tiab] OR Cyclosporins[Mesh] OR Cyclosporine[tiab] OR Adrenal cortex hormones[Mesh] OR Adrenal cortex hormones[pharmacological action] OR corticosteroids[tiab] OR corticosteroid[tiab] OR cyclooxyenase[tiab] OR COX[tiab] OR Antiinflammatory agents, non-steroidal [Mesh] OR NSAID[tiab] OR Cox-2[tiab] OR NSAIDS[tiab] OR Leukocytes[Mesh] OR Leukocyte[tiab] OR Leukocytes[tiab] OR Neutrophil[tiab] OR Neutrophils[tiab] OR Monocyte[tiab] OR Monocytes[tiab] OR Macrophage[tiab] OR Macrophages[tiab] OR Cytokines[Mesh] OR Cytokine[tiab] or Cytokines[tiab] OR Interleukin[tiab] OR Interleukins[tiab] OR Chemokine[tiab] OR Chemokines[tiab] OR Integrins[Mesh] OR Integrin[tiab] OR Integrins[tiab] OR Toll-like receptors[Mesh] OR TLR[tiab] OR Toll-like receptor[tiab] OR Toll-like receptors[tiab] OR TLRs[tiab] OR Inflammasome [tiab] Embase ‘ischemic heart disease’/exp OR ‘myocardial infarction’:ab,ti OR ‘myocardial infarctions’:ab,ti OR ‘myocardial infarct’:ab,ti OR ‘myocardial infarcts’:ab,ti OR ‘myocardial ischemia’:ab,ti OR ‘myocardial ischemias’:ab,ti OR ‘myocardial ischaemia’:ab,ti OR ‘myocardial ischaemias’:ab,ti OR ‘myocardial reperfusion’:ab,ti OR ‘coronary occlusion’:ab,ti OR ‘coronary occlusions’:ab,ti OR (myocardial:ab,ti AND reperfusion:ab,ti AND injury:ab,ti) AND ‘swine’/exp OR ‘dog’/exp OR ‘sheep’/exp OR ‘goat’/exp OR ‘large model’:ab,ti OR ‘large animal’:ab,ti OR swine:ab,ti OR pigs:ab,ti OR pig:ab,ti OR porcine:ab,ti OR suidae:ab,ti OR hog:ab,ti OR minipig:ab,ti OR minipigs:ab,ti OR dogs:ab,ti OR dog:ab,ti OR canis:ab,ti OR canine:ab,ti OR hound:ab,ti OR sheep:ab,ti OR ovis:ab,ti OR ovine:ab,ti OR goat:ab,ti OR goats:ab,ti OR capra:ab,ti OR capras:ab,ti AND ‘antiinflammatory agent’/exp OR ‘immunosuppressive agent’/exp OR ‘immunosuppressive agent’/exp OR ‘corticosteroid’/exp OR ‘anti-inflammatory’:ab,ti OR ‘antiinflammatory’:ab,ti OR ‘anti inflammatory’:ab,ti OR ‘anti-inflammatory drugs’:ab,ti OR ‘anti-inflammatory drug’:ab,ti OR ‘inflammation’:ab,ti OR ‘inflammatory’:ab,ti OR ‘pro inflammatory’:ab,ti OR ‘immunosupressive’:ab,ti OR ‘immuno supressive’:ab,ti OR ‘immuno suppressive’:ab,ti OR ‘immunosuppressant’:ab,ti OR ‘immune suppressant’:ab,ti OR ‘immunosupression’:ab,ti OR ‘radicals’:ab,ti OR ‘radical’:ab,ti OR ‘metalloproteinase’:ab,ti OR ‘complement’:ab,ti OR ‘cyclosporin’:ab,ti OR ‘corticosteroids’:ab,ti OR ‘corticosteroid’:ab,ti OR ‘cyclooxyenase’:ab,ti OR ‘cox’:ab,ti OR ‘nsaid’:ab,ti OR ‘cox 2’:ab,ti OR ‘nsaids’:ab,ti OR ‘leukocyte’:ab,ti OR ‘leukocytes’/exp OR leukocytes:ab,ti OR neutrophil:ab,ti OR neutrophils:ab,ti OR monocyte:ab,ti OR monocytes:ab,ti OR macrophage:ab,ti OR

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

macrophages:ab,ti OR cytokines:ab,ti OR cytokine:ab,ti OR chemokines:ab,ti OR chemokine:ab,ti OR interleukin:ab,ti OR interleukins:ab,ti OR integrin:ab,ti OR integrins:ab,ti OR tlr:ab,ti OR ‘toll-like receptor’/exp OR ‘toll-like receptor’:ab,ti OR ‘toll-like receptors’:ab,ti OR ‘tlrs’:ab,ti OR ‘inflammasome’:ab,ti Identify other sources for study identifications None Define search strategies for these sources Not applicable

Study selection procedure Define screening phases 1. Title/Abstract screening phase 2. Full-text screening phase Specify number of observers per screening phase 2 observers per reference per phase: G.P.J. van Hout and S.J. Jansen of Lorkeers Study selection criteria Type of study design Inclusion criteria: Controlled study / Cohort study Exclusion criteria: Other study types including non-controlled studies and case reports. Type of animal/population Inclusion criteria: Pigs, dogs, sheep and goats subjected to MI as defined previously. Co-medication that potentially influences infarct size (e.g. platelet inhibitors/β-blockade) is no exclusion criteria as long as the control group also receives the co-intervention. Exclusion criteria: Transgenic animals. All other species of animals different from the ones previously mentioned or animals from the same species but different disease models. In vitro and ex vivo studies are also excluded. Type of intervention e.g. doses, time, frequency See the previously stated definition of the intervention. The timing, delivery route and frequency of the anti-inflammatory compounds are no exclusion criteria for this systematic review. Outcome measures Inclusion criteria, studies are included in the analysis if the reported at least one of the following outcome measures:

PART ONE CHAPTER 4

1. Primary outcome measurement: Myocardial infarct size (IS) measured as a percentage of the area at risk (IS/AAR) 2. Myocardial IS measured as a percentage of the total left ventricle (IS/LV) 3. Left ventricular ejection fraction (LVEF) 4. Myocardial scar thinning given in millimetres or as a ratio, divided by the opposite, non-infarcted wall Exclusion criteria, studies are excluded in absence of the following outcome measures: 1. The absence of inclusion criteria 1 to 4: IS/AAR, IS/LV, LVEF or myocardial scar thinning. Language restrictions None. Publication date restrictions None. Other Inclusion criteria: Full text original papers only, meaning no congress abstracts. Exclusion criteria: Congress abstracts. Sort and prioritize your exclusion criteria per selection phase Selection phase title/abstract: 1. No MI 2. No anti-inflammatory compound 3. No original data (e.g. review) 4. No large animals 5. Not in vivo Selection phase Full Text 1. No full-text publication 2. No MI 3. No anti-inflammatory compound 4. No original data 5. No large animals 6. Not in vivo 7. No correct endpoints (EF, IS/AAR, IS/LV or scar thinning) 8. Lack of control group Study characteristics to be extracted Study ID First author, corresponding author, journal, publication year, source of funding.

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

Study Design characteristics Number of animals per groups and experimental groups. Animal model characteristics Species, gender, age, weight, location of infarct, method of induction of injury (e.g. ligation, balloon occlusion), model (I/R or permanent), anaesthetics, ventilation, duration of occlusion, surgical procedure (closed chest, lateral sternotomy, medial sternotomy), comorbidity. Intervention characteristics Name compound, treatment group (e.g. NSAID, free radical scavenger, corticosteroids) dosage, time of delivery, number of administrations, bolus vs. continuous administration, duration of delivery, duration of follow-up, route of delivery, co-treatment. Outcome measures Method of functional outcome assessment, method of infarct size determination. Outcome data for any of the following: 1. Myocardial infarct size (IS) measured as a percentage of the area at risk (AAR) 2. Myocardial IS measured as a percentage of the total left ventricle (LV) 3. Left ventricular ejection fraction (LVEF) 4. Myocardial scar thinning given in millimetres or as a ratio, divided by the opposite, non-infarcted wall 5. Left ventricular end diastolic volume (LVEDV) and left ventricular end systolic volume (LVESV) 6. Mortality after therapy administration Risk of bias assessment Scoring for the risk of bias is performed based on the CAMARADES checklist [Sena et al. Trends in Neuroscience, 2007]. • Publication in a peer reviewed journal • Reporting of random allocation • Reporting of blinding of the operator • Reporting of blinded assessment of outcome • Use of comorbid animals • Reporting of a sample size calculation • Reporting of compliance with animal welfare regulations • Reporting of a potential conflict of interest Data collection • Infarct size: IS/AAR in % • Infarct size: IS/LV in % • LVEF in %

PART ONE CHAPTER 4

• Scar thinning in millimetres and/or ratio • LVESV and LVEDV in mL • Mortality: number of animals that died after therapy administration Methods of data extraction/retrieval Primarily we extract data from the result section of the manuscript. When data is unavailable in text or tables, data will be extracted electronically from graphs. In absence of these data in graphs, authors will be contacted through e-mail. In case of no response after 2 weeks, we will exclude the study from the analysis. Data-analysis and -synthesis Data combination Data will be combined in a systematic review with a forest plot followed by a meta-analysis. Specify when data combination is appropriate Based on our inclusion- and exclusion criteria, we expect the models and outcome measures to be uniform enough to pool data for a combined analysis for each separate outcome measurement. Anti-inflammatory compounds have been tested for several decades in large animal models of MI and we therefore expect to include over one hundred different studies. We chose 25 as a cut off for number of studies to be included in order to allow reliable determination of publication bias and meta-analysis20. In the current systematic review, we also aim to determine whether study related parameters influence outcome (e.g. surgical approach, time of compound delivery and type of antiinflammatory compound). In our opinion, direct comparison of these subgroups is feasible when groups contain at least 5 independently conducted studies. Meta-regression will be used to explore heterogeneity between included studies. Significant predictors will be further investigated by supgroup analysis. If sufficient studies are included, within subgroup analyses will also be performed for different treatment groups of anti-inflammatory compounds (i.e. NSAIDS, free radical scavengers) to determine if study related parameters influence outcome in these particular groups of anti-inflammatory compounds. The number of parameters tested by meta-regression is based on the number of included studies per outcome measure and will be 1 parameter for every 10 studies. For the primary outcome measurement (infarct size) no correction will be applied (p<0.05 will be regarded as significant). For all the secondary outcome measures, a bonferroni correction will be applied, based on the number of parameters tested. If meta-analysis seems feasible Specify the effect measures to be used We expect the reporting of the selected outcome measurements to be very uniform since our primary outcome measures (IS/AAR, IS/LV, LVEF, scar thinning) are reported on a relative scale (ratio/percentage) in the vast majority of cases. For this reason it is assumable we will

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

be able to use the raw difference in means for these parameters. For each study group we will extract the reported mean combined with either the standard deviation or standard error of the mean, depending on the parameter provided by the different studies. Effect measures used for each individual parameter: • Mortality – Odds ratio • IS/AAR – Raw difference in means • IS/LV – Raw difference in means • LVEF – Raw difference in means • Scar thinning – Raw difference in means or Standardized difference in means, depending on the uniformity of the data • LVEDV and LVESV – Standardized difference in means Specify which study characteristics will be analysed as possible sources of heterogeneity 1. Treatment group (NSAIDs, corticosteroids, immunosuppressives, free radical scavengers, complement inhibition, cytokine/chemokine inhibitors, integrin/selectin/ cell surface inhibitors, others) 2. Ischemia-reperfusion/permanent occlusion 3. Occlusion duration 4. Timing of therapy 5. Timing of assessment 6. Year of publication 7. Animal species 8. Surgical procedure (sternotomy vs. no sternotomy) 9. Location of injury (which coronary artery was occluded) 10. Gender 11. Randomization 12. Blinding 13. Methods of infarct/function measurement 14. English/non-English articles Specify subgroups and comparisons of interests Subgroup analysis will be performed for significant predictors of outcome, based on metaregression. No sensitivity-analysis will be performed. Method of analysis Because of anticipated heterogeneity between included studies, a random effects model will be appropriate. We will also determine the extent of heterogeneity in our dataset by assessing the Tau2 and I2 statistic. Method for assessing risk of publication bias We will assess the risk of publication bias through the construction of a Funnel Plot and a

PART ONE CHAPTER 4

subsequent Egger’s regression for testing symmetry in funnel plot and detecting small study bias. Duval and Tweedie’s ‘trim and fill analysis’ will be performed to identify missing studies.

OTHER Possible limitations 1. Inclusion of studies is based on LVEF, IS and scar thinning which means that studies that do not report this will be excluded, enabling possible selection bias. 2. Our definition of ‘anti-inflammatory treatment’ is based on working mechanism and clinical usage of certain compounds, requiring arbitrary choices. In our study, possible anti-inflammatory therapies with unknown working mechanisms or major pleiotropic effects are excluded. On the other hand, compounds that will be included in the final analysis could, to some extent, also have pleiotropic effects possibly clouding the sole effect of inhibiting the inflammatory response post MI. 3. In the current study design we have not taken into account the commercialization potential of certain drug types. While clinical testing of certain compounds may seem to be very promising based on this systematic review, it should be noted that we have not taken the economic position (e.g. patent, costs) into consideration, which is also essential for eventual drug development

ANTI-INFLAMMATORY COMPOUNDS IN PRECLINICAL MI

REFERENCES 1. Christia P, Frangogiannis NG. Targeting inflammatory pathways in myocardial infarction. Eur J Clin Invest. 2013;43:986–995. 2. Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res. 2002;53:31–47. 3. Abbate A, Van Tassell BW, Biondi-Zoccai G, Kontos MC, Grizzard JD, Spillman DW, Oddi C, Roberts CS, Melchior RD, Mueller GH, Abouzaki NA, Rengel LR, Varma A, Gambill ML, Falcao RA, Voelkel NF, Dinarello CA, Vetrovec GW. Effects of interleukin-1 blockade with anakinra on adverse cardiac remodeling and heart failure after acute myocardial infarction [from the virginia commonwealth university-anakinra remodeling trial (2) (vcu-art2) pilot study]. Am J Cardiol. 2013;111:1394–1400. 4. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 5. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 6. Seropian IM, Toldo S, Van Tassell BW, Abbate A. Anti-inflammatory strategies for ventricular remodeling following St-segment elevation acute myocardial infarction. J Am Coll Cardiol. 2014;63:1593–1603. 7. Munz MR, Faria M a., Monteiro JR, Águas AP, Amorim MJ. Surgical porcine myocardial infarction model through permanent coronary occlusion. Comp Med. 2011;61:445–452. 8. Suzuki Y, Lyons JK, Yeung AC, Ikeno F. In vivo porcine model of reperfused myocardial infarction: In situ double staining to measure precise infarct area/area at risk. Catheter Cardiovasc Interv. 2008;71:100–107. 9. McCall FC, Telukuntla KS, Karantalis V, Suncion VY, Heldman AW, Mushtaq M, Williams AR, Hare JM. Myocardial infarction and intramyocardial injection models in swine. Nat Protoc. 2012;7:1479–1496. 10. Arslan F, Houtgraaf JH, Keogh B, Kazemi K, De Jong R, McCormack WJ, O’Neill L a J, McGuirk P, Timmers L, Smeets MB, Akeroyd L, Reilly M, Pasterkamp G, De Kleijn DP V. Treatment with OPN-305, a humanized anti-toll-like receptor-2 antibody, reduces myocardial ischemia/reperfusion injury in pigs. Circ Cardiovasc Interv. 2012;5:279– 287. 11. Timmers L, Henriques JPSPS, de Kleijn DPVP V, Devries JHH, Kemperman H, Steendijk P, Verlaan CWJWJ, Kerver M, Piek JJJ, Doevendans P a a., Pasterkamp G, Hoefer IEE. Exenatide Reduces Infarct Size and Improves Cardiac Function in a Porcine Model of Ischemia and Reperfusion Injury. J Am Coll Cardiol. 2009;53:501–510.

12. Suzuki Y, Yeung AC, Ikeno F. The representative porcine model for human cardiovascular disease. J Biomed Biotechnol. 2011;2011:195483. 13. Schaper W, Jageneau a, Xhonneux R. The development of collateral circulation in the pig and dog heart. Cardiologia. 1967;51:321–335. 14. Matsunaga T, Warltier DC, Weihrauch DW, Moniz M, Tessmer J, Chilian WM. Ischemia-induced coronary collateral growth is dependent on vascular endothelial growth factor and nitric oxide. Circulation. 2000;102:3098–3103. 15. Peterson MC, Syndergaard T, Bowler J, Doxey R. A systematic review of factors predicting door to balloon time in ST-segment elevation myocardial infarction treated with percutaneous intervention. Int J Cardiol. 2012;157:8–23. 16. Zhou C, Yao Y, Zheng Z, Gong J, Wang W, Hu S, Li L. Stenting technique, gender, and age are associated with cardioprotection by ischaemic postconditioning in primary coronary intervention: A systematic review of 10 randomized trials. Eur Heart J. 2012;33:3070–3077. 17. Yeh RW, Sidney S, Chandra M, Sorel M, Selby J V, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med. 2010;362:2155–2165. 18. Bui AL, Horwich TB, Fonarow GC. Epidemiology and risk profile of heart failure. Nat Rev Cardiol. 2011;8:30–41. 19. Zaragoza C, Gomez-Guerrero C, Martin-Ventura JL, Blanco-Colio L, Lavin B, Mallavia B, Tarin C, Mas S, Ortiz A, Egido J. Animal models of cardiovascular diseases. J Biomed Biotechnol. 2011;2011:497841. 20. Sena ES, Bart van der Worp H, Bath PMW, Howells DW, Macleod MR. Publication bias in reports of animal stroke studies leads to major overstatement of efficacy. PLoS Biol. 2010;8:e1000344.

chapter 5

Anti-inflammatory compounds reduce infarct size after myocardial infarction: A meta-analysis of large-animal models

In revision - Cardiovascular Research

G.P.J. van Hout1*, S.J. Jansen of Lorkeers2*, K.E. Wever3, E.S. Sena4, L.H.J.A. Kouwenberg1, W.W. van Solinge5, M.R. Macleod4, P.A. Doevendans2, G. Pasterkamp1,5, S.A.J. Chamuleau2, I.E. Hoefer1,5 * Contributed equally to this work 1

Experimental Cardiology Laboratory, University Medical Center Utrecht, Utrecht, The Netherlands

Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands

Systematic Review Centre for Laboratory Animal Experimentation, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands

Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom

Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, Utrecht

PART ONE CHAPTER 5

ABSTRACT Background The development of novel therapies is required to confine cardiac damage and prevent heart failure (HF) in patients after myocardial infarction (MI). For this purpose, numerous anti-inflammatory drugs have been tested in experimental large animal studies. However, translation of these therapeutic strategies into clinical practice has proven to be difficult. To identify promising drug groups and better understand which factors underlie the failure of transition towards the clinic, we performed a meta-analysis of large animal studies on the cardioprotective effects of anti-inflammatory drugs post-MI. Methods and Results Meta-analysis of the 183 included studies revealed a significantly reduced infarct size (IS) as a ratio of the area at risk (12.7%; 95%CI 11.1%â&#x20AC;&#x201C;14.4%, p<0.001) and also a reduced IS as a ratio of the left ventricle (3.9%;95%CI 3.1%â&#x20AC;&#x201C;4.7%, p<0.001). No difference in mortality was observed. Effect size depended on the type of drug used (p<0.001), was higher when outcome was assessed early after MI (p=0.013) and where studies included only male animals (p<0.001). Mortality in treated animals was higher in studies that blinded the investigator during the experiment (p=0.041). Conclusions Treatment with anti-inflammatory drugs leads to smaller infarct size in large animal MI models. Effect size depends on the type of drug used, which enables identification of promising compounds for future clinical testing. Timing of outcome assessment, sex and study quality are significantly associated with outcome and may explain part of the translational failure in clinical settings.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

INTRODUCTION Myocardial infarction (MI) and its consequences remain one of the greatest burdens of disease worldwide1,2. Optimized medical care has resulted in reduced acute mortality but patients surviving MI often develop diminished cardiac function and heart failure (HF)3. The progression from MI to HF occurs through adverse remodeling, a complex mechanism involving infarct expansion, myocardial scar thinning and left ventricular geometrical adaptation4. Accumulating evidence indicates that adverse remodeling arises from an exaggerated inflammatory response that is initiated during ischemia and early reperfusion5,6. The damaged myocardium attracts inflammatory cells towards the site of injury where these cells secrete cytokines, proteases and oxygen free radicals. This results in degradation of the extracellular matrix and clearing of necrotic cardiac resident cells, enabling scar formation in a later phase of cardiac wound healing7. This inflammatory effect is essential to stabilize the scar tissue8. However, it is outweighed by the acute effects of inflammatory cells on infarct expansion and cardiac function worsening early after ischemia in the reperfusion phase9–11. Indeed, multiple clinical studies have shown that elevated numbers of circulating neutrophils, monocytes and cytokines in the acute setting are associated with adverse remodeling, the development of heart failure and a worse overall prognosis12–17. Attenuation of this inflammatory response is therefore a promising strategy to limit infarct size and preserve cardiac function post-MI. Development of such anti-inflammatory strategies and transition from bench to bedside, requires their testing in clinically relevant animal models.18 Both temporal and spatial development of tissue damage post-MI, along with inflammation-related signaling pathways, differ in small animals compared to larger mammals, including humans19–21. Large animal models more closely resemble human anatomy, hemo- and pharmacodynamics and enable clinical treatment regimens, delivery routes and functional read-outs22–24. Since the early 1970s, many anti-inflammatory compounds have been shown to have efficacy in reducing reperfusion damage and post-MI remodeling in large animal models25. However, none have proved successful in clinical trials and entered routine clinical practice26,27. The reasons for this lack of success – or which class of anti-inflammatory drug has greatest efficacy in large animal MI models –are not clear. The interpretation of findings from individual studies to give an overview of the utility of this therapeutic approach is challenging because these studies use diverse experimental set-ups, animal species, timings of therapy and induction of MI. We therefore performed a systematic review and meta-analysis of large animal studies testing anti-inflammatory compounds after MI. In addition to providing an unbiased research summary, our purpose was to identify promising drug groups for future clinical testing. Moreover, we considered that meta-regression may enable us to better understand sources of heterogeneity, to identify the influence of study design factors which may be important for translational success28.

PART ONE CHAPTER 5

METHODS The protocol for this systematic review and meta-analysis has been published.29 We searched PubMed and Embase on May 1st 2014 for studies using anti-inflammatory compounds in large animal models of MI. Our inclusion criteria were: controlled study design, and the use of large animal models (defined as either pigs, dogs, sheep, goats) of myocardial infarction (coronary obstruction >30 minutes). Anti-inflammatory compounds were defined as either having FDA approval for anti-inflammatory effects or as directly targeting pivotal inflammatory mechanisms.29 Abstract publications were excluded, as were case reports and studies lacking a control group. There were no language restrictions. Studies were screened in two phases (title/abstract followed by full text) by two independent researchers (GvH and SJ). In case of disagreement, consensus was achieved by discussion in all cases. The primary outcome for meta-analysis was infarct size as percentage of the area at risk (IS/AAR). Secondary outcomes were infarct size as percentage of the left ventricle (IS/LV), mortality, left ventricular ejection fraction (EF), left ventricular end diastolic volume (LVEDV), left ventricular end systolic volume (LVESV) and wall thickness as ratio of the opposing wall (WT). Data were extracted from included studies and added to the ‘Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies’ (CAMARADES) – online international database. Statistics Because we included a variety of studies using different compounds in different animal models, any overall estimate of efficacy is of limited interest. Due to this anticipated heterogeneity, we performed a random effect meta-analysis. We used a raw difference in mean (RMD) analysis for the outcome parameters IS/AAR, IS/LV, EF and WT. For mortality, we used odds ratios30. Where individual studies reported outcome for several treatment groups (i.e. dose response studies), the size of the control group was adjusted to reflect this, as previously described28,30. To determine whether the observed results were confounded by publication bias, we used three approaches: visual inspection of funnel plots, Egger’s regression analysis for small study effects and Tweedie and Duval’s trim and fill31,32. Funnel plotting can be used to visually identify studies with small precision that overstate the effect of an intervention and are consistent with the presence of small study publication bias. Egger’s regression statistically assesses the presence of publication bias of small studies in a funnel plot by determining whether the regression line and its 95% CIs for precision versus standardized effect size intersect at the origin of the graph. Trim and fill analysis non-parametrically attempts to correct for funnel plot asymmetry by identifying and imputing theoretical missing studies. This enables recalculation of the overall treatment effect in the absence of publication bias30–32. We explored heterogeneity using meta-regression and tested which parameters were significantly associated with outcome. We planned to include 15 parameters as potential sources of heterogeneity that are relevant in either a clinical or translational perspective,

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

provided that we had at least 10 times that number of experimental comparisons30. We tested parameters in 3 separate categories: therapeutic characteristics (e.g. timing of therapy, drug group), model characteristics (e.g. species, sex, surgical approach) and risk of bias (e.g. reporting of blinding and randomization). Compounds were pooled in drug groups, based on working mechanism (supplemental table 1). We only performed metaregression for parameter values reported in at least 5 independent groups for comparison. Meta-regression is more conservative than stratification of heterogeneity. We set the statistical threshold for our predefined primary outcome measure at p<0.05, and for secondary outcome measures we applied a Bonferroni correction to give a critical value of p of 0.017. All analyses were performed using Stata version 11 (StataCorp LP, Texas, USA).

RESULTS

We found 2,530 results in PubMed and 2,778 in Embase. After merging and removal of duplicates, 4,105 unique publications remained. We excluded 3,570 studies in the first phase (title/abstract) of screening. In the next phase (full text screening), we included 183 publications in the database (Supplemental figure 1). These 183 publications reported 219 experimental comparisons for the primary outcome IS/AAR, reporting outcome from 3331 animals (1,839 treated and 1,492 control). Included studies predominantly used occlusion of the left anterior descending coronary artery (140 experiments) rather than the circumflex coronary artery (75 experiments); 4 used both. Most experiments used both male and female animals (both n=101, male n=60, female n=13, unknown n=45) and used dogs (dog n=163, pig n=54, sheep n=2). Random allocation of animals, blinded outcome assessment and blinding of the operator was reported in 143 (65%), 64 (29%) and 21 (10%) of the 219 comparisons respectively. For the secondary outcomes, 88 studies reported IS/ LV (138 experiments) and 97 reported mortality (143 experiments). LVEF was reported in 21 studies (31 experiments) and the analysis for WT included 11 studies (16 comparisons). Primary outcome For the primary outcome, treatment led to an absolute reduction in the infarct size as a percentage of the area at risk of 12.7% (95% CI 11.1% – 14.4%, p<0.001) (Figure 1A). There was substantial heterogeneity between studies (I2 80.9%, Tau2 104.5). Visual inspection of the funnel plot and Egger regression suggests a small study bias (Figure 1C&D). This was confirmed by Tweedie and Duval’s trim and fill, with 12 imputed missing studies and a corrected efficacy of 11.4% (95%CI 9.6% – 13.1%) (Figure 1C). Secondary outcomes Treatment with anti-inflammatory compounds was also associated with improved secondary outcome measurements. There was an absolute difference in mean IS/LV of 3.9% (95%CI 3.1% - 4.7%, p<0.0001) in favor of treated animals (Figure 1B). EF was higher in treated animals compared to control animals (RMD 3.4, 95% CI 0.8 – 6.1, p<0.001).

PART ONE CHAPTER 5

Figure 1.Timberplots and publication bias Timberplot of the effect of anti-inflammatory compounds on IS/AAR (A) and IS/LV (B). Dots and vertical lines represent the mean and 95% CI of individual studies. Horizontal gray bar represents the 95%CI of the mean effect size. C. Funnel plot of includes studies for the primary outcome IS/AAR. Red dots represent theoretically missing studies, identified by Tweedie and Duvalâ&#x20AC;&#x2122;s trim and fill. Vertical dotted line represents the mean effect size with (red) or without (black) correction. D. Eggerâ&#x20AC;&#x2122;s regression, showing the regression line and the 95% CI. IS/AAR=infarct size as percentage of area at risk, IS/LV=infarct size as percentage of the left ventricle.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

No difference in mortality was observed (OR 0.96, 95% CI 0.79 â&#x20AC;&#x201C;1.17, p= 1.0). Wall thickness as a percentage of the opposing wall was lower in treated animals, but of note, this outcome was predominantly assessed in studies testing the effect of non-steroid antiinflammatory drugs (NSAIDs) (11 out of 16) (WT RMD 16.3% 95%CI 8.5% â&#x20AC;&#x201C; 24.0%, p<0.001) (Supplemental figure 2). Only a small subset of studies reported LVEDV and LVESV and the extent of the dataset was too small to enable meta-analysis. Meta-regression We sought possible sources of heterogeneity using meta-regressions with effect on IS/ AAR, IS/LV and mortality as the dependent variable. The number of comparisons reporting EF and WT were insufficient for meta-regression. Drug class was a predictor of efficacy for IS/AAR and IS/LV endpoints (p<0.001, p=0.002) but not for mortality, although the point estimates for different drug classes were similar (Figure 2). For all 3 outcome measures, leukotriene inhibitors and cell adhesion molecule (CAM) inhibitors performed best, showing greater reduction in infarct size and lower mortality. The effect of anti-inflammatory compounds in improving IS/AAR and IS/LV was less apparent at later times of outcome assessment, both when measured in a categorized fashion (p=0.013 IS/AAR, p=0.001 IS/ LV) (Figure 3A and 3B) and on a continuous scale (p=0.016 IS/AAR, p=0.007 IS/LV). Furthermore, the sex of the animals used appeared to be important; for both IS/AAR and IS/LV, greater effects were seen in experiments using male animals than in those with female animals or a mixed population (IS/AAR p=0.002, IS/LV p=0.002) (Figure 4). Additionally, the effect on IS/LV was larger in temporary occlusion models compared to permanent occlusion models (p=0.014) (Figure 5A). Measures taken to preserve the internal validity of studies influenced reported outcomes. When outcome assessment was performed in a blinded fashion there was no significant difference in efficacy for IS/AAR (p=0.075). However, blinding of the operator performing the intervention was associated with increased mortality in the treatment group (p=0.041) (Figure 5B & C).

Figure 2. Meta-regression of drug group Effect size per drug group for the primary outcome IS/AAR (A), secondary outcomes IS/LV (B) and mortality (C). Dots and vertical lines represent the mean and 95% CI per drug group. IS/AAR=infarct size as percentage of area at risk, IS/LV=infarct size as percentage of the left ventricle. n=number of groups for comparison.

PART ONE CHAPTER 5

To assess if sources of heterogeneity could also be identified in specific drug groups, we performed meta-regression for sex and time of assessment for the primary outcome IS/ AAR for each of the 4 largest drug groups (NSAIDs (n=29), CAM inhibitors (n=27), Leukotriene inhibitors (n=22) and reactive oxygen species (ROS) scavengers (n=98) (see supplemental table 1 for classification). For time of assessment, results were similar to those found for the complete dataset was observed, but only time of assessment had sufficient power in the subgroup of ROS scavengers to reach statistical significance (p=0.015) (Supplemental figure 3).

Figure 3. Meta-regression of timing of therapy Effect size per timing interval after occlusion of the coronary artery, for the primary outcome IS/AAR (A) and secondary outcome IS/LV (B). IS/AAR=infarct size as percentage of area at risk, IS/LV=infarct size as percentage of the left ventricle. n=number of groups for comparison.

Figure 4. Meta-regression of sex Effect size per sex for the primary outcome IS/AAR (A) and secondary outcome IS/LV (B). IS/AAR=infarct size a percentage of area at risk, IS/LV=infarct size as percentage of the left ventricle. n=number of groups for comparison.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

DISCUSSION Efficacy of anti-inflammatory compounds Both clinical and preclinical studies suggest that the post-MI inflammatory response orchestrates cardiac wound healing and is accountable for infarct expansion 17,33. Our systematic review and meta-analysis of data from more than 180 large animal studies, including over 3,300 animals subjected to myocardial infarction, confirms that treatment with anti-inflammatory compounds reduces infarct size, thereby decreasing the risk of adverse remodeling4. Since inflammation is a complex multifactorial process, the pharmacological interventions used in different large animal studies are very heterogeneous. Our meta-regression suggests that inhibition of different inflammatory pathways has different effects on infarct size. For the first time, our analysis allows an assessment of the relative efficacy of these approaches in large animal MI models. Our findings suggest that treatment with for instance leukotriene inhibitors is more effective than NSAIDS or immunosuppressive drugs, and may therefore have greater prospects for clinical use. To the best of our knowledge, no clinical trials with compounds from the first group have been commenced in the setting of MI and therefore may be promising candidates to make their way into clinical application. Clinical Translation and preclinical study design Although comparison of the effect of different drug groups may be useful to select agents for future clinical testing, none of the compounds evaluated in the included experimental studies, that have also been tested in clinical trials, have been implemented in daily clinical practice34. This may imply critical discrepancies between the field of large animal studies of MI and clinical trials, leading to a presumable overestimation of the effect in large animal models. A possible explanation for this overestimation is publication bias. Previous reports have shown that the effect of compounds in animal research may be overestimated due to underreporting of small, neutral studies35,36. Our analysis also presents evidence of small study bias, and after correction for this bias, the overall effect of anti-inflammatory compounds is estimated to be 10% lower than the observed effect. Next, both internal and external validity of pre-clinical studies are important factors in translational medicine. Lack of blinding of both the operator and outcome assessment pose threats to internal validity37,38. Less than a third of the included studies in our systematic review reported blinding of the induction of ischemia or outcome assessment. Moreover, we observed an effect of investigator blinding; while there were no significant effects for infarct size reduction the best estimate of efficacy was lowest in high quality studies; and in studies reporting mortality, treatment was associated with significantly higher mortality in high quality studies. This underlines the importance of high methodological quality to ensure internal validity. In the perspective of external validity in large animal research, studies should resemble the clinical situation as closely as possible with regard to study design to allow for

PART ONE CHAPTER 5

extrapolation of study outcome to the clinical situation38,39. Some aspects are inherent to preclinical studies and cannot be easily altered (e.g. comorbidity, etiology of MI, anesthesia) and are thus an overall limitation of large animal MI models. However, other aspects (e.g. timing of assessment and timing of therapy) can resemble the clinical situation. Our study shows that these aspects influence outcome, since infarct size reduction in the included large animal studies is highest during the first hours post-MI and this effect is attenuated when the time of assessment is extended. Any effect limited to the first hours post-MI only, is of no clinical relevance and fails to be detected in clinical trials given their long term follow-up. Similarly, we observed sex to be a significant predictor of outcome. Since most studies used both male and female animals, translational failure is not necessarily attributable to the difference in effect size. However, the more pronounced effect in studies using male animals only, emphasizes the importance of external validity. Although the exact mechanism behind this phenomenon remains unidentified, differences between male and female subjects regarding outcome after MI have been observed in clinical and preclinical studies40â&#x20AC;&#x201C;43.

Figure 5. Meta-regression of infarction model and blinded outcome assessment Meta-regression of model of infarct creation (permanent versus temporary occlusion of the coronary artery) on the secondary outcome IS/LV (A). Meta-regression of the effect of blinded versus non-blinded outcome assessment on the primary outcome IS/AAR (B) and secondary outcome mortality (C). IS/AAR = infarct size as percentage of area at risk, IS/LV = infarct size as percentage of left ventricle. N = number of groups for comparison.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

To the best of our knowledge, this is the first study showing that attenuating the inflammatory response in male subjects is more efficacious compared to inhibiting this response in female subjects, a phenomenon that should be taken into account when designing large animal MI studies. Meta-analysis of translational studies Although our analysis shows significant effects of anti-inflammatory treatments on infarct size, it should be kept in mind that meta-analyses cannot reveal causal relationships. Hence, we do not intend to provide mechanistic insights with our study. Furthermore, the strength of a meta-analysis depends on the strength and quality of included studies. For example, reporting of several functional outcomes in pre-clinical models varies between studies, disallowing pooling of these data. Also, reporting of mortality is considered to be less rigorous compared to clinical studies, which may introduce bias into the dataset. In contrast to meta-analyses of clinical studies, heterogeneity in pre-clinical studies can be of great additional value, since parameters responsible for heterogeneity can be used as predictors for success and can help design future clinical and pre-clinical studies. The extent of our dataset was insufficient for a multivariate meta-regression analysis, therefore we cannot completely rule out co-linearity of the different sources of heterogeneity (supplemental table 2). Conclusion Our systematic review and meta-analysis of large animal MI studies show reduced infarct size in animals treated with anti-inflammatory compounds compared to control animals. This study reveals differences between the inhibitory effect of different inflammatory pathways and leukotriene inhibitors seem to be the most promising candidate to make their way to clinical application. Moreover, we stress the use of clinically relevant animal models and study designs to increase the translational value of pre-clinical research. Funding This work is partially funded by the research program of the BioMedical Materials institute, which is co-funded by the Dutch Ministry of Economic Affairs. MRM and ESS acknowledge support from the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) infrastructure award: ivSyRMAF, the CAMARADESâ&#x20AC;&#x201C;NC3Rs in vivo systematic review and meta-analysis facility. KEW was partially funded by The Netherlands Organization for Health Research and Development (ZonMW, project #104024065).

PART ONE CHAPTER 5

REFERENCES 1. Gupta A, Wang Y, Spertus J a., Geda M, Lorenze N, Nkonde-Price C, D’Onofrio G, Lichtman JH, Krumholz HM. Trends in acute myocardial infarction in young patients and differences by sex and race, 2001 to 2010. J Am Coll Cardiol. 2014;64:337–345. 2. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit J a., Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart Disease and Stroke Statistics - 2014 Update: A report from the American Heart Association. 2014. 3. Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SMJM, Hofman A, Deckers JW, Witteman JCM, Stricker BHC. Quantifying the heart failure epidemic: Prevalence, incidence rate, lifetime risk and prognosis of heart failure - The Rotterdam Study. Eur Heart J. 2004;25:1614–1619. 4. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988. 5. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 6. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 7. Christia P, Frangogiannis NG. Targeting inflammatory pathways in myocardial infarction. Eur J Clin Invest. 2013;43:986–995. 8. Timmers L, Sluijter JPG, Verlaan CWJ, Steendijk P, Cramer MJ, Emons M, Strijder C, Gründeman PF, Sze SK, Hua L, Piek JJ, Borst C, Pasterkamp G, De Kleijn DP V. Cyclooxygenase-2 inhibition increases mortality, enhances left ventricular remodeling, and impairs systolic function after myocardial infarction in the pig. Circulation. 2007;115:326–332. 9. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 10. Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res. 2002;53:31–47. 11. Arslan F, Smeets MB, O’Neill L a J, Keogh B, McGuirk P, Timmers L, Tersteeg C, Hoefer IE, Doevendans P a., Pasterkamp G, De Kleijn DP V. Myocardial ischemia/reperfusion injury is mediated by leukocytic toll-like receptor-2 and reduced by systemic administration of a novel anti-toll-like receptor-2 antibody. Circulation. 2010;121:80–90.

12. Van Der Laan AM, Hirsch A, Robbers LFHJ, Nijveldt R, Lommerse I, Delewi R, Van Der Vleuten P a., Biemond BJ, Zwaginga JJ, Van Der Giessen WJ, Zijlstra F, Van Rossum AC, Voermans C, Van Der Schoot CE, Piek JJ. A proinflammatory monocyte response is associated with myocardial injury and impaired functional outcome in patients with STsegment elevation myocardial infarction: Monocytes and myocardial infarction. Am Heart J. 2012;163:57–65.e2. 13. Núñez J, Núñez E, Bodí V, Sanchis J, Miñana G, Mainar L, Santas E, Merlos P, Rumiz E, Darmofal H, Heatta AM, Llàcer A. Usefulness of the neutrophil to lymphocyte ratio in predicting long-term mortality in ST segment elevation myocardial infarction. Am J Cardiol. 2008;101:747–752. 14. Blum a, Sclarovsky S, Rehavia E, Shohat B. Levels of T-lymphocyte subpopulations, interleukin-1 beta, and soluble interleukin-2 receptor in acute myocardial infarction. Am Heart J. 1994;127:1226– 1230. 15. Takahashi T, Hiasa Y, Ohara Y, Miyazaki S-I, Ogura R, Suzuki N, Hosokawa S, Kishi K, Ohtani R. Relationship of admission neutrophil count to microvascular injury, left ventricular dilation, and long-term outcome in patients treated with primary angioplasty for acute myocardial infarction. Circ J. 2008;72:867–872. 16. Maekawa Y, Anzai T, Yoshikawa T, Asakura Y, Takahashi T, Ishikawa S, Mitamura H, Ogawa S. Prognostic significance of peripheral monocytosis after reperfused acute myocardial infarction:a possible role for left ventricular remodeling. J Am Coll Cardiol. 2002;39:241–246. 17. Van der Laan a. M, Nahrendorf M, Piek JJ. Healing and adverse remodelling after acute myocardial infarction: role of the cellular immune response. Heart. 2012;98:1384–1390. 18. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 19. Hausenloy DJ, Baxter G, Bell R, Bøtker HE, Davidson SM, Downey J, Heusch G, Kitakaze M, Lecour S, Mentzer R, Mocanu MM, Ovize M, Schulz R, Shannon R, Walker M, Walkinshaw G, Yellon DM. Translating novel strategies for cardioprotection: The Hatter Workshop Recommendations. Basic Res Cardiol. 2010;105:677–686. 20. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker H V, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, López CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West M a, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE,

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier R V, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A. 2013;110:3507– 12. 21. Skyschally A, Van Caster P, Boengler K, Gres P, Musiolik J, Schilawa D, Schulz R, Heusch G. Ischemic postconditioning in pigs: No causal role for risk activation. Circ Res. 2009;104:15–18. 22. Timmers L, Henriques JPSPS, de Kleijn DPVP V, Devries JHH, Kemperman H, Steendijk P, Verlaan CWJWJ, Kerver M, Piek JJJ, Doevendans P a a., Pasterkamp G, Hoefer IEE. Exenatide Reduces Infarct Size and Improves Cardiac Function in a Porcine Model of Ischemia and Reperfusion Injury. J Am Coll Cardiol. 2009;53:501–510. 23. Houtgraaf JH, De Jong R, Kazemi K, De Groot D, Van Der Spoel TIG, Arslan F, Hoefer I, Pasterkamp G, Itescu S, Zijlstra F, Geleijnse ML, Serruys PW, Duckers HJ. Intracoronary infusion of allogeneic mesenchymal precursor cells directly after experimental acute myocardial infarction reduces infarct size, abrogates adverse remodeling, and improves cardiac function. Circ Res. 2013;113:153– 166. 24. Arslan F, Houtgraaf JH, Keogh B, Kazemi K, De Jong R, McCormack WJ, O’Neill L a J, McGuirk P, Timmers L, Smeets MB, Akeroyd L, Reilly M, Pasterkamp G, De Kleijn DP V. Treatment with OPN-305, a humanized anti-toll-like receptor-2 antibody, reduces myocardial ischemia/reperfusion injury in pigs. Circ Cardiovasc Interv. 2012;5:279– 287. 25. Libby P, Maroko PR, Bloor CM, Sobel BE, Braunwald E. Reduction of experimental myocardial infarct size by corticosteroid administration. J Clin Invest. 1973;52:599–607. 26. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 27. Seropian IM, Toldo S, Van Tassell BW, Abbate A. Anti-inflammatory strategies for ventricular remodeling following St-segment elevation acute myocardial infarction. J Am Coll Cardiol. 2014;63:1593–1603. 28. Jansen of Lorkeers SJ, Eding JEC, van der Spoel TIG, Vesterinen HM, Sena ES, Doevendans PA, Macleod MR, Chamuleau SAJ, van der Spoel TIG, Sena ES, Duckers HJ, Doevendans PA, Macleod MR, Chamuleau SAJ. Similar Effect of Autologous and Allogeneic Cell Therapy for Ischemic Heart Disease: Results From a Meta-Analysis of Large Animal Studies. J Am Coll Cardiol. 2014;63:A1762. 29. Van Hout GPJ, Jansen of Lorkeers SJ, Wever KE, Sena ES, van Solinge WW, Doevendans P a., Pasterkamp G, Chamuleau S a. J, Hoefer IE. Antiinflammatory compounds to reduce infarct size in large-animal models of myocardial infarction: A meta-analysis. Evidence-based Preclin Med. 2014;1:4–10.

30. Vesterinen HM, Sena ES, Egan KJ, Hirst TC, Churolov L, Currie GL, Antonic a., Howells DW, Macleod MR. Meta-analysis of data from animal studies: A practical guide. J Neurosci Methods. 2014;221:92–102. 31. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629–634. 32. Duval S, Tweedie R. Trim and fill: A simple funnelplot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56:455–463. 33. Lugrin J, Parapanov R, Rosenblatt-Velin N, RignaultClerc S, Feihl F, Waeber B, Muller O, Vergely C, Zeller M, Tardivel A, Schneider P, Pacher P, Liaudet L, Müller O, Vergely C, Zeller M, Tardivel A, Schneider P, Pacher P, Liaudet L. Cutting Edge: IL1 Is a Crucial Danger Signal Triggering Acute Myocardial Inflammation during Myocardial Infarction. J Immunol. 2014;194:499–503. 34. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, De Lemos J a., Ettinger SM, Fang JC, Fesmire FM, Franklin B a., Granger CB, Krumholz HM, Linderbaum J a., Morrow D a., Newby LK, Ornato JP, Ou N, Radford MJ, Tamis-Holland JE, Tommaso CL, Tracy CM, Woo YJ, Zhao DX. 2013 ACCF/AHA guideline for the management of stelevation myocardial infarction: A report of the American college of cardiology foundation/ american heart association task force on practice guidelines. J Am Coll Cardiol. 2013;61:e78–140. 35. MacLeod MR, Van Der Worp HB, Sena ES, Howells DW, Dirnagl U, Donnan G a. Evidence for the efficacy of NXY-059 in experimental focal cerebral ischaemia is confounded by study quality. Stroke. 2008;39:2824–2829. 36. Sena ES, Bart van der Worp H, Bath PMW, Howells DW, Macleod MR. Publication bias in reports of animal stroke studies leads to major overstatement of efficacy. PLoS Biol. 2010;8:e1000344. 37. Macleod MR, Fisher M, O’Collins V, Sena ES, Dirnagl U, Bath PMW, Buchan a., van der Worp HB, Traystman R, Minematsu K, Donnan G a., Howells DW. Good Laboratory Practice: Preventing Introduction of Bias at the Bench. Stroke. 2008;40:e50–e52. 38. Platt ML, Hayden B. Learning : Not Just the Facts , Ma ’ am , but the Counterfactuals as Well. PLoS Biol. 2011;9:1–2. 39. Heide RS Vander, Steenbergen C. Cardioprotection and myocardial reperfusion: Pitfalls to clinical application. Circ Res. 2013;113:464–477. 40. Dong M, Mu N, Ren F, Sun X, Li F, Zhang C, Yang J. Prospective Study of Effects of Endogenous Estrogens on Myocardial No-Reflow Risk in Postmenopausal Women with Acute Myocardial Infarction. J Interv Cardiol. 2014;27:437–443. 41. Zhou C, Yao Y, Zheng Z, Gong J, Wang W, Hu S, Li L. Stenting technique, gender, and age are associated with cardioprotection by ischaemic

PART ONE CHAPTER 5

postconditioning in primary coronary intervention: A systematic review of 10 randomized trials. Eur Heart J. 2012;33:3070–3077. 42. Lawesson SS, Alfredsson J, Fredrikson M, Swahn E. Time trends in STEMI--improved treatment and outcome but still a gender gap: a prospective observational cohort study from the SWEDEHEART register. BMJ Open. 2012;2:e000726–e000726. 43. Lagranha CJ, Deschamps A, Aponte A, Steenbergen C, Murphy E. Sex differences in the phosphorylation of mitochondrial proteins result in reduced production of reactive oxygen species and cardioprotection in females. Circ Res. 2010;106:1681–1691.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

SUPPLEMENTARY MATERIAL

Supplemental figure 1. Flow chart of inclusion/exclusion

PART ONE CHAPTER 5

Supplemental figure 2. Timberplot of the effect of anti-inflammatory compounds (A) Ejection fraction, (B) mortality , (C) scar thinning. Vertical lines representing 95% CI of individual studies. Horizontal line and gray bar represents the mean and 95%CI of the mean effect size. EF = Left ventricular ejection fraction.

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

Supplemental figure 3. Meta regression of timing of outcome assessment within the ROS scavengers Dots and vertical bars represent the mean effect size and 95% CI per timing interval on IS/AAR. Horizontal line and gray bar represent the overall mean and 95% CI. IS/AAR = infarct size as a percentage of the area at risk. N = number of groups for comparison.

Anti-CD18

Anti-ICAM-1 MAb

Anti-Mo1 (clone 904)

Anti-P-selectin Ab + Anti-ICAM-1 MAb

Anti-P-selectin Antibody

Bb15-42

CI-959

CY 1503

MAb to ICAM-1

MHM.23

Neutrophil Antibody

NPC 15669

PLM-2

R15.7

rPSGL-Ig

Sialyl Lewis x-OS

2-octadecylascorbic acid

Allopurinol

Allopurinol + SOD

alpha-Tocopherol

Ascorbic Acid

Ascorbic Acid + Desferrioxamine

Ascorbic Acid + Desferrioxamine + N-Acetylcysteine

Bucillamine

Catalase

Deferoxamine

Desferrioxamine

Dimethylsulfoxide

Dimethylthiourea

H290/51

Hydrogen Sulfide

Methionine

Vitamin E + Vitamin C

Vitamin E

U7006F

Trolox + Ascorbic Acid

Trolox

SOD + Catalase

SOD

SC-52608

Polyethylene glycol-conjugated SOD

PEG-SOD + Catalase

Oxypurinol

N-tert-butyl-alpha-phenylnitrone

N-Acetylcysteine

N-2-Mercaptopropionyl Glycine

CAM inhibitor

ROS Scavenger

Supplemental Table 1. Classification of drugs

TG100-115

RP 59227 (Tulopafant)

Rolipram

Poloxamer 188

OPN-305

o-Desulfated Heparin

L-659,989

Colchicine

Chymase INH (TY51469)

Chemically Modified Tetracycline-3

Adiponectin

Acetaminophen

other

REV-5901

ONO-1078

Nafazatrom

LY255283

LY233569

Glutathione

Ebselen

CI-922

BW755C

AA-861

Leukotriene inhibitor

Tyrosine sulfate

N-Acetylheparin

Mirococept

C5a receptor antagonist ADC-1004

C1-INH

Anti-C5a MAb

Complement inhibitor

Naproxen

Indomethacin

Ibuprofen

Flurbiprofen

Celecoxib

Azapropazone

NSAID

Methylprednisolone

Etanercept

Dexamethasone

Cyclosporin A

Immunosuppressive Aspirin

Aspirin

PART ONE CHAPTER 5

ANTI-INFLAMMATORY COMPOUNDS IN MI MODELS

Supplemental Table 2. Crosstabs of the number of groups for comparison for the 3 significant predictors on outcome (Sex, drug group and timing of outcome assessment) A.

Druggroup

Sex Both (n=101)

Female (n=13)

Male (n=60)

not reported (n=45)

Aspirin (n=7)

Complement Inhibitor (n=9)

ROS Scavenger (n=98)

Immunosuppressive (n=8)

CAM Inhibitor (n=27)

Leukotriene inhibitor (n=22)

NSAID (n=29)

Other (n=19)

Druggroup

Timing of assessment 0-4h (n=32)

4-6h (n=66)

6-8h (n=37)

19-50h (n=49)

>50h (n=35)

Aspirin (n=7)

Complement Inhibitor (n=9)

ROS Scavenger (n=98)

Immunosuppressive (n=8)

CAM Inhibitor (n=27)

Leukotriene inhibitor (n=22)

NSAID (n=29)

Other (n=19)

Sex

Timing of assessment 0-4h (n=32)

4-6h (n=66)

6-8h (n=37)

19-50h (n=49)

>50h (n=35)

Both (n=101)

Female (n=13)

Male (n=60)

Not reported (n=45)

part two

VALIDATION OF MEASUREMENTS AND MARKERS OF CARDIAC FUNCTION

chapter 6

Admittance based pressure volume loop measurements in a porcine model of chronic myocardial infarction

Experimental Physiology 2013 November;98(11):1565-75

G.P.J. van Hout1*, R. de Jong2*, J.E.P. Vrijenhoek1, L. Timmers1, H.J. Duckers2, I.E. Hoefer1 *Contributed equally to this work 1

Experimental Cardiology Laboratory University Medical Center Utrecht, Utrecht, The Netherlands

Molecular Cardiology Laboratory Erasmus University Medical Center, Rotterdam, The Netherlands

PART TWO CHAPTER 6

ABSTRACT Objective The aim of this study was to validate admittance based pressure volume (PV) loop measurements for the assessment of cardiac function in a porcine model of chronic myocardial infarction. Background The traditional PV-loop measurement technique requires hypertonic saline injections for parallel conductance correction prior to signal conversion into volume. Furthermore, it assumes a linear relationship between conductance and volume. More recently, an admittance-based technique has been developed, which continuously measures parallel conductance and uses a non-linear equation for volume calculation. This technique has not yet been evaluated in a large animal myocardial ischemia model. Methods Eleven pigs underwent invasive PV measurements with the admittance system (AS) and the traditional conductance system (CS) followed by 3D-echocardiography (3DE). After baseline measurements, pigs were subjected to 90 minute left anterior descending artery occlusion followed by the same measurements at 8 weeks follow-up. Results In the healthy heart, the AS showed good agreement with 3DE for LV volumes and a reasonable correlation for ejection fraction (EF) (R=0.756, p=0.007). At follow-up, an increase in end systolic volume (ESV) was observed with 3DE (+15.4±14.4mL, p=0.005) and the AS (+34.6±36.1mL, p=0.010). EF measured with 3DE (-13.2±5.2%, p<0.001) and the AS (-20.3±11.2%, p<0.001) significantly decreased. Conclusion The AS can be used to quantitatively monitor the cardiac function changes induced by myocardial infarction and provides comparable results as 3DE, rendering it a useful tool for functional testing in large animal cardiac models. New Findings • Are admittance based pressure-volume measurements useful for the assessment of cardiac function in a chronic myocardial infarction model compared to three-dimensional echocardiography and to classical conductance measurements? • Baseline admittance based measurements show good agreement with echocardiography and classical conductance in human sized porcine hearts. Myocardial infarction significantly alters ventricular dimensions and PV-loop measurements compared to 3D-Echocardiography. The novel admittance system reliably monitors cardiac function changes after myocardial infarction.

100

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

INTRODUCTION Functional cardiac parameters are considered the gold standard to evaluate therapeutic efficacy in experimental and clinical cardiovascular studies. Hence, means for reliable cardiac function measurement are essential to assess efficacy of compounds that reduce myocardial ischemia/reperfusion injury or modulate post-infarct remodeling. Several techniques are available for left ventricular (LV) function assessment, predominantly using LV volumes as a surrogate of cardiac function. This includes Magnetic Resonance Imaging (MRI) and echocardiography. MRI is considered to be the most precise technique to measure LV volumes, but it remains expensive, time consuming and does not allow intraventricular pressure measurements for real time pressure-volume relation calculations. The latter also holds true for echocardiography, which provides only limited information on fundamental aspects of ventricular contraction and relies on the use of geometric assumptions1–3. Invasive Pressure Volume (PV) measurements (PV-loops) combine real time pressure and ventricular volume measurements to determine pressure volume relationships. This provides researchers with more detailed information about systolic and diastolic function that is unique for cardiac assessment with PV studies4–6. To accurately determine these myocardial characteristics, a precise estimation of LV volumes is needed. The classical PV systems determine LV volumes by converting measured conductance signals to volume after correcting for parallel conductance by hypertonic saline injection7–9. Recently, a new technique has been designed to render hypertonic saline calibration unnecessary4,10–12. This admittance system (AS) continuously measures the phase angle to determine myocardial conductance during the cardiac cycle. Since myocardial (parallel) conductance is continuously measured, the effects of ventricular wall movement on the relative contribution of myocardial conductance to total (measured) conductance can be taken into account. Traditional systems do not allow correction for wall movement since early studies suggested that parallel conductance does not vary throughout the cardiac cycle8. However, increasing evidence emerges that the attribution of parallel conductance does vary, especially in the diseased heart. This theoretically leads to overestimating or underestimating conductance during diastole or systole respectively, affecting ventricular volume calculation5,13. Another key difference between the classical conductance and the novel admittance technique is the conversion of the measured signals into volumes. While traditional systems assume a constant, linear relationship between conductance and volume using Baan’s equation7, the AS is based on a continuous nonlinear relationship, incorporated into Wei’s equation14,15. Though the AS was validated in small animal models12,15, data from large animal studies are scarce. Recently, the admittance technique was tested in the healthy porcine heart, where it showed good agreement with transesophageal three-dimensional echocardiography16. However, the animals used in that study were only half the weight of average humans (34.4kg) and accordingly had smaller LV volumes compared to humans3. Moreover, future use in experimental and clinical studies requires thorough testing of the

101

PART TWO CHAPTER 6

system’s ability to discriminate between normal and diminished cardiac function. In the current study, we investigated this by comparing admittance based PV-loops with a classical conductance system (CS) results and 3-dimensional echocardiography (3DE) as a reference standard. The latter is a well-established technique for the assessment of LV volumes that has recently been proven to correlate very well with MRI and thermodilution, the gold standards for LV volume assessment1–3,17,18.

MATERIAL AND METHODS All animal experiments were approved by the institutional animal welfare committee and were executed conforming to the ‘Guide for the Care and Use of Laboratory Animals’. A total of 11 female landrace pigs were used in this study. Pigs (body weight 73.1±5.3 kg) were subjected to left ventricular invasive measurements with both the CS and AS, followed by epicardial 3DE in the healthy heart. After these measurements, pigs were subjected to myocardial infarction followed by invasive PV measurements and 3DE 8 weeks later. Pressure volume loop measurements in the naïve myocardium Animals were anesthetized with 10 mg/kg ketamine, 0.4 mg/kg midazolam and 0.5 mg/ kg atropine. Anesthesia was maintained with 0.5 mg/kg/h midazolam, 2.5 µg/kg/h sufentanyl and 0.1 mg/kg/h pancuronium. Pre- and post-operatively, animals received a fentanyl patch (25µ/h) and post-operatively a single injection 2mg/kg meloxacam. Venous and arterial access was obtained by placement of a 7F sheath in the jugular vein and an 8F sheath in the carotid artery after the blood vessels were surgically exposed. The thorax was opened via medial sternotomy and a snare around the inferior vena cava for preload reduction was placed. A conductance catheter was inserted into the left ventricle through the sheath in the carotid artery (CD Leycom, Zoetermeer, the Netherlands). A 6F catheter was placed in the pulmonary artery through the sheath in the jugular vein for hypertonic saline injections. After data acquisition, the PV catheter was removed and the admittance catheter was inserted (Transonic Scisense, London, Canada). After removal of the admittance catheter, epicardial 3DE was performed. All catheters were placed under fluoroscopic guidance. Infarct induction and PV measurements in the diseased heart Directly following baseline measurements, pigs were subjected to myocardial infarction (MI). The heart was exposed by opening the pericardium. This was followed by placement of a ligature around the left anterior descending (LAD) artery just distal from the origin of the first diagonal artery. The LAD was subsequently closed for 90 minutes. After reopening the vessel and observation for approximately 3 hours, the sternum was closed and the animals were weaned from anesthesia. If any cardiac arrhythmias occurred during the procedure, animals were epicardially defibrillated with 5-20 Joules. Eight weeks after infarct induction, pigs were again anesthetized and measurements were performed as described

102

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

above after lateral sternotomy due to the presence of adhesions of the heart to the sternum after primary surgery. After these measurements pigs were sacrificed by rapid bleeding and the heart was excised for the determination of infarct size. Conductance Technique A 7F conductance catheter (CA-71103-PL, CD Leycom, Zoetermeer, The Netherlands) was connected to the Sigma acquisition system (CD Leycom, Zoetermeer, The Netherlands). After ex vivo calibration, the catheter was placed in the left ventricle via a sheath in the carotid artery under fluoroscopic guidance. On average, 5 conductance segments were positioned in the left ventricle depending on the size of the heart. After insertion, baseline recordings were obtained during apnea at a rate of 250 samples/second. To correctly determine absolute LV volumes, 5 mL of 10%NaCL was directly injected into the pulmonary artery. Injections were repeated three times per animal each during simultaneous beat recording and apnea. Cardiac output required for data analyses was derived from the stroke volume (SV) measured by 3DE. Blood resistivity was assumed to be constant in all animals (150 立*cm). Subsequently, preload was reduced by temporary inferior caval vein occlusion during apnea. End systolic pressure volume relations (ESPVR) were derived from the recorded beats during vena cava occlusion that were performed three times per animal. All recordings were analyzed offline using Conductance NT 16 Software (CD Leycom, Zoetermeer, The Netherlands).

140

EDV: 106.1ml

120

100

ESV: 40.0ml

EF: 62.3%

EDV: 114.7ml 20

ESV: 70.6ml

EF: 38.5%

100

120

140

Figure 1. Methods of LV volume assessment LV volumes were determined by 3 different modalities. A. End diastolic and end systolic LV volume assessment with subsequent calculations of EF at baseline (upper half) and at follow-up (lower half). B. Admittance based pressure volume loops at baseline (red) with preload reduction in the healthy heart (green) and after myocardial infarction (blue). The lines depicted represent the ESPVR. Note that the slope of the ESPVR at follow-up compared to baseline hardly changes while a shift of V0 can be observed. C. Conductance based pressure volume loops with preload reduction in the healthy heart (green) and after myocardial infarction (blue).

103

PART TWO CHAPTER 6

Admittance technique For the admittance-based technique, a 7F tetra-polar catheter (7.0 VSL Pigtail/no lumen, Transonic Scisense, London, Canada) was used that measures admittance magnitude and phase in combination with pressure. It contains 7 platinum electrodes dividing it into 4 selectable segments. The largest segment inside the LV was used for absolute volume assessment. The catheter was connected to the ADVantage system TM (Transonic SciSense, London, Canada) linked to a multi-channel acquisition system (Iworx 404), required for real-time data acquisition. Again, the catheter was inserted via the sheath in the carotid artery under fluoroscopic guidance. After insertion, the admittance catheter measures blood and parallel conductance separating both based on phase angle with a rate of 200 samples/second. A baseline scan was performed to determine the end diastolic and end systolic blood conductance required for absolute volume calculations. The external stroke volume (SV) that is required for analyses was derived from echocardiographic measurements. Blood resistivity was assumed to be constant in all animals (150 Ω*cm). Baseline PV measurements and caval occlusions were performed under apnea as described above. Data were analyzed offline using Iworx analysis software (Labscribe V2.0). Three-dimensional echocardiography 3DE was performed in all animals using a Philips iE33 machine with an X3-1 transducer (Philips, Eindhoven, The Netherlands). The 3D-transducer (X-3, Philips, The Netherlands) was wrapped in a sterile sleeve. A pocket of gel was positioned under the transducer, to bring the complete apex in view. The transducer, together with the gel pocket, was positioned directly epicardially on the apex of the heart. The depth and sector size were adjusted to fit the complete ventricle. All data sets were acquired in real time using 7 consecutive cardiac cycles (full volume analysis). The images were analyzed offline using QLab 10.1 (3DQ advanced) analysis software. The tracing of the ventricle was performed by semi-automatic border detection as described before19. Briefly, end-diastolic and end-systolic frames are identified and on both frames the apex, anterior, lateral, inferior and septal mitral annulus are identified. The endocardial border is automatically traced. Ejection fraction is calculated by the Qlab software as (EDVESV)/ EDV*100 (figure 1a). Table 1. Cardiac parameters measured in vivo in the healthy porcine heart using the AS and CS Cardiac Parameters

AS Baseline

CS Baseline

AS Follow-up

CS Follow-up

Heart Rate (bpm)

58±15

62±15

63±12

61±10

ESP (mmHg)

117.8±21.5

110.8±23.1

92.9±23.2

89.1±18.0

EDP (mmHg)

10.5±3.2

11.7±6.5

8.6±3.9

10.4±4.5

ESPVR (slope)

2.56±2.43

2.34±1.27

3.63±4.87

3.05±2.50

-21±20

-21±61

34±52*

-14±23

All values are presented as means ± standard deviations. ESPVR – End Systolic Pressure Volume Relationship. * Significant difference vs. baseline AS. Differences with p-value <0.05 were regarded significant.

104

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

Table 2. Two-way repeated measures ANOVA between method used, time point of actual measurement and interaction between time point and measurement Source

EDV

ESV

Method

F(2,20)=6.9 p=0.005

F(2,20)=8.8 p=0.002

F(2,20)=9.9 p=0.001

Time

F(1,10)=0.7 p=0.063

F(1,10)=5.9 p=0.036

F(1,10)=23.9 p=0.001

Method*Time

F(2,20)=3.9 p=0.036

F(2,20)=6.4 p=0.007

F(2,20)=3.2 p=0.062

Differences between EDV, ESV and EF for the method used (3DE, AS or CS) the timepoint measured (baseline or follow-up) and the interaction between time point and method. *p-values <0.05 were regarded significant. Data were analyzed with a two-factor repeated measures ANOVA. Corrected EF = differences in EF corrected for infarct size.

Infarct size After the animals were sacrificed, the hearts were excised. The LV was then cut into 5 equal slices from apex to base and incubated in 1% triphenyltetrazolium chloride (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37°C 0.9%NaCL for 15 min to discriminate infarct tissue from viable myocardium in 10/11 animals. Statistical Analysis All data are expressed as mean ± standard deviation unless mentioned otherwise. Both, echocardiographic data and PV measurements were separately analyzed by two different researchers blinded to the outcome of the other technique. PV-Loop specific data (e.g. ESPVR, ESP, HR) were compared using a paired Student’s t-test after log transformation in SPSS 20.0. End diastolic volume (EDV), end systolic volume (ESV) and ejection fraction (EF) at baseline and follow-up measured by the three methods were compared using a twofactor repeated analysis of variance (ANOVA). Baseline LV volumes for the AS and CS were separately compared with 3DE measurements using a paired Student’s t-test. For follow-up measurements, a paired t-test was performed including infarct size as a covariate for EF. Follow-up and baseline measurements of the same system were compared using a paired Student’s t-test for EDV, ESV and EF. Delta EDV, ESV and EF between baseline and followup were compared among the different systems using a repeated measures ANOVA. Correlations at baseline and follow-up were tested using Pearson’s correlation test. The limits of agreement (1.96*SD = 95% confidence interval) of both PV systems compared to 3DE were determined by Bland-Altman analysis. A two-tailed F test was used to compare the size of the limits of agreement of the AS and the CS. All statistical analyses were performed in SPSS statistics version 20.0. A two-sided P-value of <0.05 was regarded statistically significant in all analyses.

105

PART TWO CHAPTER 6

RESULTS LV volumes at baseline and follow-up were measured in 11 pigs with good quality of measurements (figure 1). There were no significant differences in heart rate and pressures between measurements, indicating that measurements were performed during steady state (table 1). At follow-up, the AS’ ESPVR slope showed a non-significant increase, accompanied by a significant rightward shift of V0 (+55±52, p=0.013) (figure 1b, table 1). After absolute volume calibration, EDV, ESV and EF were measured by the two systems. Both, baseline and follow-up data were compared with 3D-echocardiographic findings by two-factor repeated measures ANOVA. There was a significant interaction between method (3DE, AS, CS) and time point (baseline, follow-up) for EDV (p=0.036) and ESV (p=0.007) and a trend towards an interaction for EF (p=0.062) indicating that cardiac volume assessment across the two time points differed for the three methods (table 2). In the naïve myocardium, none of these parameters differed between admittance, conductance and 3DE (supplemental table 1, figure 2, figure 3). Moreover, EF derived from the AS and 3DE correlated well at baseline (R=0.756, p=0.007) (table 3). However, at 8 weeks follow-up after myocardial infarction, significant differences were observed. The AS overestimated both, EDV (+28.7±41.4mL, p=0.028) and ESV (+31.3±35.4mL, p=0.014) compared to 3DE. In turn, this overestimation resulted in a significant underestimation of EF (-10.9±10.2%, p=0.009) (supplemental table 2, figure 2 and 3).

Figure 2. Left ventricular volume assessment in the healthy and infarcted porcine heart Volumes were measured by 3D-echocardiography and invasive PV-measurements with the admittance and the conductance technique in the healthy and infarcted porcine heart (n=11). A. Baseline End Diastolic Volume B. Baseline End Systolic Volume C. Baseline Ejection Fraction D. End Diastolic Volume at follow-up E. End Systolic Volume at follow-up F. Ejection Fraction at follow-up. Data are presented as mean ± 95%CI. * p<0.05; **p>0.05 after correction for infarct size.

106

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

Table 3. Correlation between 3DE, the AS and CS in the healthy porcine heart for EDV, ESV and EF LV Parameters

3DE AS

3DE CS

AS CS

EDV (mL)

0.339

0.539

0.517

ESV (mL)

0.361

0.341

0.486

EF (%)

0.756*

0.309

0.300

Numbers shown are the correlation coefficients (R). *Significant correlation between two systems (p=0.007).

For the CS, the opposite was found. Both, EDV (-11.3±23.4, p=0.021) and ESV (-15.4±14.7, p=0.018) were significantly underestimated using this technique (supplemental Table 2, figure 2 and 3). This culminated in an overestimation of EF (+10.2±13.4, p=0.037) compared to 3DE. Furthermore, infarct size correlated significantly with EF measured by 3DE at follow-up (R=-0.698, p=0.025, figure 4). Therefore, infarct size was added as a covariate in the analysis when EF at follow-up was compared among the different methods. After controlling for infarct size, the differences for EF between the AS and 3DE remained, whereas for the CS the difference in EF compared to 3DE was no longer significant (supplemental table 3). To determine the general agreement of the two systems with 3DE as a reference standard, both in the healthy heart and after 8 weeks of follow-up, Bland-Altman analyses were performed (figure 3). Next, the limits of agreement of the different PV systems versus 3DE were compared to determine differences in variance. The limits of agreement of baseline PV-loop measurements of both systems did not differ significantly. In the post infarction remodeled heart, however, the admittance system versus 3DE showed larger limits of agreement for both EDV (±37.2, p<0.05) and ESV (±34.8, p<0.05) when compared with the limits of agreement of the conductance system versus 3DE for EDV (±13.8) and ESV (±18.1) (figure 3). Cardiac function: Baseline versus follow-up Any particular system for cardiac function assessment can be considered valid when a true change in cardiac function can be detected reliably. Therefore, the primary aim of this study was to compare baseline and follow-up measurements with the AS. To monitor the degree of cardiac function changes, we used 3DE as reference. 3D-echocardiographic LV volume measurements confirmed a decline in cardiac function after MI with a significant increase in ESV (+15.4±14.4 mL, p=0.005) and a decrease in EF (-13.2±5.2%, p<0.001) (table 5), but did not reveal any significant cardiac dilatation after MI. The AS detected a significant decrease in ESV (+34.6±36.1mL, p=0.010) and EF (-20.3±11.2%, p<0.001). Similar to the echocardiographic measurements, the AS indicated no significant EDV changes after MI. Finally, delta EDV, ESV and EF were compared among the different systems. Only delta ESV (+34.6±36.0 mL vs. -3.2±32.9 mL, p=0.030) between the AS and CS significantly differed, whereas no significant differences between 3DE and the two PV techniques were observed (table 5).

107

PART TWO CHAPTER 6

Figure 3. Bland-Altman analysis of left ventricular volume assessment in the healthy and infarcted porcine heart Comparison of 3D-echocardiography (set as reference standard) and invasive PV measurements with both the admittance and the conductance technique (n=11) in the healthy and infarcted porcine heart. A. Baseline End Diastolic Volume B. Baseline End Systolic Volume C. Baseline Ejection Fraction. D. End Diastolic Volume at follow-up E. End Systolic Volume at follow-up F. Ejection Fraction at follow-up. Data are presented as mean Âą 95%CI. + = significant overestimation vs. 3D-echo; - = significant underestimation vs. 3D-echo before correction for infarct size; # = significantly larger limits of agreement versus conductance technique, p<0.05.

DISCUSSION To evaluate therapeutic efficacy of treatments for ischemic heart disease, a careful assessment of cardiac function is mandatory. Therefore, it is important to evaluate techniques for functional testing under conditions that reflect its application field as close as possible. The traditional conductance system has been used in many different studies to assess cardiac function6,20,21. However, this system structurally overestimates LV volumes due to its inherent inability to separate parallel conductance from blood conductance5,7. This can be overcome by hypertonic saline injection. However, this premises a constant parallel conductance and a linear relationship between conductance and volume, which might be too simplified and hence imprecise5,13,14. The admittance based system used in our study has been validated and implemented in multiple murine studies for the determination of cardiac function10,15,22, where it showed to be more accurate than the CS15. Recently, the AS has been used to measure LV volumes in moderately sized healthy pigs. In the study by Kutty (2013), the admittance system significantly overestimated LV volumes with a good estimation of EF in the healthy porcine heart. Furthermore, a good correlation was found between LV volumes measured by the AS and 3DE, based on repeated measurements in the same animals16. Moreover, in larger LVs, the AS showed a

108

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

trend towards a poorer agreement with 3DE16. More importantly, the effect of regional ischemia and post-infarction remodeling has not yet been evaluated. Therefore, the aim of our study was to test the ability of the AS to reliably monitor cardiac function in a human sized large animal model of chronic myocardial infarction3. To the best of our knowledge, our study is the first to evaluate the AS in a large animal ischemic heart disease model. In concordance with Kutty (2013), EF was similar at baseline and correlated significantly between AS and 3DE, whereas EDV and ESV did not. The lack of correlation for the latter two could be due to larger LV volumes in our study, which is supported by the observation that larger hearts showed a poorer agreement of 3DE and the AS in the mentioned study16. Furthermore, Bland-Altman analyses revealed modest limits of agreement at baseline. These data suggest that the AS can accurately and reliably measure cardiac function in the healthy porcine in vivo heart. Interestingly, our study reveals prominent differences at 8 weeks follow-up. At this time point, the AS overestimated LV volumes resulting in an underestimation of EF. Also, Bland-Altman analyses at follow-up showed significantly larger limits of agreement for the AS than for the CS for LV volumes, which tended to underestimate volumes. These discrepancies between baseline and follow-up measurements for the AS compared to 3DE could largely explain the interactions found between method and time point. Moreover, infarct size significantly correlated with EF measured by 3DE at follow-up. Therefore, infarct size was used as a covariate in the comparison of EF at follow-up between the different methods. After controlling for infarct size, EF measurements at follow-up remained different between the AS and 3DE indicating that the found differences between 3DE and the AS are independent of the actual infarct severity. This combination of volume over- or underestimation for the AS and CS respectively, combined with the lack of correlation of both systems with 3DE and infarct size, suggest that PV-measurements in the infarcted heart might be less accurate and reliable than in the healthy heart, supporting the importance of choosing appropriate models in the testing of novel technology. Indeed, studies on agreement and correlation between different imaging techniques (e.g. MRI, echocardiography) and volumes measured by PV methods both in small and large animal models are inconsistent15,23â&#x20AC;&#x201C;28. It should be noted however, that 3DE moderately underestimates absolute LV volumes compared to MRI1,3, which could partly explain the ASâ&#x20AC;&#x2122; overestimation of both EDV and ESV at follow-up. Nevertheless, this cannot be held responsible for the significant interaction between method and timepoint. Moreover, given the limited variation of the size of pigs used in this study, the overall variation of data -even at baseline- is considerable for all three systems.

Table 4. Correlation between 3DE, the AS and CS in the infarcted porcine heart for EDV, ESV and EF LV Parameters

3DE AS

3DE CS

AS CS

EDV (mL)

0.466

0.847*

0.240

ESV (mL)

0.252

0.247

-0.56

EF (%)

-0.050

-0.019

-0.007

Numbers shown are the correlation coeffecients (R). *Significant correlation between two systems (p=0.005).

109

PART TWO CHAPTER 6

This could indicate that the reproducibility of the different techniques is not optimal. Recently, the reproducibility of 3DE for the measurement of EF has been established. However, the inter-observer variability was considerably higher than for MRI3. To the best of our knowledge, the reproducibility of the AS has not been well established, although Kutty (2013) recently showed that repeated measurements for the same animal in the healthy heart were very similar with the AS system. In the current study, reproducibility in form of repeated measurements was not investigated and future studies should assess whether the reproducibility of the AS is influenced by MI. Furthermore, we examined the differences in volume measurements with the AS between baseline and follow-up. Importantly, EDV, ESV and EF measured with the AS at follow-up compared to baseline showed comparable changes as with 3DE. Also, comparison of delta EDV, ESV and EF between the AS and 3DE did not reveal any differences. This indicates, although induction of MI influences its accuracy, the ASâ&#x20AC;&#x2122; ability to determine functional and volume changes after MI. One of the classical features of post-infarction PV measurements is a decline of the ESPVR slope29,30. In our study however, we were unable to detect such changes with both systems. This can likely be ascribed to the fact that the previous ex vivo preload reducing experiments were performed under idealized conditions that do not necessarily represent the in vivo physiology following post MI remodeling, especially when dealing with regional ischemia. Therefore, deviations from the traditionally observed decline in the ESPVR slope may occur with a typical shift in V0, as observed by us and others6,24,31,32. The main limitation of the current study is that measurements in animals inherently have to be performed under general anesthesia. This could potentially influence the overall outcome since different levels of anesthesia might influence results. Additionally, the sternotomy could interfere with conductance signals resulting in less accurate measurements. Furthermore, the sequence of measurements was not random, meaning that LV function was first measured with the CS and only thereafter with the AS in all animals. Although it is imaginable that catheter insertion and extraction could influence cardiac function on a short-term basis, it is probably not sufficient to explain the differences found in LV volumes.

Figure 4. Correlation between infarct size and 3DE Correlation plot showing a significant correlation between infarct size and EF determined at follow-up by 3DE. R=0.698, p=0.025 (n=10).

110

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

Table 5. Differences in LV volumes between follow-up and baseline measurements for 3DE, the AS and CS LV Parameters

3DE

EDV (mL)

7.6±24.0

23.3±44.8

-9.1±37.8

ESV (mL)

15.4±14.4*

34.6±36.1*

-3.2±32.9†

EF (%)

-13.2±5.2*

-20.3±11.2*

-6.0±21.0

All values are presented as mean differences ± standard deviations. *Difference of Paired T-test for measurements at follow-up vs. baseline for the same system. †Difference between delta measurements vs. the AS. Differences with p-value <0.05 was regarded significant.

In this perspective, hypertonic saline injections could also temporarily alter the resistivity of blood in the LV. Since the circulatory volume of the animals exceeds the amount of hypertonic saline injected by far, it seems unlikely that the admittance measurements were influenced, also considering the time between the injections and the insertion of the admittance catheter. Finally, the 3DE recordings and the PV measurements were not simultaneously performed. Although no major hemodynamic changes between the measurements were observed, the observed discrepancies could partly be due to moderate changes between the consecutive measurements. In conclusion, LV volumes in the healthy heart can be accurately and reliably measured by the AS with a good correlation for EF with 3DE. However, post-MI remodeling influences these measurements, making PV-loop computations less accurate. Nevertheless, like 3DE, the AS can successfully identify a decrease in cardiac function in a porcine model of chronic myocardial infarction. These data show that the admittance-based technique is valid for the assessment of cardiac function in large animal models. Given the lack of correlation between echo- and PV-loop-based volume measurements with the current systems, future studies ideally should combine the strength of either echocardiographic or MRI based volume measurements with PV-loop specific parameters as functional endpoints in cardiac large animal models. Acknowledgements This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs, Agriculture and Innovation. The authors gratefully acknowledge Cees Verlaan, Marlijn Jansen, Merel Schurink and Joyce Visser for their excellent technical support.

111

PART TWO CHAPTER 6

REFERENCES 1. Dorosz JL, Lezotte DC, Weitzenkamp D a., Allen L a., Salcedo EE. Performance of 3-dimensional echocardiography in measuring left ventricular volumes and ejection fraction: A systematic review and meta-analysis. J Am Coll Cardiol. 2012;59:1799–1808. 2. Santos-Gallego C, Vahl T, Gaebelt H, Lopez M, Ares-Carrasco S, Sanz J, Goldman M, Hajjar R, Fuster V, Badimon J. 3D-Echocardiography Demonstrates Excellent Correlation With Cardiac Magnetic Resonance for Assessment of Left Ventricular Function and Volumes in a Model of Myocardial Infarction. J Am Coll Cardiol. 2012;59:E1564. 3. Greupner J, Zimmermann E, Grohmann A, Dübel HP, Althoff T, Borges AC, Rutsch W, Schlattmann P, Hamm B, Dewey M. Head-to-head comparison of left ventricular function assessment with 64-row computed tomography, biplane left cineventriculography, and both 2- and 3-dimensional transthoracic echocardiography: Comparison with magnetic resonance imaging as the reference s. J Am Coll Cardiol. 2012;59:1897– 1907. 4. Raghavan K, Porterfield JE, Kottam ATG, Feldman MD, Escobedo D, Valvano JW, Pearce J a. Electrical conductivity and permittivity of murine myocardium. IEEE Trans Biomed Eng. 2009;56:2044–2053. 5. Wei CL, Valvano JW, Feldman MD, Nahrendorf M, Peshock R, Pearce J a. Volume catheter parallel conductance varies between end-systole and enddiastole. IEEE Trans Biomed Eng. 2007;54:1480– 1489. 6. Burkhoff D, Mirsky I, Suga H. Assessment of systolic and diastolic ventricular properties via pressurevolume analysis: a guide for clinical, translational, and basic researchers. Am J Physiol Heart Circ Physiol. 2005;289:H501–H512. 7. Baan J, van der Velde ET, de Bruin HG, Smeenk GJ, Koops J, van Dijk a D, Temmerman D, Senden J, Buis B. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation. 1984;70:812– 823. 8. Lankford EB, Kass D a, Maughan WL, Shoukas a a. Does volume catheter parallel conductance vary during a cardiac cycle? Am J Physiol. 1990;258:H1933–H1942. 9. Krenz M. Conductance, admittance, and hypertonic saline: should we take ventricular volume measurements with a grain of salt? J Appl Physiol. 2009;107:1683–1684. 10. Kottam ATG, Porterfield J, Raghavan K, Fernandez D, Feldman MD, Valvano W, Pearce J a. Measurement and Implications. New York. 2006;l:4336–4339.

112

11. Raghavan K, Wei C, Kottam a, Altman D, Fernandez D, Reyes M, Valvano J, Feldman M, Pearce J. Design of intstrumentation and data-acquisition for complex admittance measurements. Biomed Sciense Instrum. 2004;40:453–457. 12. Clark JE, Kottam A, Motterlini R, Marber MS. Measuring left ventricular function in the normal, infarcted and CORM-3-preconditioned mouse heart using complex admittance-derived pressure volume loops. J Pharmacol Toxicol Methods. 2009;59:94–99. 13. Kornet L, Schreuder JJ, Van Der Velde ET, Jansen JRC. The volume-dependency of parallel conductance throughout the cardiac cycle and its consequence for volume estimation of the left ventricle in patients. Cardiovasc Res. 2001;51:729– 735. 14. Wei CL, Valvano JW, Feldman MD, Pearce J a. Nonlinear conductance-volume relationship for murine conductance catheter measurement system. IEEE Trans Biomed Eng. 2005;52:1654– 1661. 15. Porterfield JE, Kottam ATG, Raghavan K, Escobedo D, Jenkins JT, Larson ER, Treviño RJ, Valvano JW, Pearce J a, Feldman MD. Dynamic correction for parallel conductance, GP, and gain factor, alpha, in invasive murine left ventricular volume measurements. J Appl Physiol. 2009;107:1693– 1703. 16. Meyer a, Zoll J, Charles a L, Charloux a, de Blay F, Diemunsch P, Sibilia J, Piquard F, Geny B. Skeletal muscle mitochondrial dysfunction during chronic obstructive pulmonary disease: central actor and therapeutic target. Exp Physiol. 2013;98:1063– 1078. 17. Shimada YJ, Ishikawa K, Kawase Y, Ladage D, Tilemann L, Shiota T, Hajjar RJ. Comparison of left ventricular stroke volume assessment by two- and three-dimensional echocardiography in a swine model of acute myocardial infarction validated by thermodilution method. Echocardiography. 2012;29:1091–1095. 18. Jenkins C, Bricknell K, Chan J, Hanekom L, Marwick TH. Comparison of Two- and Three-Dimensional Echocardiography With Sequential Magnetic Resonance Imaging for Evaluating Left Ventricular Volume and Ejection Fraction Over Time in Patients With Healed Myocardial Infarction. Am J Cardiol. 2007;99:300–306. 19. Soliman OIIII, Krenning BJJ, Geleijnse MLL, Nemes A, van Geuns R-JJ, Baks T, Anwar AMM, Galema TWW, Vletter WBB, ten Cate FJ, Cate FJT. A comparison between QLAB and tomtec full volume reconstruction for real time three-dimensional echocardiographic quantification of left ventricular volumes. Echocardiography. 2007;24:967–974. 20. Jegger D, Jeanrenaud X, Nasratullah M, Chassot P-G, Mallik A, Tevaearai H, von Segesser LK, Segers P, Stergiopulos N. Noninvasive Doppler-

ADMITTANCE-BASED PRESSURE-VOLUME LOOP MEASUREMENTS

derived myocardial performance index in rats with myocardial infarction: validation and correlation by conductance catheter. Am J Physiol Heart Circ Physiol. 2006;290:H1540–H1548. 21. Timmers L, Henriques JPSPS, de Kleijn DPVP V, Devries JHH, Kemperman H, Steendijk P, Verlaan CWJWJ, Kerver M, Piek JJJ, Doevendans P a a., Pasterkamp G, Hoefer IEE. Exenatide Reduces Infarct Size and Improves Cardiac Function in a Porcine Model of Ischemia and Reperfusion Injury. J Am Coll Cardiol. 2009;53:501–510. 22. Tabima DM, Hacker T a, Chesler NC. Measuring right ventricular function in the normal and hypertensive mouse hearts using admittancederived pressure-volume loops. Am J Physiol Heart Circ Physiol. 2010;299:H2069–H2075. 23. Lin HY, Freed D, Lee TWR, Arora RC, Ali A, Almoustadi W, Xiang B, Wang F, Large S, King SB, Tomanek B, Tian G. Quantitative assessment of cardiac output and left ventricular function by noninvasive phase-contrast and cine MRI: Validation study with invasive pressure-volume loop analysis in a swine model. J Magn Reson Imaging. 2011;34:203–210. 24. Winter EM, Grauss RW, Atsma DE, Hogers B, Poelmann RE, Van Der Geest RJ, Tschöpe C, Schalij MJ, Gittenberger-De Groot a. C, Steendijk P. Left ventricular function in the post-infarct failing mouse heart by magnetic resonance imaging and conductance catheter: A comparative analysis. Acta Physiol. 2008;194:111–122. 25. Feldman MD, Erikson JM, Mao Y, Korcarz CE, Lang RM, Freeman GL. Validation of a mouse conductance system to determine LV volume: comparison to echocardiography and crystals. Am J Physiol Heart Circ Physiol. 2000;279:H1698–H1707. 26. Amirhamzeh MMR, Dean D a., Jia CX, Cabreriza SE, Yano OJ, Burkhoff D, Spotnitz HM. Validation of right and left ventricular conductance and echocardiography for cardiac function studies. Ann Thorac Surg. 1996;62:1104–1109. 27. Jacoby C, Molojavyi A, Flögel U, Merx MW, Ding Z, Schrader J. Direct comparison of magnetic resonance imaging and conductance microcatheter in the evaluation of left ventricular function in mice. Basic Res Cardiol. 2006;101:87–95. 28. Nielsen JM, Kristiansen SB, Ringgaard S, Nielsen TT, Flyvbjerg A, Redington AN, Bøtker HE. Left ventricular volume measurement in mice by conductance catheter: evaluation and optimization of calibration. Am J Physiol Heart Circ Physiol. 2007;293:H534–H540. 29. Suga H, Sagawa K, Shoukas a a. Load independence of the instantaneous pressurevolume ratio of the canine left ventricle and effects of epinephrine and heart rate on the ratio. Circ Res. 1973;32:314–322. 30. Suga H, Sagawa K. Instantaneous pressure-volume relationships and their ratio in the excised, supported canine left ventricle. Circ Res.

1974;35:117–126. 31. Sunagawa K, Maughan WL, Sagawa K. Effect of regional ischemia on the left ventricular endsystolic pressure-volume relationship of isolated canine hearts. Circ Res. 1983;52:170–178. 32. Steendijk P, Baan J, Van Der Velde ET, Baan J. Effects of critical coronary stenosis on global systolic left ventricular function quantified by pressure-volume relations during dobutamine stress in the canine heart. J Am Coll Cardiol. 1998;32:816–826.

113

chapter 7

Admittance based pressure volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction

Physiological Reports 2014 April 22;2(4)e00287

G.P.J. van Hout1*, S.J. Jansen of Lorkeers2*, J.M.I.H. Gho2, P.A. Doevendans1,2, W.W. van Solinge3, G. Pasterkamp1, S.A.J. Chamuleau2, I.E. Hoefer1 * Contributed equally to this work 1

Experimental Cardiology Laboratory, University Medical Center Utrecht, The Netherlands

Department of Cardiology, University Medical Center Utrecht, The Netherlands

Department of Clinical Chemistry and Haematology, University Medical Center Utrecht, The Netherlands

PART TWO CHAPTER 7

ABSTRACT Background A novel admittance based pressure volume system (AS) has recently been developed and introduced. Thus far, the new technique has been validated predominantly in small animals. In large animals it has only been compared to three-dimensional echocardiography (3DE) where the AS showed to overestimate left ventricular (LV) volumes. To fully determine the accuracy of this device, we compared the AS with gold standard cardiac magnetic resonance imaging (CMRI). Objective The aim of this study was to directly compare LV volume measurements obtained by the AS with CMRI in a porcine model of chronic myocardial infarction (MI). Methods Fourteen pigs were subjected to 90 minutes closed chest balloon occlusion of the left anterior descending artery. After 8 weeks of follow-up, pigs were consecutively subjected to LV volume measurements by the AS, CMRI and 3DE under general anesthesia. Results The AS overestimated end diastolic volume (EDV; +20.9±30.6mL, p=0.024) and end systolic volume (ESV; +17.7±29.4mL, p=0.042) but not ejection fraction (EF; +2.46±6.16%, p=NS) compared to CMRI. Good correlations of EDV (R=0.626, p=0.017) and EF (R=0.704, p=0.005) between the AS and CMRI were observed. EF measured by the AS and 3DE also correlated significantly (R=0.624, p=0.030). Conclusion After subjection of pigs to MI, the AS very moderately overestimates LV volumes and shows accurate measurements for EF compared to CMRI. This makes the AS a useful tool to determine cardiac function and dynamic changes in large animal models of cardiac disease.

116

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

INTRODUCTION Cardiac function is a main endpoint in efficacy testing of novel therapeutics in clinical and translational cardiovascular research1,2. This requires objective and reliable tools for its determination. Many current methods use left ventricular (LV) volumes as a surrogate of cardiac function. Apart from volume assessment, invasive pressure-volume (PV) measurements (PV-Loops) provide researchers with more specific information on both systolic and diastolic myocardial function and dynamic changes which are not measurable with most imaging modalities3–5. However, reliable analysis of various PV-Loop parameters depends on adequate LV volume calibration, which requires accurate LV volume assessment. Recently, a novel method of measuring PV loops has been introduced. This admittance based PV Loop system (AS) differs from the classical conductance system in several ways. First, the new method does not assume a constant contribution of parallel conductance during the cardiac cycle6,7. Secondly, it differs in the way parallel conductance is separated from blood conductance making the new method less time-consuming and the system easier to use. Finally, the AS applies a non-linear relationship between conductance and volume for the conversion of blood conductance into volume7–9. This novel method has been extensively validated in small animal models10–16. Only recently, the AS has been evaluated in large animal models17,18. The large animal studies demonstrated good performance, but show that the AS moderately overestimates EDV and ESV. In both studies, however, three-dimensional echocardiography (3DE) was used as a reference standard. 3DE has been proven to be superior to the more conventional twodimensional echocardiography for volume measurements and is relatively inexpensive and useful for quick and non-invasive assessment of LV volumes19–22. However, 3DE has been observed to be less precise in estimating EF compared to cardiac magnetic resonance imaging (CMRI), the gold standard for volume determination19,20. 3DE is also known to consistently underestimate both EDV and ESV and this underestimation increases in diseased hearts23,24. Therefore, direct comparison with CMRI is especially mandatory after post-infarction remodeling. Hence, the aim of this study was to compare the AS derived EDV, ESV and EF to CMRI derived measurements in a closed chest porcine model of chronic myocardial infarction.

MATERIAL AND METHODS All animal experiments were approved by the institutional animal welfare committee of the UMC Utrecht and were executed conforming to the ‘Guide for the Care and Use of Laboratory Animals’. A total of 14 specific pathogen free female landrace pigs were evaluated in this study (Van Beek Lelystad, the Netherlands). Pigs (body weight 79.1±5.9 kg) were subjected to myocardial infarction followed by invasive PV measurements, CMRI and 3DE at 8 weeks follow-up. Before any procedural intervention, animals were housed

117

PART TWO CHAPTER 7

in pairs. Because of health concerns, animals were housed individually after myocardial infarction until the end of the study. Pigs were fed twice a day, water was available ad libitum. Infarct induction All animals were pre-treated with amiodaron for 10 days (1200mg loading dose, 800mg/ day maintenance), clopidogrel for 3 days (75 mg/day) and acetylsalicylic acid for 1 day (320 mg loading dose, 80 mg/day maintenance). All medication was continued until the end of the study. Animals were anesthetized in their cage with an intramuscular injection of 10 mg/kg ketamine, 0.4 mg/kg midazolam and 0.5 mg/kg atropine. Anesthesia was maintained with intravenous infusion of 0.5 mg/kg/h midazolam, 2.5 Âľg/kg/h sufentanyl and 0.1 mg/kg/h pancuronium and pigs were mechanically ventilated. Pre-operatively, animals received a fentanyl patch (25Âľg/h). Arterial access was obtained by introduction of an 8F sheath into the carotid artery after surgical exposure. A coronary angiogram of the left coronary tree was acquired using an 8F JL4 guiding catheter (Boston scientific, Natick, USA). The diameter of the left anterior descending artery (LAD) was measured directly distally to the second diagonal artery. An adequately sized balloon was placed

Short axis apex

Short axis pap

Short axis basal

Long axis

End systolic

End diastolic

Apical 4 chamber

Apical 2 chamber

Short axis

3D Reconstruction

End systolic

End diastolic

Figure 1. Different techniques for LV volume measurements EDV, ESV and EF were measured with 3 different modalities. A. Representative recordings of PV Loops with the AS. 10 beats were recorded during apnea. B. Representative images of a pig subjected to CMRI measurements at the end of systole and diastole. 4 different views are depicted. Short axis at the apex, papillary muscle (pap) and at a basal level combined with a long axis view. C. Representative images 3DE measurements at the end of systole and diastole. Three different views are depicted. Apical 4 chamber, apical 2 chamber and a short axis view, together resulting in a 3D reconstruction of the left ventricle.

118

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

Figure 2. Measured LV volumes by the AS, CMRI and 3DE in the infarcted porcine heart LV dimensions were obtained by the three different modalities. A. EDV 8 weeks after MI. B. ESV 8 weeks after MI. C. EF 8 weeks after MI. Data are presented as mean ± 95%CI. * p<0.05. LV = left ventricular; AS = admittance based PV system; CMRI = cardiac magnetic resonance imaging; 3DE = three-dimensional echocardiography.

distal from the second diagonal branch and inflated for 90 minutes. After reperfusion and observation for approximately 2 hours, the surgical wound was closed and animals were weaned from anesthesia. Animals were defibrillated in case of ventricular fibrillation (VF). Invasive PV-Loop measurements Admittance based PV-loop measurements were performed as recently described. 17 In short, animals were again anesthetized according to the protocol described above 8 weeks after infarct induction. Arterial access was obtained by introduction of an 8F sheath into the carotid artery. The 7F tetra-polar admittance catheter (7.0 VSL Pigtail/no lumen, Transonic Scisense, London, Canada) was inserted into the left ventricle through the sheath in the carotid artery under fluoroscopic guidance. The catheter measures admittance magnitude and phase in combination with pressure. It contains 7 platinum electrodes dividing it into 4 selectable segments. The largest segment inside the LV was used for absolute volume assessment. The catheter was connected to the ADVantage system TM (Transonic SciSense, London, Canada) linked to a multi-channel acquisition system (Iworx 404), required for real-time data acquisition. A baseline scan was performed to determine the end diastolic and end systolic blood conductance required for absolute volume calculations. The external stroke volume (SV) required for volume calibration and analysis was derived from CMRI measurements. Blood resistivity was assumed to be constant in all animals (150 Ω*cm). All measurements were performed during apnea. Data were offline analyzed using Iworx analysis software (Labscribe V2.0). Cardiac magnetic resonance imaging (CMRI) Immediately after PV measurements, pigs were transported to the CMRI scanner. CMRI images were obtained with a 3T CMRI scanner (Achieva TX, Philips Healthcare). Animals were placed on the CMRI table in a supine position, under continuous anesthesia. A 32-channel receiver coil was placed over the chest. ECG-gated steady-state free precesion cine imaging was obtained in a short axis and a two-chamber long axis view (Voxelsize acquisition = 2*2.1mm, recon voxelsize = 1.25*1.25mm, Slice thickness = 8mm,

119

PART TWO CHAPTER 7

Bandwith = 1243Hz, echo time = 1.62ms, repetition time = 3.2ms, balanced gradient echo readout = 20. Offline imaging analysis was performed in Qmass MR 7.4 enterprise solutions (Medis medical imaging systems BV, Leiden). Three-dimensional echocardiography 3DE was performed with a X3-1 transducer on an iE33 ultrasound device (Philips, Eindhoven, The Netherlands) directly after CMRI as previously described17. After medial sternotomy, a gel-filled flexible sleeve was placed directly on the apex of the heart. The depth and sector size were adjusted to fit the complete ventricle. All data sets were acquired in real time using 7 consecutive cardiac cycles (full volume analysis). The images were analyzed offline using QLab 10.1 (3DQ advanced) analysis software. Ventricle tracing was performed by semi-automatic border detection as described before (Gain 50%, compression 50%, frame rate = 20-30 frames/second, 1 dataset = 7 beats). 25 Because of incomplete capture of the left ventricle on echocardiographic recordings, 2 out 14 pigs were excluded from the analysis. Infarct size Animals were sacrificed by exsanguinations under anesthesia. The hearts were excised and the LV was then cut into 5 equal slices from apex to base. Slices were incubated in 1% triphenyltetrazolium chloride (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37°C 0.9%NaCl for 15 min to discriminate infarct tissue from viable myocardium. Statistical Analysis All data are expressed as mean ± standard deviation unless stated otherwise. CMRI data and PV measurements were separately analyzed by two different researchers blinded to the outcome of the other technique. End diastolic volume (EDV), end systolic volume (ESV) and ejection fraction (EF) measured by the AS and 3DE were compared with CMRI values using a paired Student’s t-test. Correlations were tested using Pearson’s correlation test. The limits of agreement (1.96*SD = 95% confidence interval) of the AS compared to CMRI were determined by Bland-Altman analysis. All statistical analyses were performed in SPSS statistics version 20.0. A two-sided P-value of <0.05 was regarded statistically significant in all analyses.

Table 1. Assessment of LV dimensions in the infarcted porcine heart with the AS, CMRI and 3DE LV Parameters

CMRI

3DE

EDV (mL)

186±39

165±25

79±15

ESV (mL)

116±30

99±15

46±11

EF (%)

37.5±8.6

40.0±7.0

41.9±9.2

All values are presented as mean ± standard deviations. N=14 for AS and CMRI measurements, n=12 for 3DE measurements.

120

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

Table 2. Correlations between the AS, CMRI and 3DE for EDV, ESV and EF in the infarcted porcine heart LV Parameters

CMRI vs. AS

CMRI vs. 3DE

3DE vs. AS

EDV (mL)

0.626*

0.160

0.235

ESV (mL)

0.271

0.320

-0.111

EF (%)

0.704*

0.982*

0.624*

Numbers shown are the correlation coefficients (R). *Significant correlation between two techniques (p<0.05).

RESULTS At 8 weeks follow-up after MI, LV volumes were assessed in 14 pigs with both the AS and CMRI (figure 1A & 1B). 7 out of 14 animals showed at least one episode of VF. Left ventricular end systolic pressure (122±27mmHg) and end diastolic pressure (13±8mmHg) were in a physiological range during the experiment. Heart rate did not show any significant difference between AS measurement and CMRI measurements (54±18 vs. 58±16 bpm, p=NS). Moreover, heart rate correlated significantly (R=0.747, p=0.002) between AS and CMRI derived measurements, indicating that both measurements were performed under a steady state. 3DE vs. CMRI To test the degree of LV volume underestimation by 3DE compared to CMRI, we measured LV volumes with 3DE in 12 out of 14 animals (Figure 1C). EDV (79.1±14.9mL (3DE) vs. 165.3±24.8mL (CMRI), p<0.001) and ESV (45.8±10.6mL (3DE) vs. 98.6±15.2mL (CMRI), p<0.001) were underestimated by 3DE compared to CMRI as expected (Figure 2A & 2B, table 1). EF measured by MRI was slightly overestimated by 3DE (41.9±9.23% (3DE) vs. 40.0±7.0% (CMRI), p=0.018), with an almost perfect correlation (R=0.982, p<0.001) (Figure 2C & 3C, table 2). We also determined whether LV size influenced the accuracy of the measurements by 3DE. Indeed Bland-Altman analyses revealed a trend towards a lower accuracy for EDV, ESV and EF in the larger sized hearts (Figure 4A – 4C). AS vs. CMRI Next we determined whether EDV, ESV and EF measured with the AS deviated from the CMRI values. In the present study, the AS significantly overestimated both EDV (186.2±39.2mL (AS) vs. 165.3±24.8mL (CMRI), p=0.024) and ESV (116.4±29.7mL (AS) vs. 98.6±15.2mL (CMRI), p=0.042) compared to CMRI (Figure 2A & 2B, table 1). However, EF measured by the AS was not significantly different from MRI derived EF (37.5±8.6% (AS) vs. 40.0±7.0% (CMRI), p=NS) (Figure 2C, table 1). Importantly, we observed a good correlation of EDV (R=0.626, p=0.017) and EF (R=0.704, p=0.005) between the AS and CMRI (Figure 3A & 3B, table 2) and a good correlation of EF between AS and 3DE (R=0.624, p=0.030) (Figure 3D). Subsequently, the limits of agreement of the AS opposed to CMRI were determined by Bland-Altman analysis. Both EDV and ESV measured by the AS showed a moderate to good agreement and EF showed a good to excellent agreement with CMRI

121

PART TWO CHAPTER 7

(Figure 4). The mean percentage inter-method difference was 12% for EDV, 17% for ESV and 6% for EF. No particular trend was detected between the accuracy of measurements and increasing values for ventricular volumes by CMRI. Correlation with infarct size As LV volumes are often measured as surrogates of cardiac function, it is essential to test whether the volume measurements of the various systems reflected the actual extent of myocardial damage. Therefore, infarct size was determined at 8 weeks follow-up. Mean infarct size was 13.8Âą2.9%. We found a good correlation of CMRI based ESV (R=0.685, p=0.007) and EF (R=-0.608, p=0.021) with infarct size (Figure 5A & 5B, table 3). For 3DE, a similar correlation for ESV (R=0.659, p=0.020) and EF (R=-0.650, p=0.022) compared to infarct size was found, indicating that both systemsâ&#x20AC;&#x2122; measurement reflect the decay in cardiac function after myocardial infarction (Figure 5C & 5D). With the AS, however, we failed to establish a significant correlation between either ESV or EF with infarct size.

DISCUSSION Real-time PV loops provide researchers with more detailed information on cardiac function than most imaging modalities.3,26 Although some of these parameters are volumeindependent, many PV-loop parameters depend on accurate determination of LV volumes. Traditional PV-systems have shown to accurately estimate LV volumes but require additional calibration steps that are susceptible to technical error.4,9,27 The novel AS is easier to use and does not require these calibration steps that could potentially increase variability.11 To the best of our knowledge, this is the first study to directly compare the AS with CMRI, the gold standard for in vivo LV volume assessment. Using CMRI as a reference standard in a closed chest porcine model, the AS accurately measures EF with only modest overestimation of EDV and ESV. The AS shows small mean percentage inter-method differences and a better accuracy for EDV, ESV and EF than 3DE in this study. Furthermore, our study shows that there is a good correlation for EDV and EF between AS and CMRI and a good correlation for EF between AS and 3DE. Both CMRI and 3DE are reflecting the extent of cardiac damage since ESV and EF of the two methods correlate well with

Table 3. Correlations between ESV and EF for the AS, CMRI and 3DE and infarct size (IS) in the infarcted porcine heart LV Parameters

AS vs. IS

CMRI vs. IS

3DE vs. IS

EDV (mL)

-0.146

0.239

0.365

ESV (mL)

-0.072

0.685*

0.659*

EF (%)

-0.174

-0.608*

-0.650*

Numbers shown are the correlation coefficients (R). *Significant correlation between two techniques (p<0.05).

122

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

7 Figure 3. Correlation plots of LV volumes between the AS, CMRI and 3DE in the infarcted porcine heart A. EDV correlates significantly between the AS and CMRI. R=0.626 p=0.017, n=14. B. EF correlates significantly between the AS and CMRI. R=0.704, p=0.005, n=14. C. EF correlates significantly between CMRI and 3DE. R=0.982, p<0.001, n=12. D. EF correlates significantly between 3DE and the AS. R=0.624, p=0.030, n=12. LV = left ventricular; AS = admittance based PV system; CMRI = cardiac magnetic resonance imaging; 3DE = three-dimensional echocardiography.

infarct size, even though scar remodeling has already occurred after such an extensive follow-up period and infarct size analysis has become less representative due to possible scar shrinkage and resorption.28 Presumably this phenomenon causes the lack of correlation between infarct size and EDV measured by any method in our study. Very recently the performance of the AS in a large animal model has also been investigated by us and others. 17,18. Here, an overestimation of EDV and ESV at baseline was found. This overestimation was more pronounced after MI and led to an underestimation of EF in the diseased heart compared to the reference standard. Both large animal studies were however limited by the use of 3DE as the reference standard. It has recently been shown that EDV and ESV are increasingly underestimated by 3DE in more diseased left ventricles, which is in line with our findings in the present study (figure 4a-4c).23,24 The underestimation observed in our study is slightly more severe compared to clinical data. Mor-Avi et al. found an underestimation for EDV of -67Âą46mL when comparing 3DE to CMRI in a heterogeneous population of patients with both healthy

123

PART TWO CHAPTER 7

and failing hearts. 24 In the present study we only performed measurements in infarcted hearts, which presumably is the reason for this relatively severe underestimation. These data suggest that comparison of the AS with CMRI is warranted to fully determine the accuracy of this novel technique. The current study has several limitations. First of all, measurements were performed under general anesthesia. Since different levels of anesthesia could influence cardiac performance, anesthesia could be a confounding factor in comparing the AS with CMRI despite a consistent heart rate. Secondly, during CMRI, arterial pressure was not registered. Although heart rates did not indicate a change in hemodynamic state, theoretically a difference in arterial pressure between data assessment with the different systems could occur, possibly influencing the outcome. In this study CMRI was preceded by measurements with the AS. Ideally this should be the other way round since the more invasive procedure could influence measurements with the less invasive procedure. Unfortunately, this was proven to be impossible due to logistical difficulties. Nevertheless, measurements with the AS only take up 15 to 20 minutes and the surgical intervention to reach the carotid artery is only minor without any substantial blood loss. Swine used in the present study had relatively large hearts compared to averagely sized human hearts. 20 We failed to observe a trend in decreased accuracy towards the larger sized hearts, so this is unlikely to be of major influence. In the current study we did not perform measurements with a classical conductance system. Therefore, no conclusion can be drawn on the possible superiority of either one of the systems above the other. Lastly, we did not take the effect of the sternotomy during 3DE on physiologic parameters into account. Animals underwent sternotomy after AS and CMRI measurements so the direct comparison of these two methods is not affected by this limitation.

Figure 4. Bland-Altman plots of LV volumes in the infarcted porcine heart Bland-Altman plots of 3DE (A to C) and the AS (D to E) with CMRI as reference standard. The green line represents the mean. The red lines represent the limits of agreement. A. End diastolic volume. B. End systolic volume. C. Ejection fraction. AS = admittance based PV system; CMRI = cardiac magnetic resonance imaging; 3DE = three-dimensional echocardiography.

124

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

7 Figure 5. Correlation plots of LV volumes with infarct size in the infarcted porcine heart Significant correlations of LV dimensions of CMRI and 3DE with infarct size. A. ESV (CMRI) correlates significantly with infarct size. R=0.685, p=0.007, n=14. B. EF (CMRI) correlates significantly with infarct size. R=-0.608, p=0.021, n=14. C. ESV (3DE) correlates significantly with infarct size. R=0.659, p=0.020, n=12. D. EF (3DE) correlates significantly with infarct size. R=0.650, p=0.022. CMRI = cardiac magnetic resonance imaging; 3DE = three-dimensional echocardiography.

Acknowledgements The authors have nothing to declare and gratefully acknowledge Marlijn Jansen, Joyce Visser, Martijn van Nieuwburg, Grace Croft and Evelyn Velema for their excellent technical support and Joep van Oorschot for acquiring the MRI images. This research forms part of the Project P1.04 SMARTCARE and Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs.

125

PART TWO CHAPTER 7

REFERENCES 1. Van Der Spoel TIG, Jansen Of Lorkeers SJ, Agostoni P, Van Belle E, Gyngysi M, Sluijter JPG, Cramer MJ, Doevendans P a., Chamuleau S a J. Human relevance of pre-clinical studies in stem cell therapy: Systematic review and meta-analysis of large animal models of ischaemic heart disease. Cardiovasc Res. 2011;91:649–658. 2. Gho JMIH, Kummeling GJM, Koudstaal S, Jansen Of Lorkeers SJ, Doevendans P a, Asselbergs FW, Chamuleau S a J. Cell therapy, a novel remedy for dilated cardiomyopathy? A systematic review. J Card Fail. 2013;19:494–502. 3. Burkhoff D, Mirsky I, Suga H. Assessment of systolic and diastolic ventricular properties via pressurevolume analysis: a guide for clinical, translational, and basic researchers. Am J Physiol Heart Circ Physiol. 2005;289:H501–H512. 4. Krenz M. Conductance, admittance, and hypertonic saline: should we take ventricular volume measurements with a grain of salt? J Appl Physiol. 2009;107:1683–1684. 5. Lips DJ, van der Nagel T, Steendijk P, Palmen M, Janssen BJ, van Dantzig JM, de Windt LJ, Doevendans P a. Left ventricular pressure-volume measurements in mice: Comparison of closedchest versus open-chest approach. Basic Res Cardiol. 2004;99:351–359. 6. Kornet L, Schreuder JJ, Van Der Velde ET, Jansen JRC. The volume-dependency of parallel conductance throughout the cardiac cycle and its consequence for volume estimation of the left ventricle in patients. Cardiovasc Res. 2001;51:729– 735. 7. Wei CL, Valvano JW, Feldman MD, Nahrendorf M, Peshock R, Pearce J a. Volume catheter parallel conductance varies between end-systole and enddiastole. IEEE Trans Biomed Eng. 2007;54:1480– 1489. 8. Porterfield JE, Kottam ATG, Raghavan K, Escobedo D, Jenkins JT, Larson ER, Treviño RJ, Valvano JW, Pearce J a, Feldman MD. Dynamic correction for parallel conductance, GP, and gain factor, alpha, in invasive murine left ventricular volume measurements. J Appl Physiol. 2009;107:1693– 1703. 9. Baan J, van der Velde ET, de Bruin HG, Smeenk GJ, Koops J, van Dijk a D, Temmerman D, Senden J, Buis B. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation. 1984;70:812– 823. 10. Tabima DM, Hacker T a, Chesler NC. Measuring right ventricular function in the normal and hypertensive mouse hearts using admittancederived pressure-volume loops. Am J Physiol Heart Circ Physiol. 2010;299:H2069–H2075.

126

11. Kottam ATG, Porterfield J, Raghavan K, Fernandez D, Feldman MD, Valvano W, Pearce J a. Measurement and Implications. New York. 2006;l:4336–4339. 12. Kottam A, Dubois J, McElligott A, Henderson KK. Novel approach to admittance to volume conversion for ventricular volume measurement. Proc Annu Int Conf IEEE Eng Med Biol Soc EMBS. 2011;2011:2514–2517. 13. Clark JE, Kottam A, Motterlini R, Marber MS. Measuring left ventricular function in the normal, infarcted and CORM-3-preconditioned mouse heart using complex admittance-derived pressure volume loops. J Pharmacol Toxicol Methods. 2009;59:94–99. 14. Raghavan K, Kottam ATG, Valvano JW, Pearce J a. Rats. Area. 2008;l:993–996. 15. Raghavan K, Porterfield JE, Kottam ATG, Feldman MD, Escobedo D, Valvano JW, Pearce J a. Electrical conductivity and permittivity of murine myocardium. IEEE Trans Biomed Eng. 2009;56:2044–2053. 16. Trevino RJ, Jones DL, Escobedo D, Porterfield J, Larson E, Chisholm GB, Barton A, Feldman MD. Validation of a new micro-manometer pressure sensor for cardiovascular measurements in mice. Biomed Instrum Technol. 2010;44:75–83. 17. Van Hout GPJ, de Jong R, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittance-based pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75. 18. Kutty S, Kottam AT, Padiyath A, Bidasee KR, Li L, Gao S, Wu J, Lof J, Danford D a, Kuehne T. Validation of admittance computed left ventricular volumes against real-time three-dimensional echocardiography in the porcine heart. Exp Physiol. 2013;98:1092–101. 19. Dorosz JL, Lezotte DC, Weitzenkamp D a., Allen L a., Salcedo EE. Performance of 3-dimensional echocardiography in measuring left ventricular volumes and ejection fraction: A systematic review and meta-analysis. J Am Coll Cardiol. 2012;59:1799–1808. 20. Greupner J, Zimmermann E, Grohmann A, Dübel HP, Althoff T, Borges AC, Rutsch W, Schlattmann P, Hamm B, Dewey M. Head-to-head comparison of left ventricular function assessment with 64-row computed tomography, biplane left cineventriculography, and both 2- and 3-dimensional transthoracic echocardiography: Comparison with magnetic resonance imaging as the reference s. J Am Coll Cardiol. 2012;59:1897– 1907. 21. Kawamura R, Seo Y, Ishizu T, Atsumi A, Yamamoto M, Machino-Ohtsuka T, Nakajima H, Sakai S, Tanaka YO, Minami M, Aonuma K. Feasibility of left ventricular volume measurements by threedimensional speckle tracking echocardiography depends on image quality and degree of left

ADMITTANCE PV LOOPS VERSUS CARDIAC MRI

ventricular enlargement: Validation study with cardiac magnetic resonance imaging. J Cardiol. 2013;m:1–9. 22. Santos-Gallego C, Vahl T, Gaebelt H, Lopez M, Ares-Carrasco S, Sanz J, Goldman M, Hajjar R, Fuster V, Badimon J. 3D-Echocardiography Demonstrates Excellent Correlation With Cardiac Magnetic Resonance for Assessment of Left Ventricular Function and Volumes in a Model of Myocardial Infarction. J Am Coll Cardiol. 2012;59:E1564. 23. Moceri P, Doyen D, Bertora D, Cerboni P, Ferrari E, Gibelin P. Real time three-dimensional echocardiographic assessment of left ventricular function in heart failure patients: underestimation of left ventricular volume increases with the degree of dilatation. Echocardiography. 2012;29:970–7. 24. Mor-Avi V, Jenkins C, Kühl HP, Nesser HJ, Marwick T, Franke A, Ebner C, Freed BH, SteringerMascherbauer R, Pollard H, Weinert L, Niel J, Sugeng L, Lang RM. Real-Time 3-Dimensional Echocardiographic Quantification of Left Ventricular Volumes. Multicenter Study for Validation With Magnetic Resonance Imaging and Investigation of Sources of Error. JACC Cardiovasc Imaging. 2008;1:413–423. 25. Soliman OIIII, Krenning BJJ, Geleijnse MLL, Nemes A, van Geuns R-JJ, Baks T, Anwar AMM, Galema TWW, Vletter WBB, ten Cate FJ, Cate FJT. A comparison between QLAB and tomtec full volume reconstruction for real time three-dimensional echocardiographic quantification of left ventricular volumes. Echocardiography. 2007;24:967–974. 26. Suga H, Sagawa K. Instantaneous pressure-volume relationships and their ratio in the excised, supported canine left ventricle. Circ Res. 1974;35:117–126. 27. Winter EM, Grauss RW, Atsma DE, Hogers B, Poelmann RE, Van Der Geest RJ, Tschöpe C, Schalij MJ, Gittenberger-De Groot a. C, Steendijk P. Left ventricular function in the post-infarct failing mouse heart by magnetic resonance imaging and conductance catheter: A comparative analysis. Acta Physiol. 2008;194:111–122. 28. Holmes JW, Yamashita H, Waldman LK, Covell JW. Scar remodeling and transmural deformation after infarction in the pig. Circulation. 1994;90:411–420.

127

chapter 8

Elevated mean neutrophil volume represents altered neutrophil composition and reflects damage after myocardial infarction

In revision - Basic Research in Cardiology

G.P.J. van Hout1, W.W. van Solinge2, C.M. Gijsberts1,3, M.P.J. Teuben4, P.H.C. Leliefeld4, M. Heeres4, F. Nijhoff5, S. de Jong5, L. Bosch1, S.C.A. de Jager1, A. Huisman2, P.R. Stella5, G. Pasterkamp1,2, L.J. Koenderman4, I.E. Hoefer1,2 1

Experimental Cardiology Laboratory, University Medical Center Utrecht, The Netherlands

Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, The Netherlands

ICIN - Netherlands Heart Institute, the Netherlands

Department of Respiratory Medicine, University Medical Center Utrecht, The Netherlands

Department of Cardiology, University Medical Center Utrecht, The Netherlands

PART TWO CHAPTER 8

ABSTRACT Background Myocardial infarction (MI) induces an exaggerated inflammatory response in which neutrophils fulfill a prominent role. Mean neutrophil volume (MNV) represents the average size of the circulating neutrophil population. Our goal was to determine the effect of MI on MNV and investigate the mechanisms behind MNV elevation. Methods MNV of 84 MI patients was compared with the MNV of 209 stable angina patients and correlated to simultaneously measured CK levels. Fourteen pigs were subjected to temporary coronary balloon occlusion and blood was sampled at multiple time points to measure MNV. Echocardiography was performed followed by ex vivo infarct size assessment after 72 hours. Results MNV was higher in MI patients compared to stable angina patients (602SD26AU vs. 580SD20AU, p<0.0001) and correlated with simultaneously measured CK levels (R=0.357, p<0.0001). In pigs, MNV was elevated post-MI (451SD11AU vs. 469SD12AU), p<0.0001). MNV correlated with infarct size (R=0.705, p=0.007) and inversely correlated with left ventricular ejection fraction (R=-0.718, p=0.009). Cell sorting revealed an increased presence of banded neutrophils after MI, which have a higher MNV compared to mature neutrophils post-MI (495 SD14AU vs. 478 SD11AU, p=0.012). MNV from coronary sinus blood was higher than MNV of neutrophils from simultaneously sampled arterial blood (463SD7.6AU vs. 461SD8.6AU, p=0.013) post-MI. Conclusions The current study shows MNV is elevated and reflects cardiac damage post-MI. MNV increases due to altered neutrophil composition and systemic neutrophil activation. MNV may be an interesting parameter for prognostic assessment in MI and provide new insights into pathological innate immune responses evoked by ischemia-reperfusion.

130

ELEVATED MNV REFLECTS DAMAGE AFTER MI

INTRODUCTION Ischemic damage after myocardial infarction (MI) induces a detrimental inflammatory response1. The best treatment to salvage myocardium post-MI is to restore myocardial reperfusion through percutaneous coronary intervention (PCI)2. Apart from myocardial salvage, reperfusion also allows for immediate interaction between the damaged myocardium and circulating cells, among which neutrophils are the first responders3–5. After myocardial reperfusion, circulating cells migrate to the infarcted tissue to clear out necrotic cells and orchestrate cardiac wound healing4. Paradoxically, the influx of inflammatory cells into the myocardium can result in the elimination of viable cardiomyocytes, thereby inducing infarct expansion. Many clinical studies have shown that the severity of this inflammatory response is reflected by systemically measurable inflammatory parameters6. Especially circulating neutrophils have been shown to represent the amount of inflicted damage, since neutrophil numbers and ratios are associated with a worse prognosis and more adverse events after MI7–15. Currently, prognostic value of neutrophils to predict the outcome after MI is limited to neutrophil numbers or derivatives hereof, like neutrophil/lymphocyte ratio16,17. Studies that focus on morphological changes of neutrophils, or use e.g. membrane proteins as a prognostic marker are scarce. The reasons may be manifold. Amongst technical and cost issues, neutrophils have classically been regarded as a uniform population with detrimental effects in MI related cardiac injury. However, new data show the existence of neutrophil subtypes, like banded and hypersegmented neutrophils, with distinct functional capacity that is reflected by morphological differences and receptor expression patterns18,19. Morphological changes in the neutrophil population may therefore represent an altered circulating neutrophil composition during MI and investigating these characteristics will contribute to the understanding of phenotypic and functional differences of neutrophil subsets in MI. Mean neutrophil volume (MNV) represents the average size of the circulating neutrophil population20. In contrast to many other morphological characteristics, MNV can be easily determined by clinically implemented automated hematological cell analyzers21. Moreover, changes in MNV have been proven to be associated with an increased inflammatory response and is a marker of disease severity in several infectious diseases and trauma, independent of neutrophil numbers20,22–27. Since MI induces an inflammatory response, and MNV is known to be elevated in inflammation-related diseases, in the current study we hypothesize that MNV changes after myocardial infarction in patients suffering from MI. We also hypothesize that possible MNV elevation reflects the extent of cardiac damage and is caused by an altered circulating neutrophil composition. To this extent, we investigated the effect of myocardial ischemiareperfusion injury on MNV in a porcine model of MI and determined if MNV could be a possible marker in the setting of MI.

131

PART TWO CHAPTER 8

METHODS Patient sampling To determine MNV after MI, blood samples from ST-elevation MI (STEMI) patients were compared with samples from stable angina patients. Eighty-four STEMI patients included in the DEB-AMI trial (ClinicalTrials.gov identifier NCT00856765) were retrospectively selected. The exact in- and exclusion criteria of the DEB-AMI study cohort are described elsewhere 28. Most importantly, patients were eligible for inclusion in the DEB-AMI trial only when having evidence of a single culprit lesion in the target vessel and received a PCI within 12 hours after the onset of complaints. This minimizes the chances for confounding effects of previous ischemia on MNV. Patients with peri-procedural cardiac arrest were excluded. Patients between 18 and 80 years of age, suffering from STEMI (diagnosed by the presence of anginal complaints and >1mm ST elevation in >2 contiguous leads or new left bundle branch block) between 2009 and 2013, from whom a white blood cell count was available within 72 hours after the onset of complaints, were included. As a non-ischemic control group, stable angina patients with significant coronary artery disease were selected from the UCORBIO cohort (clinicaltrials.gov identifier: NCT02304744), a biobank of patients undergoing coronary angiography in the University Medical Center in Utrecht, the Netherlands. From 2011 to 2013, patients presenting with stable complaints (either stable angina, dyspnea complaints or silent ischemia) were enrolled from the catheterization laboratories (n = 209)29. In both STEMI and stable patients, MNV was measured by the Cell-Dyn Sapphire (CDSapphire), an automated hematological analyzer (see below). Creatine kinase (CK) was measured routinely after PCI in STEMI patients. Data for this study were obtained from the Utrecht Patient Oriented Database (UPOD). UPOD comprises information on patient demographics, hospital discharge diagnoses, medical procedures, medication orders and laboratory tests for all patients treated at the UMC Utrecht. The UPOD database is described in detail elsewhere 30. All patients provided written informed consent. This study conforms to the declaration of Helsinki. Medical Review Board approval for this research project was obtained at the UMC Utrecht. Porcine sampling All animal experiments were approved by the institutional animal welfare committee and were executed conforming to the ‘Guide for the Care and Use of Laboratory Animals’. A total of 19 female landrace pigs were used in this study. Fourteen pigs (body weight 70.1 SD3.8 kg) were subjected to closed-chest left anterior descending artery (LAD) occlusion for 75 minutes followed by 3 days of reperfusion. Arterial blood was collected at baseline, at the end of the ischemic period and at multiple time points after reperfusion (0, 15, 30, 60 and 120 minutes) and was collected in ethylenediaminetetraacetic (EDTA) containing vacutainers followed by the measurements of MNV with the same type of hematology analyzer that was used for the analysis of human samples. From 4 pigs, additional blood samples were drawn at 4 and 8 hours reperfusion. Furthermore, coronary

132

ELEVATED MNV REFLECTS DAMAGE AFTER MI

sinus blood sampling was performed 15 minutes after reperfusion simultaneously with arterial blood sampling. After 72 hours follow-up, pigs were re-anesthetized and 3D-echocardiography was performed after sternotomy. Five pigs were used for sham procedures, undergoing the same procedures except myocardial ischemia induction. Surgical procedure Pre-treatment and anesthesia protocols have been described in detail elsewhere 31,32. In short, all animals were pre-treated with acetylsalicylic acid for 1 day (320mg loading dose, 80mg/day maintenance), clopidogrel for 3 days (75mg/day) and amiodaron for 10 days (1200mg loading dose, 800mg/day maintenance). All medication was continued until the end of the 72-hour follow-up period. Animals were anesthetized with an intramuscular injection of 0.4mg/kg midazolam, 10mg/kg ketamine and 0.014mg/kg atropine. Venous access was obtained by insertion of an 18G cannula in the ear vein for intravenous administration of 5mg/kg sodiumthiopental. Anesthesia was maintained with intravenous infusion of 0.5mg/kg/h midazolam, 2.5µg/kg/h sufentanyl and 0.1mg/kg/h pancuronium. Pre-operatively, animals received a fentanyl patch (25µg/h). Arterial access was obtained by introduction of an 8F sheath into the carotid artery after surgical exposure. A coronary angiogram of the left coronary tree was acquired using an 8F JL4 guiding catheter (Boston scientific, Natick, MA, USA). An adequately sized balloon was placed distal to the second or third diagonal branch, depending on the anatomy of the coronary artery, and inflated for 75 minutes. Animals were observed for 3 hours post-reperfusion and a permanent catheter was placed in the jugular vein in 4 pigs to allow venous blood sampling at 4 and 8 hours reperfusion. The surgical wound was closed and animals were weaned from anesthesia. Animals were defibrillated in case of ventricular fibrillation (VF). Echocardiography, infarct size and troponin Three-dimensional echocardiography was performed as described before31,32. In short, pigs were re-anesthetized according to the same protocol after 72 hours. Medial sternotomy was performed and a gel-filled flexible sleeve was placed directly on the apex of the heart. An X3-1 transducer on an iE33 ultrasound device (Philips, Eindhoven, The Netherlands) was used to perform the echocardiogram. Images were analyzed offline using QLab 10.1 (3DQ advanced) analysis software. Due to incomplete capture of the LV, one animal was excluded from the analysis. After 3D-echocardiography, animals were sacrificed by exsanguination under anesthesia. The heart was excised and the LV was cut into 5 equal slices from apex to base. Slices were incubated in 1% TTC (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37°C 0.9%NaCl for 10 minutes to discriminate between infarct tissue and viable myocardium. After incubation, photographs of the slices were made and the infarct size as a ratio of the left ventricle (LV) was quantified using ImageJ software (NIH, Bethesda, MD, USA). Troponin was measured in plasma isolated from blood drawn after 2 hours reperfusion, using a UniCel DxI Immunoassay system (Beckman Coulter, Brea, CA, USA) with a paramagnetic particle, chemiluminescent immunoassay (Beckman Coulter, Brea, CA, USA).

133

PART TWO CHAPTER 8

Sham experiments Five animals were subjected to sham experiments. Pre-treatment and anesthesia were similar to the protocol described above. Arterial access was again obtained by introduction of an 8F sheath into the carotid artery after surgical exposure. A coronary angiogram of the left coronary tree was acquired using an 8F JL4 guiding catheter (Boston scientific, Natick, MA, USA). Animals were observed for 6 hours (corresponding to the observation period of the animals subjected to MI), followed by exsanguination. Cell-Dyn Sapphire â&#x20AC;&#x201C; MNV measurements For the measurements of MNV in both patients and pigs, the Cell-Dyn Sapphire (Abbott diagnostics, Santa-Clara, CA, USA) was used. This device is a routine hematology analyzer, which allows for whole-blood analysis using spectrophotometry, electrical impedance and laser light scattering (Multi angle polarized scatter separation, M.A.P.S.S.) to classify blood cells (platelets, erythrocytes and leukocytes). The device comprises detectors that are used for optical light scattering to measure different parameters that represent morphological characteristics (e.g. cellular volume, granularity, nuclear shape) of cells. Based on combinations of these parameters, cells are automatically classified into different subpopulations (neutrophils, monocytes, lymphocytes, eosinophils and basophils) according to pre-set gates incorporated into the software of the device21,33,34. For the measurements of MNV, a routine white blood cells count was performed for both patient and pig samples. Flow cytometry measurements in STEMI patients To assess if STEMI patients showed increased circulating neutrophil subsets, flow cytometry analysis in two patients was performed. Blood was collected in EDTA filled vacutainers and co-incubated with a mouse anti-human CD62L-ECD monoclonal antibody (Beckman Coulter, Brea, CA, USA) and a monoclonal CD16-APC-A700 antibody (Beckman Coulter, Brea, CA, USA) for 30 minutes. After incubation, blood was lysed for 10 minutes using Optilyse C (Beckman Coulter, Brea, CA, USA) followed by flow-cytometry measurements on the GalliosTM system (Beckman Coulter, Brea, CA, USA). Identification of neutrophil subsets in pigs For the identification of neutrophil subsets in pigs, measurements of CD62L/CD16 expression on porcine neutrophils was performed using the Cell-Dyn Sapphire in samples drawn from a subset of 4 pigs. A supplementary analyzer processing option normally used for the automated determination of CD3+ CD4+ T-Helper and CD3+ CD8+ T-Suppressor cells, uses a modified analytical approach that combines simultaneous measurements of optical (necessary to determine MNV) and fluorescent (FL1/FL2, for CD62L/CD16 expression analysis) characteristics 35,36. Whole blood samples were co-incubated for 20 minutes with a commercially obtained directly labeled mouse-anti pig monoclonal antibodies against Fc_RIII (CD16- phycoerythrin (PE), clone G7, AbdSerotec, Germany)) and, recently by our laboratory developed, mouse anti-pig monoclonal antibody against

134

ELEVATED MNV REFLECTS DAMAGE AFTER MI

L-selectin (CD62L)(Abmart Inc. Shanghai, China). The unconjugated CD62L antibody was labeled with Alexa-488 (Life Technologies, Carlsbad, CA, USA) followed by automated measurements on the hematology analyzer. Porcine cells sorting and visualization For morphological examination of neutrophil subsets, additional blood samples (n=4) were collected in EDTA vacutainers. Red blood cells were lysed using 4ºC isotonic NH4Cl. Remaining white blood cells were washed with phosphate buffered saline supplemented with sodium citrate (0.4% wt/vol) and pasteurized plasma protein solution (10% vol/vol) (PBS2+). After washing and centrifugation at 1500 rpm for 5 minutes at 4oC, cells were resuspended in PBS2+ and kept on ice until incubation with antibodies. Resuspended cells were incubated at 4oC for 45 minutes with commercial obtained directly labeled mouse anti-pig monoclonal antibodies against FcyRIII (CD16- phycoerythrin (PE), clone G7, AbDSerotec, Germany) as well as the same mouse anti-pig monoclonal antibody against L-selectin (CD62L) as described before. Thereafter cells were washed and sorted on a MoFloAstrios (Beckman Coulter, Brea, CA, USA). Blood neutrophils were identified by their distinct forward and sideward scatter profiles. This yielded a neutrophil purity over 98%. Thereafter, subsets were sorted based on their CD16 and CD62L expression patterns. In order to determine morphological characteristics of PMN subsets, a total of 100,000 cells per subset were isolated and used for cytospin slides and stained with May-Gruenwald & Giemsa. Determination of MNV after local activation of neutrophils Porcine whole blood from healthy pigs (n=4) was stimulated in vitro with either different concentrations of lipopolysaccharide (LPS, Sigma, Escherichia coli 055:B5) (10ng, 100ng, 1µg) or phosphate buffered saline (control). After 2 hours of incubation, samples were automatically measured by the automated hematological analyzer to determine MNV as described before. To assess if the damaged myocardium could influence MNV levels, we performed coronary sinus sampling 15 minutes after reperfusion in pigs subjected to MI. A 4F-catheter was temporarily placed in the coronary sinus through an introducer sheath in the jugular vein to draw blood 15 minutes after reperfusion while blood was drawn simultaneously from the aorta as a reference. Data-analysis and statistics Files from human samples were analyzed by the internal Cell-Dyn Sapphire algorithm according to pre-set gates for different leukocyte subsets. Data from porcine samples were extracted and analyzed with manually set gates (Kaluza, Beckman Coulter, Brea, CA, USA), since the incorporated pre-set gates of the hematological analyzer were not suitable for porcine circulating cells. Data are expressed as means +/- standard deviations unless mentioned otherwise. Groups were compared using Student’s t-test for unrelated measurements in case of comparisons of two separate groups and a paired Student’s t-test for related measurements. Correlations were tested with Pearson’s correlation test.

135

PART TWO CHAPTER 8

RESULTS MNV is elevated in patients post-MI MNV was significantly higher in STEMI patients (n=84) compared to stable angina patients (n=209) that served as a control group (602 SD26 arbitrary units (AU) vs. 580 SD20AU, p<0.0001) (figure 1A). MNV also correlated significantly with simultaneously measured CK levels (R=0.357, p<0.0001) in STEMI patients (figure 1B). MNV is elevated in pigs post-MI To investigate the change of MNV in a controlled, prospective setting, 14 pigs were subjected to MI. One pig died due to resistant VF. In the remaining 13 pigs, MNV was elevated in a time-dependent manner post-MI (451 SD11AU (baseline, n=13) vs. 460 SD9AU, p=0.011 (15 min. reperfusion, n=13) vs. 469 SD12AU, p<0.0001 (2hr reperfusion, n=13) vs. 483 SD11AU, p=0.009 (4h reperfusion, n=4) vs. 490 SD17AU, p=0.014 (8h reperfusion, n=4) (figure 1C). Moreover, Î&#x201D;MNV between baseline and follow-up measurements differed significantly when comparing pigs subjected to MI and sham operated pigs after a similar follow-up period of 4 hours (20.4 SD11.1AU vs. 7.0 SD7.8AU, p=0.026) (figure 1D). A

700

680

R=0.357

640

MNV (AU)

650 600

560

550 500

Stable AP

520

STEMI

CK (Log scale AU)

520

MNV (AU)

480

** 30

ď &#x201E;MNV (AU)

500 **

460

20 10

440 ischemia

420

600

reperfusion

200

400

600

Sham

Time from the onset of ischemia (min)

Figure 1. MNV is elevated after MI A. MNV is higher in STEMI patients compared to stable angina patients. B. MNV correlates with logtransformed CK levels in STEMI patients. C. MNV is elevated in pigs after subjection to myocardial infarction. D. MNV is higher in pigs subjected to MI than in pigs undergoing sham operation after a similar observation period. * = P<0.0001, ** p<0.05.

136

ELEVATED MNV REFLECTS DAMAGE AFTER MI

B 480

480

470

470 MNV (AU)

MNV (AU)

460 450 440

450 R=0.705

Infarct size (as % of LV)

440

R=0.685

60 70 80 ESV(mL) at 3 days reperfusion

D 480

480

470

MNV (AU)

460

460 450 440

460 450

R=-0.718

40 45 EF(%) at 3 days reperfusion

440

R=0.673

1.0 10 6 2.0 10 6 TnI (pg/mL)

3.0 10 6

Figure 2. MNV correlates to cardiac damage in a porcine model of MI A. MNV correlates with infarct size as a percentage of the left ventricle. B. MNV correlates with ESV measured by 3D-echocardiography 3 days after reperfusion. C. MNV correlates with EF measured by 3D-echocardiography 3 days after reperfusion. D. MNV correlates with TnI levels measured 2 hours after reperfusion. Pg = pictograms, mL = milliliter.

To assess whether MNV related to the amount of myocardial damage, MNV was correlated to infarct size as a percentage of the left ventricle (IS/LV) and cardiac function as assessed by 3D-echocardiography 72 hours post-MI, combined with correlation to troponin I (TnI) levels measured in plasma 2 hours post-MI. MNV measured from blood drawn at 15 minutes reperfusion correlated significantly with IS/LV (R=0.705, p=0.007)(figure 2A), end systolic volume (ESV) (R=0.685, p=0.014) (figure 2B) and inversely correlated with left ventricular ejection fraction (LVEF) (R=-0.718, p=0.009) (figure 2C) measured by 3D-echocardiography on day 3. MNV also positively correlated with TnI (R=0.673, p=0.023) (figure 2D). Altered composition of circulating neutrophils contributes to MNV elevation To investigate whether the increased MNV was due to the appearance of circulating neutrophil subsets, in particular banded and hypersegmented neutrophils, we performed CD62L/CD16 flow cytometry analysis of circulating neutrophils of a healthy control (figure 3A) and two STEMI patients (figure 3B&3C). Since both patients revealed the presence of a more heterogeneous population, we also determined the presence of neutrophils subsets

137

PART TWO CHAPTER 8

at baseline (figure 3D) and post-MI (figure 3E) in pigs. Consecutive sorting and visualization of the different subtypes confirmed the presence of banded (CD16dim/CD62Lhigh), mature (CD16high/CD62Lhigh) and hypersegmented (CD16high/CD62Ldim) neutrophils in pigs post-MI (figure 3E). Quantification of subset numbers revealed an absolute increase in mature neutrophils (3.3 SD0.93*106cells/mL vs. 15.3 SD2.97*106cell/mL, p<0.0001), banded neutrophils (0.36 SD0.16*106cell/mL vs. 5.6 SD2.36*106cells/ml, p=0.001) and hypersegmented neutrophils (0.22 SD0.08*106cells/ml vs. 0.82 SD0.39cells/ml, p=0.011) between baseline and 8 hours reperfusion in pigs subjected to MI (figure 4A). Moreover, relative contribution of subsets also changed over time. At 8 hours reperfusion, mature neutrophils contributed less to the total population of circulating neutrophils than at baseline (70.8 SD5.6% vs. 84.4 SD5.5%, p=0.008). On the other hand, the contribution of banded neutrophils increased when 8 hours reperfusion was compared to baseline (25.2 SD6.4% vs. 9.78 SD5.6%, p=0.006), while the contribution of hypersegmented neutrophils remained unchanged (3.93 SD1.91% vs. 5.78 SD1.56%, p=0.152) (figure 4B). To assess whether this altered composition of circulating neutrophils could, in part, be responsible for an altered MNV, the MNV of different neutrophil subtypes was determined by an automated hematological analyzer. Post-MI, MNV of banded neutrophils was significantly higher than MNV of mature neutrophils (495 SD14 AU vs. 478 SD11 AU, p=0.012) (figure 4C). Activation of neutrophils results in higher mean neutrophil volume Not only banded neutrophils, but also mature neutrophils after MI showed a higher MNV than mature neutrophils at baseline, indicating the presence of another mechanism underlying these changes. Therefore, we assessed if local activation of neutrophils could alter MNV by stimulating whole blood of healthy pigs with 3 different doses of LPS (10ng, 100ng, 1µg) or PBS in vitro. After incubation of 2 hours at 37ºC, a dose-dependent increase of MNV was seen (454 SD6.8 AU (control) vs. 470 SD8.7 AU (LPS 1 µg), p=0.013) (figure 5A). Moreover, we determined if MNV was also altered due to the interaction between neutrophils and the damaged myocardium in vivo. To this extent, coronary sinus sampling was performed simultaneously with aortic sampling 15 minutes after reperfusion. MNV of neutrophils measured from coronary sinus blood was significantly higher than MNV of neutrophils from simultaneously collected arterial blood (463 SD7.6 AU vs 461 SD8.6 AU, p=0.013) (figure 5B).

DISCUSSION The detrimental effects of the inflammatory response post-MI are believed to outweigh the positive effects37. Indeed, multiple clinical studies have shown that different parameters of inflammation correlate with prognosis and adverse events after MI6,9,38,39. Since neutrophils play a prominent role in myocardial ischemia-reperfusion injury40, and MNV is known to be elevated in inflammation-related diseases23,24, we hypothesized that MNV

138

ELEVATED MNV REFLECTS DAMAGE AFTER MI

CD16

CD62L

CD16

Figure 3. Scatterplots based on CD62L and CD16 expression of circulating neutrophils of STEMI patients and pigs subjected to MI A. Scatterplot of circulating neutrophils of a healthy control. B&C. Scatterplot of circulating neutrophils of patients within 72 hours post-MI. D. Scatterplot of circulating neutrophils of a pig before subjection to MI. E. Scatterplot of a pig 4 hours after reperfusion with representative pictures of morphological differences between banded neutrophils (green box), mature neutrophils (gray box) and hypersegmented neutrophils (red box).

could serve as a possible marker in the setting of MI. In this perspective, the current study sought to investigate whether MNV is elevated post-MI and what the possible explanation behind this elevation could be. To the best of our knowledge, this is the first study to show that MNV is elevated in STEMI patients in response to myocardial infarction. Moreover, MNV correlates with CK levels, suggesting possible reflection of the amount of cardiac damage post-MI. Similar to the patient setting, MNV is elevated in pigs and MNV levels early after MI correlate with multiple parameters of cardiac damage measured after 3 days follow-up. These findings propose that MNV could withhold prognostic value in the setting of MI. Additionally we investigated for the first time, whether a change in MNV was the result of an altered neutrophil composition, or the activation of circulating neutrophils. It has recently been corroborated that migration of specific neutrophil subsets into the systemic circulation occurs in inflammation related diseases41â&#x20AC;&#x201C;44. However, ours is the first study that identifies this phenomenon after MI in patients and morphologically confirmed these subsets in a porcine model of myocardial infarction.

139

PART TWO CHAPTER 8

Although previously suggested by others26, we here provide the first direct evidence that this altered circulating neutrophil composition contributes to the elevated MNV. We have shown that banded neutrophils have a significantly higher MNV than mature neutrophils and account for more than a quarter of the circulating neutrophils post-reperfusion in pigs. Similar levels of banded neutrophils have been found in human subjects after in vivo LPS infusion, suggesting that our porcine model is representative for the human situation as well41. In contrast to the previous assumption that the increased MNV in inflammation-related diseases is only due to transmigration of banded neutrophils22,25,26, we show that an altered subset composition is not the sole mechanism, since both in vitro and in vivo neutrophil activation results in significant MNV elevation. This phenomenon underscores that morphological characteristics of neutrophils could withhold great prognostic value post-MI, being directly and imminently influenced by the damaged myocardium. To what extent both processes are exactly responsible for the change in MNV is hard to determine since they are very dynamic and time-dependent45. From our data and those A

Figure 4 Banded neutrophils (relatively) increase during reperfusion and have a higher MNV in pigs subjected to MI A. Absolute increase in banded (green), mature (gray) and hypersegmented (red) neutrophils during reperfusion. B. Relative increase of banded neutrophils (green) and decrease of mature neutrophils (gray) during reperfusion. C. Banded neutrophils (green) have a higher MNV compared to mature neutrophils (gray), dotted line represents average MNV at baseline. R=reperfusion, Mt=mature, Bd=banded, HS=hypersegmented neutrophils, *p<0.05

140

ELEVATED MNV REFLECTS DAMAGE AFTER MI

MNV (AU)

0 0

Control

LPS 10ng LPS 100ng

LPS 1g

Aorta

Coronary Sinus

Figure 5. Local activation of neutrophils results in an increased MNV A. In vitro stimulation with LPS dose dependently increases MNV. B. MNV in blood drawn from the coronary sinus is higher than MNV from simultaneously drawn blood from the aorta. *p<0.05, **p=0.09

of others, however, it is clear that the migration of subsets towards the systemic circulation occurs within a few hours from the initial event41. We also show that activation of neutrophils by the damaged myocardium already occurs during ischemia and early reperfusion. The contribution of banded neutrophils to the total neutrophil population is limited within this timeframe. This suggests that both processes overlap but that activation is probably playing a more dominant role in the first phase after cardiac ischemia. This seems logical since many different danger molecules are released from the damaged myocardium into the systemic circulation from the onset of ischemia onwards4,40,46. The current study advocates MNV as a possible prognostic marker after MI and may lead to further insight into the mechanism behind reperfusion damage and myocardial remodeling post-MI. Since neutrophils are prominent players in myocardial reperfusion injury40, it remains to be elucidated whether MNV is influenced by cardioprotective strategies, or could even be causally involved in the process of infarct expansion. In our study, we established correlations with cardiac damage early after reperfusion. Additionally, the association of MNV with adverse events after MI, may not only depend on cardiac damage but also on comorbidity and risk factors not present in the large animal model we used. The prognostic value should therefore be validated in a large prospective patient cohort. Apart from morphological differences of different neutrophil subsets, also phenotypical differences have been identified19,41. What the distinctive role of the different neutrophil subsets in the setting of myocardial injury is, remains to be elucidated. Whether the presence of a heterogeneous population of neutrophils is associated with future adverse events after MI will be part of future studies and falls beyond the scope of our study. In conclusion, the current study shows that MNV is elevated after MI and provides evidence that MNV predicts cardiac damage post-MI. Moreover, we show that MNV not only increases due to altered neutrophil composition, but also systemic neutrophil activation.

141

PART TWO CHAPTER 8

Given the fast and inexpensive nature of MNV measurements with already clinically implemented devices, MNV may be an interesting parameter for future prognostic assessment in MI. Acknowledgements The authors gratefully acknowledge Marlijn Jansen, Joyce Visser, Martijn van Nieuwburg, Grace Croft, Evelyn Velema, Marjon Wesseling and Ron Stokwielder for their excellent technical support. All institutional and national guidelines for the care and use of laboratory animals were followed and approved by the appropriate institutional committees. Grant support: This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs. Conflict of interest: The authors report no relationships that could be construed as a conflict of interest.

142

ELEVATED MNV REFLECTS DAMAGE AFTER MI

REFERENCES 1. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 2. Kristensen SD, Laut KG, Fajadet J, Kaifoszova Z, Kala P, Di Mario C, Wijns W, Clemmensen P, Agladze V, Antoniades L, Alhabib KF, De Boer MJ, Claeys MJ, Deleanu D, Dudek D, Erglis A, Gilard M, Goktekin O, Guagliumi G, Gudnason T, Hansen KW, Huber K, James S, Janota T, Jennings S, Kajander O, Kanakakis J, Karamfiloff KK, Kedev S, Kornowski R, Ludman PF, Merkely B, Milicic D, Najafov R, Nicolini F a., Noč M, Ostojic M, Pereira H, Radovanovic D, Sabaté M, Sobhy M, Sokolov M, Studencan M, Terzic I, Wahler S, Widimsky P. Reperfusion therapy for ST elevation acute myocardial infarction 2010/2011: Current status in 37 ESC countries. Eur Heart J. 2014;35:1957–1970. 3. Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res. 2002;53:31–47. 4. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 5. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 6. Van der Laan a. M, Nahrendorf M, Piek JJ. Healing and adverse remodelling after acute myocardial infarction: role of the cellular immune response. Heart. 2012;98:1384–1390. 7. Dragu R, Huri S, Zuckerman R, Suleiman M, Mutlak D, Agmon Y, Kapeliovich M, Beyar R, Markiewicz W, Hammerman H, Aronson D. Predictive value of white blood cell subtypes for long-term outcome following myocardial infarction. Atherosclerosis. 2008;196:405–412. 8. Husser O, Bodi V, Sanchis J, Nunez J, Mainar L, Chorro FJ, Lopez-Lereu MP, Monmeneu JV, Chaustre F, Forteza MJ, Trapero I, Dasi F, Benet I, Riegger G a J, Llacer A. White blood cell subtypes after STEMI: temporal evolution, association with cardiovascular magnetic resonance--derived infarct size and impact on outcome. Inflammation. 2011;34:73–84. 9. Horne BD, Anderson JL, John JM, Weaver A, Bair TL, Jensen KR, Renlund DG, Muhlestein JB. Which white blood cell subtypes predict increased cardiovascular risk? J Am Coll Cardiol. 2005;45:1638–1643. 10. Tamhane UU, Aneja S, Montgomery D, Rogers EK, Eagle K a., Gurm HS. Association Between Admission Neutrophil to Lymphocyte Ratio and Outcomes in Patients With Acute Coronary Syndrome. Am J Cardiol. 2008;102:653–657. 11. Núñez J, Núñez E, Bodí V, Sanchis J, Miñana G, Mainar L, Santas E, Merlos P, Rumiz E, Darmofal H, Heatta AM, Llàcer A. Usefulness of the neutrophil to lymphocyte ratio in predicting long-term

mortality in ST segment elevation myocardial infarction. Am J Cardiol. 2008;101:747–752. 12. Duffy BK, Gurm HS, Rajagopal V, Gupta R, Ellis SG, Bhatt DL. Usefulness of an elevated neutrophil to lymphocyte ratio in predicting long-term mortality after percutaneous coronary intervention. Am J Cardiol. 2006;97:993–996. 13. Papa A, Emdin M, Passino C, Michelassi C, Battaglia D, Cocci F. Predictive value of elevated neutrophil-lymphocyte ratio on cardiac mortality in patients with stable coronary artery disease. Clin Chim Acta. 2008;395:27–31. 14. Liebetrau C, Hoffmann J, Dorr O, Gaede L, Blumenstein J, Biermann H, Pyttel L, Thiele P, Troidl C, Berkowitsch a., Rolf a., Voss S, Hamm CW, Nef H, Mollmann H. Release Kinetics of Inflammatory Biomarkers in a Clinical Model of Acute Myocardial Infarction. Circ Res. 2014;116:867–875. 15. Palmerini T, Généreux P, Mehran R, Dangas G, Caixeta A, Riva D Della, Mariani A, Xu K, Stone GW. Association among leukocyte count, mortality, and bleeding in patients with non-st-segment elevation acute coronary syndromes (from the Acute catheterization and urgent intervention triage StrategY [ACUITY] Trial). Am J Cardiol. 2013;111:1237–1245. 16. Azab B, Zaher M, Weiserbs KF, Torbey E, Lacossiere K, Gaddam S, Gobunsuy R, Jadonath S, Baldari D, McCord D, Lafferty J. Usefulness of neutrophil to lymphocyte ratio in predicting short-and long-term mortality after NonST-elevation myocardial infarction. Am J Cardiol. 2010;106:470–476. 17. Sen N, Afsar B, Ozcan F, Buyukkaya E, Isleyen A, Akcay AB, Yuzgecer H, Kurt M, Karakas MF, Basar N, Hajro E, Kanbay M. The neutrophil to lymphocyte ratio was associated with impaired myocardial perfusion and long term adverse outcome in patients with ST-elevated myocardial infarction undergoing primary coronary intervention. Atherosclerosis. 2013;228:203–210. 18. Beyrau M, Bodkin J V, Nourshargh S. Neutrophil heterogeneity in health and disease: a revitalized avenue in inflammation and immunity. OpenBiol. 2012;2:120134. 19. Scapini P, Cassatella M a. Social networking of human neutrophils within the immune system. Blood. 2014;124:710–719. 20. Bagdasaryan R, Zhou Z, Tierno B, Rosenman D, Xu D. Neutrophil VCS parameters are superior indicators for acute infection. Lab Hematol. 2007;13:12–16. 21. De Smet D, Van Moer G, Martens G a., Nanos N, Smet L, Jochmans K, De Waele M. Use of the celldyn sapphire hematology analyzer for automated counting of blood cells in body fluids. Am J Clin Pathol. 2010;133:291–299. 22. Mardi D, Fwity B, Lobmann R, Ambrosch a. Mean cell volume of neutrophils and monocytes compared with C-reactive protein, Interleukin-6 and white blood cell count for prediction of sepsis

143

PART TWO CHAPTER 8

and nonsystemic bacterial infections. Int J Lab Hematol. 2010;32:410–418. 23. Bhargava M, Saluja S, Sindhuri U, Saraf a., Sharma P. Elevated mean neutrophil volume+CRP is a highly sensitive and specific predictor of neonatal sepsis. Int J Lab Hematol. 2014;36. 24. Lee a. J, Kim SG. Mean cell volumes of neutrophils and monocytes are promising markers of sepsis in elderly patients. Blood Res. 2013;48:193–197. 25. Celik IH, Demirel G, Aksoy HT, Erdeve O, Tuncer E, Biyikli Z, Dilmen U. Automated determination of neutrophil VCS parameters in diagnosis and treatment efficacy of neonatal sepsis. Pediatr Res. 2012;71:121–125. 26. Zhu Y, Cao X, Zhang K, Xie W, Xu D, Zhong C. Delta Mean Neutrophil Volume (ΔMNV) is Comparable to Procalcitonin for Predicting Postsurgical Bacterial Infection. J Clin Lab Anal. 2014;28:301–305. 27. Lam SW, Leenen LPH, van Solinge WW, Hietbrink F, Huisman A. Comparison between the prognostic value of the white blood cell differential count and morphological parameters of neutrophils and lymphocytes in severely injured patients for 7-day in-hospital mortality. Biomarkers. 2012;17:642–647. 28. Belkacemi A, Agostoni P, Nathoe HM, Voskuil M, Shao C, Van Belle E, Wildbergh T, Politi L, Doevendans P a., Sangiorgi GM, Stella PR. First results of the DEB-AMI (Drug Eluting Balloon in Acute ST-segment elevation myocardial infarction) Trial: A multicenter randomized comparison of drug-eluting balloon plus bare-metal stent versus bare-metal stent versus drug-eluting stent in primary p. J Am Coll Cardiol. 2012;59:2327–2337. 29. Gijsberts CM, Gohar A, Ellenbroek GHJM, Hoefer IE, de Kleijn DPV, Asselbergs FW, Nathoe HM, Agostoni P, Rittersma SZH, Pasterkamp G, Appelman Y, den Ruijter HM. Severity of stable coronary artery disease and its biomarkers differ between men and women undergoing angiography. Atherosclerosis. 2015; 30. Ten Berg MJ, Huisman A, Van Den Bemt PML a, Schobben AF a M, Egberts ACG, Van Solinge WW. Linking laboratory and medication data: New opportunities for pharmacoepidemiological research. Clin Chem Lab Med. 2007;45:13–19. 31. De Jong R, van Hout GPJ, Houtgraaf JH, Kazemi K, Wallrapp C, Lewis a., Pasterkamp G, Hoefer IE, Duckers HJ. Intracoronary Infusion of Encapsulated Glucagon-Like Peptide-1-Eluting Mesenchymal Stem Cells Preserves Left Ventricular Function in a Porcine Model of Acute Myocardial Infarction. Circ Cardiovasc Interv. 2014;7:673–683. 32. Van Hout GPJ, Jansen Of Lorkeer SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE, Jansen SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE. Admittance-based pressure-volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014;2:e00287.

144

33. Hedberg P, Lehto T. Aging stability of complete blood count and white blood cell differential parameters analyzed by Abbott CELL-DYN Sapphire hematology analyzer. Int J Lab Hematol. 2009;31:87–96. 34. Müller R, Mellors I, Johannessen B, Aarsand AK, Kiefer P, Hardy J, Kendall R, Scott CS. European multi-center evaluation of the Abbott Cell-Dyn sapphire hematology analyzer. Lab Hematol. 2006;12:15–31. 35. Johannessen B, Roemer B, Flatmoen L, Just T, Aarsand a. K, Scott CS. Implementation of monoclonal antibody fluorescence on the Abbott CELL-DYN Sapphire haematology analyser: Evaluation of lymphoid, myeloid and platelet markers. Clin Lab Haematol. 2006;28:84–96. 36. Groeneveld KM, Heeres M, Leenen LPH, Huisman A, Koenderman L. Immunophenotyping of posttraumatic neutrophils on a routine haematology analyser. Mediators Inflamm. 2012;2012:509513. 37. Seropian IM, Toldo S, Van Tassell BW, Abbate A. Anti-inflammatory strategies for ventricular remodeling following St-segment elevation acute myocardial infarction. J Am Coll Cardiol. 2014;63:1593–1603. 38. Takahashi T, Hiasa Y, Ohara Y, Miyazaki S-I, Ogura R, Suzuki N, Hosokawa S, Kishi K, Ohtani R. Relationship of admission neutrophil count to microvascular injury, left ventricular dilation, and long-term outcome in patients treated with primary angioplasty for acute myocardial infarction. Circ J. 2008;72:867–872. 39. Maekawa Y, Anzai T, Yoshikawa T, Asakura Y, Takahashi T, Ishikawa S, Mitamura H, Ogawa S. Prognostic significance of peripheral monocytosis after reperfused acute myocardial infarction:a possible role for left ventricular remodeling. J Am Coll Cardiol. 2002;39:241–246. 40. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 41. Pillay J, Kamp VM, Van Hoffen E, Visser T, Tak T, Lammers JW, Ulfman LH, Leenen LP, Pickkers P, Koenderman L. A subset of neutrophils in human systemic inflammation inhibits T cell responses through Mac-1. J Clin Invest. 2012;122:327–336. 42. Hao S, Andersen M, Yu H. Detection of immune suppressive neutrophils in peripheral blood samples of cancer patients. Am J Blood Res. 2013;3:239–45. 43. Welin A, Amirbeagi F, Christenson K, Björkman L, Björnsdottir H, Forsman H, Dahlgren C, Karlsson A, Bylund J. The Human Neutrophil Subsets Defined by the Presence or Absence of OLFM4 Both Transmigrate into Tissue In Vivo and Give Rise to Distinct NETs In Vitro. PLoS One. 2013;8:e69575. 44. Tsuda Y, Takahashi H, Kobayashi M, Hanafusa T, Herndon DN, Suzuki F. Three different neutrophil subsets exhibited in mice with different

ELEVATED MNV REFLECTS DAMAGE AFTER MI

susceptibilities to infection by methicillin-resistant Staphylococcus aureus. Immunity. 2004;21:215– 226. 45. Wirths S, Bugl S, Kopp HG. Neutrophil homeostasis and its regulation by danger signaling. Blood. 2014;123:3563–3566. 46. Lefer DJ. Do neutrophils contribute to myocardial reperfusion injury? Basic Res Cardiol. 2002;97:263– 267.

145

part three

THE SEARCH FOR NOVEL THERAPEUTIC STRATEGIES

chapter 9

Targeting danger associated molecular patterns after myocardial infarction

Expert Opinion on Therapeutic Targets - Accepted August 2015

G.P.J. van Hout1, F. Arslan1, G. Pasterkamp1,2 , I.E. Hoefer1,2 1 2

Experimental Cardiology Laboratory, University Medical Center Utrecht, The Netherlands Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, The Netherlands

PART THREE CHAPTER 9

ABSTRACT Introduction Myocardial infarction (MI) provokes an intense inflammatory response that can lead to left ventricular adverse remodeling and heart failure (HF). The prognosis of HF patients is poor and related to a decreased quality of life and considerable health care costs. Hence, targeting the early inflammatory response after MI provides an interesting target to attenuate left ventricular remodeling and prevent HF. Areas covered In the current review we discuss the theory that our immune system does not distinguish between self and non-self, but rather senses danger. So-called danger-associated molecular patterns (DAMPs) serve as ligands for pattern recognition receptors (PRRs), which act as signal transduction molecules to induce a pro-inflammatory state. Many different DAMPs and PRRs have been identified recently. Here, we provide a concise overview of their interactions as well as their role in the inflammatory response after MI. Expert opinion Interference with Toll-like receptor (TLR) 2, TLR4 and NLRP3-inflammasome signaling has consistently shown to reduce infarct size and preserve cardiac function post-MI in experimental animal models. Since clinically applicable inhibitors have been developed for these pathways, the path has been cleared to assess whether these promising results can be translated into the human situation. Article highlights • Myocardial infarction induces a sterile inflammatory immune response that is detrimental to cardiac function. • Upstream interaction between DAMPs and pattern recognition receptors drives the cardiac inflammatory response. • Many different DAMPs have been identified that can be subdivided into constitutive and inducible. • TLRs, NOD-like receptors and RAGE have so far been identified as receptors that are able to recognize DAMPs and facilitate signal transduction to induce a pro-inflammatory state. • Especially TLR2, TLR4 and NLRP3-inflammasome mediated signaling are promising targets for clinical therapeutics and selective inhibitors have been developed to enable future clinical testing and investigate the efficacy of blocking these pathways in post-MI patients. • Since many anti-inflammatory compounds were successful in experimental studies but failed in clinical trials, optimization of clinically relevant experimental study design is needed to increase the chance of translational success and future validation of these mechanisms.

150

TARGETING DAMPS AFTER MI

INTRODUCTION Survival rates after acute myocardial infarction (MI) have significantly increased over the last decades due to improved medical treatment and reperfusion strategies1,2. However, patients surviving MI often have a deteriorated cardiac function and are therefore at risk for the development of heart failure (HF). HF is a chronic disorder with a poor prognosis that is comparable to or worse than most cancers, a decreased quality of life and considerable health care costs3. Unfortunately, the intrinsic regenerative capacity of the human heart is very limited4. After MI, the damaged cardiac tissue is replaced by collagen-based scar tissue that influences cardiac physiology. The extent and character of the physiological consequences is directly proportional and related to the degree of tissue injury (i.e. infarct size) and quality of wound healing processes. To this extent, fast and thorough repair of the damaged myocardium is mandatory. The process of cardiac repair is often divided in three, partially overlapping, chronological phases: The inflammatory phase, the proliferative phase and the maturation phase5. The inflammatory phase is required to initiate cardiac wound healing. Inflammation, through e.g. macrophage infiltration, enables the removal of necrotic tissue and promotes angiogenesis. There is accumulating evidence that in the acute setting, directly following cardiac ischemia, the inflammatory response is detrimental to cardiac function and overall prognosis6â&#x20AC;&#x201C;9. Immediately after ischemic injury, the damaged tissue releases danger associated molecular patterns (DAMPs). DAMPs binding to pattern recognition receptors (PRRs) induce inflammatory pathways and stimulate cytokine secretion. Amongst others, these cytokines promote upregulation of selectins and integrins by endothelial and circulating cells, which infiltrate the myocardium (figure 1). Though being attracted to remove cellular debris, leukocytes also attack damaged but still viable cardiomyocytes after MI. This results in secondary infarct extension, leading to wall thinning and progressive geometrical adaptation of the left ventricle. Eventually, this process, referred to as adverse cardiac remodeling, can culminate in HF10. Since the acute inflammatory response can have detrimental effects, inhibition of postischemia interaction of DAMPs and PRRs may serve as a clinical target to reduce infarct size and preserve cardiac function. In the current review we provide a detailed overview of the known interactions of DAMPs and PRRs, evaluate how modulating these pathways could preserve cardiac function and prevent HF and identify the most promising targets for clinical translation. Pattern recognition PRRs act as sensors for danger molecules such as DAMPs and PAMPs (pathogen associated molecular patterns) and generally induce the production of pro-inflammatory cytokines. DAMPs can act on multiple PRRs and multiple PRRs can share the same ligand. Also, activation of different PRRs may lead to the same final common pathway and a similar inflammatory response11â&#x20AC;&#x201C;13.

151

PART THREE CHAPTER 9

Toll-like receptors Among the most studied PRRs in cardiovascular disease are the toll-like receptors (TLRs)14. Mammalian TLRs were identified by Medzhitov et al. as an important part of the immune system in 199715, and the interaction between TLRs and DAMPs was first described in 2000 by Ohashi et al16. TLRs contain an ectodomain with leucine-rich repeats for ligand recognition. The ectodomain is linked to transmembrane domains and intracellular Tollinterleukin (TIR) domains for intracellular signal transduction17. Overall, TLRs, which are expressed both on circulating and cardiac resident cells, can be divided into two groups based on their subcellular locations. The TLRs that are predominantly present on the cell surface are TLR1, TLR2, TLR4, TLR5, TLR6 and TLR11. These TLRs act as sensors for a potentially hostile extracellular environment and react to extracellular DAMPs (HSPs, HMGB1, FN-EDA). Intracellular TLRs are TLR3, TLR7, TLR8 and TLR9. These TLRs are responsible for the recognition of danger signals such as microbial nucleic acids or self-DNA.

C D

DAMPs Capillary

Figure 1. Schematic overview of the occurrence of infarct extension after MI Due to ischemia-reperfusion injury the infarct becomes larger according to a wave front principle (dashed small black arrows), with early subendocardial involvement and progression towards a transmural infarction. Due to ischemia-reperfusion, DAMPs are released from the damaged myocardium and enter the systemic circulation (A). Here they activate circulating leukocytes, which subsequently transmigrate out of the circulation into the damaged tissue (B). Once resided, these inflammatory cells cause damage, thereby inducing infarct enlargement (C). D. Histological sample of infarct tissue containing neutrophils in porcine myocardium 72 hours after ischemia-reperfusion injury.

152

TARGETING DAMPS AFTER MI

The stimulation of TLRs by DAMPs leads to the dimerization of the intracellular TIR domain and subsequent activation of myeloid differentiation primary response gene (MyD)88 and nuclear factor (NF)-kB. During normal homeostasis, NF-kB is bound to the inhibitor molecule nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha (IκBα) in the cytoplasm. When activated, IκBα is rapidly phosphorylated, enabling translocation of NF-kB to the nucleus. Once inside the nucleus, NF-kB binds to the DNA, thereby mediating the transcription of pro-inflammatory genes14. Although the MyD88 pathway is the most common pathway, MyD88-independent signaling through activation of predominantly TLR3 or TLR7 has also been observed through the adaptor molecule TIR-domain-containing adaptor inducing interferon-β (TRIF). Eventually, this also leads to NF-kB signaling as well as production of interferon (IFN) by the intracellular signaling molecule interferon regulatory factor (IRF)3 (figure 2)17,18. In cardiovascular disease, and particularly in MI models, TLR2 and TLR4 have received most attention12. TLR2 knockout mice show improved cardiac function, reduced infarct size and myocardial inflammation due to a reduced TLR2 mediated leucocyte influx. Furthermore, TLR2 mediates adverse ventricular remodeling through enhancing inflammation and cardiac fibrosis in both small and large animals19–23. Similar to TLR2, a reduction of TLR4 mediated signaling in mice results in reduced infarct size, decreased inflammation and reduced adverse remodeling20,24–29. Both, knockout and pharmacological studies indicate that TLRs 2 and 4 are promising therapeutic targets to improve cardiac repair and reduce infarct size after MI. The exact role of the other TLRs in the inflammatory response after MI has not been fully established yet. There is increasing evidence that TLR3 also contributes to the adverse effects caused by inflammation after MI. Since TLR3 signaling is different from other TLRmediated inflammation due to its MyD88-independent signal transduction, it represents other down-stream effects17. TLR3 can be bound by RNA and several studies have now shown that blocking this interaction reduces apoptosis and protects against myocardial ischemia-reperfusion injury30–33. Additionally, it has been reported that TLR9 serves as a PRR for DNA, thereby promoting NF-kB mediated pro-inflammatory cytokine secretion34. Although convincing evidence exists that mitochondrial DNA (mt)DNA binds to TLR9 and contributes to the adverse effects caused by inflammation after MI35, there is no consensus whether stimulating this receptor by pretreatment is beneficial after cardiac inflammation36–38. Nod-like receptors Nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs) are a group of cytosolic PRRs that are also known to be involved in inflammatory signaling after activation. All NLRs share a highly conserved structure containing either a pyrin domain (PYD) or caspase recruitment domain (CARD) combined with a central nucleotide binding (NACHT) domain followed by a C-terminal leucine-rich repeat (LRR) domain. While the CARD and PYD domain enable protein-protein interaction, the LRR domain is thought to serve as a sensor for danger signals. The NACHT domain functions as a link between the two and can acquire an active state to facilitate oligomerization39.

153

PART THREE CHAPTER 9

The NLRs can be subdivided into 3 subsets based on evolutionary similarities. The NOD subfamily predominantly involves receptors that are membrane associated and induce NFkB mediated signaling as well as autophagy related pathways. So far, only NOD-1 and NOD-2 have shown to be involved in the induction of cardiac dysfunction after stimulation, without the actual discovery of an endogenous ligand for these receptors 40,41. Similar to the NOD subfamily, the interleukin (IL)-1b converting enzyme protease-activating factor (IPAF) subfamily responds to PAMPs, while no DAMPs have been identified that act on these PRRs. The NOD-like, PYD (NLRP) subfamily is the third group of NLRs. This group, together with some NLRs from the IPAF group, is able to form intracellular macromolecular structures upon activation, termed inflammasomes. Inflammasomes are activated by the accumulation of adenosine triphosphate (ATP) and cell debris after tissue injury 42. Although 14 NLRPs have been identified, only 1 of these NRLPs, NLRP3, has been linked to cardiovascular disease39,43. The formation and activation of the NLRP3-inflammasome involves the NLR protein, the apoptotic speck-containing protein (ASC) and pro-caspase-1. Preceding activation, the NLRP3-inflammasome requires priming through TLR-mediated NF-kB signals 42. Following priming, extracellular ATP and cellular debris results in inflammasome activation and amplification of inflammatory responses and tissue injury by cleavage and activation of caspase-1 13. Caspase-1 then induces the conversion of pro-interleukin (IL)-1b to mature IL-1b and proIL-18 to IL-18 44,45.

RAGE

Extracellular DNA/RNA

HMGB1 / HSPs

MyD88

TRIF

NF-кß

IRF3

TLR2, TLR4

Endosome Pro-inflammatory cytokines

Nucleus TLR9, TLR3 NF-кß

Extracellular ATP

IRF3

mtDNA

IFN

Pro-inflammatory Response

Mitochondrion Cardiolipin

pro-IL-1ß

IL-1ß

pro-IL-18

IL-18

P2X7 receptor

Potassium

NLRP3-inflammsome

Figure 2. Overview of upstream intracellular signaling pathways that lead to a pro-inflammatory response after MI DAMPs (ATP, (mt)DNA, HSPs and HMGB1) activate PPRs (TLRs, RAGE, NLRP3-inflammasome) and induce MyD88 and TRIF signaling. This leads to nuclear migration of NF-κβ and IRF3, resulting in pro-inflammatory cytokine synthesis. IL-1β and IL-18 are separately activated through assembly of the NLRP3-inflammasome that has been primed through initial TLR signaling.

154

TARGETING DAMPS AFTER MI

Reduction of NLRP3-inflammasome signaling results in attenuation of the inflammatory response following MI, since IL-1b and IL-18 are strong mediators that play a role in the early immune reaction. Reduced caspase-1 signaling results in decreased myocardial fibrosis in humans and protects from cardiac damage after MI in caspase-1 knockout mice 46,47 . Similarly, ASC and NLRP3 knockout studies also showed a marked reduction of infarct size and preservation of cardiac function 44,47,48. The responsible cell type for inflammasome mediated signaling is controversial. While cardiac fibroblasts and circulating cells have both been proven to be potent secretors of IL-1b and IL-18, they seem to be equally important in infarct extension triggered by inflammation 47,48. Moreover, inflammasome activation in cardiomyocytes leads to a caspase-1 dependent programmed cell death, referred to as pyroptosis 43,44. Recently, small-molecule inhibitors against the NLRP3-inflammasome have been developed, enabling pharmacological intervention 49,50. Future translational studies will have to determine whether selective inhibition with these compounds are effective and determine if NLRP3-inflammasome inhibition is a promising target for the treatment of patients suffering from MI. Receptor for advanced glycation end-products The receptor for advanced glycation end-products (RAGE) also serves as a multiligand PRR and belongs to the immunoglobulin gene superfamily. RAGE contains a V-type domain to interact with its ligands, two C-type domains, a transmembrane spinning helix and a C-terminal cytosolic domain (ctRAGE), to enable intracellular signal transduction51,52. An isoform lacking the V-type domain, making it irresponsive to stimuli and an isoform lacking the C-terminal cytosolic domain, thereby acting as a decoy receptor, have also been discovered51,53. As its name implies, RAGE acts primarily as a receptor for advanced glycation end-products (AGEs). AGEs have predominantly been under investigation for their role in diabetes and atherosclerosis and can bind to multiple receptors54,55. Unlike RAGE, which serves as an actual signal transduction receptor, these AGE-binding receptors have in common that they contribute to the break down and endocytotis of AGEs. On the other hand, RAGE can also bind other ligands than AGEs, including high mobility group box (HMGB)1 and S100/calgranulins, which are calcium-binding polypeptides, accumulating extracellularly during chronic inflammation54. Once activated, RAGE enables the translocation of NF-kB to the nucleus for transcription of pro-inflammatory cytokines. Additionally, ligation of RAGE induces a positive feedback loop resulting in RAGE receptor upregulation and NF-kB synthesis56. The administration of recombinant HMGB1 aggravates myocardial ischemia reperfusion injury, suggesting that RAGE plays a role in the post-MI immune reaction57. Moreover, macrophages and cardiac fibroblasts exposed to hypoxic conditions in vitro show increased S100A8/A9 mediated RAGE activation58. The same study showed that RAGE knockout mice are protected after myocardial ischemia-reperfusion injury and RAGE mediated signaling is essential for circulating cells to migrate to the myocardium and exert detrimental effects. Accordingly, combined blockade of RAGE signaling by siRNA and soluble RAGE synergistically protected from cardiac damage after MI59.

155

PART THREE CHAPTER 9

Table 1. Known DAMPs and their PRRs DAMPs

PRR

Role in MI

Ref

α and β defensins

TLR4

172,173

Amyloid β

TLR2, TLR4/6, NLRP3

174

APLAs

TLR2

175,176

ATP

NLRP3

Yes

50,54

β defensin 2

TLR4

177

β defensin 3

TLR2

178

Biglycan

TLR2, TLR4

Yes

179,180

Cardiac myosin

TLR2, TLR8

Yes

Complement membrane attack complex

NLRP3

Probable

181,182

Calcium pyrophosphate dihydrate

NLRP3

(mt)DNA

TLR9, NLRP3, AIM2

Yes

41,76

Eosinophil derived neurotoxins

TLR2

183

Fetuin A

TLR4

184

Fibrinogen

TLR4

185

Fibronectin-EDA

TLR2, TLR4

Yes

117

Glucose

NLRP3

Yes

186,187

Gp96

TLR2, TLR4

HMGB1

TLR2, TLR4, TLR9, RAGE

Yes

31,111

Heparan sulfate

TLR4

188

HSP22

TLR4

HSP60

TLR2, TLR4

Yes

91,95

HSP70

TLR2, TLR4

Yes

97,99

HSP72

TLR4

Yes

Hyaluronan (fragments)

TLR2, TLR4, NLRP3

Probable

189,190

IgG chromatin complexes

TLR9

191

Lactoferrin

TLR4

192

mRNA

TLR3

Yes

Neutrophil elastase

TLR4

Probable

193,194

Osteopontin

TLR9

Probable

195,196

Oxidized LDL

TLR2, TLR4, TLR2/6

197,198

Oxidized phospholipids

TLR2, TLR4

199

S100A1

RAGE

Yes

119,128,130

S100A12

RAGE

120

S100A6

RAGE

120

S100A8/A9

TLR4, RAGE

Yes

124,126

S100B

RAGE

120

Saturated fatty acids

TLR4

200

Serum amyloid A

TLR2, TLR4

201

Surfactant protein A/D

TLR2, TLR4

202

ssRNA

TLR7, TRL8

203

Tenascin-C

TLR4

Probable

204

TNF-α

NLRP3

Yes

138,205

Uric acid (crystals)

TLR2, TLR4, NLRP3

206,207

Versican

TLR2/6

Probable

208,209

156

TARGETING DAMPS AFTER MI

Danger-associated molecular patterns The hypothesis that the immune system does not distinguish between self and non-self, but rather senses danger signals regardless of their source, was first proposed by Matzinger in 199460,61. She proposed that an inflammatory response is evoked by exogenous pathogens through PAMPs, or upon tissue damage by the release of DAMPs60. The currently identified DAMPs are a heterogeneous group of molecules that induce the activation of innate immune cells and can be divided into constitutive and inducible DAMPs. Constitutive DAMPs are not actively released by damaged cells but are the direct result of cellular destruction or necrosis. These molecules can only be found in high concentrations in the extracellular compartments after substantial tissue damage, such as (mitochondrial) DNA, RNA, and cardiac myosin62â&#x20AC;&#x201C;64. Inducible DAMPs, on the other hand, are actively secreted by damaged tissue. Among these, the best studied in the context of MI are heat shock proteins (HSPs), HMGB1 and fibronectin-end domain A (EDA). Although many different DAMPs exist, here we discuss the DAMPs that have been proven or are likely to play a role in the post-MI inflammatory response (table 1). Constitutive DAMPs Danger molecules that normally reside in the cytoplasm of cardiac resident cells are known to act as ligands for various PRRs upon tissue damage. An important danger molecule that has been shown to provoke a sterile inflammatory response is ATP. High extracellular ATP levels activate the P2X 7 membrane receptor and induce an efflux of intracellular potassium65. This leads to release of the mitochondrial lipid cardiolipin and subsequent NLRP3inflammasome activation, enabling the secretion of IL-1Î˛ and IL-18 66. Cardiac myosin is part of the sarcomeric complex and contributes to force and motion generation through its cyclic interaction with actin67. Recently, it has been shown that cardiomyocyte damage leads to the release of cardiac myosin and subsequent activation of the innate immune system via TLR2 and TLR8 on circulating monocytes, thereby possibly aggravating cardiac dysfunction in a mouse model of myocarditis63. Since the release of cardiac myosin also occurs after MI, it can be hypothesized that this process also plays a role in the post-MI inflammatory response. DNA has shown to serve as a powerful activator of the innate immune system68. It has recently been established that binding of extracellular DNA to TLR9 occurs at a site where an unmethylated cytosine residue is followed by guanosine (CpG)34. CpG sites occur much more common in prokaryotic than eukaryotic DNA, indicating that exogenous bacterial DNA is more powerful in stimulating the immune system than endogenous DNA 69. Mitochondria contain their own DNA and evolutionarily descend from bacteria. Mitochondrial DNA (mtDNA) therefore contains considerable numbers of CpG sites and extracellular mtDNA serves as a powerful DAMP to activate the innate immune system through inducing the activation, migration and degranulation of neutrophils through binding to TLR970. Next to extracellular DNA, intracellular DNA can also be detected since certain pathogens (e.g. viruses) reside intracellularly. When escaping degradation, intracellular mtDNA has been shown to be detrimental for cardiomyocyte life span and

157

PART THREE CHAPTER 9

enhances heart failure in a pressure overload heart failure model35. Although no mechanistic studies have been performed so far that investigate the role of mtDNA after MI, it has been postulated that mtDNA binding to TLR9 may play an essential role in post-MI inflammation71. Additionally, circulating mtDNA levels increase endothelial permeability and are elevated in patients suffering from ischemic heart disease72–74, indicating that targeting mtDNA mediated signaling could be an interesting target for alleviating the post-MI inflammatory reaction. Circulating extracellular RNA also serves as a powerful danger molecule and binds to TLR3 to stimulate MyD-88 independent TRIF-signaling and induce the secretion of IFN30,31. When subjected to coronary artery occlusion, TRIF-knockout mice show a decreased infarct size and preserved cardiac function. Unlike in TLR2 or TLR4 mediated signaling, these mice do not show a decreased expression of cardiac cytokines, but less cardiac apoptosis, indicating independent signaling pathways for TLR3. RNase markedly reduced cardiomyocyte cytokine release after hypoxia in vitro and RNase administration attenuated myocardial ischemia-reperfusion injury in vivo in mice subjected to MI 30,33. These data indicate that combined inhibition of TLR3 and TLR2/4 signaling may synergistically attenuate the inflammatory response and could possibly serve as a powerful tool to confine cardiac damage after MI. Heat shock proteins Heat shock proteins (HSPs) are part of a protein family originally described to be synthesized in response to heat shock and were first discovered in the salivary glands of Drosophila spp. in 196275,76. Since then, several HSPs have been discovered and are divided into families based on molecular weight77. Under normal physiological conditions, HSPs are abundantly present in the cytosol and function as chaperones in protein folding and translocation. Specifically, HSP70 and HSP90 are involved in the expression and presentation of antigens by transmembrane major histocompatibility complexes (MHCs)77,78. Together, the HSP family represents the most abundant intracellular proteins77. When these proteins are released, either as a result of necrotic cell death, or due to active secretion of cells, certain HSPs can be recognized by the adaptive immune system79–81. Roelofs et al. identified HSP22 as a ligand for TLR4 in rheumatoid arthritis82. HSP60, HSP70, HSP72 and Gp96 have also been identified as ligands for either TLR2 or TLR483–86. The function of HSPs in the presence of ischemic cardiac injury has also been investigated. HSP27, HSP60 and HSP72 are known to be upregulated in a rat model of MI87. HSP60 is strongly associated with the development of heart failure after MI, since it activates leukocytes that induce infarct extension in mice and rats88,89. HSP72 seems to act opposite to HSP60 and protects the rodent heart against ischemic cardiac injury90. Increased HSP70 levels also promote cardiomyocyte survival91 and attenuate cardiac damage in a pretreated rat model of MI92, although opposite effects have also been reported, where HSP70 stimulates proinflammatory responses93. Given this inconsistent evidence, future mechanistic studies are needed to elucidate the exact role of HSP70 in the inflammatory response post-MI.

158

TARGETING DAMPS AFTER MI

High mobility group box 1 The function of the HMGB1 protein depends on its nuclear, cellular or subcellular location.94. In the nucleus, HMGB1 is the most abundant non-histone DNA binding protein. Here, HMGB1 enables DNA transcription, repair and replication95,96. In the cytoplasm, HMGB1 regulates many different processes, among which autophagy and apoptosis. HMGB1 is also a critical regulator of mitochondrial function97. Within the cell, HMGB1 also senses nucleic acids and contributes to innate immune activation. Cytoplasmic recognition of nucleic acids occurs through binding to TLR3, TLR7 and TLR969. In this setting, HMGB1 is required to enhance the immune response and thus serves as a DAMP98,99. Apart from its intracellular activity, HMGB1 is thought to be a down-stream mediator of sepsis since it can be released by exogenous cellular activation100. Moreover, delayed administration of neutralizing antibodies targeted against HMGB1 protects against death in a mouse sepsis model101. HMGB1 is released by hypoxic or necrotic cells and acts on several PRRs including TLR2, TLR4, TLR9 and RAGE. It is able to form hetero-complexes with interleukin (IL)-1, IL-12, nucleic acids and histones. These complexes synergistically aggravate the inflammatory response compared to single ligands binding to their PRR102. When secreted by the damaged myocardium, HMGB1 activates NF-kB signaling through binding to TLR2, TLR4 and RAGE103. The role of HMGB1 post-MI seems to be pivotal. It has shown to be a mediator of inflammation and organ damage in the acute setting after myocardial reperfusion injury24,104. However, other reports suggest that HMGB1 mediates favorable effects of cardiac healing in mice post-MI105,106. Stimulation of HMGB1 signaling is effective in permanent ligation models, presumably due to its positive effect on angiogenesis105, while administration in ischemia-reperfusion models does not seem to be beneficial62,107. Accordingly, the effects of HMGB1 seem to be dose dependent, with beneficial effects at low doses and detrimental, heart failure promoting effects at high doses107, underscoring the complexity of the post-MI inflammatory response post-MI. Fibronectin Fibronectin is an essential component of the extracellular matrix. Alterations to this glycoprotein occur through an altered microenvironment due to e.g. ischemia-reperfusion injury. As a consequence, alternative splicing occurs by which the fibronectin-end domain A (FN-EDA) is upregulated. FN-EDA has been shown to enhance leukocyte and fibroblast activation through binding to and activation of α4β1 and α9β1 integrin receptors, while activating TLR2 and TLR4 has also been postulated 108–110. Importantly, FN-EDA is upregulated in mice subjected to MI111. Fibronectin-EDA knockout mice show improved survival rate, preserved cardiac function and left ventricular geometrics. Furthermore, lack of FN-EDA leads to suppression of inflammation in the post-infarcted myocardium and reduced fibrosis without affecting adequate scar formation111. Calgranulins The calgranulin family consists of 24 members that function as intracellular Ca2+ sensors. Calgranulins, also referred to as S100 proteins, have a high degree of similarity but are not

159

PART THREE CHAPTER 9

interchangeable112. S100 proteins are involved in many different processes such as cell differentiation and proliferation, but also migration, inflammation and myocyte contraction112,113. Their structure and the corresponding cellular responses of S100 proteins depend on Ca2+ levels. The S100 proteins can be subdivided into three main groups based on functionality: molecules that only serve as intracellular regulators, those that exert both intra- and extracellular effects and finally those that predominantly act extracellularly114. The extracellular S100 molecules serve as intercellular communicators, exert paracrine and autocrine actions and can enter the circulation to induce systemic effects. It is still under debate whether these proteins are actively secreted by resident cells or passively released as a result of tissue damage112. Among the numerous functions of the S100 proteins, they are also known to serve as ligands for PRRs. S100B, S100A6 and S100A12 are known to bind to RAGE. S100A8 and S100A9 are able to form hetero- and homodimers. These complexes can interact with RAGE and TLR4 and stimulate NF-kB signaling, thereby influencing tumor growth or creating a metastatic niche115â&#x20AC;&#x201C;117. The affinity of S100A8, S100A9 or its heterodimers for RAGE and TLR4 seems to be a point of discussion. Some investigators found that S100A8 binds to TLR-4 and is the active component in the S100A8/S100A9 heterodimer118, while others reported that S100A9 is a ligand for TLR4 and the binding sites of S100A9 to TLR4 are similar to those required for RAGE binding119. RAGE predominantly binds S100A9 while the affinity of S100A8 or the heterodimer for RAGE seems much lower119. Increasing evidence exists that calgranulins play a role in cardiovascular disease. Especially S100A8 and S100A9 levels increase in the presence of certain cardiovascular risk factors like hypertension, smoking, diabetes and atherosclerosis120. More importantly, both calgranulins rapidly increase during myocardial infarction and seem to originate from the site of injury121. Experimental data reveal that both proteins play a causative role in the setting of MI and attenuate the post-MI inflammatory response through RAGE-dependent NF-B signaling58. A more complex role is ascribed to S100A1. This S100 protein serves as a Ca2+-dependent inotropic factor in the heart and therefore positively influences cardiac function in rodent models of heart failure113,122,123. Moreover, recent evidence has also shown that S100A1 is released from cardiomyocytes and specifically internalized by cardiac fibroblast after MI. This results in the induction of an immunomodulatory phenotype by which cardiac fibroblasts promote an anti-fibrotic state and protect the heart from damage after myocardial infarction. In contrast to other DAMPs, S100A1 therefore seems to be protective after MI, possibly also by attenuating early inflammatory actions directly post-MI124. Cytokines The activation of PRRs by endogenous ligands leads to the synthesis of pro-inflammatory cytokines that are secreted by cardiac resident cells and circulating inflammatory cells. Cytokines play a crucial role in the amplification of the inflammatory response by attracting more circulating leukocytes to the damaged myocardium post-MI125. Additionally, cytokines stimulate cardiomyocyte apoptosis126, induce the expression of matrix-degrading proteases,

160

TARGETING DAMPS AFTER MI

leading to dilative cardiac remodeling127 and exert negative inotropic actions on the myocardium128. On the other hand, pro-inflammatory mediators can also induce key reparative signals. In addition to their role in the clearance of necrotic tissue, cytokines stimulate expression of fibrogenic and angiogenic growth factors, thereby initiating scar formation and promoting revascularization in the proliferation and maturation phase of cardiac wound healing129. Tumor necrosis factor (TNF)-α has a dual role in the progression and pathogenesis of myocardial ischemia reperfusion injury and heart failure130. The downstream effects of TNF-α greatly depend on whether this cytokine binds to TNF receptor 1 or 2. TNF-α signaling is therefore a very complex process and does not merely result in a proinflammatory response following cardiac injury131. A high dose of TNF-α is detrimental to the myocardium after MI, while a low dose of TNF-α improves myocardial function and decreases cellular injury in rodents129,132. Since contradictory results exist regarding TNF-α antagonism133,134, interference with TNF-α signaling seems to be unsuitable for the preservation of cardiac function post-MI. The IL-1 family consists of 11 members among which IL-1α, IL-1β, IL-18 and IL33. These cytokines possess a similar gene structure and signal through a group of closely related IL-receptors135. There is growing evidence that members of the IL-1 family play an essential role in the pathogenesis of the inflammatory response post-MI136. Recently, is has been discovered that IL-1α serves as a potent mediator of the post-MI inflammatory response137. IL-1β is also an essential player, since it is increased in the infarcted myocardium and drives the local inflammatory reaction138. Pharmacological reduction of IL-1β levels or IL-1 receptor blockade reduces cardiomyocyte apoptosis and attenuates cardiac remodeling after experimental acute myocardial infarction in rodents126,138–140. IL-18 is another key player in the inflammatory response post-MI141. Administration of an IL-18 neutralizing antibody or stem cells overexpressing IL-18 binding protein before ischemia lead to a reduction of infarct size in mice and rats respectively142,143. IL-18 also seems to orchestrate IL-1β mediated cardiac dysfunction144. Moreover, IL-18 levels are associated with the risk of adverse events post-MI145–147, making it an interesting target for future clinical testing. IL-10 is a potent anti-inflammatory cytokine, a suppressor of various pro-inflammatory mediators and a deactivator of monocytes148. IL-10 contributes to improved left ventricular function and attenuates LV remodeling by suppressing the immune response through STAT3 signaling in mice149. Many other cytokines and chemokines play a role in the inflammatory reaction after myocardial infarction and therefore are potentially therapeutic targets150. These cytokine levels are strongly associated with prognosis after myocardial infarction, which could implicate a causal relation between post-MI remodeling and the different cytokines 151. However, mechanistic and translational studies yet have to be conducted to determine if these cytokines can serve as viable therapeutic options.

161

PART THREE CHAPTER 9

Conclusions The primary treatment for acute MI is to re-establish coronary blood flow. Myocardial ischemia-reperfusion induces an intense inflammatory response that is required for the clearance of necrotic cardiac tissue and enables formation of a collagen-based scar. However, the inflammatory response plays a crucial role in enhancing acute injury post-MI and is responsible for infarct enlargement, eventually leading to left ventricular adverse remodeling and HF. Here, we therefore focused on the inhibition of early interactions between DAMPs and PRRs as a possible target to confine cardiac damage directly after MI. Immediately after MI, damaged or necrotic cardiomyocytes release several, functionally unrelated DAMPS, that act as ligands for PRRs, which are predominantly localized on cardiac fibroblasts and circulating leukocytes. All identified ligands are able to bind to a variety of PRRs and vice versa one PRR can bind multiple ligands. Together with heterodimerization and synergistic actions, this cross-talk makes the interaction between DAMPs and PRRs very complex. Nevertheless, intracellular signaling pathways often converge at a nuclear level where NF-kB activation leads to a pro-inflammatory profile of cardiac fibroblasts and cardiomyocytes. This eventually leads to attraction of circulating leukocytes to the damaged myocardium, increased cardiac cell death and dysfunction of surviving cardiomyocytes. Inhibition of this upstream inflammatory response therefore represents a promising therapeutic target for future drug development. Well-designed translational studies and clinical trials are now required to investigate whether the promising results seen in mechanistic studies can be confirmed in patients. Expert opinion In the early 1970s, investigators discovered for the first time that corticosteroid administration led to reduced cardiac cell death and altered cardiac wound healing in dogs subjected to MI152. From there onwards, numerous clinical and experimental studies have been performed to carefully map the course of the inflammatory response and cardiac wound healing after myocardial infarction6,11,125,153. However, it was not until 20 years ago that there was a true paradigm shift in the way this sterile immune response was approached60,61. With the discovery that molecules that are directly released upon tissue damage act on membrane bound receptors, a new group of molecules became subject for therapeutic interventions. The interaction of these DAMPs with their receptors seems to orchestrate the early inflammatory response after MI. The last few years, numerous different DAMPs have been discovered that interact with only a handful of PRRs. Given the extensive tissue damage post-MI, it is likely that multiple ligands act on the same PRR. Therefore it would be more feasible to inhibit signaling through blocking PRRs rather than trying to reduce the expression of certain DAMPs. Indeed, when looking at experimental studies, most investigators seem to follow this approach154. The most promising and consistent preservation of cardiac function and infarct size reduction have been accomplished with inhibition of TLR2, TLR4 and NLRP3inflammasome mediated signaling. These particular studies have not only provided proof of concept in rodents, but also have shown that pharmacological inhibition is feasible.

162

TARGETING DAMPS AFTER MI

Due to the promising experimental data with inhibition of these specific PRRs, antagonists for TLR2 (OPN-305) and NLRP3-inflammasome (16673-34-0, MCC950) signaling have now been developed with suitable pharmacodynamics and –kinetics for clinical testing23,49,50,155. Moreover, interference with IL-1β signaling has also shown consistent, pronounced beneficial effects in experimental models. Since potent clinical inhibitors already exist for other disease entities, clinical trials are currently underway to assess the effectiveness of IL-1β inhibition in patients after MI and first results seem promising156,157. As mentioned above, both DAMPs and PRRs are multifunctional. Inhibition of these signaling pathways might therefore also affect other pathways during the early inflammatory response after MI that could either damage or protect the heart. Future studies need to determine whether TLR or inflammasome inhibition affects these pathways to assess the exact effects on long-term cardiac repair and better predict the efficacy in clinical studies. The ultimate goal of clinical trials translating experimental findings into clinical applications is to improve survival rates and reduce the number of adverse events. Since the extent of cardiac damage is a very strong predictor for mortality and HF, this goal can be accomplished by aiming at the reduction of infarct size and preservation of cardiac function. The medical care for MI patients has improved tremendously the past decades and timely reperfusion has resulted in increased survival and less adverse events after MI158. Short-term administration of a single compound after MI to further reduce cardiac damage and adverse events is therefore not an easily obtainable goal but should nevertheless be pursued since a lot of gain still can be accomplished given the burden of ischemic heart disease on society. The research field investigating anti-inflammatory compounds in patients following MI has experienced many setbacks in the perspective of clinical translation of promising therapeutics. A wide range of mechanisms have been found to play a crucial role in the inflammatory response after MI and inhibition of these mechanisms, like complement signaling and leukocyte transmigration, in experimental animal models leads to a consistent reduction of infarct size159–161. However, subsequent clinical studies have failed to reproduce these findings in patients162–164. The failure of these compounds in clinical studies can be attributed to the presence of side-effects and a lack of effect on infarct size, cardiac function or adverse event162,164–166. Potentially, the latter may be based on a less prominent role of reperfusion injury in patients compared to experimental MI models. Another essential factor is the timing of therapy. Since the hyper-acute and acute phases of inflammation commence at ischemia and early reperfusion, the inflammatory response needs to be targeted before or directly after reperfusion167. However, inflammation also plays a prominent role in the subacute phase after MI. Targeting the inflammatory response could therefore also provide a second window of opportunity for therapy during the first hours post-reperfusion, although experimental data on the efficacy of the targeting the inflammatory response at this time point seem to be controversial168. Inhibition of the inflammatory response beyond this time frame has also been performed to directly target adverse remodeling and prevent geometrical changes of the left ventricle.

163

PART THREE CHAPTER 9

However, the infiltration of inflammatory cells and their relative contribution to overall cell numbers are very dynamic during cardiac wound healing. During the first hours to days after MI, the infarcted myocardium is mainly infiltrated by pro-inflammatory neutrophils and M1-macrophages that are generally considered to be detrimental to cardiac repair169. After several days, however, these cells go into apoptosis and are replaced by other immune cells (e.g. M2-macrophages) that orchestrate reparative processes, such as angiogenesis and scar formation. Profound inhibition of inflammation during the proliferation and maturation phase of cardiac wound healing could hence be detrimental to cardiac performance and result in adverse effects, like cardiac aneurysm formation5. Hence, the efficacy and safety of a particular anti-inflammatory compound may depend largely on the timing of its administration. It is therefore essential to take this factor into consideration when designing clinical trials to translate findings from experimental MI studies. A third possible reason for the previously observed translational failure of anti-inflammatory therapeutics is that the complete abolishment of the acute inflammatory response does not result in reduced cardiac damage and improved function in MI patients in the long term, since leukocytes are also required to induce proper cardiac wound healing. Pronounced immune-suppression in the acute phase could negatively influence the proliferation and maturation phase and thereby alter the long-term effects in clinical studies. Indeed, several experimental studies that tested the effects of corticosteroids and non-steroid anti-inflammatory drugs have shown that pronounced immune-suppression can be detrimental to cardiac repair170â&#x20AC;&#x201C;172. In contrast, targeting DAMPs and PRRs might result in â&#x20AC;&#x2DC;fine tuningâ&#x20AC;&#x2122; the inflammatory response rather than completely abolishing it, thereby possibly exerting beneficial effects in MI patients. However, clinical trials yet need to confirm this hypothesis. Importantly, translational failure in general can also be attributed to (1) clinically irrelevant experimental study design and selection of inappropriate animal models, (2) the testing of inconclusive therapies and (3) poor clinical study design168 . These factors should be taken into account in future drug development. To improve the chance of translational success, clinically relevant translational studies should therefore always be conducted before commencing clinical testing173. Although knockout and pretreatment studies are absolutely essential to determine a possible causative relationship on a mechanistic level, they do not mimic clinical treatment regimens. Importantly, pretreatment with agonists often has similar effects as treatment with antagonists due to the effect of cardiac preconditioning174. Studies using such a pretreatment approach obscure the true effect of certain signaling pathways and hamper translation of these therapeutic strategies toward clinical approaches. Animal studies should therefore resemble a clinical treatment regimen as closely as possible and studies applying drug administration preceding cardiac ischemia have to be interpreted with caution. Moreover, findings from rodent studies also need to be validated in more translational large animal models175.

164

TARGETING DAMPS AFTER MI

Despite the chance of translational failure, future clinical studies should continue the quest for a beneficial effect of anti-inflammatory compounds in patients after MI, since additional strategies are required to further reduce the chance of adverse events and mortality. Moreover, many recently discovered mechanisms could play an essential role in patients and effective treatment strategies blocking these pathways have now been developed. Mechanistic studies should concentrate on further identifying essential interactions between DAMPs and PRRs that drive the inflammatory response after MI. Since many of the PRR signal transduction pathways converge intracellularly, future therapies targeting a common upstream pathway could serve as an interesting approach in the search for a novel antiinflammatory therapeutic target to reduce infarct size, preserve cardiac function and prevent adverse remodeling after MI.

165

PART THREE CHAPTER 9

REFERENCES 1. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit J a., Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart Disease and Stroke Statistics - 2014 Update: A report from the American Heart Association. Circulation. 2014;129:e28–e292. 2. Mendis S, Abegunde D, Yusuf S, Ebrahim S, Shaper G, Ghannem H, Shengelia B. WHO study on Prevention of REcurrences of Myocardial Infarction And StrokE (WHO-PREMISE). Bull World Health Organ. 2005;83:820–828. 3. Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SMJM, Hofman A, Deckers JW, Witteman JCM, Stricker BHC. Quantifying the heart failure epidemic: Prevalence, incidence rate, lifetime risk and prognosis of heart failure - The Rotterdam Study. Eur Heart J. 2004;25:1614–1619. 4. Williams AR, Hare JM. Mesenchymal stem cells: Biology, pathophysiology, translational findings, and therapeutic implications for cardiac disease. Circ Res. 2011;109:923–940. 5. Frangogiannis NG. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 2014;11:255–65. 6. Van der Laan a. M, Nahrendorf M, Piek JJ. Healing and adverse remodelling after acute myocardial infarction: role of the cellular immune response. Heart. 2012;98:1384–1390. 7. Maekawa Y, Anzai T, Yoshikawa T, Asakura Y, Takahashi T, Ishikawa S, Mitamura H, Ogawa S. Prognostic significance of peripheral monocytosis after reperfused acute myocardial infarction:a possible role for left ventricular remodeling. J Am Coll Cardiol. 2002;39:241–246. 8. Takahashi T, Hiasa Y, Ohara Y, Miyazaki S-I, Ogura R, Suzuki N, Hosokawa S, Kishi K, Ohtani R. Relationship of admission neutrophil count to microvascular injury, left ventricular dilation, and long-term outcome in patients treated with primary angioplasty for acute myocardial infarction. Circ J. 2008;72:867–872. 9. Van Der Laan AM, Hirsch A, Robbers LFHJ, Nijveldt R, Lommerse I, Delewi R, Van Der Vleuten P a., Biemond BJ, Zwaginga JJ, Van Der Giessen WJ, Zijlstra F, Van Rossum AC, Voermans C, Van Der Schoot CE, Piek JJ. A proinflammatory monocyte response is associated with myocardial injury and impaired functional outcome in patients with STsegment elevation myocardial infarction: Monocytes and myocardial infarction. Am Heart J. 2012;163:57–65.e2.

166

10. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988. 11. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 12. Feng Y, Chao W. Toll-like receptors and myocardial inflammation. Int J Inflam. 2011;2011:170352. 13. Lamkanfi M, Dixit VM. Inflammasomes and Their Roles in Health and Disease. Annu Rev Cell Dev Biol. 2012;28:137–161. 14. Akira S, Takeda K. Toll-like receptor signalling. Nat Rev Immunol. 2004;4:499–511. 15. Medzhitov R, Preston-Hurlburt P, Janeway C a. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature. 1997;388:394–397. 16. Ohashi K, Burkart V, Flohé S, Kolb H. Cutting edge: heat shock protein 60 is a putative endogenous ligand of the toll-like receptor-4 complex. J Immunol. 2000;164:558–561. 17. Kawai T, Akira S. The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol. 2010;11:373–384. 18. Akira S, Uematsu S, Takeuchi O. Pathogen Recognition and Innate Immunity. Cell. 2006;124:783–801. 19. Mersmann J, Iskandar F, Latsch K, Habeck K, Sprunck V, Zimmermann R, Schumann RR, Zacharowski K, Koch A. Attenuation of Myocardial Injury by HMGB1 Blockade during Ischemia/ Reperfusion Is Toll-Like Receptor 2-Dependent. Mediators Inflamm. 2013;2013:1–8. 20. Jin C, Cleveland JC, Ao L, Li J, Zeng Q, Fullerton D a. Human Myocardium Releases Heat Shock Protein 27 ( HSP27 ) after Global Ischemia : The Proinflammatory Effect of Extracellular HSP27 through Toll-like Receptor ( TLR ) -2 and TLR4. Mol Med. 2014;27:280–289. 21. Arslan F, Smeets MB, O’Neill L a J, Keogh B, McGuirk P, Timmers L, Tersteeg C, Hoefer IE, Doevendans P a., Pasterkamp G, De Kleijn DP V. Myocardial ischemia/reperfusion injury is mediated by leukocytic toll-like receptor-2 and reduced by systemic administration of a novel anti-toll-like receptor-2 antibody. Circulation. 2010;121:80–90. 22. Shishido T, Nozaki N, Yamaguchi S, Shibata Y, Nitobe J, Miyamoto T, Takahashi H, Arimoto T, Maeda K, Yamakawa M, Takeuchi O, Akira S, Takeishi Y, Kubota I. Toll-Like Receptor-2 Modulates Ventricular Remodeling after Myocardial Infarction. Circulation. 2003;108:2905–2910. 23. Arslan F, Houtgraaf JH, Keogh B, Kazemi K, De Jong R, McCormack WJ, O’Neill L a J, McGuirk P, Timmers L, Smeets MB, Akeroyd L, Reilly M, Pasterkamp G, De Kleijn DP V. Treatment with OPN-305, a humanized anti-toll-like receptor-2 antibody, reduces myocardial ischemia/reperfusion injury in pigs. Circ Cardiovasc Interv. 2012;5:279–287.

TARGETING DAMPS AFTER MI

24. Ding HS, Yang J, Chen P, Yang J, Bo SQ, Ding JW, Yu QQ. The HMGB1-TLR4 axis contributes to myocardial ischemia/reperfusion injury via regulation of cardiomyocyte apoptosis. Gene. 2013;527:389–393. 25. Ding HS, Yang J, Gong FL, Yang J, Ding JW, Li S, Jiang YR. High mobility box 1 mediates neutrophil recruitment in myocardial ischemia-reperfusion injury through toll like receptor 4-related pathway. Gene. 2012;509:149–153. 26. Timmers L, Sluijter JPG, Van Keulen JK, Hoefer IE, Nederhoff MGJ, Goumans MJ, Doevendans P a., Van Echteld CJ a, Joles J a., Quax PH, Piek JJ, Pasterkamp G, De Kleijn DP V. Toll-like receptor 4 mediates maladaptive left ventricular remodeling and impairs cardiac function after myocardial infarction. Circ Res. 2008;102:257–264. 27. Shimamoto A, Chong AJ, Yada M, Shomura S, Takayama H, Fleisig AJ, Agnew ML, Hampton CR, Rothnie CL, Spring DJ, Pohlman TH, Shimpo H, Verrier ED. Inhibition of toll-like receptor 4 with eritoran attenuates myocardial ischemiareperfusion injury. Circulation. 2006;114:I270–4. 28. Xu Y, Huang Y, Wu C, Zhang J, Liu Q. Effect of carvedilol on cardiomyocyte apoptosis in a rat model of myocardial infarction: A role for toll-like receptor 4. Indian J Pharmacol. 2013;45:458. 29. Oyama J -i. Reduced Myocardial IschemiaReperfusion Injury in Toll-Like Receptor 4-Deficient Mice. Circulation. 2004;109:784–789. 30. Chen C, Feng Y, Zou L, Wang L, Chen HH, Cai J-Y, Xu J-M, Sosnovik DE, Chao W. Role of extracellular RNA and TLR3-Trif signaling in myocardial ischemia-reperfusion injury. J Am Heart Assoc. 2014;3:e000683. 31. Cavassani KA, Ishii M, Wen H, Schaller MA, Lincoln PM, Lukacs NW, Hogaboam CM, Kunkel SL. TLR3 is an endogenous sensor of tissue necrosis during acute inflammatory events. J Exp Med. 2008;205:2609–21. 32. Karikó K, Ni H, Capodici J, Lamphier M, Weissman D. mRNA is an endogenous ligand for Toll-like receptor 3. J Biol Chem. 2004;279:12542–50. 33. Lu C, Ren D, Wang X, Ha T, Liu L, Lee EJ, Hu J, Kalbfleisch J, Gao X, Kao R, Williams D, Li C. Tolllike receptor 3 plays a role in myocardial infarction and ischemia/reperfusion injury. Biochim Biophys Acta. 2014;1842:22–31. 34. Rutz M, Metzger J, Gellert T, Luppa P, Lipford GB, Wagner H, Bauer S. Toll-like receptor 9 binds single-stranded CpG-DNA in a sequence- and pHdependent manner. Eur J Immunol. 2004;34:2541– 50. 35. Oka T, Hikoso S, Yamaguchi O, Taneike M, Takeda T, Tamai T, Oyabu J, Murakawa T, Nakayama H, Nishida K, Akira S, Yamamoto A, Komuro I, Otsu K. Mitochondrial DNA that escapes from autophagy causes inflammation and heart failure. Nature. 2012;485:251–5.

36. Mathur S, Walley KR, Boyd JH. The Toll-like receptor 9 ligand CPG-C attenuates acute inflammatory cardiac dysfunction. Shock. 2011;36:478–83. 37. Ohm IK, Gao E, Belland Olsen M, Alfsnes K, Bliksøen M, Øgaard J, Ranheim T, Nymo SH, Holmen YD, Aukrust P, Yndestad A, Vinge LE. Tolllike receptor 9-activation during onset of myocardial ischemia does not influence infarct extension. PLoS One. 2014;9:e104407. 38. Cao Z, Ren D, Ha T, Liu L, Wang X, Kalbfleisch J, Gao X, Kao R, Williams D, Li C. CpG-ODN, the TLR9 agonist, attenuates myocardial ischemia/ reperfusion injury: involving activation of PI3K/Akt signaling. Biochim Biophys Acta. 2013;1832:96– 104. 39. Mason DR, Beck PL, Muruve DA. Nucleotidebinding oligomerization domain-like receptors and inflammasomes in the pathogenesis of nonmicrobial inflammation and diseases. J Innate Immun. 2012;4:16–30. 40. Fernández-Velasco M, Prieto P, Terrón V, Benito G, Flores JM, Delgado C, Zaragoza C, Lavin B, Gómez-Parrizas M, López-Collazo E, Martín-Sanz P, Boscá L. NOD1 activation induces cardiac dysfunction and modulates cardiac fibrosis and cardiomyocyte apoptosis. PLoS One. 2012;7:e45260. 41. Li X, Li F, Chu Y, Wang X, Zhang H, Hu Y, Zhang Y, Wang Z, Wei X, Jian W, Zhang X, Yi F. NOD2 deficiency protects against cardiac remodeling after myocardial infarction in mice. Cell Physiol Biochem. 2013;32:1857–66. 42. Latz E, Xiao TS, Stutz A. Activation and regulation of the inflammasomes. Nat Rev Immunol. 2013;13:397–411. 43. Takahashi M. NLRP3 Inflammasome as a Novel Player in Myocardial Infarction. Int Heart J. 2014;55:101–105. 44. Mezzaroma E, Toldo S, Farkas D, Seropian IM, Van Tassell BW, Salloum FN, Kannan HR, Menna a. C, Voelkel NF, Abbate a. The inflammasome promotes adverse cardiac remodeling following acute myocardial infarction in the mouse. Proc Natl Acad Sci. 2011;108:19725–19730. 45. Van Tassell BW, Toldo S, Mezzaroma E, Abbate A. Targeting interleukin-1 in heart disease. Circulation. 2013;128:1910–1923. 46. Pomerantz BJ, Reznikov LL, Harken AH, Dinarello CA. Inhibition of caspase 1 reduces human myocardial ischemic dysfunction via inhibition of IL-18 and IL-1beta. Proc Natl Acad Sci U S A. 2001;98:2871–6. 47. Kawaguchi M, Takahashi M, Hata T, Kashima Y, Usui F, Morimoto H, Izawa A, Takahashi Y, Masumoto J, Koyama J, Hongo M, Noda T, Nakayama J, Sagara J, Taniguchi S, Ikeda U. Inflammasome activation of cardiac fibroblasts is essential for myocardial ischemia/reperfusion injury. Circulation. 2011;123:594–604.

167

PART THREE CHAPTER 9

48. Sandanger Ø, Ranheim T, Vinge LE, Bliksøen M, Alfsnes K, Finsen A V., Dahl CP, Askevold ET, Florholmen G, Christensen G, Fitzgerald K a., Lien E, Valen G, Espevik T, Aukrust P, Yndestad A. The NLRP3 inflammasome is up-regulated in cardiac fibroblasts and mediates myocardial ischaemiareperfusion injury. Cardiovasc Res. 2013;99:164– 174. 49. Marchetti C, Chojnacki J, Toldo S, Mezzaroma E, Tranchida N, Rose SW, Federici M, Van Tassell BW, Zhang S, Abbate A. A novel pharmacologic inhibitor of the NLRP3 inflammasome limits myocardial injury after ischemia-reperfusion in the mouse. J Cardiovasc Pharmacol. 2014;63:316–22. 50. Coll RC, Robertson AAB, Chae JJ, Higgins SC, Muñoz-Planillo R, Inserra MC, Vetter I, Dungan LS, Monks BG, Stutz A, Croker DE, Butler MS, Haneklaus M, Sutton CE, Núñez G, Latz E, Kastner DL, Mills KHG, Masters SL, Schroder K, Cooper M a, O’Neill LAJ. A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases. Nat Med. 2015;21:248–55. 51. Xie J, Méndez JD, Méndez-Valenzuela V, AguilarHernández MM. Cellular signalling of the receptor for advanced glycation end products (RAGE). Cell Signal. 2013;25:2185–97. 52. Schmidt AM, Vianna M, Gerlach M, Brett J, Ryan J, Kao J, Esposito C, Hegarty H, Hurley W, Clauss M. Isolation and characterization of two binding proteins for advanced glycosylation end products from bovine lung which are present on the endothelial cell surface. J Biol Chem. 1992;267:14987–97. 53. Zhao J, Randive R, Stewart JA. Molecular mechanisms of AGE/RAGE-mediated fibrosis in the diabetic heart. World J Diabetes. 2014;5:860–7. 54. Bierhaus A, Humpert PM, Morcos M, Wendt T, Chavakis T, Arnold B, Stern DM, Nawroth PP. Understanding RAGE, the receptor for advanced glycation end products. J Mol Med. 2005;83:876– 886. 55. Basta G, Lazzerini G, Massaro M, Simoncini T, Tanganelli P, Fu C, Kislinger T, Stern DM, Schmidt AM, De Caterina R. Advanced glycation end products activate endothelium through signaltransduction receptor RAGE: a mechanism for amplification of inflammatory responses. Circulation. 2002;105:816–22. 56. Bierhaus A, Schiekofer S, Schwaninger M, Andrassy M, Humpert PM, Chen J, Hong M, Luther T, Henle T, Klöting I, Morcos M, Hofmann M, Tritschler H, Weigle B, Kasper M, Smith M, Perry G, Schmidt AM, Stern DM, Häring HU, Schleicher E, Nawroth PP. Diabetes-associated sustained activation of the transcription factor nuclear factor-kappaB. Diabetes. 2001;50:2792–808. 57. Andrassy M, Volz HC, Igwe JC, Funke B, Eichberger SN, Kaya Z, Buss S, Autschbach F, Pleger ST, Lukic IK, Bea F, Hardt SE, Humpert PM, Bianchi ME, Mairbäurl H, Nawroth PP, Remppis A, Katus H a.,

168

Bierhaus A. High-mobility group box-1 in ischemiareperfusion injury of the heart. Circulation. 2008;117:3216–3226. 58. Volz HC, Laohachewin D, Seidel C, Lasitschka F, Keilbach K, Wienbrandt AR, Andrassy J, Bierhaus A, Kaya Z, Katus HA, Andrassy M. S100A8/A9 aggravates post-ischemic heart failure through activation of RAGE-dependent NF-κB signaling. Basic Res Cardiol. 2012;107:250. 59. Ku SH, Hong J, Moon H-H, Jeong JH, Mok H, Park S, Choi D, Kim SH. Deoxycholic acid-modified polyethylenimine based nanocarriers for RAGE siRNA therapy in acute myocardial infarction. Arch Pharm Res. 2015; 60. Matzinger P. The danger model: a renewed sense of self. Science. 2002;296:301–305. 61. Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol. 1994;12:991–1045. 62. De Haan JJ, Smeets MB, Pasterkamp G, Arslan F. Danger signals in the initiation of the inflammatory response after myocardial infarction. Mediators Inflamm. 2013;2013:206039. 63. Zhang P, Cox CJ, Alvarez KM, Cunningham MW. Cutting edge: cardiac myosin activates innate immune responses through TLRs. J Immunol. 2009;183:27–31. 64. Cruz CM, Rinna A, Forman HJ, Ventura ALM, Persechini PM, Ojcius DM. ATP activates a reactive oxygen species-dependent oxidative stress response and secretion of proinflammatory cytokines in macrophages. J Biol Chem. 2007;282:2871–9. 65. Mariathasan S, Weiss DS, Newton K, McBride J, O’Rourke K, Roose-Girma M, Lee WP, Weinrauch Y, Monack DM, Dixit VM. Cryopyrin activates the inflammasome in response to toxins and ATP. Nature. 2006;440:228–32. 66. Iyer S, He Q, Janczy J, Elliott E, Zhong Z, Olivier A, Sadler J, Knepper-Adrian V, Han R, Qiao L, Eisenbarth S, Nauseef W, Cassel S, Sutterwala F. Mitochondrial cardiolipin is required for Nlrp3 inflammasome activation. Immunity. 2013;39:311– 323. 67. Yin Z, Ren J, Guo W. Sarcomeric protein isoform transitions in cardiac muscle: a journey to heart failure. Biochim Biophys Acta. 2015;1852:47–52. 68. Pisetsky DS. The origin and properties of extracellular DNA: From PAMP to DAMP. Clin Immunol. 2012;144:32–40. 69. Krieg AM. CpG motifs in bacterial DNA and their immune effects. Annu Rev Immunol. 2002;20:709– 60. 70. Zhang Q, Raoof M, Chen Y, Sumi Y, Sursal T, Junger W, Brohi K, Itagaki K, Hauser CJ. Circulating mitochondrial DAMPs cause inflammatory responses to injury. Nature. 2010;464:104–7. 71. Konstantinidis K, Kitsis RN. Cardiovascular biology: Escaped DNA inflames the heart. Nature. 2012;485:179–80.

TARGETING DAMPS AFTER MI

72. Sun S, Sursal T, Adibnia Y, Zhao C, Zheng Y, Li H, Otterbein LE, Hauser CJ, Itagaki K. Mitochondrial DAMPs increase endothelial permeability through neutrophil dependent and independent pathways. PLoS One. 2013;8:e59989. 73. Bliksøen M, Mariero LH, Ohm IK, Haugen F, Yndestad A, Solheim S, Seljeflot I, Ranheim T, Andersen GØ, Aukrust P, Valen G, Vinge LE. Increased circulating mitochondrial DNA after myocardial infarction. Int J Cardiol. 2012;158:132–4. 74. Wang L, Xie L, Zhang Q, Cai X, Tang Y, Wang L, Hang T, Liu J, Gong J. Plasma nuclear and mitochondrial DNA levels in acute myocardial infarction patients. Coron Artery Dis. 2015; 75. Tissières a, Mitchell HK, Tracy UM. Protein synthesis in salivary glands of Drosophila melanogaster: relation to chromosome puffs. J Mol Biol. 1974;84:389–398. 76. Ritossa F. A new puffing pattern induced by temperature shock and DNP in Drosophila. Experientia. 1962; 77. Binder RJ. Functions of heat shock proteins in pathways of the innate and adaptive immune system. J Immunol. 2014;193:5765–71. 78. Callahan MK, Garg M, Srivastava PK. Heat-shock protein 90 associates with N-terminal extended peptides and is required for direct and indirect antigen presentation. Proc Natl Acad Sci U S A. 2008;105:1662–7. 79. Young JC. Mechanisms of the Hsp70 chaperone system. Biochem Cell Biol. 2010;88:291–300. 80. Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. Necrotic but not apoptotic cell death releases heat shock proteins, which deliver a partial maturation signal to dendritic cells and activate the NF-kappa B pathway. Int Immunol. 2000;12:1539– 46. 81. Miyake Y, Yamasaki S. Sensing necrotic cells. Adv Exp Med Biol. 2012;738:144–52. 82. Roelofs MF, Boelens WC, Joosten LAB, AbdollahiRoodsaz S, Geurts J, Wunderink LU, Schreurs BW, van den Berg WB, Radstake TRDJ. Identification of small heat shock protein B8 (HSP22) as a novel TLR4 ligand and potential involvement in the pathogenesis of rheumatoid arthritis. J Immunol. 2006;176:7021–7. 83. Warger T, Hilf N, Rechtsteiner G, Haselmayer P, Carrick DM, Jonuleit H, von Landenberg P, Rammensee H-G, Nicchitta C V, Radsak MP, Schild H. Interaction of TLR2 and TLR4 ligands with the N-terminal domain of Gp96 amplifies innate and adaptive immune responses. J Biol Chem. 2006;281:22545–53. 84. Vabulas RM, Ahmad-Nejad P, Ghose S, Kirschning CJ, Issels RD, Wagner H. HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J Biol Chem. 2002;277:15107–12. 85. Cohen-Sfady M, Nussbaum G, Pevsner-Fischer M, Mor F, Carmi P, Zanin-Zhorov A, Lider O, Cohen IR. Heat shock protein 60 activates B cells via the TLR4-

MyD88 pathway. J Immunol. 2005;175:3594–602. 86. Chase MA, Wheeler DS, Lierl KM, Hughes VS, Wong HR, Page K. Hsp72 induces inflammation and regulates cytokine production in airway epithelium through a TLR4- and NF-kappaBdependent mechanism. J Immunol. 2007;179:6318–24. 87. Tanonaka K, Toga W, Yoshida H, Takeo S. Myocardial heat shock protein changes in the failing heart following coronary artery ligation. Hear Lung Circ. 2003;12:60–65. 88. Li Y, Si R, Feng Y, Chen HH, Zou L, Wang E, Zhang M, Warren HS, Sosnovik DE, Chao W. Myocardial ischemia activates an injurious innate immune signaling via cardiac heat shock protein 60 and tolllike receptor 4. J Biol Chem. 2011;286:31308– 31319. 89. Kim S-C, Stice JP, Chen L, Jung JS, Gupta S, Wang Y, Baumgarten G, Trial J, Knowlton A a. Extracellular heat shock protein 60, cardiac myocytes, and apoptosis. Circ Res. 2009;105:1186–1195. 90. Tanonaka K, Furuhama KI, Yoshida H, Kakuta K, Miyamoto Y, Toga W, Takeo S. Protective effect of heat shock protein 72 on contractile function of perfused failing heart. Am J Physiol Heart Circ Physiol. 2001;281:H215–H222. 91. Martin JL, Mestril R, Hilal-Dandan R, Brunton LL, Dillmann WH. Small heat shock proteins and protection against ischemic injury in cardiac myocytes. Circulation. 1997;96:4343–4348. 92. Lubbers NL, Polakowski JS, Wegner CD, Burke SE, Diaz GJ, Daniell KM, Cox BF. Oral bimoclomol elevates heat shock protein 70 and reduces myocardial infarct size in rats. Eur J Pharmacol. 2002;435:79–83. 93. Asea a, Kraeft SK, Kurt-Jones E a, Stevenson M a, Chen LB, Finberg RW, Koo GC, Calderwood SK. HSP70 stimulates cytokine production through a CD14-dependant pathway, demonstrating its dual role as a chaperone and cytokine. Nat Med. 2000;6:435–442. 94. Andersson U, Antoine DJ, Tracey KJ. The functions of HMGB1 depend on molecular localization and post-translational modifications. J Intern Med. 2014;276:420–4. 95. Bustin M, A. Lehn D, Landsman D. Structural features of the HMG chromosomal proteins and their genes. Biochim Biophys Acta - Gene Struct Expr. 1990;1049:231–243. 96. Bustin M, Reeves R. High-Mobility-Group Chromosomal Proteins: Architectural Components That Facilitate Chromatin Function. Prog Nucleic Acid Res Mol Biol. 1996;54:35–100b. 97. Tsung A, Tohme S, Billiar TR. High-mobility group box-1 in sterile inflammation. J Intern Med. 2014;276:425–43. 98. Krieg AM, Yi AK, Matson S, Waldschmidt TJ, Bishop GA, Teasdale R, Koretzky GA, Klinman DM. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature. 1995;374:546–9.

169

PART THREE CHAPTER 9

99. Ishii KJ, Coban C, Kato H, Takahashi K, Torii Y, Takeshita F, Ludwig H, Sutter G, Suzuki K, Hemmi H, Sato S, Yamamoto M, Uematsu S, Kawai T, Takeuchi O, Akira S. A Toll-like receptorindependent antiviral response induced by doublestranded B-form DNA. Nat Immunol. 2006;7:40–8. 100. Qin S, Wang H, Yuan R, Li H, Ochani M, Ochani K, Rosas-Ballina M, Czura CJ, Huston JM, Miller E, Lin X, Sherry B, Kumar A, Larosa G, Newman W, Tracey KJ, Yang H. Role of HMGB1 in apoptosis-mediated sepsis lethality. J Exp Med. 2006;203:1637–42. 101. Yang H, Ochani M, Li J, Qiang X, Tanovic M, Harris HE, Susarla SM, Ulloa L, Wang H, DiRaimo R, Czura CJ, Wang H, Roth J, Warren HS, Fink MP, Fenton MJ, Andersson U, Tracey KJ. Reversing established sepsis with antagonists of endogenous highmobility group box 1. Proc Natl Acad Sci U S A. 2004;101:296–301. 102. Harris HE, Andersson U, Pisetsky DS. HMGB1: a multifunctional alarmin driving autoimmune and inflammatory disease. Nat Rev Rheumatol. 2012;8:195–202. 103. Park JS, Svetkauskaite D, He Q, Kim J-Y, Strassheim D, Ishizaka A, Abraham E. Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein. J Biol Chem. 2004;279:7370–7377. 104. Hu G, Zhang Y, Jiang H, Hu X. Exendin-4 attenuates myocardial ischemia and reperfusion injury by inhibiting high mobility group box 1 protein expression. Cardiol J. 2013;20:600–604. 105. Kitahara T, Takeishi Y, Harada M, Niizeki T, Suzuki S, Sasaki T, Ishino M, Bilim O, Nakajima O, Kubota I. High-mobility group box 1 restores cardiac function after myocardial infarction in transgenic mice. Cardiovasc Res. 2008;80:40–46. 106. Takahashi K, Fukushima S, Yamahara K, Yashiro K, Shintani Y, Coppen SR, Salem HK, Brouilette SW, Yacoub MH, Suzuki K. Modulated inflammation by injection of high-mobility group box 1 recovers post-infarction chronically failing heart. Circulation. 2008;118:S106–14. 107. Lotze MT, Tracey KJ. High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Nat Rev Immunol. 2005;5:331–342. 108. Schoneveld a. H, Hoefer I, Sluijter JPG, Laman JD, de Kleijn DP V, Pasterkamp G. Atherosclerotic lesion development and Toll like receptor 2 and 4 responsiveness. Atherosclerosis. 2008;197:95–104. 109. Okamura Y, Watari M, Jerud ES, Young DW, Ishizaka ST, Rose J, Chow JC, Strauss JF. The Extra Domain A of Fibronectin Activates Toll-like Receptor 4. J Biol Chem. 2001;276:10229–10233. 110. Liao Y-F, Gotwals PJ, Koteliansky VE, Sheppard D, Van De Water L. The EIIIA segment of fibronectin is a ligand for integrins alpha 9beta 1 and alpha 4beta 1 providing a novel mechanism for regulating cell adhesion by alternative splicing. J Biol Chem. 2002;277:14467–14474.

170

111. Arslan F, Smeets MB, Riem Vis PW, Karper JC, Quax PH, Bongartz LG, Peters JH, Hoefer IE, Doevendans P a., Pasterkamp G, De Kleijn DP. Lack of fibronectin-EDA promotes survival and prevents adverse remodeling and heart function deterioration after myocardial infarction. Circ Res. 2011;108:582–592. 112. Bresnick AR, Weber DJ, Zimmer DB. S100 proteins in cancer. Nat Rev Cancer. 2015;15:96–109. 113. Most P, Seifert H, Gao E, Funakoshi H, Völkers M, Heierhorst J, Remppis A, Pleger ST, DeGeorge BR, Eckhart AD, Feldman AM, Koch WJ. Cardiac S100A1 protein levels determine contractile performance and propensity toward heart failure after myocardial infarction. Circulation. 2006;114:1258–68. 114. Donato R, Cannon BR, Sorci G, Riuzzi F, Hsu K, Weber DJ, Geczy CL. Functions of S100 proteins. Curr Mol Med. 2013;13:24–57. 115. Ichikawa M, Williams R, Wang L, Vogl T, Srikrishna G. S100A8/A9 activate key genes and pathways in colon tumor progression. Mol Cancer Res. 2011;9:133–48. 116. Källberg E, Vogl T, Liberg D, Olsson A, Björk P, Wikström P, Bergh A, Roth J, Ivars F, Leanderson T. S100A9 interaction with TLR4 promotes tumor growth. PLoS One. 2012;7:e34207. 117. Hiratsuka S, Watanabe A, Sakurai Y, AkashiTakamura S, Ishibashi S, Miyake K, Shibuya M, Akira S, Aburatani H, Maru Y. The S100A8-serum amyloid A3-TLR4 paracrine cascade establishes a premetastatic phase. Nat Cell Biol. 2008;10:1349–55. 118. Vogl T, Tenbrock K, Ludwig S, Leukert N, Ehrhardt C, van Zoelen MAD, Nacken W, Foell D, van der Poll T, Sorg C, Roth J. Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock. Nat Med. 2007;13:1042–9. 119. Björk P, Björk A, Vogl T, Stenström M, Liberg D, Olsson A, Roth J, Ivars F, Leanderson T. Identification of human S100A9 as a novel target for treatment of autoimmune disease via binding to quinoline-3-carboxamides. PLoS Biol. 2009;7:e97. 120. Schiopu A, Cotoi OS. S100A8 and S100A9: DAMPs at the crossroads between innate immunity, traditional risk factors, and cardiovascular disease. Mediators Inflamm. 2013;2013:828354. 121. Altwegg LA, Neidhart M, Hersberger M, Müller S, Eberli FR, Corti R, Roffi M, Sütsch G, Gay S, von Eckardstein A, Wischnewsky MB, Lüscher TF, Maier W. Myeloid-related protein 8/14 complex is released by monocytes and granulocytes at the site of coronary occlusion: a novel, early, and sensitive marker of acute coronary syndromes. Eur Heart J. 2007;28:941–8. 122. Most P, Remppis A, Pleger ST, Katus HA, Koch WJ. S100A1: a novel inotropic regulator of cardiac performance. Transition from molecular physiology to pathophysiological relevance. Am J Physiol

TARGETING DAMPS AFTER MI

Regul Integr Comp Physiol. 2007;293:R568–77. 123. Pleger ST, Remppis A, Heidt B, Völkers M, Chuprun JK, Kuhn M, Zhou R-H, Gao E, Szabo G, Weichenhan D, Müller OJ, Eckhart AD, Katus HA, Koch WJ, Most P. S100A1 gene therapy preserves in vivo cardiac function after myocardial infarction. Mol Ther. 2005;12:1120–9. 124. Rohde D, Schön C, Boerries M, Didrihsone I, Ritterhoff J, Kubatzky KF, Völkers M, Herzog N, Mähler M, Tsoporis JN, Parker TG, Linke B, Giannitsis E, Gao E, Peppel K, Katus HA, Most P. S100A1 is released from ischemic cardiomyocytes and signals myocardial damage via Toll-like receptor 4. EMBO Mol Med. 2014;6:778–94. 125. Frangogiannis NG. The Immune System and the Remodeling Infarcted Heart. J Cardiovasc Pharmacol. 2014;63:185–195. 126. Abbate A, Salloum FN, Vecile E, Das A, Hoke NN, Straino S, Giuseppe GL, Biondi-Zoccai, Houser JE, Qureshi IZ, Ownby ED, Gustini E, Biasucci LM, Severino A, Capogrossi MC, Vetrovec GW, Crea F, Baldi A, Kukreja RC, Dobrina A. Anakinra, a recombinant human interleukin-1 receptor antagonist, inhibits apoptosis in experimental acute myocardial infarction. Circulation. 2008;117:2670–2683. 127. Li YY, Feng YQ, Kadokami T, McTiernan CF, Draviam R, Watkins SC, Feldman a M. Myocardial extracellular matrix remodeling in transgenic mice overexpressing tumor necrosis factor alpha can be modulated by anti-tumor necrosis factor alpha therapy. Proc Natl Acad Sci U S A. 2000;97:12746– 12751. 128. Schulz R, Panas DL, Catena R, Moncada S, Olley PM, Lopaschuk GD. The role of nitric oxide in cardiac depression induced by interleukin-1 beta and tumour necrosis factor-alpha. Br J Pharmacol. 1995;114:27–34. 129. Kurrelmeyer KM, Michael LH, Baumgarten G, Taffet GE, Peschon JJ, Sivasubramanian N, Entman ML, Mann DL. Endogenous tumor necrosis factor protects the adult cardiac myocyte against ischemic-induced apoptosis in a murine model of acute myocardial infarction. Proc Natl Acad Sci U S A. 2000;97:5456–5461. 130. Kleinbongard P, Heusch G, Schulz R. TNFα in atherosclerosis, myocardial ischemia/reperfusion and heart failure. Pharmacol Ther. 2010;127:295– 314. 131. Hamid T, Gu Y, Ortines R V., Bhattacharya C, Wang G, Xuan Y-T, Prabhu SD. Divergent Tumor Necrosis Factor Receptor-Related Remodeling Responses in Heart Failure: Role of Nuclear Factor- B and Inflammatory Activation. Circulation. 2009;119:1386–1397. 132. Asgeri M, Pourafkari L, Kundra A, Javadzadegan H, Negargar S, Nader ND. Dual effects of tumor necrosis factor alpha on myocardial injury following prolonged hypoperfusion of the heart. Immunol Invest. 2015;44:23–35.

133. Chung ES. Randomized, Double-Blind, PlaceboControlled, Pilot Trial of Infliximab, a Chimeric Monoclonal Antibody to Tumor Necrosis Factor- , in Patients With Moderate-to-Severe Heart Failure: Results of the Anti-TNF Therapy Against Congestive Heart failure (ATTACH. Circulation. 2003;107:3133–3140. 134. Padfield GJ, Din JN, Koushiappi E, Mills NL, Robinson SD, Cruden NLM, Lucking AJ, Chia S, Harding S a, Newby DE. Cardiovascular effects of tumour necrosis factor α antagonism in patients with acute myocardial infarction: a first in human study. Heart. 2013;99:1330–5. 135. Sims JE, Smith DE. The IL-1 family: regulators of immunity. Nat Rev Immunol. 2010;10:89–102. 136. Bujak M, Frangogiannis NG. The role of IL-1 in the pathogenesis of heart disease. Arch Immunol Ther Exp (Warsz). 2009;57:165–176. 137. Lugrin J, Parapanov R, Rosenblatt-Velin N, RignaultClerc S, Feihl F, Waeber B, Muller O, Vergely C, Zeller M, Tardivel A, Schneider P, Pacher P, Liaudet L, Müller O, Vergely C, Zeller M, Tardivel A, Schneider P, Pacher P, Liaudet L. Cutting Edge: IL1 Is a Crucial Danger Signal Triggering Acute Myocardial Inflammation during Myocardial Infarction. J Immunol. 2014;194:499–503. 138. Bujak M, Dobaczewski M, Chatila K, Mendoza LH, Li N, Reddy A, Frangogiannis NG. Interleukin-1 receptor type I signaling critically regulates infarct healing and cardiac remodeling. Am J Pathol. 2008;173:57–67. 139. Van Tassell BW, Varma A, Salloum FN, Das A, Seropian IM, Toldo S, Smithson L, Hoke NN, Chau VQ, Robati R, Abbate A. Interleukin-1 trap attenuates cardiac remodeling after experimental acute myocardial infarction in mice. J Cardiovasc Pharmacol. 2010;55:117–122. 140. Salloum FN, Chau V, Varma A, Hoke NN, Toldo S, Biondi-Zoccai GGL, Crea F, Vetrovec GW, Abbate A. Anakinra in experimental acute myocardial infarction-does dosage or duration of treatment matter? Cardiovasc Drugs Ther. 2009;23:129–135. 141. Abbate A, Van Tassell BW, Biondi-Zoccai GGL. Blocking interleukin-1 as a novel therapeutic strategy for secondary prevention of cardiovascular events. BioDrugs. 2012;26:217–33. 142. Venkatachalam K, Prabhu SD, Reddy VS, Boylston WH, Valente AJ, Chandrasekar B. Neutralization of interleukin-18 ameliorates ischemia/reperfusioninduced myocardial injury. J Biol Chem. 2009;284:7853–7865. 143. Wang M, Tan J, Wang Y, Meldrum KK, Dinarello C a, Meldrum DR. IL-18 binding protein-expressing mesenchymal stem cells improve myocardial protection after ischemia or infarction. Proc Natl Acad Sci U S A. 2009;106:17499–17504. 144. Toldo S, Mezzaroma E, O’Brien L, Marchetti C, Seropian IM, Voelkel NF, Van Tassell BW, Dinarello C a., Abbate a. Interleukin-18 mediates interleukin1-induced cardiac dysfunction. AJP Hear Circ

171

PART THREE CHAPTER 9

Physiol. 2014;306:H1025–H1031. 145. Blankenberg S, Tiret L, Bickel C, Peetz D, Cambien F, Meyer J, Rupprecht HJ. Interleukin-18 is a strong predictor of cardiovascular death in stable and unstable angina. Circulation. 2002;106:24–30. 146. Blankenberg S, Luc G, Ducimetière P, Arveiler D, Ferrières J, Amouyel P, Evans A, Cambien F, Tiret L. Interleukin-18 and the Risk of Coronary Heart Disease in European Men: The Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation. 2003;108:2453–2459. 147. Hartford M, Wiklund O, Hultén LM, Persson A, Karlsson T, Herlitz J, Hulthe J, Caidahl K. Interleukin-18 as a predictor of future events in patients with acute coronary syndromes. Arterioscler Thromb Vasc Biol. 2010;30:2039–2046. 148. Frangogiannis NG, Mendoza LH, Lindsey ML, Ballantyne CM, Michael LH, Smith CW, Entman ML. IL-10 is induced in the reperfused myocardium and may modulate the reaction to injury. J Immunol. 2000;165:2798–2808. 149. Krishnamurthy P, Rajasingh J, Lambers E, Qin G, Losordo DW, Kishore R. IL-10 inhibits inflammation and attenuates left ventricular remodeling after myocardial infarction via activation of STAT3 and suppression of HuR. Circ Res. 2009;104:e9–18. 150. Franz-Josef Neumann, MD; Ilka Ott, MD; Meinrad Gawaz, MD; Gert Richardt, MD; Heidi Holzapfel; Marianne Jochum, PhD; Albert Schömig M. Cardiac Release of Cytokines and Inflammatory Responses in Acute Myocardial Infarction. Circulation. 1995; 151. Kaptoge S, Seshasai SRK, Gao P, Freitag DF, Butterworth AS, Borglykke A, Di Angelantonio E, Gudnason V, Rumley A, Lowe GDO, Jørgensen T, Danesh J. Inflammatory cytokines and risk of coronary heart disease: New prospective study and updated meta-analysis. Eur Heart J. 2014;35:578– 89. 152. Libby P, Maroko PR, Bloor CM, Sobel BE, Braunwald E. Reduction of experimental myocardial infarct size by corticosteroid administration. J Clin Invest. 1973;52:599–607. 153. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 154. Arslan F, Keogh B, McGuirk P, Parker AE. TLR2 and TLR4 in ischemia reperfusion injury. Mediators Inflamm. 2010;2010:704202. 155. Reilly M, Miller RM, Thomson MH, Patris V, Ryle P, McLoughlin L, Mutch P, Gilboy P, Miller C, Broekema M, Keogh B, McCormack W, van de Wetering de Rooij J. Randomized, double-blind, placebo-controlled, dose-escalating phase I, healthy subjects study of intravenous OPN-305, a humanized anti-TLR2 antibody. Clin Pharmacol Ther. 2013;94:593–600. 156. Abbate A, Kontos MC, Abouzaki NA, Melchior RD, Thomas C, Van Tassell BW, Oddi C, Carbone S, Trankle CR, Roberts CS, Mueller GH, Gambill ML,

172

Christopher S, Markley R, Vetrovec GW, Dinarello CA, Biondi-Zoccai G. Comparative Safety of Interleukin-1 Blockade With Anakinra in Patients With ST-Segment Elevation Acute Myocardial Infarction (from the VCU-ART and VCU-ART2 Pilot Studies). Am J Cardiol. 2015;115:288–292. 157. Sonnino C, Christopher S, Oddi C, Toldo S, Falcao RA, Melchior RD, Mueller GH, Abouzaki NA, Varma A, Gambill ML, Van Tassell BW, Dinarello CA, Abbate A. Leukocyte activity in patients with STsegment elevation acute myocardial infarction treated with anakinra. Mol Med. 2014;20:486–9. 158. Steg PG, James SK, Atar D, Badano LP, Lundqvist CB, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, Van’T Hof A, Widimsky P, Zahger D, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, FunckBrentano C, Hasdai D, Hoes A, Kirchhof P, Kolh P, McDonagh T, Moulin C, Popescu B a., Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Astin F, Åström-Olsson K, Budaj A, Clemmensen P, Collet JP, Fox K a., Fuat A, Gustiene O, Hamm CW, Kala P, Lancellotti P, Maggioni A Pietro, Merkely B, Neumann FJ, Piepoli MF, Van De Werf F, Verheugt F, Wallentin L, Blömstrom-Lundqvist C, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, van ’t Hof A, Widimsky P, Zahger D. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 2012;33:2569– 2619. 159. Tanaka M, Brooks SE, Richard VJ, FitzHarris GP, Stoler RC, Jennings RB, Arfors KE, Reimer K a. Effect of anti-CD18 antibody on myocardial neutrophil accumulation and infarct size after ischemia and reperfusion in dogs. Circulation. 1993;87:526–535. 160. Näslund U, Häggmark S, Johansson G, Marklund SL, Reiz S. Limitation of myocardial infarct size by superoxide dismutase as an adjunct to reperfusion after different durations of coronary occlusion in the pig. Circ Res. 1990;66:1294–1301. 161. Vakeva a P, Agah a, Rollins S a, Matis L a, Li L, Stahl GL. Myocardial infarction and apoptosis after myocardial ischemia and reperfusion: role of the terminal complement components and inhibition by anti-C5 therapy. Circulation. 1998;97:2259– 2267. 162. Mahaffey KW, Granger CB, Nicolau JC, Ruzyllo W, Weaver WD, Theroux P, Hochman JS, Filloon TG, Mojcik CF, Todaro TG, Armstrong PW. Effect of pexelizumab, an anti-C5 complement antibody, as adjunctive therapy to fibrinolysis in acute myocardial infarction: The COMPlement inhibition in myocardial infarction treated with thromboLYtics

TARGETING DAMPS AFTER MI

(COMPLY) trial. Circulation. 2003;108:1176–1183. 163. Armstrong PW, Granger CB, Adams PX, Hamm C, Holmes D, O’Neill WW, Todaro TG, Vahanian A, Van de Werf F. Pexelizumab for acute ST-elevation myocardial infarction in patients undergoing primary percutaneous coronary intervention: a randomized controlled trial. JAMA. 2007;297:43– 51. 164. Rusnak JM, Kopecky SL, Clements IP, Gibbons RJ, Holland a E, Peterman HS, Martin JS, Saoud JB, Feldman RL, Breisblatt WM, Simons M, Gessler CJ, Yu a S. An anti-CD11/CD18 monoclonal antibody in patients with acute myocardial infarction having percutaneous transluminal coronary angioplasty (the FESTIVAL study). Am J Cardiol. 2001;88:482– 487. 165. Granger CB, Mahaffey KW, Weaver WD, Theroux P, Hochman JS, Filloon TG, Rollins S, Todaro TG, Nicolau JC, Ruzyllo W, Armstrong PW. Pexelizumab, an anti-C5 complement antibody, as adjunctive therapy to primary percutaneous coronary intervention in acute myocardial infarction: The COMplement inhibition in Myocardial infarction treated with Angioplasty (COMMA) trial. Circulation. 2003;108:1184–1190. 166. Faxon DP, Gibbons RJ, Chronos N a F, Gurbel P a., Sheehan F. The effect of blockade of the CD11/ CD18 integrin receptor on infarct size in patients with acute myocardial infarction treated with direct angioplasty: The results of the HALT-MI study. J Am Coll Cardiol. 2002;40:1199–1204. 167. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med. 2007;357:1121–1135. 168. Hausenloy DJ, Yellon DM. Myocardial ischemiareperfusion injury: a neglected therapeutic target. J Clin Invest. 2013;123:92–100. 169. N ahrendorf M, Swirski FK. Monocyte and macrophage heterogeneity in the heart. Circ Res. 2013;112:1624–33. 170. Jugdutt BI, Basualdo C a. Myocardial infarct expansion during indomethacin or ibuprofen therapy for symptomatic post infarction pericarditis. Influence of other pharmacologic agents during early remodelling. Can J Cardiol. 1989;5:211–221. 171. Kloner R a, Fishbein MC, Lew H, Maroko PR, Braunwald E. Mummification of the infarcted myocardium by high dose corticosteroids. Circulation. 1978;57:56–63. 172. Hammerman H, Kloner R a, Schoen FJ, Brown EJ, Hale SL, Braunwald E. The effects of early indomethacin administration on late scar formation following myocardial infarction. Trans Assoc Am Physicians. 1982;95:104–109. 173. H eusch G. Cardioprotection: Chances and challenges of its translation to the clinic. Lancet. 2013;381:166–175. 174. H ausenloy DJ, Yellon DM. “Conditional Conditioning” in cardiac bypass surgery. Basic Res Cardiol. 2012;107:258.

175. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 176. Burzyn D, Rassa JC, Kim D, Nepomnaschy I, Ross SR, Piazzon I. Toll-like receptor 4-dependent activation of dendritic cells by a retrovirus. J Virol. 2004;78:576–84. 177. Chen Q, Jin Y, Zhang K, Li H, Chen W, Meng G, Fang X. Alarmin HNP-1 promotes pyroptosis and IL-1β release through different roles of NLRP3 inflammasome via P2X7 in LPS-primed macrophages. Innate Immun. 2014;20:290–300. 178. Halle A, Hornung V, Petzold GC, Stewart CR, Monks BG, Reinheckel T, Fitzgerald KA, Latz E, Moore KJ, Golenbock DT. The NALP3 inflammasome is involved in the innate immune response to amyloid-beta. Nat Immunol. 2008;9:857–65. 179. Mulla MJ, Brosens JJ, Chamley LW, Giles I, Pericleous C, Rahman A, Joyce SK, Panda B, Paidas MJ, Abrahams VM. ORIGINAL ARTICLE: Antiphospholipid Antibodies Induce a ProInflammatory Response in First Trimester Trophoblast Via the TLR4/MyD88 Pathway. Am J Reprod Immunol. 2009;62:96–111. 180. Satta N, Kruithof EKO, Fickentscher C, DunoyerGeindre S, Boehlen F, Reber G, Burger D, de Moerloose P. Toll-like receptor 2 mediates the activation of human monocytes and endothelial cells by antiphospholipid antibodies. Blood. 2011;117:5523–31. 181. Biragyn A, Ruffini PA, Leifer CA, Klyushnenkova E, Shakhov A, Chertov O, Shirakawa AK, Farber JM, Segal DM, Oppenheim JJ, Kwak LW. Toll-like receptor 4-dependent activation of dendritic cells by beta-defensin 2. Science. 2002;298:1025–9. 182. Guo B-Y, Xie G-C, Duan Z-J, Xia L-J. [The effects of recombinant human beta-defensin-3 on expression of interleukin-17A and interleukin-22 in BEAS-2B cell]. Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi. 2013;27:260–2. 183. Westermann D, Mersmann J, Melchior A, Freudenberger T, Petrik C, Schaefer L, LüllmannRauch R, Lettau O, Jacoby C, Schrader J, BrandHerrmann S-M, Young MF, Schultheiss HP, Levkau B, Baba HA, Unger T, Zacharowski K, Tschöpe C, Fischer JW. Biglycan is required for adaptive remodeling after myocardial infarction. Circulation. 2008;117:1269–76. 184. Schaefer L, Babelova A, Kiss E, Hausser H-J, Baliova M, Krzyzankova M, Marsche G, Young MF, Mihalik D, Götte M, Malle E, Schaefer RM, Gröne H-J. The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest.

173

PART THREE CHAPTER 9

2005;115:2223–33. 185. Laudisi F, Spreafico R, Evrard M, Hughes TR, Mandriani B, Kandasamy M, Morgan BP, Sivasankar B, Mortellaro A. Cutting edge: the NLRP3 inflammasome links complement-mediated inflammation and IL-1β release. J Immunol. 2013;191:1006–10. 186. Lindberg S, Pedersen SH, Mogelvang R, Galatius S, Flyvbjerg A, Jensen JS, Bjerre M. Soluble form of membrane attack complex independently predicts mortality and cardiovascular events in patients with ST-elevation myocardial infarction treated with primary percutaneous coronary intervention. Am Heart J. 2012;164:786–92. 187. Yang D, Chen Q, Su SB, Zhang P, Kurosaka K, Caspi RR, Michalek SM, Rosenberg HF, Zhang N, Oppenheim JJ. Eosinophil-derived neurotoxin acts as an alarmin to activate the TLR2-MyD88 signal pathway in dendritic cells and enhances Th2 immune responses. J Exp Med. 2008;205:79–90. 188. Pal D, Dasgupta S, Kundu R, Maitra S, Das G, Mukhopadhyay S, Ray S, Majumdar SS, Bhattacharya S. Fetuin-A acts as an endogenous ligand of TLR4 to promote lipid-induced insulin resistance. Nat Med. 2012;18:1279–85. 189. Smiley ST, King JA, Hancock WW. Fibrinogen stimulates macrophage chemokine secretion through toll-like receptor 4. J Immunol. 2001;167:2887–94. 190. Zhou R, Tardivel A, Thorens B, Choi I, Tschopp J. Thioredoxin-interacting protein links oxidative stress to inflammasome activation. Nat Immunol. 2010;11:136–40. 191. Su H, Ji L, Xing W, Zhang W, Zhou H, Qian X, Wang X, Gao F, Sun X, Zhang H. Acute hyperglycaemia enhances oxidative stress and aggravates myocardial ischaemia/reperfusion injury: role of thioredoxin-interacting protein. J Cell Mol Med. 2013;17:181–91. 192. Johnson GB, Brunn GJ, Kodaira Y, Platt JL. Receptor-mediated monitoring of tissue well-being via detection of soluble heparan sulfate by Toll-like receptor 4. J Immunol. 2002;168:5233–9. 193. Yamasaki K, Muto J, Taylor KR, Cogen AL, Audish D, Bertin J, Grant EP, Coyle AJ, Misaghi A, Hoffman HM, Gallo RL. NLRP3/cryopyrin is necessary for interleukin-1beta (IL-1beta) release in response to hyaluronan, an endogenous trigger of inflammation in response to injury. J Biol Chem. 2009;284:12762– 71. 194. Jiang D, Liang J, Fan J, Yu S, Chen S, Luo Y, Prestwich GD, Mascarenhas MM, Garg HG, Quinn DA, Homer RJ, Goldstein DR, Bucala R, Lee PJ, Medzhitov R, Noble PW. Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nat Med. 2005;11:1173–9. 195. Leadbetter EA, Rifkin IR, Hohlbaum AM, Beaudette BC, Shlomchik MJ, Marshak-Rothstein A. Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature.

174

2002;416:603–7. 196. Curran CS, Demick KP, Mansfield JM. Lactoferrin activates macrophages via TLR4-dependent and -independent signaling pathways. Cell Immunol. 2006;242:23–30. 197. Devaney JM, Greene CM, Taggart CC, Carroll TP, O’Neill SJ, McElvaney NG. Neutrophil elastase upregulates interleukin-8 via toll-like receptor 4. FEBS Lett. 2003;544:129–32. 198. Akiyama D, Hara T, Yoshitomi O, Maekawa T, Cho S, Sumikawa K. Postischemic infusion of sivelestat sodium hydrate, a selective neutrophil elastase inhibitor, protects against myocardial stunning in swine. J Anesth. 2010;24:575–81. 199. Shinohara ML, Lu L, Bu J, Werneck MBF, Kobayashi KS, Glimcher LH, Cantor H. Osteopontin expression is essential for interferon-alpha production by plasmacytoid dendritic cells. Nat Immunol. 2006;7:498–506. 200. Mohamed IA, Mraiche F. Targeting osteopontin, the silent partner of NA(+) /H(+) exchanger isoform 1 in cardiac remodeling. J Cell Physiol. 2015; 201. Bae YS, Lee JH, Choi SH, Kim S, Almazan F, Witztum JL, Miller YI. Macrophages generate reactive oxygen species in response to minimally oxidized low-density lipoprotein: toll-like receptor 4- and spleen tyrosine kinase-dependent activation of NADPH oxidase 2. Circ Res. 2009;104:210–8, 21p following 218. 202. Stewart CR, Stuart LM, Wilkinson K, van Gils JM, Deng J, Halle A, Rayner KJ, Boyer L, Zhong R, Frazier WA, Lacy-Hulbert A, El Khoury J, Golenbock DT, Moore KJ. CD36 ligands promote sterile inflammation through assembly of a Toll-like receptor 4 and 6 heterodimer. Nat Immunol. 2010;11:155–61. 203. Erridge C, Kennedy S, Spickett CM, Webb DJ. Oxidized phospholipid inhibition of toll-like receptor (TLR) signaling is restricted to TLR2 and TLR4: roles for CD14, LPS-binding protein, and MD2 as targets for specificity of inhibition. J Biol Chem. 2008;283:24748–59. 204. Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. J Biol Chem. 2001;276:16683–9. 205. Cheng N, He R, Tian J, Ye PP, Ye RD. Cutting edge: TLR2 is a functional receptor for acute-phase serum amyloid A. J Immunol. 2008;181:22–6. 206. Montalbano AP, Hawgood S, Mendelson CR. Mice deficient in surfactant protein A (SP-A) and SP-D or in TLR2 manifest delayed parturition and decreased expression of inflammatory and contractile genes. Endocrinology. 2013;154:483–98. 207. Scheel B, Teufel R, Probst J, Carralot J-P, Geginat J, Radsak M, Jarrossay D, Wagner H, Jung G, Rammensee H-G, Hoerr I, Pascolo S. Toll-like receptor-dependent activation of several human blood cell types by protamine-condensed mRNA.

Eur J Immunol. 2005;35:1557–66. 208. Midwood K, Sacre S, Piccinini AM, Inglis J, Trebaul A, Chan E, Drexler S, Sofat N, Kashiwagi M, Orend G, Brennan F, Foxwell B. Tenascin-C is an endogenous activator of Toll-like receptor 4 that is essential for maintaining inflammation in arthritic joint disease. Nat Med. 2009;15:774–80. 209. Franchi L, Eigenbrod T, Núñez G. Cutting edge: TNF-alpha mediates sensitization to ATP and silica via the NLRP3 inflammasome in the absence of microbial stimulation. J Immunol. 2009;183:792–6. 210. Liu-Bryan R, Scott P, Sydlaske A, Rose DM, Terkeltaub R. Innate immunity conferred by Toll-like receptors 2 and 4 and myeloid differentiation factor 88 expression is pivotal to monosodium urate monohydrate crystal-induced inflammation. Arthritis Rheum. 2005;52:2936–46. 211. Martinon F, Pétrilli V, Mayor A, Tardivel A, Tschopp J. Gout-associated uric acid crystals activate the NALP3 inflammasome. Nature. 2006;440:237–41. 212. Toeda K, Nakamura K, Hirohata S, Hatipoglu OF, Demircan K, Yamawaki H, Ogawa H, Kusachi S, Shiratori Y, Ninomiya Y. Versican is induced in infiltrating monocytes in myocardial infarction. Mol Cell Biochem. 2005;280:47–56. 213. Kim S, Takahashi H, Lin W-W, Descargues P, Grivennikov S, Kim Y, Luo J-L, Karin M. Carcinomaproduced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature. 2009;457:102–6.

175

chapter 10

The selective NLRP3-inflammasome inhibitor MCC950 reduces infarct size and preserves cardiac function in a pig model of myocardial infarction

In revision - European Heart Journal

G.P.J. van Hout1, L. Bosch1, G.H.J.M. Ellenbroek1, J.J. de Haan1, W.W. van Solinge2, M.A. Cooper3, F. Arslan1, S.C.A. de Jager1, A.A.B. Robertson3, G. Pasterkamp1,2, I.E. Hoefer1,2 1 2

Experimental Cardiology Laboratory, University Medical Center Utrecht, The Netherlands Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, The Netherlands

Institute for Molecular Bioscience, The University of Queensland, Brisbane, Australia

PART THREE CHAPTER 10

ABSTRACT Background Myocardial infarction (MI) triggers an intense inflammatory response that is associated with infarct expansion and is detrimental for cardiac function. Interleukin (IL)-1β and IL-18 are key players in this response and are controlled by the NLRP3-inflammasome. In the current study we therefore hypothesized that selective inhibition of the NLRP3-inflammasome reduces infarct size and preserves cardiac function in a highly translational porcine MI model. Methods and Results Thirty female landrace pigs were subjected to 75 minutes transluminal balloon occlusion and treated with the NLRP3-inflammasome inhibitor MCC950 (6mg/kg or 3mg/kg) or placebo for 7 days in a randomized, blinded fashion. After 7 days, 3D-echocardiography was performed to assess cardiac function and Evans blue / TTC double staining was executed to assess the area at risk (AAR) and infarct size (IS). The IS/AAR was lower in the 6 mg/kg group (64.6±8.8%, p=0.004) and 3 mg/kg group (69.7±7.2%, p=0.038) compared to the control group (77.5±6.3%). MCC950 treatment markedly preserved left ventricular ejection fraction in treated animals (6 mg/kg 47±8%, p=0.001; 3 mg/kg 45±7%, p=0.031; control 37±6%). Myocardial neutrophil influx was attenuated in treated compared to non-treated animals (6 mg/kg 132±72 neutrophils/ mm2, p=0.035; 3 mg/kg 207±210 neutrophils/mm2, p=ns; control 266±158 neutrophils/ mm2). Myocardial IL-1β levels showed a dose-dependent trend among experimental groups. Conclusions NLRP3-inflammasome inhibition reduces infarct size and preserves cardiac function in a randomized, blinded translational large animal MI model. Hence, NLRP3-inflammasome inhibition may have therapeutic potential in acute MI patients.

178

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

INTRODUCTION Myocardial infarction (MI) is one of the most important causes of death worldwide1,2. Improved treatment strategies for MI, including percutaneous coronary intervention (PCI), have led to better survival3. However, patients with deteriorated cardiac function are at increased risk for heart failure (HF) and the improved survival of MI patients potentiates its incidence. HF is accountable for a large fraction of cardiovascular deaths, is associated with a poor quality of life and extensive health care costs4. Novel therapeutics that prevent HF post-MI through preserving cardiac function are therefore crucial. Cardiac ischemia-reperfusion triggers a sterile inflammatory reaction. Though essential for wound healing, this intense inflammatory response also deteriorates cardiac function5,6morbidity related to myocardial infarction is increasing in Western societies. Acute and chronic immune responses elicited by myocardial ischemia have an important role in the functional deterioration of the heart. Research on modulation of the inflammatory responses was focused on effector mediators such as leukocytes. However, increasing evidence indicates that various endogenous ligands that act as ‘danger signals’, also called danger-associated molecular patterns (DAMPs. Upon reperfusion, danger molecules secreted by damaged cardiac resident cells attract leukocytes to the injured myocardium7. Next to removing cell debris, these cells exert detrimental effects, thereby contributing to myocardial reperfusion injury and infarct expansion8,9arterial wall, and cardiomyocytes in a very well-choreographed manner. Neutrophils have been implicated as primary and secondary mediators of lethal injury after reperfusion to coronary vascular endothelium and cardiomyocytes. The involvement of neutrophils in the pathogenesis of lethal myocardial injury has been inferred from (1. Infarct size and cardiac function are long-term predictors of adverse remodeling and HF10. Immediate damage control by attenuating the post-MI inflammatory response could therefore be of great benefit in MI patients. The interleukin (IL)-1 family is strongly involved in the inflammatory response after MI11. In this context, interleukin (IL)-1β and IL-18 are particularly important12. IL-1β levels are upregulated in the infarcted myocardium and IL-18 and IL-1β have been described to directly impair cardiac contractility in a synergistic way13. Moreover, their levels predict the occurrence of adverse events in patients after MI14–17however, whether elevations of circulating IL-18 precede the onset of coronary events in apparently healthy individuals. METHODS AND RESULTS: We evaluated the relationship between baseline plasma levels of IL-18 and the subsequent incidence of coronary events over a 5-year follow-up in the Prospective Epidemiological Study of Myocardial Infarction (PRIME and attenuation of IL-1β or IL-18 signaling has been shown to reduce infarct size and preserve cardiac function in rodent studies18–21. Recently, it has become evident that IL-1β and IL-18 signaling is regulated by the NODLike-Receptor, pyrin containing domain 3 (NLRP3)-inflammasome22. Activation of the NLRP3-inflammasome occurs upon tissue injury and leads to cleavage of caspase-1, which in turn converts pro-IL-1β and pro-IL-18 into their mature form. Interference with NLRP3inflammasome signaling results in reduced IL-1β and IL-18 signaling, combined with

179

PART THREE CHAPTER 10

decreased caspase-1 induced cell death (pyroptosis)23. Consequently, NLRP3-inflammasome inhibition leads to pronounced infarct size reduction, attenuation of adverse remodeling and preservation of cardiac function in small animal MI models24–27but also by endogenous danger signals released during ischemia and necrosis. As triggers for the innate immune NLRP3 inflammasome protein complex appear to overlap with those for cardiac ischemiareperfusion (I/R. Although these studies provide important mechanistic insights, they do not reflect clinical application since they either involved knockout models or pharmacological treatment preceding MI induction25,27,28the mechanism by which myocardial I/R induces inflammation remains unclear. Recent evidence indicates that a sterile inflammatory response triggered by tissue damage is mediated through a multiple-protein complex called the inflammasome. Therefore, we hypothesized that the inflammasome is an initial sensor for danger signal(s. To successfully translate these findings into clinical applications, preclinical testing in clinically relevant models is mandatory29. Until recently, this was precluded by the lack of selective pharmacological NLRP3inflammasome inhibitors28associated with hypoglycemia. The aim of this study was to measure the effects on the NLRP3 inflammasome of 16673-34-0, an intermediate substrate free of the cyclohexylurea moiety, involved in insulin release.\\n\\nMETHODS AND RESULTS: We synthesized 16673-34-0 (5-chloro-2-methoxy-N-[2-(4-sulfamoylphenyl. MCC950 is a novel, selective small-molecule NLRP3-inflammasome inhibitor with powerful in vitro and in vivo inhibitory effects30, enabling clinically relevant testing in large animal models. We therefore hypothesized that pharmacological interference with NLRP3inflammasome signaling by administration of MCC950 results in infarct size reduction and preservation of cardiac function in a highly translational porcine MI model.

METHODS In vitro assay To measure MCC950 efficacy in pigs, porcine peripheral blood mononuclear cells (PBMCs) were isolated from baseline blood samples (n=6) using Ficoll density-gradient centrifugation. Cells were kept in RPMI medium containing 1% fetal bovine serum at 37˚C and were stimulated with 10µg/ml LPS (Sigma-Aldrich, St. Louis, MO, USA). After one hour, 5mM adenosine triphosphate (ATP) (Sigma-Aldrich, St. Louis, MO, USA) and different doses of MCC950 (0µM, 0.3µM, 75µM or 450µM) were added. After 3 hours total incubation time, IL-1β synthesis was measured in the supernatant using a luminex immunoassay specific for porcine IL-1β (Procarta™ Simplex, eBioscience, San Diego, CA, USA), according to the manufacturer’s protocol. Measured IL-1β concentrations were corrected for the number of PBMCs. Animal Study design All animal experiments were approved by the institutional animal welfare committee and were executed conforming to the ‘Guide for the Care and Use of Laboratory Animals’.

180

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

A total of 30 female landrace pigs (body weight 66±4.0 kg) were subjected to transluminal closed-chest left anterior descending coronary artery (LAD) balloon occlusion for 75 minutes followed by 7 days of reperfusion. One hour prior and two hours post-occlusion, pigs were subjected to transesophageal three-dimensional echocardiography (3D TEE). Fifteen minutes prior to reperfusion, pigs were randomly assigned to intravenous infusion with either a high dose of MCC950 (6 mg/kg in 40mL phosphate buffered saline (PBS)), a low dose of MCC950 (3 mg/kg in 40mL PBS) or PBS alone at a rate of 80mL/hour. Intravenous infusion was repeated daily on day 1 to day 6. At the moment of balloon deflation an additional gift of 6mg (high dose, in 5mL PBS), 3mg (low dose, in 5mL PBS) or PBS alone (5mL) was selectively injected into the LAD. On day 7, animals were again anesthetized, subjected to 3D TEE and invasive pressure-volume (PV) measurements. This was followed by transthoracic echocardiography and dobutamine stress-echocardiography to assess regional contractility and cardiac reserve capacity. We then performed in vivo determination of the area at risk (AAR) through injection of Evans blue. Finally, the heart was explanted for the determination of infarct size and the myocardial inflammatory response. The investigators were blinded to which treatment group the animals were allocated during both the experiments and the analysis of results. Allocation of animals to the different experimental groups occurred randomly. Detailed methods of echocardiography, PV measurements and infarct size assessment are described in the supplemental methods. Circulating levels of MCC950 In vivo MCC950 concentrations were measured in plasma samples using mass-spectrometry. 500 µL methanol was added to 100 µL of sample and then centrifugation at 20200g for 5 minutes. The supernatant was evaporated at 40°C under gentle nitrogen stream and dissolved in 100 µl 0.1% formic acid in H2O. Analytes were resolved by reversed phase liquid chromatography using a Waters Acquity UPLC with a HSS T3 (2.1 mm × 100 mm × 1.8 µm) column hold at 30°C. We quantified the MCC950 content using negative ionization mode and multiple reaction monitoring on a Waters Xevo TQ mass spectrometer (MRM transition: 403.10 > 204.06). Instrument specific cone and collision parameters were derived experimentally. Total instrument analysis time was 10 minutes per injection. Calibration levels of MCC950 were made in methanol in the range of 0.0002-13.8 µM and mixed with the same amount of pooled plasma, followed by the same sample preparation procedure. measurements. Neutrophil numbers, IL-1β assay & circulating markers Circulating neutrophil numbers at different time points after reperfusion were measured by whole-blood analysis using an automated hematological cell-counter (Cell-Dyn Sapphire, Abbott, Santa Clara, CA, USA). The Cell-Dyn Sapphire is a routine hematology analyzer, which uses spectophotometry, electrical impedance and laser light scattering to classify blood cells (platelets, erythrocytes and leukocytes).

181

PART THREE CHAPTER 10

Plasma samples were obtained by whole blood centrifugation at 1850g and were immediately stored at -80ºC. Troponin I (TnI), aspartate transaminase (AST), Alanine aminotransferase (ALT) and C-reactive protein (CRP) levels were measured using a clinical chemistry analyzer (AU5811, Beckman Coulter). Neutrophil numbers in myocardial tissue were quantified in paraffin-embedded histological biopsies that were conserved in 4% formaldehyde for at least 7 days. Histological samples were cut into 5μm sections. Sections were deparaffinized followed by incubation with a monoclonal mouse anti-pig antibody against porcine neutrophils (Clone PM1, BMA Biomedicals, Augst, Switzerland) for 60 minutes and secondary incubation with Brightvision Poly-AP-anti-mouse (ImmunoLogic, Duiven, the Netherlands) for 30 minutes. For microscopic visualization, incubation with liquid permanent red (DAKO, Heverlee, Belgium) for 10 minutes was performed. For every animal, semi-automated analysis was performed at 200x magnification using CellSens software (Center Valley, PA, USA). IL-1β was measured in the myocardium using a luminex immunoassay (Procarta™ Simplex, eBioscience, San Diego, CA, USA) according to the manufacturer’s instructions and corrected for total protein concentration. Statistics All data are expressed as mean ± standard deviation unless stated otherwise. Myocardial infarct size, left ventricular ejection fraction (LVEF) and all other outcomes were compared using a One-way ANOVA followed by post-hoc analysis to compare individual groups with each other. Blood parameters levels and leukocyte levels in the three treatment groups at different time points were analyzed using mixed models. Not normally (gausian) distributed parameters were transformed with the natural logarithm (LN). The mixed models include group and time point as fixed factors and a random intercept for each pig. To determine whether the course in time of the parameters was different for the groups the interaction group*time point of measurement was also taken into the model. When the P-value for the test for intraction was greater than 0.25 it was excluded from the model and also baseline measurements were not taken into the model.All statistical analyses were performed in SPSS statistics version 20.0. A two-sided P-value of <0.05 was regarded statistically significant in all analyses.

RESULTS In vitro assay In vitro stimulation of isolated porcine mononuclear cells with LPS and ATP led to a pronounced secretion of IL-1β (figure 1A). Administration of 0.3 µM MCC950 significantly attenuated this response (-45±21%, p=0.002). Increasing MCC950 concentrations further reduced IL-1β secretion (75 µM: -51±24%, p=0.001 compared to control, p=0.023 compared to 0.3 µM; 450 µM: -64±21%, p<0.001 compared to control, p=0.011 compared to 75 µM) (figure 1A).

182

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

Survival, hemodynamics and MCC950 levels Three out of 30 pigs subjected to MI died before the follow-up duration of 7 days was completed. Two animals died before reperfusion (1 in control group, 1 in low dose group) due to persistent VF. One pig died on day one after MI induction (high dose group), without preceding clinical signs of acute cardiac failure. Echocardiography prior to MI induction revealed a congenital anomaly (ventricle septum defect) in 1 pig (low dose group), which was therefore excluded from the study. This allowed analysis of 9 animals in the control group, 8 animals in the low dose group and 9 animals in the high dose group. MCC950 did not exert any direct chronotropic or inotropic effects in these animals since heart rate (HR) and mean arterial blood pressure (MAP) were similar during the first 3 hours after the induction of myocardial infarction and subsequent compound administration (supplemental table 1). To verify that daily intravenous administration of MCC950 led to continuous inhibition of the NLRP3-inflammasome, in vivo levels were measured at 15 minutes of reperfusion, at 4 hours reperfusion and just prior to the second intravenous administration of the compound (24 hours reperfusion) (figure 1B). At 15 minutes reperfusion, the high dose group had significantly higher circulating MCC950 levels compared to the low dose group (101±15 µM vs. 55±6 µM, p<0.001). These levels remained significantly different at 4 hours reperfusion (18±6.8 µM vs. 9.5±5 µM, p=0.009). Just prior to the second daily intravenous infusion, MCC950 levels were still detectable and were within the range that proved efficacious during in vitro inhibition (supplemental table 2), suggesting continuous inhibition of the NLRP3-inflammasome.

Figure 1. MCC950 dose-dependently attenuates IL-1β signaling in porcine PBMCs A. Isolated porcine PBMCs secrete IL-1β after administration of LPS and ATP in vitro. Addition of MCC950 after LPS/ATP stimulation dose-dependently reduced IL-1 β release (n=6). B. In vivo MCC950 levels during the first 24 hours of the experiment were within therapeutic range. MCC950 levels were measured at 15 minutes, 4 hours and 24 hours of reperfusion (n=8-9 per group). *=p<0.05

183

PART THREE CHAPTER 10

Figure 2. MCC950 administration dose-dependently reduces infarct size A. Representative pictures of myocardial segments of the three experimental groups. The blue area represents the remote area, the AAR is stained red and the infarcted myocardium is stained white. B. The AAR was similar in all groups. C. The IS/AAR was significantly lower in the high dose group and the low dose group compared to the control group. D. The IS/LV was significantly lower in the high dose group but not in the low dose group compared to the control group. AAR = area at risk, LV = left ventricle, IS = infarct size. * p<0.05

Infarct size At 7 days follow-up, infarct size was assessed by Evans blue/TTC double staining. Figure 2A shows representative pictures of myocardial segments from the 6 mg/kg group, the 3 mg/kg group and the placebo treated group with the remote area in blue, the area at risk (AAR) in red and infarcted tissue in white. The AAR as a percentage of the LV (AAR/LV) was similar in all three groups (high dose group 21.3±3.2%, low dose group 23.0±6.9%, control group 22.6±4.4%, p=0.7) (figure 2B). Infarct size (IS) as a percentage of the AAR (IS/AAR) was significantly higher in the control group compared to both treatment groups (high dose group 64.6±8.8%, p=0.004; low dose group 69.7±7.2%, p=0.038; control group 77.5±6.3%) (figure 2C). The IS as a percentage of the LV (IS/LV) of the control group was significantly higher compared to the high dose group, but not to the low dose group (high dose group 13.7±2.8%, p=0.023; low dose group 15.8±4.2%, p=0.5; control group 17.2±3.1%) (figure 2D). Global cardiac function Global cardiac function was measured by 3D TEE at baseline, at 2 hours reperfusion and after 7 days of follow-up. Baseline cardiac function was similar in all groups. At 2 hours

184

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

Table 1. Left ventricular geometrical parameters at baseline, 2 hours reperfusion and 7 days reperfusion Parameter

Baseline

2 hours reperfusion

7 days reperfusion

Con

3mg

6mg

Con

3mg

6mg

Con

3mg

6mg

125±14

122±18

121±13

96±15

96±16

104±19

129±13

127±20

131±15

ESV (mL)

51±7

50±14

48±10

52±10

53±12

57±16

82±13

70±13†

69±12*

SV (mL)

74±11

72±7

73±6

44±7

48±11

47±8

57±14†

62±8*

EF (%)

59±4

60±6

61±4

46±5

46±6

46±8

37±6

45±7*

48±5*

EDV (mL)

EDV = end diastolic volume, ESV = end systolic volume, SV = stroke volume, EF = ejection fraction. 6mg = 6 mg/kg group, 3mg = 3 mg/kg group, Con = control group. * = p<0.05, † = 0.05<p<0.1

reperfusion, global cardiac function did not significantly differ among groups (table 1). At follow-up, no difference in end diastolic volume (EDV) was observed between the different groups (figure 3A). End systolic volume at 7 days was significantly lower in the high dose group and a trend was observed for the low dose group compared to the control group (high dose group 69±12 mL, p=0.042; low dose group 70±13 mL, p=0.085; control group 82±13 mL) (figure 3B). Stroke volume (SV) was significantly higher at 7 days in the high dose group and again a trend was observed for the low dose group (high dose group 62±8 mL, p=0.001; low dose group 57±14 mL, p=0.096; control group 47±8 mL) (table 1). LV ejection fraction (EF) was higher in both the high dose group and the low dose group compared to the control group after 7 days follow-up (high dose group 47±8%, p=0.001; low dose group 45±7%, p=0.031; control group 37±6%) (figure 3C). The end systolic pressure volume relationship (ESVPR), measured by invasive real-time PV loops, showed a higher contractility for the high dose group, but not for the low dose group compared to the control group at 7 days follow-up (high dose group 4.7±2.0 mmHg/mL, p=0.043; low dose group 2.7±1.4 mmgHg/mL, p=0.8; control group 2.5±1.7 mmHg/mL) (figure 3D). Regional cardiac function and dobutamine echocardiography At 7 days follow-up, regional cardiac function was assessed at a midventricular and apical level by transthoracic echocardiography, before and after dobutamine infusion. The contractility of the AAR (septal wall) was assessed through measuring SWT, which was significantly higher in both treatment groups compared to the control group (figure 4A & 4B). SWT of the remote myocardium (lateral wall) was also significantly higher at a midventricular level in animals treated with the high dose, while apical lateral SWT did not differ among groups (figure 4C & 4D). FS and FAS were significantly different between treated and non-treated animals (figure 4E - 4H). Dobutamine stress-echocardiography revealed dose-dependent significant differences in cardiac reserve capacity (table 2). Midventricular septal and lateral WT were significantly higher in animals treated with MCC950 compared to control animals after dobutamine administration (Septal: high dose group 1.5±0.2 cm, p=0.005; low dose group 1.6±0.3 cm, p=0.011, control group 1.3±0.2 cm) (Lateral: high dose group 1.8±0.3 cm, p=0.010;

185

PART THREE CHAPTER 10

low dose group 2.0±0.4 cm, p=0.003; 1.5±0.1 cm) (figure 5A & 5B). Stress-echocardiography also revealed an increased septal WT in the high dose group (1.5±0.2 cm, p=0.043) and a trend in the low dose group (1.5±0.3 cm, p=0.076) compared to control animals (1.2±0.3 cm) at the apical level, while the difference in lateral WT between groups during stress echocardiography did not reach statistical significance at this level (figure 5C & 5D). Serological and histological read-outs The extent of cardiac damage was also reflected by systemic cardiac marker concentrations. A trend towards lower troponin I levels was observed at 4 hours reperfusion in the high dose group (7870±3809 pg/mL, p=0.080), but not in the low-dose group (9006±2024 pg/ mL, p=0.2) compared to the control group (12453±6304 pg/mL)(figure 6A). The circulating levels of AST differed significantly among groups (p=0.049) and a trend was observed for differences in ALT levels (p=0.063) during the 7-day follow-up period (figure 6B & 6C). To

Figure 3. NLRP3-inflammasome inhibition dose-dependently preserves global cardiac function A. EDV did not differ among the experimental groups. B. ESV was lower in the high dose group and a trend was observed for the low dose group compared to the control group. C. EF was significantly lower in the high dose group and the low dose group compared to the control group. D. The ESPVR in the high dose group but not the low dose group differed from the control group. EDV = end diastolic volume, ESV = end systolic volume, EF = ejection fraction, ESPVR = end systolic pressure volume relationship. * p<0.05, † = 0.05<p<0.1

186

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

assess if inflammasome inhibition also reduced systemic inflammatory markers post MI, systemic neutrophil numbers and CRP were measured. Again a trend was observed for a difference in circulating neutrophil levels (p=0.056), while the interaction between treatment group and time point was significant for CRP (p=0.030), indicating MCC950 influenced CRP levels in a time-dependent manner post-MI (figure 6D & 6E). Intra-myocardial IL-1β levels showed a trend towards higher levels in the control group compared to the treatment groups (high dose group 131±82 pg/mg, p=0.076; low dose group 156±64 pg/mg, p=0.2; control group 211±104 pg/mg) (figure 6F). A dosedependent effect was observed for the influx of neutrophils into the infarcted myocardium (high dose group 132±72 neutrophils/mm2, p=0.035; low dose group 207±210 neutrophils/ mm2, p=0.5; control group 266±158 neutrophils/mm2) (figure 6G & 6H).

Figure 4. Regional cardiac function is preserved after NLRP3-inflammasome inhibition Midventricular and apical septal SWT were significantly higher in the high dose and low dose group compared to the control group (A&B). Lateral SWT at a midventricular level was higher in the high dose group while lateral SWT at an apical level was equal among groups (C&D). FS at a midventricular level was higher after MCC950 administration while no differences were observed at an apical level (E&F). FAS was higher in MCC950 compared to placebo treated animals at a midventricular and apical level (G&H). SWT = septal wall thickening, FS = fractional shortening, FAS = fractional area shortening. * p<0.05, † = 0.05<p<0.1

187

PART THREE CHAPTER 10

Table 2. Left ventricular regional systolic parameters during dobutamine stress-echocardiography Parameters

Septal WTES (cm)

Mid ventricular

Apex

Con

3mg

6mg

Con

3mg

6mg

1.3±0.2

1.6±0.3*

1.5±0.2*

1.2±0.3

1.5±0.3†

1.5±0.2*

9±16

24±22

26±10*

7±10

17±16

22±12*

Lateral WTES (cm)

1.5±0.1

2.0±0.4*

1.8±0.3*

1.3±0.3

1.6±0.4

1.5±0.2

Lateral SWT (%)

39±22

73±36*

73±38*

12±14

39±20*

48±22*

LViDES (cm)

3.3±0.7

2.7±0.5*

2.3±0.6

2.0±0.4

1.8±0.5†

28±8

40±13*

38±7*

29±11

32±9

36±11

9.9±2.9

6.7±2.2*

7.0±1.8*

4.6±1.3

3.2±0.9*

2.0±1.0*

45±9

56±17

56±8*

46±10

53±12

63±13*

Septal SWT (%)

FS (%) LViaES (cm2) FAS (%)

WTES = End systolic wall thickness, SWT = systolic wall thickening, LViDES = End systolic left ventricular internal diameter, FS = fractional shortening, LViaES = End systolic left ventricular internal area, FAS = fractional area shortening. 6mg = 6 mg/kg group, 3mg = 3 mg/kg group, Con = control group. * = p<0.05, † = 0.05<p<0.1

DISCUSSION Mechanistic studies have shown an essential role for the NLRP3-inflammasome in IL-18 and IL-1β driven inflammation in both cardiac resident and circulating cells after MI31. NLRP3-inflammasome activation also leads to caspase-1-dependent cell death (pyroptosis) in rodent cardiomyocytes23. Both IL-18 and IL-1β signaling as well as pyroptosis induce amplification of the initial ischemic damage, culminating in infarct size expansion and decreased cardiac contractility, thereby increasing the risk of HF11,13,32,33. Our study is in line with these findings and for the first time provides in vivo efficacy data for infarct size reduction and cardiac function preservation by NLRP3-inflammasome inhibition in a clinically relevant large animal myocardial infarction model. MCC950 has recently been shown to selectively inhibit NLRP3-inflammasome formation and reduce pyroptosis, IL-18 and IL-1β signaling30. Given the size and specificity of MCC950, this compound could potentially serve as a clinically relevant inhibitor for NLRP3inflammasome driven diseases. In the current study we show that daily intravenous administration of MCC950 can maintain pharmacologically active circulating concentration as evident from our and others’ in-vitro studies30. Importantly, we studied for the first time the effect of MCC950 in an animal MI model. Transgenic mouse models and pretreatment protocols (pharmacological preconditioning) in rodent studies have elucidated the mechanisms of the post-MI inflammatory response. However, inflammation-related signaling pathways significantly differ between small and larger mammals34,35. Moreover, large animal models allow application of minimally invasive techniques, thereby avoiding major traumatic injury that could confound the possible effect of anti-inflammatory treatment strategies36,37. Transgenic mouse models and pretreatment protocols (pharmacological preconditioning) in rodent studies have elucidated the

188

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

mechanisms of the post-MI inflammatory response. However, inflammation-related signaling pathways significantly differ between small and larger mammals34,35. Moreover, large animal models allow application of minimally invasive techniques, thereby avoiding major traumatic injury that could confound the possible effect of anti-inflammatory treatment strategies36,37. Hence, large animal model testing is an essential step in bringing NLRP3-inflammasome targeted therapies from bench into clinical application29,38â&#x20AC;&#x201C;41. In the current study we show that selective inhibition of the NLRP3-inflammasome dosedependently reduces infarct size in a porcine MI model according to a treatment regimen that is clinically feasible. A pronounced difference in LV function was found in treated compared to non-treated animals at 7 days post-MI. This effect found in our study is markedly higher than the effects found in most large animal cardioprotection studies41. At two hours reperfusion, no differences in cardiac function was detected, possibly indicating that the primary effect of NLRP3-inflammasome inhibition lies in attenuating the inflammatory response in the subacute phase after MI, something that has also been observed by others42.

Figure 5. MCC950 administration preserves the cardiac reserve capacity A. Septal WT at a midventricular level was higher prior to dobutamine infusion and remained higher after dobutamine infusion in pigs treated with MCC950 compared to control animals. B. Midventricular lateral WT prior to dobutamine infusion was similar among groups. Dobutamine infusion lead to significantly higher Midventricular lateral WT in MCC950 treated animals compared to control animals. C. Apical septal WT after dobutamine infusion was higher in MCC950 treated animals compared to control animals. D. Apical lateral WT differed non-significantly after dobutamine infusion among groups. WT = wall thickness. * p<0.05, â&#x20AC; = 0.05<p<0.1

189

PART THREE CHAPTER 10

Assessment of local cardiac contractility revealed that not only the cardiac AAR responded to treatment, but also the remote myocardium of treated animals showed increased contractility compared to control animals. This could be explained by the observation that IL-18 and IL-1β directly suppress cardiac performance13. Additionally, the reserve capacity of the myocardium was higher in MCC950 treated animals compared to placebo treated animals after dobutamine infusion. This suggests that NLRP3-inflammasome inhibition attenuates the decrease in exercise capacity that is often observed in patients post-MI43. NLRP3-inflammasome inhibition could possibly also lead to the prevention of LV geometrical changes, however we did not observe excessive cardiac dilatation in control animals after the follow-up duration of 7 days. Since infarct resorption occurs at longer follow-up duration, Evans Blue / TTC double staining becomes less reliable, thereby obscuring any effects on infarct size reduction44. Moreover, IL-1β levels are mainly elevated during the (sub)acute phase post-MI25. A longer follow-up period would therefore preclude the assessment of myocardial IL-1β content. For these reasons we decided to limit the follow-up to 7 days in the current study.

Figure 6. Inhibition of NLRP3-inflammasome signaling reduces inflammation A. A dose-dependent trend was observed in cardiac troponin levels at 4 hours reperfusion. B. Circulating levels of AST were significantly different among groups. C. A trend towards different circulating levels of ALT was observed. D. A trend towards different circulating levels of neutrophils during the 7-day follow-up period in the different experimental groups was observed. E. Circulating levels of CRP were significantly lower in the high dose group compared to the control group over time. F. A dose-dependent trend was seen towards lower myocardial IL-1β content 7 days post-MI. G. MCC950 administration led to a reduced influx of neutrophils after MI in the high dose. H. Representative pictures of neutrophils (in red) in the infarcted myocardium of the different experimental groups. AST= aspartate transaminase, ALT = alanine transaminase, CRP = c-reactive protein. Data are depicted as mean±SEM. * p<0.05, † = 0.05<p<0.1

190

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

Our study reveals that circulating markers of damage and inflammation were lower in animals treated with a high dose of MCC950. Myocardial infiltration of circulating neutrophils was lower and a trend was observed for decreased myocardial IL-1β levels in the high dose group. Since the low dose group also showed preservation of cardiac function, possible explanations for the discrepancy between the effects on cardiac function and inflammation could be that partial inhibition of IL-1β and IL-18 does not affect systemic inflammatory parameters in large animals, but directly improves cardiac contractility13. In our model, circulating IL-1β or IL-18 levels were undetectable after MI. Therefore we were unable to assess the direct effect of MCC950 on these cytokines. The inhibition of the NLRP3-inflammasome may have advantages over the inhibition of IL-1β alone, since the targeted blockade of IL-1β does not reduce mortality in previously reported mouse models, whereas MCC950 administration does30,45. Moreover, it remains unclear to what extent IL-1β signaling, IL-18 signaling or pyroptosis are responsible for the detrimental effects post-MI. At least some of this effect may be attributed to IL-18 signaling and pyroptosis20,23,46,47. Additionally, complete blockade of IL-1β signaling may lead to unwanted effects. Although having been proven effective in many auto-immune diseases12, complete inhibition may prevent effective scar formation, which requires a certain degree of inflammation to remove necrotic tissue9. By selectively blocking the NLRP3inflammasome, other inflammasomes that play an essential role in pathogen mediated IL-1β release remain intact. This may prevent the occurrence of infections, another sideeffect that has been observed in selective IL-1β blockade in patients48. In conclusion, our study reveals that continuous inhibition of the NLRP3-inflammasome mediated signaling decreases the post-MI inflammatory response. After 7 days, daily intravenous infusion of MCC950, a selective small-molecule inhibitor of the NLRP3inflammasome, culminates in a pronounced effect on cardiac function preservation and infarct size reduction in a randomized, blinded porcine MI study. Interference with NLRP3inflammasome mediated signaling therefore has become a promising target to reduce infarct size and preserve cardiac function in MI patients. Acknowledgements The authors gratefully acknowledge Marlijn Jansen, Joyce Visser, Martijn van Nieuwburg, Grace Croft, Evelyn Velema, Renée de Rooij, Petra van der Kraak and Ron Stokwielder for their excellent technical support. All institutional and national guidelines for the care and use of laboratory animals were followed and approved by the appropriate institutional committees. Sources of Funding This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs. Disclosures None.

191

PART THREE CHAPTER 10

REFERENCES 1. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit J a., Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart Disease and Stroke Statistics - 2014 Update: A report from the American Heart Association. 2014. 2. Yeh RW, Sidney S, Chandra M, Sorel M, Selby J V, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med. 2010;362:2155–2165. 3. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, De Lemos J a., Ettinger SM, Fang JC, Fesmire FM, Franklin B a., Granger CB, Krumholz HM, Linderbaum J a., Morrow D a., Newby LK, Ornato JP, Ou N, Radford MJ, Tamis-Holland JE, Tommaso CL, Tracy CM, Woo YJ, Zhao DX. 2013 ACCF/AHA guideline for the management of stelevation myocardial infarction: A report of the American college of cardiology foundation/ american heart association task force on practice guidelines. J Am Coll Cardiol. 2013;61:e78–140. 4. Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SMJM, Hofman A, Deckers JW, Witteman JCM, Stricker BHC. Quantifying the heart failure epidemic: Prevalence, incidence rate, lifetime risk and prognosis of heart failure - The Rotterdam Study. Eur Heart J. 2004;25:1614–1619. 5. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 6. Timmers L, Pasterkamp G, De Hoog VC, Arslan F, Appelman Y, De Kleijn DP V. The innate immune response in reperfused myocardium. Cardiovasc Res. 2012;94:276–283. 7. De Haan JJ, Smeets MB, Pasterkamp G, Arslan F. Danger signals in the initiation of the inflammatory response after myocardial infarction. Mediators Inflamm. 2013;2013:206039. 8. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 9. Frangogiannis NG. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 2014;11:255–65. 10. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988. 11. Bujak M, Frangogiannis NG. The role of IL-1 in the pathogenesis of heart disease. Arch Immunol Ther Exp (Warsz). 2009;57:165–176.

192

12. Sims JE, Smith DE. The IL-1 family: regulators of immunity. Nat Rev Immunol. 2010;10:89–102. 13. Toldo S, Mezzaroma E, O’Brien L, Marchetti C, Seropian IM, Voelkel NF, Van Tassell BW, Dinarello C a., Abbate a. Interleukin-18 mediates interleukin1-induced cardiac dysfunction. AJP Hear Circ Physiol. 2014;306:H1025–H1031. 14. Blankenberg S, Luc G, Ducimetière P, Arveiler D, Ferrières J, Amouyel P, Evans A, Cambien F, Tiret L. Interleukin-18 and the Risk of Coronary Heart Disease in European Men: The Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation. 2003;108:2453–2459. 15. Hartford M, Wiklund O, Hultén LM, Persson A, Karlsson T, Herlitz J, Hulthe J, Caidahl K. Interleukin-18 as a predictor of future events in patients with acute coronary syndromes. Arterioscler Thromb Vasc Biol. 2010;30:2039–2046. 16. Blankenberg S, Tiret L, Bickel C, Peetz D, Cambien F, Meyer J, Rupprecht HJ. Interleukin-18 is a strong predictor of cardiovascular death in stable and unstable angina. Circulation. 2002;106:24–30. 17. Van Tassell BW, Raleigh JMV, Abbate A. Targeting Interleukin-1 in Heart Failure and Inflammatory Heart Disease. Curr Heart Fail Rep. 2014;12:33–41. 18. Salloum FN, Chau V, Varma A, Hoke NN, Toldo S, Biondi-Zoccai GGL, Crea F, Vetrovec GW, Abbate A. Anakinra in experimental acute myocardial infarction-does dosage or duration of treatment matter? Cardiovasc Drugs Ther. 2009;23:129–135. 19. Abbate A, Salloum FN, Vecile E, Das A, Hoke NN, Straino S, Giuseppe GL, Biondi-Zoccai, Houser JE, Qureshi IZ, Ownby ED, Gustini E, Biasucci LM, Severino A, Capogrossi MC, Vetrovec GW, Crea F, Baldi A, Kukreja RC, Dobrina A. Anakinra, a recombinant human interleukin-1 receptor antagonist, inhibits apoptosis in experimental acute myocardial infarction. Circulation. 2008;117:2670–2683. 20. Venkatachalam K, Prabhu SD, Reddy VS, Boylston WH, Valente AJ, Chandrasekar B. Neutralization of interleukin-18 ameliorates ischemia/reperfusioninduced myocardial injury. J Biol Chem. 2009;284:7853–7865. 21. Wang M, Tan J, Wang Y, Meldrum KK, Dinarello C a, Meldrum DR. IL-18 binding protein-expressing mesenchymal stem cells improve myocardial protection after ischemia or infarction. Proc Natl Acad Sci U S A. 2009;106:17499–17504. 22. Latz E, Xiao TS, Stutz A. Activation and regulation of the inflammasomes. Nat Rev Immunol. 2013;13:397–411. 23. Takahashi M. NLRP3 Inflammasome as a Novel Player in Myocardial Infarction. Int Heart J. 2014;55:101–105. 24. Zuurbier CJ, Jong WMC, Eerbeek O, Koeman A, Pulskens WP, Butter LM, Leemans JC, Hollmann MW. Deletion of the innate immune NLRP3 receptor abolishes cardiac ischemic preconditioning and is associated with decreased

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

IL-6/STAT3 signaling. PLoS One. 2012;7:e40643. 25. Sandanger Ø, Ranheim T, Vinge LE, Bliksøen M, Alfsnes K, Finsen A V., Dahl CP, Askevold ET, Florholmen G, Christensen G, Fitzgerald K a., Lien E, Valen G, Espevik T, Aukrust P, Yndestad A. The NLRP3 inflammasome is up-regulated in cardiac fibroblasts and mediates myocardial ischaemiareperfusion injury. Cardiovasc Res. 2013;99:164– 174. 26. Mezzaroma E, Toldo S, Farkas D, Seropian IM, Van Tassell BW, Salloum FN, Kannan HR, Menna a. C, Voelkel NF, Abbate a. The inflammasome promotes adverse cardiac remodeling following acute myocardial infarction in the mouse. Proc Natl Acad Sci. 2011;108:19725–19730. 27. Kawaguchi M, Takahashi M, Hata T, Kashima Y, Usui F, Morimoto H, Izawa A, Takahashi Y, Masumoto J, Koyama J, Hongo M, Noda T, Nakayama J, Sagara J, Taniguchi S, Ikeda U. Inflammasome activation of cardiac fibroblasts is essential for myocardial ischemia/reperfusion injury. Circulation. 2011;123:594–604. 28. Marchetti C, Chojnacki J, Toldo S, Mezzaroma E, Tranchida N, Rose SW, Federici M, Van Tassell BW, Zhang S, Abbate A. A novel pharmacologic inhibitor of the NLRP3 inflammasome limits myocardial injury after ischemia-reperfusion in the mouse. J Cardiovasc Pharmacol. 2014;63:316–22. 29. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 30. Coll RC, Robertson AAB, Chae JJ, Higgins SC, Muñoz-Planillo R, Inserra MC, Vetter I, Dungan LS, Monks BG, Stutz A, Croker DE, Butler MS, Haneklaus M, Sutton CE, Núñez G, Latz E, Kastner DL, Mills KHG, Masters SL, Schroder K, Cooper MA, O’Neill LAJ. A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases. Nat Med. 2015; 31. Grundmann S, Bode C, Moser M. Inflammasome activation in reperfusion injury: Friendly fire on myocardial infarction? Circulation. 2011;123:574– 576. 32. O’brien LC, Mezzaroma E, Van Tassell BW, Marchetti C, Carbone S, Abbate A, Toldo S. Interleukin-18 as a therapeutic target in acute myocardial infarction and heart failure. Mol Med. 2014;20:221–9. 33. Abbate A, Van Tassell BW, Biondi-Zoccai GGL. Blocking interleukin-1 as a novel therapeutic strategy for secondary prevention of cardiovascular events. BioDrugs. 2012;26:217–33. 34. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker H V, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, López CM, Honari

S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West M a, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier R V, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A. 2013;110:3507–12. 35. Skyschally A, Van Caster P, Boengler K, Gres P, Musiolik J, Schilawa D, Schulz R, Heusch G. Ischemic postconditioning in pigs: No causal role for risk activation. Circ Res. 2009;104:15–18. 36. Gross GJ, Baker JE, Moore J, Falck JR, Nithipatikom K. Abdominal surgical incision induces remote preconditioning of trauma (RPCT) via activation of bradykinin receptors (BK2R) and the cytochrome P450 epoxygenase pathway in canine hearts. Cardiovasc Drugs Ther. 2011;25:517–522. 37. Keith Jones W, Fan GC, Liao S, Zhang JM, Wang Y, Weintraub NL, Kranias EG, Schultz J El, Lorenz J, Ren X. Peripheral nociception associated with surgical incision elicits remote nonischemic cardioprotection via neurogenic activation of protein kinase C signaling. Circulation. 2009;120:S1–9. 38. Hausenloy DJ, Yellon DM. Myocardial ischemiareperfusion injury: a neglected therapeutic target. J Clin Invest. 2013;123:92–100. 39. Heusch G. Cardioprotection: Chances and challenges of its translation to the clinic. Lancet. 2013;381:166–175. 40. Sluijter JPG, Condorelli G, Davidson SM, Engel FB, Ferdinandy P, Hausenloy DJ, Lecour S, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Van Laake LW. Novel therapeutic strategies for cardioprotection. Pharmacol Ther. 2014;144:60–70. 41. Jansen of Lorkeers SJ, Eding JEC, Vesterinen HM, van der Spoel TIG, Sena ES, Duckers HJ, Doevendans PA, Macleod MR, Chamuleau SAJ. Similar Effect of Autologous and Allogeneic Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-Analysis of Large Animal Studies. Circ Res. 2014;116:80–86. 42. Jong WMC, Leemans JC, Weber NC, Juffermans NP, Schultz MJ, Hollmann MW, Zuurbier CJ. Nlrp3 plays no role in acute cardiac infarction due to low cardiac expression. Int J Cardiol. 2014;177:41–3. 43. Mcmurray JJ V, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, Falk V, Filippatos G, Fonseca C, Gomez-Sanchez MA, Jaarsma T, Kober L, Lip GYH, Maggioni A Pietro, Parkhomenko A, Pieske BM, Popescu B a., Ronnevik PK, Rutten FH, Schwitter J, Seferovic P, Stepinska J, Trindade PT, Voors A a., Zannad F, Zeiher A, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Kirchhof P, Knuuti J, Kolh P, Mcdonagh T, Moulin C, Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Bonet LA,

193

PART THREE CHAPTER 10

Avraamides P, Ben Lamin H a., Brignole M, Coca A, Cowburn P, Dargie H, Elliott P, Flachskampf FA, Guida GF, Hardman S, Iung B, Merkely B, Mueller C, Nanas JN, Nielsen OW, Orn S, Parissis JT, Ponikowski P. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012. Eur J Heart Fail. 2012;14:803–869. 44. Holmes JW, Yamashita H, Waldman LK, Covell JW. Scar remodeling and transmural deformation after infarction in the pig. Circulation. 1994;90:411–420. 45. Brydges SD, Mueller JL, McGeough MD, Pena CA, Misaghi A, Gandhi C, Putnam CD, Boyle DL, Firestein GS, Horner AA, Soroosh P, Watford WT, O’Shea JJ, Kastner DL, Hoffman HM. Inflammasome-mediated disease animal models reveal roles for innate but not adaptive immunity. Immunity. 2009;30:875–87. 46. Dinarello C a., Novick D, Kim S, Kaplanski G. Interleukin-18 and IL-18 binding protein. Front Immunol. 2013;4:289. 47. Brydges SD, Broderick L, McGeough MD, Pena CA, Mueller JL, Hoffman HM. Divergence of IL-1, IL-18, and cell death in NLRP3 inflammasomopathies. J Clin Invest. 2013;123:4695–705. 48. Dinarello CA, van der Meer JWM. Treating inflammation by blocking interleukin-1 in humans. Semin Immunol. 2013;25:469–84.

194

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

SUPPLEMENTAL METHODS Surgical protocol of infarct induction Pre-treatment and anesthesia protocols have been described in detail elsewhere1. In short, all animals were pre-treated with acetylsalicylic acid for 1 day (320 mg loading dose, 80 mg/day maintenance), clopidogrel for 3 days (75 mg/day) and amiodaron for 10 days (1200 mg loading dose, 800mg/day maintenance). All medication was continued until the end of the 7-day follow-up period. Pre-operatively, animals also received a fentanyl patch (25Âľg/h). Animals were anesthetized with an intramuscular injection of 0.4 mg/kg midazolam, 10 mg/kg ketamine and 0.014 mg/kg atropine. Venous access was obtained by insertion of an 18G cannula in the ear vein for intravenous administration of 5 mg/kg sodiumthiopental. Anesthesia was maintained with intravenous infusion of 0.5 mg/kg/h midazolam, 2.5 Âľg/kg/h sufentanyl and 0.1 mg/kg/h pancuronium. Arterial access was obtained by introduction of an 8F sheath into the carotid artery after surgical exposure. A coronary angiogram of the left coronary tree was acquired using a 7F JL4 guiding catheter (Boston scientific, Natick, MA, USA). An adequately sized balloon was placed distal to the first diagonal branch and inflated for 75 minutes. Animals were observed for 3 hours postreperfusion and a permanent catheter was placed in the jugular vein to allow daily intravenous medication and venous blood sampling. The surgical wound was closed and animals were weaned from anesthesia. Animals were defibrillated in case of ventricular fibrillation (VF). Heart rate and arterial blood pressure were measured continuously and documented every 30 minutes. Echocardiography, PV measurements and dobutamine stress-echocardiography 3D TEE was performed at baseline, 2 hours post-MI and after 7 days of follow-up. An X72t transducer on an iE33 ultrasound device (Philips, Eindhoven, The Netherlands) was used to perform a transesophageal echocardiogram. The depth and sector size were adjusted to fit the complete left ventricle. All data sets were acquired in real time using 7 consecutive cardiac cycles (full volume analysis). Images were analyzed offline using QLab 10.0 (3DQ advanced) analysis software. Ventricle tracing was performed by semi-automatic border detection as described elsewhere2. Pressure-volume measurements were performed after 3D TEE on day 7 as recently described 1,3. Arterial and venous access was obtained under general anesthesia according to the protocol described above. A 7F tetra-polar admittance catheter (7.0 VSL Pigtail/no lumen, Transonic SciSense, London, Canada) was inserted into the left ventricle (LV) through an 8F sheath in the carotid artery under fluoroscopic guidance. The catheter was connected to the ADVantage systemTM (Transonic SciSense, London, Canada) linked to a multi-channel acquisition system (Iworx 404), required for real-time data acquisition. An 8F Fogarty catheter was inserted through a 9F introducer sheath in the jugular vein into the inferior vena cava. The Fogarty catheter was inflated to induce preload reduction and 10-12 consecutive heartbeats were recorded during apnea. Data were analyzed offline using Iworx analysis software (Labscribe V2.0).

195

PART THREE CHAPTER 10

Following PV measurements, grey scale short-axis transthoracic echocardiographic measurements were obtained in parasternal position using a broadband S5-1 transducer (Philips, Eindhoven, The Netherlands). Mid ventricular and apical images were obtained by acquiring three successive cardiac cycles. Regional left ventricular function was analyzed in Xcelera Ultrasound version 4.1 (Philips, Eindhoven, The Netherlands). Systolic wall thickening (SWT) was calculated by measuring end diastolic (ED) and end systolic (ES) wall thickness (WT) (SWT=(WTES-WTED)/WTES*100) for both the septal and lateral wall. Fractional shortening (FS) was assessed by comparing the ES and ED LV internal diameter (LVid) (FS=(LVidED-LVidES)/LVidED*100). Fractional area shortening (FAS) was calculated by determining the change in LV ES and ED diastolic internal area (LVia) (FAS=(LViaED-LViaES)/ LViaED*100). After images were obtained, dobutamine was infused through the sheath in the jugular vein at a rate of 2.5ug/kg/min. Once a plateau phase was reached, based on a constant arterial blood pressure, short axis images were again obtained to assess regional contractility. Area at risk & infarct size Before exsanguination, an additional 8F introducer sheath was inserted in the ipsilateral carotid artery. After medial sternotomy, an 8F JL4 guiding catheter (Boston Scientific, Natick, MA, USA) was inserted through the first sheath and a coronary angiogram was performed to ensure normal angiographic flow. An adequately sized balloon was placed distal to the first diagonal branch at the same site as the initial occlusion. A 7F JL4 guiding catheter (Boston Scientific, Natick, MA, USA) was placed in the right coronary artery. A 50 mL Luer lock syringe filled with 2% Evans blue dissolved in 30 mL 0.9%NaCl was attached to the 8F guiding catheter and another syringe containing 20 mL Evans blue solution to the 7F guiding catheter. Evans blue was then simultaneously infused at a rate of 10 mL/s in the left and right coronary artery. Immediately following Evans blue infusion, animals were sacrificed by exsanguination under anesthesia by opening the inferior caval vein. The heart was excised and the left ventricle (LV) was cut into 5 equal slices from apex to base. Slices were photographed and then incubated in 1% tetrazolium trichloride (TTC) (SigmaAldrich Chemicals, Zwijndrecht, the Netherlands) in 37째C 0.9%NaCl for 10 minutes to discriminate between infarct tissue and viable myocardium. After incubation, slices were again photographed. The remote area, area at risk (AAR) and infarct area were quantified using ImageJ software (NIH, Bethesda, MD, USA). Following quantification, infarcted tissue, tissue from the border zone and tissue from the remote area were collected and either conserved in 4% formaldehyde for histological quantification of neutrophils, or snap frozen in liquid nitrogen for cytokine measurements.

196

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

REFERENCES 1. Van Hout GPJ, Jansen Of Lorkeer SJ, Gho JMIH, Doevendans P a, van Solinge WW, Pasterkamp G, Chamuleau S a J, Hoefer IE. Admittance-based pressure-volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014;2:e00287. 2. Soliman OIIII, Krenning BJJ, Geleijnse MLL, Nemes A, van Geuns R-JJ, Baks T, Anwar AMM, Galema TWW, Vletter WBB, ten Cate FJ, Cate FJT. A comparison between QLAB and tomtec full volume reconstruction for real time three-dimensional echocardiographic quantification of left ventricular volumes. Echocardiography. 2007;24:967–974. 3. Van Hout GPJ, de Jong R, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittance-based pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75.

197

198

69±10

62±9

3 mg/kg

6 mg/kg

84±9

85±10

84±15

67±9

66±9

66±8 87±21

89±9

87±14

MAP (mmHg)

30 min

HR (bpm)

68±17

68±11

64±12 89±27

94±18

91±18

MAP (mmHg)

60 min HR (bpm)

71±17

59±12

62±7 99±25

102±17

103±21

MAP (mmHg)

90 min HR (bpm)

72±12

67±21

65±9 94±21

87±13

95±18

MAP (mmHg)

120 min HR (bpm)

HR = heart rate, bpm = beats per minute. MAP = mean arterial blood pressure, mmHg = millimeters mercury.

68±13

Control

MAP (mmHg)

0 min

HR (bpm)

Treatment group

Supplemental table 1. Hemodynamic parameters during ischemia and reperfusion among different treatment arms

80±12

82±25

74±9 90±18

88±16

97±16

MAP (mmHg)

150 min HR (bpm)

75±11

75±15

83±15

85±18

90±18

96±18

MAP (mmHg)

180 min HR (bpm)

PART THREE CHAPTER 10

INFLAMMASOME INHIBITION IN A PORCINE MI MODEL

Supplementary table 2. In vivo levels of MCC950 during the first 24 hours post-MI 15m reperfusion MCC050 (µM)

4 hours reperfusion MCC050 (µM)

24 hours reperfusion MCC050 (µM)

6 mg/kg

101.3±14.6

18.3±6.8

0.24±0.13

3 mg/kg

54.7±6*

9.5±4.8*

0.21±0.29

* Significantly lower than MCC950 levels in 6 mg/kg group (p<0.05).

199

chapter 11

Intracoronary infusion of encapsulated GLP-1 eluting mesenchymal stem cells preserves left ventricular function in a porcine model of acute myocardial infarction

Circulation Cardiovascular Interventions 2014 October;7(5):673-83

R. de Jong1, G.P.J. van Hout2*, J.H. Houtgraaf1*, K. Kazemi1, C. Wallrapp3, A. Lewis4, G. Pasterkamp2; I.E. Hoefer2; H J. Duckers1 * Contributed equally to this work 1

Department Cardiology, Thorax Center, Erasmus University Medical Center, Rotterdam, The Netherlands

Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands

BTG International Germany GmbH, Alzenau, Germany

Biocompatibles UK Ltd, a BTG International group company, Farnham, UK

PART THREE CHAPTER 11

ABSTRACT Background Engraftment and survival of stem cells in the infarcted myocardium remain problematic in cell-based therapy for cardiovascular disease. To overcome these issues, encapsulated mesenchymal stem cells (eMSC) were developed that were transfected to produce glucagon-like peptide-1, an incretin hormone with known cardioprotective effects, alongside MSC endogenous paracrine factors. This study was designed to investigate the efficacy of different doses of intracoronary infusion of eMSC in a porcine model of acute myocardial infarction (AMI). Methods and Results One-hundred pigs were subjected to a moderate AMI (posterolateral AMI; n=50) or a severe AMI (anterior AMI; n=50), whereupon surviving animals (n=36 moderate, n=33 severe) were randomized to receive either intracoronary infusion of 3 incremental doses of eMSC or Ringers’ Lactate control. Cardiac function was assessed using invasive hemodynamics, echocardiography and histological analysis. A trend was observed in the moderate AMI model, whereas in the severe AMI model, left ventricular ejection fraction improved by +9.3% (p=0.004) in the best responding eMSC group, due to a preservation of left ventricular end-systolic volume. Arteriolar density increased 3-fold in the infarct area (8.4 ± 0.9/mm2 in controls, versus 22.2±2.6/mm2 in eMSC group; p<0.001). Although not statistically significant, capillary density was 30% higher in the border zone (908.1±99.7/mm2 in control versus 1209.0±64.6/mm2 in eMSC group; p=ns) Conclusion Encapsulated MSC enable sustained local delivery of cardio-protective proteins to the heart, thereby enhancing angiogenesis and preserving contractile function in an animal AMI model.

202

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

INTRODUCTION Regenerative stem cell therapy to promote cardiac repair has been a target of interest in the last 10 years to prevent heart failure after an acute myocardial infarction (AMI)1,2. Various different stem cells have been investigated for their ability to repair the heart3. Mesenchymal stem cells (MSC) seem to be a potent candidate to date. The mechanism of action of MSC is primarily based on the release of paracrine factors to the myocardium4. However, retention and survival of stem cells in the myocardium after intracoronary infusion (IC) remains an issue in cell-based therapy, since only a limited number of surviving stem cells remain in the myocardium, thereby limiting the potential benefit of the therapy5–7. CellBeads™, that consist of alginate-encapsulated MSC (BTG International Germany GmbH, Alzenau, Germany), were developed to improve survival of cells in the myocardium, thereby elongating the release of cardio-protective proteins into the infarcted myocardium. These encapsulated MSC (eMSC) secrete endogenous paracrine factors that include vascular endothelial growth factor (VEGF), monocyte chemotactic protein-1 (MCP-1), interleukin (IL)-6, IL-8, glial-derived neurotrophic factor (GDNF) and neurotrophin-3 (NT-3) and are genetically modified to produce glucagon-like peptide-1 (GLP-1) fusion protein which comprises two GLP-1 molecules bound by an intervening peptide, extending its half-life in vivo8–12. Amongst its beneficial effects in type 2 diabetes, GLP-1 has anti-apoptotic and cardio-protective properties13–16. Infusion of a GLP-1 analogue, exenatide, after AMI resulted in a reduction of infarct size, thereby improving cardiac function in a preclinical and clinical setting14–20. However, GLP-1 has a short half-life in vivo. Therefore infusion directly at target site or, in case of the GLP-1 eluting eMSC, production directly on-site could render a long-term release of GLP-1 especially due to the prolonged half-life of the GLP-1 fusion protein. Encapsulated MSC have a diameter of 170 micron resulting in entrapment in the coronary system following IC infusion10,21. The alginate shell surrounding the cells allows diffusion of oxygen and nutrients through the pores of the alginate shell into the MSC as well as diffusion of paracrine factors out of the bead. Moreover, the alginate shell protects the MSC against a host immune respons21. Previously, IC infusion of up to 60,000 eMSC in naïve and infarcted porcine myocardium was well tolerated without any sign of microvascular obstruction 10. The alginateencapsulated MSC remain viable for at least 7 days, still secreting the recombinant GLP-1 and MSC paracrine factors10. In this study, we aimed to explore the long term safety, feasibility, and efficacy of incremental doses of intracoronary delivered encapsulated MSC in a moderate and severe porcine AMI model. The primary endpoint in this study was cardiac function as assessed by echocardiography. The secondary endpoints were cardiac contractile function as measured by PV-loop analysis, infarct size, collagen density, capillary density, arteriole density, apoptosis and cardiomyocyte size.

203

PART THREE CHAPTER 11

MATERIALS AND METHODS A total of 100 female Landrace pigs (Van Beek, Lelystad, The Netherlands; 70 ± 5 kg) were randomized in this study. All animal experiments were performed according to the “Guide for care and the use of laboratory animals” and all experiments were previously approved by the institutional animal welfare committee of the University of Utrecht, Utrecht, the Netherlands. The efficacy of IC administered eMSC was investigated in a moderate size infarct (posterolateral AMI (LCx-model; study 1) and in the second phase in a severe anterior AMI model (LAD-model; study 2). Experimental design The design of this study is summarized in Figure 1. Briefly, 100 female pigs underwent an AMI (50 in each group. Pigs that survived infarct induction (n=73; n=36 in the moderate infarct study and n=37 in the severe AMI study), were divided into 4 groups in each study to receive either IC infusion of eMSC or Ringers’ Lactate control solution. The safety up to one week after IC infusion of GLP-1 eMSC was previously described10. The safety up to 8 weeks was investigated in this study and was defined as mortality, occurrence of ventricular arrhythmias and the occurrence of heart failure (fluid retention, need for treatment with heart failure medication). Cardiac function was assessed by echocardiography and pressurevolume-loop (PV-loop) analysis (in the severe anterior model only). Eight weeks after infarct induction animals were terminated and the hearts were excised for infarct size calculations and histological analysis. Encapsulated Mesenchymal stem cells The eMSC used in this study had an outer diameter of 170 µm and contained 75 human MSC stably lentivirally transfected to release a GLP-1 fusion protein, which comprises two GLP-1 molecules bound by an intervening peptide, giving it an extended half-life in vivo (CellBeads™, BTG Germany, Alzenau, Germany)8,10,12. Animals were randomized to a treatment group 60 minutes after the onset of ischemia. Encapsulated MSC were then thawed and dissolved in 100 ml Ringers’ Lactate (RL). A final concentration of 200 eMSC/ mL to 600 eMSC/mL, depending on the dose, or 100 ml of RL were IC infused in 50 minutes at an infusion rate of 2 mL/min via a micro-catheter (Twin Pass catheter, Vascular Solutions, Minneapolis, USA) that was inserted through an 8F JL4 guiding catheter through a cannulated carotid artery. The eMSC were infused at the exact site of the previous occlusion 30 minutes after inducing reperfusion in both MI models. All solutions were color coded and administered in a blinded fashion. Study 1 A moderate posterolateral infarct All pigs were prepared, anesthetized, intubated and ventilated according to a standardized protocol described in the supplemental data section. An 8F sheath (Cordis, Miami, USA) was introduced into the carotid artery and an angiogram of the left coronary tree was

204

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

acquired using an 8F JL4 guiding catheter (Boston scientific, Natick, USA). A posterolateral myocardial infarction was induced by inflation of an angioplasty balloon in the proximal LCx for 90 minutes (Trek, 3.5-4.0x12, Abbott, Illinois, USA). Animals were randomized into 4 groups: 20,000 eMSC (n= 8), 40,000 eMSC (n=9), 60,000 eMSC (n=10) or RL control solution (n=9) after 60 minutes of ischemia. The Thrombolysis in myocardial infarction (TIMI) flow in epicardial coronary arteries was registered pre-, during and post-infusion of eMSC to rule out potential microvascular obstruction (MVO). Both antegrade flow of contrast as well as outwash of contrast were observed. To further quantify MVO, coronary flow reserve (CFR) was measured before and after eMSC infusion and at 8 week follow-up (FU) (ComboWire, Vulcano, Zaventem, Belgium, see supplement). All animals received a Revealâ&#x201E;˘ event recorder for the detection of arrhythmic events (Medtronic, Tilburg, The Netherlands). Cardiac function was assessed using 2D-echocardiography at baseline, after infarct induction and at 8 week FU. Left ventricular ejection fraction (LVEF) and LV volumes derived from 2D-echocardiography were calculated by the modified Simpson rule (see supplement)22. All analyses were performed by an investigator blinded for the therapy allocation. At 8 weeks (Âą 3 days) post MI, animals were anesthetized, intubated and ventilated according to institutional protocol (see supplement). The RevealTM event recorder was interrogated. After assessment of cardiac function as described above, animals were terminated and the heart was excised. Study 2 Severe anterior AMI model A severe anterior AMI was induced by a sternotomy and ligation of the mid LAD distal to the first diagonal for 90 minutes using a prolene ligature. An open chest procedure was applied in this phase of the study to reduce peri-operative mortality. Previous studies that applied a closed chest LAD occlusion experienced a mortality rate of 40% due to VF during infarct induction, the animal welfare committee did therefore not allow a closed chest procedure and we had to change the protocol to an open chest procedure in which defibrillation directly on the heart was possible, thereby enhancing peri-procedural survival. Due to the open chest procedure, epicardial 3D-echocardiography was performed after reperfusion and animals with an LVEF higher than 45% were excluded from the study (n=4). The remaining animals were randomized into 4 groups: 10,000 eMSC (n=6); 20,000 eMSC (n=9); 40,000 eMSC (n=8) or RL buffer as control solution (n=10). Based on the results of study 1, we decided to infuse lower doses of eMSC in this study. A Twin pass catheter was then positioned in the target vessel and placebo or cell solutions were administered at a fixed infusion rate of 2 mL/min. TIMI flow was assessed before during and after infusion of eMSC or placebo solution. Three-D-echocardiography was performed in this study at baseline, after infarct induction and at 8 week FU. Due to the anatomical position of the porcine ribs, it is not possible to obtain clear 3D-echocardiography images in a close chest model nor is it possible to obtain 2D-echocardiographic images directly after open chest surgery. We therefore decided to perform 3D-echocardiography in this study at baseline, post AMI and at 8 week FU, alongside 2D-echocardiography at baseline and 8 week FU

205

PART THREE CHAPTER 11

(see supplement). A RevealTM recorder was implanted for the detection of arrhythmic events. At 8 weeks FU, functional measurements were performed, whereupon the heart was excised and processed as described above. Adding to the 2D and 3D-echocardigraphy data , PVloop analysis was performed to obtain data regarding cardiac contractility at week FU in this study. The exact protocol has been described elsewhere (Supplemental methods)23.

Figure 1. Study design Flowchart of both studies. B: image of intracoronary infusion of encapsulated mesenchymal stem cells (eMSC) via a micro-catheter into the target vessel. Following infusion, eMSC will be retained in the vascular bed, behaving like micro-factories that release paracrine factors for cardiac repair. The image of the formation of a blood clot reflects the human situation of an AMI, in the case of this study, myocardial infarction was induced by obstructing the coronary artery with a balloon or ligature. IC indicates intracoronary; LAD: left anterior descending artery; LCX; Left circumflex artery; PV-loop: pressure-volume loop; qPCR: quantitative polymerase chain reaction. This figure is used with permission of BMM, Rogier Trompert Medical Art.

206

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Tissue collection and infarct size analysis At 8 weeks FU, the hearts were excised and processed for determination of infarct size and (see Supplemental data), as previously described14,24. Assessment of vascular density, collagen density and cardiomyocyte apoptosis Paraffin embedded biopsies were sectioned into 5 µm slices (see supplement). Subsequently, the slides were stained for determination of vascular density, collagen density and cardiomyocyte apoptosis using appropriate antibodies (see supplement) Arterioles and capillaries were expressed as number per mm2. cardiomyocyte apoptosis was expressed as percentage apoptotic cardiomyocytes per view. Quantitative PCR analysis of GLP-1 and human housekeeping genes Quantitative PCR analysis was performed to quantify expression of human GLP-1 and BNP (online supplement for detailed description). Statistical analysis Continuous data are presented as mean ± standard error of the mean (SEM). The delta represents the difference between post AMI and 8 week FU and was first calculated for each animal individually whereupon the average per group was calculated. Comparison of means between groups was performed using a one-way-ANOVA, followed by Dunnett’s test to detect differences between treatment groups and control. Changes over LVEF, LVEDV, LVESV and CFR over time were assessed using a linear mixed effects repeated measures model. P-values < 0.05 were considered significant. All analyses were performed using IBM SPPS Statistics 20 (Chicago, USA).

RESULTS Study 1 Moderate Posterolateral AMI model Mortality and arrhythmias One animal in the control group died of VF one day after the infarct procedure. Two animals in the 60.000 group died of VF, 1 and 5 days following the infarct procedures respectively. None of the animals needed to be treated for heart failure during follow-up and ventricular arrhythmias were not detected by the RevealTM event recorder during the 8 week follow-up period in both groups. Coronary flow Infusion of the 170 µm eMSC did not visually impede antegrade coronary flow up to infusion of 60.000 eMSC since all animals exhibited TIMI 3 flow directly after infusion. Outwash of contrast was slower in 2/8 pigs in the 60.000 group and 1/9 animals in the 40.000 group but this was still within the definition of TIMI 325. CFR did not change

207

PART THREE CHAPTER 11

following eMSC infusion confirming that MVO did not occur to a significant level directly following infusion (Figure 2). Based on these results, it was decided to omit CFR measurements in study 2. At 8 weeks follow-up, flow remained within TIMI 3 range in all groups. CFR did not change significantly over time within the groups (Figure 2).

Figure 2. Coronary flow Reserve Coronary flow reserve after AMI before and after placebo or encapsulated MSC (eMSC) infusion and at 8 week FU.

Figure 3. Cardiac function as measured by 2D-echocardiography A-C. Left ventricular ejection fraction (LVEF) and volumes in posterolateral AMI model over time. D-F LVEF and LV volumes over time in the anterior AMI model. eMSC: encapsulated mesenchymal stem cells; LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume; PMI: post myocardial infarct; FU: follow-up.

208

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Left ventricular ejection fraction and left ventricular volumes After infarct induction, no differences were observed on LVEF between groups, indicating comparable infarct size. At 8 weeks follow up, LVEF was a higher EF was observed in the 20,000 eMSC group opposed to control although this did not reach statistical significance. (+6%; p=NS; Figure 3; Supplemental Table 1). Infarct size Infarct size was only 9.6 ± 1.3% in the control group opposed to 7.6 ± 1.2% in the 20,000 group (p=NS). The 40,000 and 60,000 group however, showed similar infarcts sizes as the ringer lactate control (9.1 ± 1.2%; 9.3 ± 1.8% respectively; Figure 4A) Vascular density, collagen density and cardiomyocyte apoptosis Capillary density in border and remote areas was not enhanced in all groups (Supplemental Table II, Supplemental Figure I). However arteriolar density increased by almost 200% in the infarct area in the 20,000 group, this change was not statistically significant (Supplemental Figure I C and D, Supplemental Table II), whereas eMSC did not enhance arteriole formation in the border and remote segments. Moreover collagen deposition and cardiomyocyte apoptosis were not affected (Supplemental Figure IE-F). Study 2 Severe Anterior AMI model Mortality and arrhythmias One animal in the control group died of VF one day after the infarct procedure. None of the animals needed to be treated for heart failure during follow-up and ventricular arrhythmias were not detected by the Reveal™ event recorder (Figure 1). Outwash of contrast was decreased (TIMI 2) in 43% of the animals in the 40.000 group and none of the animals in the other groups directly post infusion and at 8 weeks FU.

Figure 4. Infarct size A: infarct size in posterolateral infarct model in different groups. B: Although not statistically significant, infarct size in study 2 was approximately 20% in control and was reduced by 25% in the optimal dose group. eMSC: encapsulated mesenchymal stem cell.

209

PART THREE CHAPTER 11

Improvement in left ventricular ejection fraction In this phase of the study both 2D-echocardiography and 3D-echocardiography were applied. On 3D-echocardiography, LVEF at baseline and following infarct induction was comparable between the groups (Supplemental Table I, Figure 5). At 8 weeks follow up, LVEF increased by +0.2 ± 1.0% in the control group to 38.9 ± 1.5%. 20,000 eMSC improved LVEF by +9.3% to 44.7 ± 1.2 % (Anova: p=0.004; Control vs 20.000 p=0.011; Supplemental Table I). The increase in LVEF was mainly due to a preservation of LVESV, which decreased by -2.9 ± 4.8 ml (20,000 eMSC) compared to an increase of +18.1 ± 4.5 ml in control animals (Figure 5). LVEF was comparable between 2D-echocardiography and 3D-echocardiography. LV volumes were 15 ml higher when measured by 2D-echocardiography. The same significant effect in LVEF was seen on 2D-echocardiography. Cardiac contractility End-systolic pressure volume relationship (ESPVR), which reflects systolic contractile function, was dramatically increased by 230% in all treatment groups as opposed to control animals (6.5 ± 1.3 mmHg/ml in 20,000 eMSC group vs. 1.9 ± 0.5 mmHg/ml in control; anova: p=0.03, p=0.007 vs control). No differences were observed in other PV-loop derived parameters (Figure 6, Supplemental Table III). Infarct size

Figure 5. 3D-echocardiography anterior infarct model A: Left ventricular ejection fraction (LVEF) measured by 3D-echo over time. B: Left ventricular end-diastolic volume (LVEDV) over time. C: Left ventricular end-systolic volume (LVESV) over time. D: delta LVEF represents the absolute difference between LVEF at 8 weeks and directly following AMI. There is a remarkable 9.3% increase in LVEF in the 20.000 group. eMSC indicates encapsulated mesenchymal stem cells. *p=0.011 vs control

210

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Figure 6. Pressure volume loop analysis of ESPVR A: End-systolic pressure volume relation (ESPVR) is enhanced in all encapsulated MSC (eMSC) groups. B: Actual vena cava occlusion of a control animal (blue) and 20.000 eMSC animal. *P=0.007 versus control, + P=0.03 versus control

Although not significant, infarct size decreased by -20% in the 20,000 eMSC group (Figure 4B, Supplemental Table III). Vascular density and collagen density Capillary formation was not enhanced in border areas, nor increased the capillary density in remote areas. However, arteriolar density improved in the infarct area in all doses by +200-300% (22.2 ± 2.6/mm2 versus 8.4 ± 0.9/mm2; Anova: p<0.001; Figure 7; Supplemental Table II). Collagen density was not reduced in all dose groups. Cardiomyocyte apoptosis and hypertrophy Cardiomyocyte apoptosis was not influenced by eMSC treatment, which might be related to the late time point. However, the surface area of the cardiomyocytes in the infarct border zone of animals treated with 20.000 eMSC was almost 50% lower than in control and the other dose groups (417.1 ± 110.2 µm2 in 20.000 group versus 823.1 ± 74.0 µm2 in control; p=0.009; Figure 8). The same was observed in the remote myocardial segments. Location and survival of encapsulated MSC Encapsulated MSC were found on regular bases in the infarct-border zone. No shedding was macroscopically and microscopically observed to remote myocardial segments or remote organs. Encapsulated MSC were still intact (Supplemental Figure II). Unfortunately, neither human housekeeping gene GAPDH nor human GLP-1 expression could be detected

211

PART THREE CHAPTER 11

by qPCR, indicating that the MSC inside the beads do not elute paracrine factors at 8 week FU (Supplemental Figure II) and are most likely dead. Surrounding the eMSC, an inflammatory response occurred (Supplemental Figure III). This response was comparable between the dose groups. Cells included in this response were macrophages and granulocytes.

DISCUSSION In the current study, we investigated the efficacy and the optimal dose of encapsulated immune-protected MSC in a porcine model of AMI. This is the first time that the efficacy of microencapsulated MSC was investigated for treatment of AMI. In the moderate infarct study, only a trend towards an improvement of cardiac function was observed. In the severe infarct study, LVEF was remarkably improved following eMSC infusion. Moreover, contractility was enhanced and neovascularization occurred.

Figure 7. Angiogenesis A-D: Trichrome stain for detection of collagen. A: In the infarct area, no significant difference towards less collagen deposition was observed in all animals treated with encapsulated mesenchymal stem cells (eMSC; P=0.09). B: Example of Trichrome stain infarct area. Pictures are taken at a 10 times magnification. Blue represents collagen, pink viable myocardium. Collagen deposition is more dens in the animals in the control group opposed to animals in the 20,000 eMSC group. C: Collagen density border area was not enhanced in treated animals s. D: Example of Trichrome stain in border area. E-F: Arterioles were stained by smooth muscle actin stain. E: In the infarct area, arteriole density was enhanced in treated animals (*p=0.04; +p=0.0001, =p=0.002 vs control). F: Representative images of smooth muscle actin stain in infarct area in a control animal and in a 20,000 eMSC animal at 10 times magnification. G-H: Capillary density G: A trend in a decreased capillary density was observed in 20.000 eMSC group. H: Representative images of Isolectin stain in the border area of a control animal and a 20,000 eMSC animal respectively at 20 times magnification.

212

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Figure 8. Cardiomyocyte apoptosis, cardiomyocyte size and cardiomyocyte nuclear density A-C: Cardiomyocyte apoptosis via TUNEL stain in the border area and remote areas. C: Example of Fluorescent TUNEL stain. Blue cells are alive, green cells are apoptotic. D-F: Cardiomyocyte size in the border area was significantly lower in the 20.000 group (D). In the remote myocardial segments, the surface area was lower in all animals treated with eMSC (E). This finding corresponds with an increased nuclear density in the highest dose groups in remote areas (F). *p=0.009, +P=0.01, =p=0.006, -p= 0.0007, **p=0.03, ++p= 0.003 versus control . eMSC indicates encapsulated mesenchymal stem cells

Safety In line with our previous results, eMSC infusion did not show side-effects in a preclinical setting10. Previous studies suggested a reduction in TIMI flow and CFR following intracoronary infusion of mesenchymal stem cells6,26,27. In this study, CFR did not significantly decrease directly after infusion, nor at 8 week FU, indicating that MVO is not significant. Studies that report MVO after infusion of MSC, usually apply high doses, which could eventually result in obstruction. Several other preclinical and clinical studies have shown that under carefully controlled and monitored conditions, e.g. slow infusion rate and low cell number, infusion of MSC or MSC like stem cells or cardiospheres is well tolerated24,28,29. In the case of eMSC the same assumption could be made. None of the animals developed heart failure during the follow-up period, nor increased the incidence of arrhythmic events, indicating no side-effects of eMSC therapy were observed during 8 week FU period. Efficacy When given IC, 20,000 alginate encapsulated MSC significantly improve LVEF after a severe AMI. In the moderate infarction model we detected a trend, which was further explored in a more severe AMI model. As expected, the treatment effect turned out to be more pronounced in animals with more severe AMI, presumably due to a larger pharmacological window in the anterior model. This corresponds with a large meta-analysis by Jeevanantham et al who concluded that patients with a LVEF blow 43% benefit more from cell-based

213

PART THREE CHAPTER 11

cardiac repair.1 However, another meta-analysis failed to show a more prominent treatment effect of BMMNC therapy in patients with lower LVEF3. If a treatment effect is already noticeable in a small infarct, the benefit will be expanded in a large infarct. In absolute terms, IC infusion of eMSC improved LVEF by +9.3% in severe anterior AMI model. The most effective dose in our study was 20,000 eMSC, which equals 1.6 million MSC, and is several orders of magnitude less than used in most preclinical trials using cultured MSC. However, eMSC are better retained into the myocardium than unprotected stem cells culminating in prolonged protein release. This complicates direct comparison. Infusing less than 20,000 eMSC was not effective, suggesting that this dose is the lower limit. On the other hand, 40,000 eMSC were not superior to lower doses, which does not necessarily preclude a dose dependent effect but could also reflect unfavorable effects of incremental MVO undetectable by our CFR measurements or the adverse effects of a possible xenogeneic immune response which could be more severe in the high dose group. Clinical applicability Before possible translation of this product to a clinical treatment strategy, several issues should be addressed in future studies. First, lentiviral transduction for cell based gene therapy remains subject to debate since insertional mutagenesis and thus tumorgenicity might be of increased risk30. This could influence the clinical applicability of the transduced eMSC. However, the MSC are contained by the alginate and are unable to migrate to the hostâ&#x20AC;&#x2122;s tissue. Furthermore, the alginate exceeds the lifespan of the cells, with no exception observed in our study, making the risk for possible tumorgenicity rather low, in our opinion. Second, as described in previous studies, an increased inflammatory response around the eMSC has been detected in the present study, which can most likely be attributed to a xenogenic reaction. Future studies should determine if this response is due to xenogenecity or a direct result of e.g. cell death inside the alginate microsphere. If the latter is the case then this would also influence the clinical applicability of this product since excessive inflammation is detrimental to infarct healing post-MI31. Working mechanism of encapsulated GLP-1 eluting MSC Several preclinical studies showed that MSC therapy after AMI enhanced the formation of arterioles and capillaries by a release of paracrine factors24,32. MSC inside the alginate beads also have a pro-angiogenic effect33. Moreover, eMSC also improved angiogenesis in a hind-limb ischemia model and in porcine interposition grafts11,33,34. This pro-angiogenic effect might explain the observed preservation of cardiac function33,34. Furthermore, the eMSC group shows a trend towards reduced fibrosis and improved myocardial viability in the infarct zone was observed, which could result in a reduction in total infarct size4,24,35. However, this effect was not statistically significant in our study, most likely based on limited numbers of animals. The trend towards less fibrosis could be explained by the secretion of immune-modulatory cytokines by MSC. IL-6 has been shown to increase the lifespan of neutrophils in the hostile post-AMI environment and improves healing of the infarct wound36. In addition, MSC trigger the transition of classical M1 to

214

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

anti-inflammatory M2 macrophages, thereby enhancing infarct healing via increased angiogenesis36. Alongside pro-angiogenic factors and immunomodulatory cytokines, MSCs secrete antiapoptotic factors that improve cell survival24,35. Additionally, eMSC in this study were transfected to secrete recombinant GLP-1. Exenatide, a GLP-1 analogue, has shown to reduce infarct size and improve cardiac function in a pig AMI model14. Moreover, in 2 recent clinical trials in which exenatide was injected in AMI patients (TIMI 0-1 flow) before PCI, infarct size was reduced by 50%, indicating that Exenatide prevents reperfusion mediated cell death15,16. As eMSC are infused within 30 minutes after reperfusion, the effect on infarct size could be related to limitation of reperfusion damage. Next to its anti-apoptotic effects, GLP-1 has been shown to directly improve cardiac contractility by increasing intracellular cyclic-AMP concentrations37. This latter effect could be responsible for the significant ESPVR improvement in all eMSC groups. The observed preservation in contractility could also be explained by strengthening of the heartâ&#x20AC;&#x2122;s matrix by the infusion of alginate thereby preventing LV remodeling as previously was shown38,39. However, in these studies liquid alginate was used that diffuses through the vessel wall38. Although eMSC cannot leave the vessel lumen, they seem to provide structural support to the myocardium. The heart contains a population of resident cardiac stem cells40. It is hypothesized that MSC can activate these stem cells in order to stimulate infarct repair following an ischemic event. Suzuki et al concluded from a porcine study, in which MSC were IC injected following AMI that MSC stimulate endogenous cardiac stem cells to home to the site of injury and differentiate into cardiomyocytes32. In addition to this, the MSC in their study differentiated into cardiac stem cells. Moreover, cardiomyocytes were stimulated to proliferate in animals that were treated with MSC. All these effects combined resulted in an increased cardiomyocyte nuclear density. In our study, cardiomyocyte nuclear density is increased in border and remote areas possibly suggesting that eMSC stimulate myocardial salvage, which in turn results in a reduction of compensatory hypertrophy. Myocardial salvage in this study is based on a reduction of apoptotis by GLP1 and most likely proliferation of adult cardiomyocytes. It was recently shown, that IL-6 secreted by MSC-like stem cells, stimulated cardiomyocyte proliferation41. As the eMSC produce IL-6, this would be the proposed working mechanism of cardiomyocyte proliferation in this study. Differentiation of MSC inside the beads could not have contributed to the increased number of cardiomyocytes, because MSC do not leave their shell. Activation of endogenous stem cells also remains a proposed mechanism of action. As was shown by Suzuki et al and Houtgraaf et al activation of resident cardiac stem cells is not detectable after 6 weeks following transplantation, the c-kit stain was omitted from current protocol24,32. Moreover, cardiomyocyte proliferation could be not detected at 8 week FU in a previous study that used MSC24. Therefore ki-67 stain was not executed. Study limitations Despite our best efforts, this study has some limitations. First, cardiac function was not assessed by golden-standard cardiac MRI. However, echocardiographic measurements

215

PART THREE CHAPTER 11

have shown excellent correlations with MRI. We are therefore confident that our findings would be corroborated by MRI42. In the two sub-studies we used slightly different protocols. In the more severe anterolateral model we used an open chest procedure to minimize high peri-procedural mortality following mid LAD occlusion. The advantage of this open chest procedure enabled us to use epicardial 3D-echocardiography alongside transthoracic 2D-echocardiography in the anterior model. Although the two methods have shown to be in good concordance with one another, values should not be directly compared. Therefore, the 2 methods were both shown in the second phase43,44. Second, in this study we did not observe an increased MVO, as assessed by CFR, after eMSC infusion. However, variability in heart rate and arterial blood pressure might interfere with reliable CFR measurement. Since we did not do CFR measurements under atrial pacing, these parameters might have varied in the present study, thereby possibly influence the sensitivity of the CFR measurements. Third, we did not include all possible controls including eMSC not transduced to produce GLP-1, empty beads or MSC only. This decision was based on our previous pilot data where we found beneficial effects on apoptosis and inflammatory response of eMSC opposed to empty alginate beads without MSC. Moreover, in a proof of concept study by Wright et al., IC infusion of macro-eMSC resulted in preserved LVEF and reduced infarct size when compared to empty alginate beads without cells and encapsulated human MSC that were not transduced to express GLP-111. Since both the animal model and the eMSC preparation differed in this study, no firm conclusion can be drawn on the superiority of GLP-1 eluting eMSC used in the current study compared to non-transduced eMSC. Furthermore, a side-by-side comparison between eMSC and MSC would have been very interesting. However, in this phase of the study, we decided that we first wanted to evaluate the efficacy of this new product before we performed a side-by-side comparison. Moreover, we did not know the optimal eMSC dose. If we would like to compare the efficacy of eMSC end MSC, the amount of MSC should be comparable between the groups. Here, the MSC are derived from humans. This makes sense as we aimed to test the efficacy of the clinical product. Further research is needed to test whether allogeneic encapsulated mesenchymal stem cells are equal or even have a superior effect. As the GLP-1 analogue produced by the eMSC consists of a GLP-1 fusion protein with a short half-life, a direct comparison with native GLP-1 or GLP-1 analogues with prolonged half-life would be difficult to interpret. There is no sign of degradation of the encapsulated MSC at 8 weeks FU. More research is needed to investigate how and when the encapsulated MSC are degraded. Conclusions IC infusion of encapsulated GLP-1 eluting MSC resulted in a preserved LVEF in a porcine AMI model and is well tolerated up to 8 weeks FU. Infusion of eMSC improves cardiac function by preservation of end-systolic volume and cardiac contractility. Encapsulated MSC are able to prevent cardiomyocyte hypertrophy, enhance adaptive neovascularization, thereby limiting left ventricular remodeling. Importantly, encapsulated MSC are an

216

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

interesting platform that enables sustained paracrine delivery of MSC proteins and recombinant proteins to the damaged myocardium. Acknowledgements We would like to thank J. Visser, M. Janssen, C. Verlaan, E. Velema, M. Schurink, G. Croft, E. van de Kamp , J. Huizingh and L. Bosman for their excellent technical assistance. Sources and funding This research forms part of the Project P5.02 CellBeads of the research program of the BioMedical Materials institute, co-funded by the Dutch Ministry of Economic Affairs. Disclosures R. de Jong G.P.J. van Hout J.H. Houtgraaf K. Kazemi C. Wallrapp A. Lewis G. Pasterkamp I. Hoefer H.J. Duckers

None None None None employee BTG International Germany employee Biocompatibles UK Ltd None None None

217

CHAPTER XXX

REFERENCES 1. Jeevanantham V, Butler M, Saad A, Abdel-Latif A, Zuba-Surma EK, Dawn B. Adult bone marrow cell therapy improves survival and induces long-term improvement in cardiac parameters: a systematic review and meta-analysis. Circulation. 2012;126:551–68. 2. Strauer B-E, Steinhoff G. 10 Years of Intracoronary and Intramyocardial Bone Marrow Stem Cell Therapy of the Heart From the Methodological Origin To Clinical Practice. J Am Coll Cardiol. 2011;58:1095–104. 3. De Jong R, Houtgraaf JH, Samiei S, Boersma E, Duckers HJ. Intracoronary Stem Cell Infusion After Acute Myocardial Infarction: A Meta-Analysis and Update on Clinical Trials. Circ Cardiovasc Interv. 2014;7:156–167. 4. Williams AR, Hare JM. Mesenchymal Stem Cells: Biology, Pathophysiology, Translational Findings, and Therapeutic Implications for Cardiac Disease. Circ Res. 2011;109:923–940. 5. Van der Spoel TIG, Lee JC-T, Vrijsen K, Sluijter JPG, Cramer MJM, Doevendans P a, van Belle E, Chamuleau S a J. Non-surgical stem cell delivery strategies and in vivo cell tracking to injured myocardium. Int J Cardiovasc Imaging. 2011;27:367–83. 6. Freyman T, Polin G, Osman H, Crary J, Lu M, Cheng L, Palasis M, Wilensky RL. A quantitative, randomized study evaluating three methods of mesenchymal stem cell delivery following myocardial infarction. Eur Heart J. 2006;27:1114–22. 7. Hou D, Youssef EA-S, Brinton TJ, Zhang P, Rogers P, Price ET, Yeung AC, Johnstone BH, Yock PG, March KL. Radiolabeled cell distribution after intramyocardial, intracoronary, and interstitial retrograde coronary venous delivery: implications for current clinical trials. Circulation. 2005;112:I150–6. 8. Weber C, Pohl S, Poertner R. Production process for stem cell based therapeutic implants: expansion of the production cell line and cultivation of encapsulated cells. Bioreact Syst …. 2010;123:143– 162. 9. Trouche E, Girod Fullana S, Mias C, Ceccaldi C, Tortosa F, Seguelas MH, Calise D, Parini a, Cussac D, Sallerin B. Evaluation of alginate microspheres for mesenchymal stem cell engraftment on solid organ. Cell Transplant. 2010;19:1623–33. 10. Houtgraaf JH, de Jong R, Monkhorst K, Tempel D, van de Kamp E, den Dekker WK, Kazemi K, Hoefer I, Pasterkamp G, Lewis AL, Stratford PW, Wallrapp C, Zijlstra F, Duckers HJ. Feasibility of intracoronary GLP1 eluting CellBeadTM infusion in acute myocardial infarction. Cell Transplant. 2013;22:535–43. 11. Wright EJ, Farrell KA, Malik N, Kassem M, Lewis AL, Wallrapp C, Holt CM. Encapsulated glucagonlike peptide-1-producing mesenchymal stem cells have a beneficial effect on failing pig hearts. Stem Cells Transl Med. 2012;1:759–69.

218

12. Wallrapp C, Thoenes E, Thürmer F, Jork A, Kassem M, Geigle P. Cell-based delivery of glucagon-like peptide-1 using encapsulated mesenchymal stem cells. J Microencapsul. 2013;30:315–24. 13. Nikolaidis L a, Mankad S, Sokos GG, Miske G, Shah A, Elahi D, Shannon RP. Effects of glucagon-like peptide-1 in patients with acute myocardial infarction and left ventricular dysfunction after successful reperfusion. Circulation. 2004;109:962–5. 14. Timmers L, Henriques JPS, de Kleijn DP V, Devries JH, Kemperman H, Steendijk P, Verlaan CWJ, Kerver M, Piek JJ, Doevendans P a, Pasterkamp G, Hoefer IE. Exenatide reduces infarct size and improves cardiac function in a porcine model of ischemia and reperfusion injury. J Am Coll Cardiol. 2009;53:501–10. 15. Lønborg J, Kelbæk H, Vejlstrup N, Bøtker HE, Kim WY, Holmvang L, Jørgensen E, Helqvist S, Saunamäki K, Terkelsen CJ, Schoos MM, Køber L, Clemmensen P, Treiman M, Engstrøm T. Exenatide reduces final infarct size in patients with STsegment-elevation myocardial infarction and shortduration of ischemia. Circ Cardiovasc Interv. 2012;5:288–95. 16. Woo JS, Kim W, Ha SJ, Kim JB, Kim S-J, Kim W-S, Seon HJ, Kim KS. Cardioprotective effects of exenatide in patients with ST-segment-elevation myocardial infarction undergoing primary percutaneous coronary intervention: results of exenatide myocardial protection in revascularization study. Arterioscler Thromb Vasc Biol. 2013;33:2252–60. 17. Bose AK, Mocanu MM, Carr RD, Brand CL, Yellon DM, Glp- G. Glucagon-like Petide 1 can Directly Protect the Heart Against Ischemia / Reperfusion Injury. Diabetes. 2005;54:146–151. 18. Nikolaidis L, Doverspike A. Glucagon-like peptide-1 limits myocardial stunning following brief coronary occlusion and reperfusion in conscious canines. … Pharmacol …. 2005;312:303–308. 19. Nikolaidis L a, Mankad S, Sokos GG, Miske G, Shah A, Elahi D, Shannon RP. Effects of glucagon-like peptide-1 in patients with acute myocardial infarction and left ventricular dysfunction after successful reperfusion. Circulation. 2004;109:962–5. 20. Sokos GG, Nikolaidis L a, Mankad S, Elahi D, Shannon RP. Glucagon-like peptide-1 infusion improves left ventricular ejection fraction and functional status in patients with chronic heart failure. J Card Fail. 2006;12:694–9. 21. Caplan A. Mesenchymal stem cells. J Orthop Res. 1991;30:315–324. 22. Folland ED, Parisi a. F, Moynihan PF, Jones DR, Feldman CL, Tow DE. Assessment of left ventricular ejection fraction and volumes by real- time, twodimensional echocardiography. A comparison of cineangiographic and radionuclide techniques. Circulation. 1979;60:760–766. 23. Van Hout GPJ, de Jong R, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittance-based

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75. 24. Houtgraaf JH, de Jong R, Kazemi K, de Groot D, van der Spoel TIG, Arslan F, Hoefer I, Pasterkamp G, Itescu S, Zijlstra F, Geleijnse ML, Serruys PW, Duckers HJ. Intracoronary infusion of allogeneic mesenchymal precursor cells directly after experimental acute myocardial infarction reduces infarct size, abrogates adverse remodeling, and improves cardiac function. Circ Res. 2013;113:153–66. 25. Kern MJ, Moore JA, Aguirre F V, Bach RG, Caracciolo EA, Wolford T, Khoury AF, Mechem C, Donohue TJ. Determination of Angiographic (TIMI Grade) Blood Flow by Intracoronary Doppler Flow Velocity During Acute Myocardial Infarction. Circ . 1996;94 :1545–1552. 26. Vulliet PR, Greeley M, Halloran SM, MacDonald K a, Kittleson MD. Intra-coronary arterial injection of mesenchymal stromal cells and microinfarction in dogs. Lancet. 2004;363:783–4. 27. Lim SY, Kim YS, Ahn Y, Jeong MH, Hong MH, Joo SY, Nam K Il, Cho JG, Kang PM, Park JC. The effects of mesenchymal stem cells transduced with Akt in a porcine myocardial infarction model. Cardiovasc Res. 2006;70:530–42. 28. Makkar RR, Smith RR, Cheng K, Malliaras K, Thomson LEJ, Berman D, Czer LSC, Marbán L, Mendizabal A, Johnston P V, Russell SD, Schuleri KH, Lardo AC, Gerstenblith G, Marbán E. Intracoronary cardiosphere-derived cells for heart regeneration after myocardial infarction (CADUCEUS): a prospective, randomised phase 1 trial. Lancet. 2012;379:895–904. 29. Houtgraaf JH, den Dekker WK, van Dalen BM, Springeling T, de Jong R, van Geuns RJ, Geleijnse ML, Fernandez-Aviles F, Zijlsta F, Serruys PW, Duckers HJ. First Experience in Humans Using Adipose Tissue-Derived Regenerative Cells in the Treatment of Patients With ST-Segment Elevation Myocardial Infarction. J Am Coll Cardiol. 2012;59:539–540. 30. Mostoslavsky G, Kotton DN, Fabian AJ, Gray JT, Lee J-S, Mulligan RC. Efficiency of transduction of highly purified murine hematopoietic stem cells by lentiviral and oncoretroviral vectors under conditions of minimal in vitro manipulation. Mol Ther. 2005;11:932–40. 31. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 32. Suzuki G, Iyer V, Lee T-C, Canty JM. Autologous mesenchymal stem cells mobilize cKit+ and CD133+ bone marrow progenitor cells and improve regional function in hibernating myocardium. Circ Res. 2011;109:1044–54. 33. Katare R, Riu F, Rowlinson J, Lewis A, Holden R, Meloni M, Reni C, Wallrapp C, Emanueli C, Madeddu P. Perivascular Delivery of Encapsulated Mesenchymal Stem Cells Improves Postischemic Angiogenesis Via Paracrine Activation of VEGF-A.

Arterioscler Thromb Vasc Biol. 2013;33:1872–80. 34. Huang W-C, Newby GB, Lewis AL, Stratford PW, Rogers C a, Newby AC, Murphy GJ. Periadventitial human stem cell treatment reduces vein graft intimal thickening in pig vein-into-artery interposition grafts. J Surg Res. 2013;183:33–9. 35. Zhao Y, Li T, Wei X, Bianchi G, Hu J, Sanchez PG, Xu K, Zhang P, Pittenger MF, Wu ZJ, Griffith BP. Mesenchymal stem cell transplantation improves regional cardiac remodeling following ovine infarction. Stem Cells Transl Med. 2012;1:685–95. 36. Van den Akker F, Deddens JC, Doevendans P a, Sluijter JPG. Cardiac stem cell therapy to modulate inflammation upon myocardial infarction. Biochim Biophys Acta. 2013;1830:2449–58. 37. Grieve DJ, Cassidy RS, Green BD. Emerging cardiovascular actions of the incretin hormone glucagon-like peptide-1 : potential therapeutic benefits beyond glycaemic control ? Br J Pharmacol. 2009;:1340–1351. 38. Leor J, Tuvia S, Guetta V, Manczur F, Castel D, Willenz U, Petneházy O, Landa N, Feinberg MS, Konen E, Goitein O, Tsur-Gang O, Shaul M, Klapper L, Cohen S. Intracoronary injection of in situ forming alginate hydrogel reverses left ventricular remodeling after myocardial infarction in Swine. J Am Coll Cardiol. 2009;54:1014–23. 39. Bai XP, Zheng HX, Fang R, Wang TR, Hou XL, Li Y, Chen XB, Tian WM. Fabrication of engineered heart tissue grafts from alginate/collagen barium composite microbeads. Biomed Mater. 2011;6:045002. 40. Leri A, Kajstura J, Anversa P. Role of cardiac stem cells in cardiac pathophysiology: a paradigm shift in human myocardial biology. Circ Res. 2011;109:941–61. 41. Przybyt E, Krenning G, Brinker MGL, Harmsen MC. Adipose stromal cells primed with hypoxia and inflammation enhance cardiomyocyte proliferation rate in vitro through STAT3 and Erk1/2. J Transl Med. 2013;11:39. 42. Jenkins C, Bricknell K, Hanekom L, Marwick TH. Reproducibility and accuracy of echocardiographic measurements of left ventricular parameters using real-time three-dimensional echocardiography. J Am Coll Cardiol. 2004;44:878–86. 43. Dorosz JL, Lezotte DC, Weitzenkamp DA, Allen LA, Salcedo EE. Performance of 3-Dimensional Echocardiography in Measuring Left Ventricular Volumes and Ejection Fraction. J Am Coll Cardiol. 2012;59:1799–1808. 44. Greupner J, Zimmermann E, Grohmann A, Dübel H-P, Althoff T, Borges AC, Rutsch W, Schlattmann P, Hamm B, Dewey M. Head-to-Head Comparison of Left Ventricular Function Assessment with 64Row Computed Tomography, Biplane Left Cineventriculography, and Both 2- and 3-Dimensional Transthoracic Echocardiography: Comparison With Magnetic Resonance Imaging as the Reference S. J Am Coll Cardiol. 2012;59:1897– 907.

219

PART THREE CHAPTER 11

SUPPLEMENTAL DATA Detailed Material and Methods Medication and Anesthesia All animals were pre-medicated starting ten days prior to infarct induction with dual antiplatelet therapy ((acetylsalicylic Acid: 300 mg loading followed by 80 mg qd (Centrafarm, Etten-Leur, The Netherlands) and clopidogrel 75 mg qd(Sanofi-Aventis, Paris, France)) and anti-arrhythmic therapy (amiodarone, loading dose of 1200 mg followed by 800 mg qd; Sanofi-Aventis, Paris, France). Dual antiplatelet and anti-arrhythmic therapy was continued during follow up. General anesthesia was induced with 0.4 mg/ kg midazolam (Actavis, Zug, Switzerland), 10 mg/kg ketamine (Narketan,Vétoquinol, Lure Cedex, France) and 1 mg of Atropine (Pharmachemie BV., The Netherlands) and maintained with intravenous infusion of midazolam 0.5 mg/kg/h, sufentanil 2.5 µg/kg/h (Janssen-Cilag B.V., Tilburg, The Netherlands) and pancuronium 0.1 mg/kg/h (Inresa, Battenheim, Germany). Upon infarct induction and Cellbead infusion, all animals were therapeutically heparinized with 2 doses of 5000 IE (Leo pharma, Ballerup, Denmark) IV and received intravenous infusion of eptifibatide (bolus of 180 μg/kg and 2 μg/kg/min (GlaxoSmithKline BV, Zeist, The Netherlands). A fentanyl plaster 25 µg/h (Janssen-Cilag B.V., Tilburg, The Netherlands) was applied before and after the first procedure for analgesia. The animals were treated with one doses of Augmentin intravenously (1000/100 mg, Sandoz, Holzkirchen, Denmark) before the infarct procedure. All animals in received meloxacam 2 times 0.5 mg/kg daily (Produlab-Pharma B.V. Raamsdonksveer, The Netherlands) for two days after index procedure for additional analgesia. Coronary flow In the first study, coronary flow was assessed after myocardial infarct,after the complete infusion of eMSC and 8 weeks following infusion by coronary flow reserve measurements (CFR). Every 25cc infusion was stopped and TIMI flow was assessed according to the TIMI criteria1. During CFR measurements, Average peak velocity (APV; cm/sec) was assessed in normal conditions after AMI and during maximum vasodilation following injection of 140 mcg/kg/min adenosine IV. Peak APV was assessed and CFR was calculated. Echocardiography Two-dimensional grey scale images at a frame rate of 60-90 frames/s were obtained in parasternal position using a Philips iE33 (Phillips, Eindhoven, The Netherlands), equipped with a broadband S5-1 transducer. Transthoracic echocardiographs were acquired at baseline, post myocardial infarct and eight weeks follow-up in the posterolateral infarct model and at baseline and 8 week FU in the anterior infarct model. A long axis view and

220

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Primers for qPCR Porcine housekeeping gene B actin forward

AAGAGCTACGAGCTGCCCGAC

B actin reverse

GTGTTGGCGTAGAGGTCCTTC

Human housekeeping gene GAPDH forward

GCTCATTTCCTGGTATGACAAC

GAPDH reverse

GAGGGTCTCTCTCTTCCTCTT

Recombinant GLP-1 CM-2 forward

GTGAGCTCTTATCTGGAAGGCC

CM-2 reverse

AGATAAGAGCTCACATCGCTGG

BNP BNP forward

GCAGCAGCCTCTATCCTCTC

BNP reverse

TCCTGTATCCCTGGCAGTTC

three levels of short axis view (basal, mid ventricular and apical levels) were obtained by acquiring three successive cardiac cycles. Echocardiographic data was transferred to and analyzed using Image Arena 4.1 (Tomtec Imaging Systems, Uterschleissheim, Germany). This analysis was performed by an investigator blinded for the treatment groups. Regional left ventricular function was assessed by measuring the change in LV cavity area (fractional area change) during systole and diastole at basal (mitral valve level), mid ventricular (papillary muscle level) and apical level. Left ventricular volumes were then calculated by the modified Simpson rule: LV end diastolic volume (LVEDV)= (AbED) * L/3 + (AmED+ApED)/2) * L/3 + 1/3(ApED) * L/3; LV end systolic volume (LVESV)= (AbES) * L/3 + (AmES+ApES)/2) * L/3 + 1/3(ApES) * L/3, in which Ab is the area at basal level, whereas Am and Ap are the areas at mid and apical level respectively2. L is defined as the length of the ventricle from apex to base. The length was obtained of 4-chamber view 3D echo in severe AMI study at baseline and follow-up. In the LCx study, the length of the ventricle in the long axis view was used. LVEF was calculated following standardized formula: ((LVEDV-LVESV)/LVEDV)*100. 3D-echocardiography In study 2, epicardial 3D-echocardiographs were acquired with X-3 transducer using the iE33 ultrasound machine (Philips, Eindhoven, The Netherlands). The 3D-transducer (X-3, Philips, Eindhoven, The Netherlands) was wrapped in a sterile sleeve. A pocket of gel was positioned under the transducer, to bring the complete apex a vu. The transducer was positioned directly epicardially on the apex of the heart. We positioned the transducer in all animals in the same direction so we obtained a 4-chamber view. The depth and sector size were adjusted to fit the complete ventricle. In each pig, the data sets were acquired in real time using 7 consecutive cardiac cycles (full volume analysis).

221

PART THREE CHAPTER 11

The images were offline analyzed using QLab 10.1 (3DQ advanced) analysis software. The tracing of the ventricle was perfomed by semi-automatic border detection as described before by Soliman and coworkers3. Briefly, LV quantification starts by proper 4-chamber view and orthogonal views. The end-diastolic volume and end-systolic frames are identified and on both frames and the apex, anterior, lateral, inferior and septal mitral annulus is identified. Qlab 10.1 automatically traces the endocardial border. Traces that are unsatisfactory can be manually adjusted. The ejection fraction is calculated by the Qlab software as (LVEDV-LVESV)/LVEDV*100. PV-loop measurement and analysis In the severe infarct model, a 7F tetrapolar admittance catheter (7.0 VSL Pigtail, no lumen, Transonic, Scisence, London, Canada) was introduced in the left ventricle via the aortic valve, and the pigtail tip of the catheter was positioned in the apex. This catheter measures admittance magnitude and phase in combination with pressure. The catheter exists of 7 electrodes, that divide it into 4 segments. The largest segment that was positioned into the LV was used for the measurements. The catheter was connected to the ADVantage sytem (Transonic, Scisence, London, Canada) for real-time data assessment4â&#x20AC;&#x201C;6. PV loop measurements were performed at eight weeks follow up. To obtain data regarding contractility, a caval vein occlusion was performed. The thorax was opened, the inferior vena cava was located whereupon a prolene suture was placed around it. During breath hold, the inferior caval vein was occluded by the suture until pressure drop was >50%. The suture was released and blood flow to the heart was restored. In this study, contractility parameters and relaxation parameters were used in the analysis. LVEF and LV-volumes were assessed by 2D-echocardiography and 3D-echocardiography. Systolic PV-loopderived parameters that were assessed and analysed in current study were: dP/dt max (maximum increase in pressure/s), end-systolic-pressure-volume-relationship (ESPVR), prerecruitable stroke work (PRSW). Diastolic parameters that were determined were: Tau (an isovolumic relaxance constance), dP/dt min (maximum relaxation over time), end-diastolic pressure volume relationship (EDPVR). Tissue collection and infarct size calculations At 8 weeks follow-up, the left ventricle was separated from the right ventricle and sliced into 5-6 slices of approximately 1 cm thickness. Slices were incubated in 1% triphenyltetrazolium chloride (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37Â°C 0.9%NaCl for 15 min to discriminate infarct tissue from viable myocardium followed by inspection for the presence of potential ectopic micro-infarctions in non-infarcted segments. Slices were weighed and photographed for infarct size calculations as described before.7 Biopsies of infarct area, border zone and remote areas were obtained and embedded in paraffin for further histological analysis. Additional biopsies were snap frozen for RNA retrieval and qPCR analysis. The liver, spleen, lungs and kidneys were excised and macroscopically analyzed for abnormalities, whereupon multiple random biopsies were taken to exclude shedding of eMSC to remote organs. Biopsies were embedded in paraffin,

222

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

cut and stained with Hematoxilin and Eosin and microscopically evaluated to a technician blinded for the allocated therapy and evaluated regarding the occurrence of eMSC in remote organs. Immuno-histochemical stainings Collagen content, cardiomyocyte size and myocardial salvage index Collagen content in infarct, border and remote myocardial segments was assessed via trichrome stain. Briefly, all sections were deparaffinised, fixated in Buoin’s fixative (SigmaAldreich, St. Louis, USA) at 56º for 15 minutes, Nuclei were stained with haematoxylin for 3 minutes. The slides were submerged in Trichrome-AB solution for 5 minutes after which they were treated with 0.5% acetic acid for 1 minute. Slides were mounted with Entallan (Merck, Darmstadt, Germany). Three random pictures were made at a 10X magnification and collagen content was calculated as percentage collagen of total surface area using automated analysis software (Clemex, Quebec, Canada). To calculate cardiomyocyte size, 3 random pictures of all slides of the border area and remote area were obtained at 40X magnification. The surface area was determined by the automated analysis software. Only transversely cut cardiomyocytes containing a nucleus were analysed. Capillary and arteriole density Arteriole density was obtained in infarct, border and remote myocardial segments, using alpha smooth muscle actin (SMA, clone 1a4, Sigma-Aldreich, St Louis, USA). Endogenous peroxidase activity was blocked by 3% Methanol/H2O2 solution for 30% and incubated with SMA 1:1000 overnight. Subsequently, the slides were incubated with secondary HRPconjugated goat-anti-mouse dilution 1:200 (DAKO, Glostrup, Denmark) for 90%. All slides were immersed in DAB solution for 2 minutes (DAKO, Glostrup, Denmark) and mounted with Entallan. A technician that was blinded for the dose groups took 3 random pictures at 10 times magnification. Arterioles per pictures were counted and expressed as number per mm2. Capillary density was only assessed in border en remote areas, because almost all capillaries are destroyed after AMI. All sections were dewaxed and pre-treated with trypsin EDTA (Lonza, Verviers, Belgium). Endogenous peroxidase activity was blocked as described above. All slides were incubated with Isolectin B4 (Bandeiraea simplicifolia Isolectin B4, Dako, Glostrup, Germany) diluted 1/50 overnight. After that the slides were immersed in DAB solution at mounted with Entallan. Photographs were taken at 20X magnification and number per mm2 was calculated. Apoptosis Apoptosis was quantified in all regions using In-Situ cell detection kit-FITC labelled (second study) or In-situ cell detection kit-HRP labelled (first study) (Roche, Basel, Swiss). The manufacturer’s instructions were followed. After mounting with Vectashield with DAPI, 3 random photographs at 40X magnification were made using an Olympus IX55 fluorescence

223

PART THREE CHAPTER 11

microscope and the number of apoptotic cells were counted. Apoptosis is depicted as percentage of apoptotic cells. Quantitative PCR The expression of human household genes, GLP-1 and BNP as markers for hypertrophy were investigated using qPCR analysis. RNA was isolated from snap frozen biopsies of infarct area, border zone and remote myocardial using RNA-Bee™ RNA Isolation Solvent (Tel-Test Inc., Friendswood, USA) according to the manufacturer’s protocol. CDNA was created using Bio-Rad iScript™ cDNA Synthesis Kit (Bio-Rad, Veenendaal, The Netherlands), according to the manufacturer’s instructions. The primers that were investigated can be found in the table below. qPCR was performed using SensiMix™ SYBR & Fluorescein Kit (Bioline, Boston, USA) and expression was detected by Bio-Rad MyiQ System Software of MyiQ™ Optical Module. The threshold cycle (Ct) values from interested genes were normalized to porcine Beta Actin expression(housekeeping gene) or human GAPDH (GLP-1).

RESULTS Brain Natriuretic Peptide (BNP) analysis Blood was sampled at baseline and at 8 week follow up. The clinical chemical laboratory of University medical center in Utrecht and in Rotterdam, both tried to measure BNP levels using an ELISA. Unfortunately, this was not successful, therefore BNP levels were detected using qPCR analysis. There were no differences in qPCR levels of BNP between groups. Necropsy There were no signs of anatomical malformations by macroscopic analysis of the heart, liver, spleen, lungs and kidneys. There were no signs of macro-infarcts in remote areas of the heart. Left ventricular weight No differences were observed regarding left ventricular weight at 8 week follow-up in all groups in both studies.

224

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

11 Figure I. Fibrosis, vascular density and apoptosis A-B: Collagen density in infarct and border area in the first phase of the study. C: Arteriole density increased in infarct area in all dose groups opposed to control. D: Capillary density. E-F: Cardiomyocyte apoptosis in border and remote segments respectively. eMSC indicates encapsulated mesenchymal stem cells

225

PART THREE CHAPTER 11

Figure II. eMSC at 8 week FU A; Example of qPCR reaction using porcine housekeeping gene actin, human housekeeping gene GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) en GLP-1 (Glucagon-like peptide-1) in a border sample of an animal treated with eMSCs. There was no signal detected of GLP-1 or human housekeeping genes. B. TUNEL stain of an eMSC at 8 week FU. Cell density inside the eMSC is low and the cells are positive (brown) for apoptosis, indicating that the MSC have died during the 8 week FU.

226

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

Figure III. Inflammatory reaction around the eMSC at 8 week FU The same response occurred around all eMSC groups equally and consisted of granulocytes and macrophages.

227

PART THREE CHAPTER 11

Supplemental Table I. Volumes and ejection fraction assessed by 2D and 3D echocardiography Intermediate infarct 2D-echo

Control

20.000 CB

40.000 CB

60.000 CB

P -value ANOVA

LVEF Baseline

51.2 ± 0.6

LVEDV Baseline

102.2 ± 1.0

52.0 ± 0.2

52.7 ± 0.6

56.3 ± 0.5

90.7 ± 0.7

101.5 ± 2.2

90.2 ± 0.8

LVESV Baseline

50.0 ± 1.0

43.5 ± 0.4

48.4 ± 1.5

39.4 ± 0.5

LVEF PMI

43.7 ± 0.8

42.6 ± 0.4

42.1 ± 0.3

43.6 ± 0.7

LVEDV PMI

105.2 ± 1.4

100.3 ± 2.0

111.9 ± 1.1

95.4 ± 1.0

LVESV PMI

59.3 ± 1.2

59.0 ± 1.0

55.7 ± 1.0

LVEF 8W FU

42.6 ± 0.8

47.5 ± 0.7

46.0 ± 0.8

47.3 ± 1.0

0.09

LVEDV 8W FU

123.4 ± 3.5

116.1 ± 1.4

122.0 ± 2.7

109.2 ± 2.3

0.08

LVESV 8W FU

70.8 ± 2.2

61.2 ± 1.4

69.5 ± 2.2

55.5 ± 2.1

0.09

Severe infarct 2D-echo

Control

10.000 CB

20.000 CB

40.000 CB

P-value ANOVA

LVEF Baseline

52.8 ± 2.6

54.6 ± 3.5

55.5 ± 1.7

51.7 ± 3.3

LVEDV Baseline

97.4 ± 2.4

93.9 ± 4.7

91.3 ± 4.5

93.9 3.3

LVESV Baseline

40.6 ± 5.3

43.6 ± 3.9

40.8 ± 2.8

45.0 ± 3.4

LVEF 8W FU

40.6 ± 2.0

44.7 ± 2.4

48.1 ± 0.9

47.1 ± 2.8

0.04

LVEDV 8W FU

123.8 ± 3.7

125.2 ± 3.5

112.5 ± 3.8

118.5 ± 8.6

LVESV 8W FU

70.3 ± 3.4

70.4 ± 3.8

58.6 ± 3.0

61.8 ± 8.3

Control

10.000 CB

20.000 CB

40.000 CB

P-value ANOVA

LVEF Baseline

57.6 ± 1.0

58.5 ± 1.8

58.4 ± 1.0

55.8 ± 1.4

LVEDV Baseline

93.1 ± 3.4

87.7 ± 7.9

84.6 ± 6.7

86.5 ± 4.1

LVESV Baseline

39.7 ± 1.2

36.3 ± 2.5

35.1 ± 3.0

38.1 ± 1.5

LVEF PMI

38.9 ± 1.1

41.9 ± 1.5

35.3 ± 1.2

39.9 ± 1.3

LVEDV PMI

83.4 ± 5.5

81.9 ± 4.1

81.8 ± 4.5

71.8 ± 2.9

LVESV PMI

51.0 ± 3.6

47.0 ± 2.6

52.0 ± 2.7

43.2 ± 2.4

NS 0.032

3D-echo

LVEF 8W FU

38.9 ± 1.5

41.6 ± 1.7

44.7 ± 1.2

42.3 ± 2.1

LVEDV 8W FU

108.1 ± 6.4

116.4 ± 5.8

89.2 ± 7.1

99.2 ± 13.1

LVESV 8W FU

65.4 ± 3.2

68.0 ± 4.1

49.5 ± 4.2

58.9 ± 8.9

0.03

Delta LVEF

0.2 ± 1.0

0.6 ± 0.9

9.3 ± 1.6

3,5 ± 5,1

0.004

Delta LVEDV

30.9 ± 8.2

38.9 ± 8.1

6.4 ± 7.5

21.9 ± 11.3

Delta LVESV

18.1 ± 4.5

20.8 ± 3.9

2.9 ± 4.8

12.8 ± 7.7

Supplemental Table I represents echocardiographic derived volumes and ejection fraction. 2D-echocardiographic volumes are calculated by the modified Simpson method in both studies and 3D-echocardiography in study 2. P-values are corrected using a post-hoc Bonferroni correction. LCx indicates left circumflex artery; LVEF: left ventricular ejection fraction; LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume; PMI: post myocardial infarct; 8W FU: eight week follow-up; LAD: left anterior descending artery; delta: indicates the absolute difference between post myocardial infarct measurement and 8 week FU. eMSC indicated encapsulated MSC

228

3.7 ± 0.5

10.000 eMSC

62.6 ± 7.6 3.4 ± 0.6 2.0 ± 0.3

12.8 ± 1.1 14.7 ± 1.5 10.1 ± 0.9

855.5 ± 83.7 1131 ± 196.2

7.0 ± 0.7 5.4 ± 0.6

Control

% collagen % collagen % collagen

arterioles/mm2 arterioles/mm2 arterioles/mm2

capillaries/mm2 capillaries/mm2

% apoptosis/view

Infarct

Border

Remote

Infarct

Border

Remote

Border

Remote

Border

Remote

20.5 ± 1.4

95.1 ± 2.1 9.2 ± 3.1 7.6 ± 2.6

% collagen % collagen % collagen

Infarct

Border

Remote

Collagen density

Infarct size

Study 2: Severe infarct

Cardiomyocyte apoptosis

Capillary density

Arteriole density

Collagen density

2.5 ± 0.7

4.7 ± 1.3

76.1 ± 8.7

5.2 ± 0.9

890.7 ± 142.6

992.0 ± 147.4

10.4 ± 2.5

18.3 ± 3.3

22.7 ± 2.9

1.4 ± 0.2

4.4 ± 1.1

59.1 ± 7.8

7.6 ± 1.2

9.6 ± 1.3

Infarct size

20,000 eMSC

Control

measure

Study 1: Moderate infarct parameter

Supplemental Table II. Histological analysis

3.2 ± 0.6

3.2 ± 1.3

78.2 ± 6.2

16.7± 0.2

20.000 eMSC

4.5 ± 0.9

5.5 ± 0.4

1140.0 ± 1160.4

889.3 ± 128.9

11.2 ± 1.7

16.7 ± 1.5

20.8 ± 3.4

1.8 ± 0.3

3.5 ± 0.7

71.5 ± 5.0

9.1 ± 1.2

40,000 eMSC

1.8 ± 0.9

5.2 ± 1.6

72.6 ± 9.0

40.000 eMSC

3.9 ± 0.7

4.2 ± 0.5

1157.0 ± 196.5

966.7 ± 145.2

9.6 ± 2.3

14.2 ± 1.3

17.3 ± 1.8

1.7 ± 0.5

2.8 ± 0.6

69.6 ± 5.0

9.3 ± 1.8

60,000 eMSC

0.05

0.04

0.09

P-value

P- value

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

229

230 arterioles/mm2 arterioles/mm2

Border

Remote

capillaries/mm2

Remote

% apoptotis/view

Remote

4.8 ± 0.9

7.5 ± 2.0

1131.0 ± 119.9

908.1 ± 99.6

3.8 ± 0.5

6.6 ± 0.4

8.4 ± 0.9

1.9 ± 1.0

3.7 ± 1.5

899.8 ± 76.8

1066.0 ± 152.0

4.2 ± 0.3

6.0 ± 0.4

15.5± 4.4

4.0 ± 1.5

2.4 ± 0.7

1232.0 ± 130.9

1209 ± 64.6

5.7 ± 1.1

9.7 ± 1.4

22.2 ± 2.6

mmHg/s

mmHg/mL

mmHg

dP/dt + 8 week FU

ESPVR 8 week FU

PRSW 8 week FU

mmHg/s

mmHg/ml

dP/dt - 8 week FU

EDPVR Baseline

0.184 ± 0.07

-624.2 ± 129.7

68.8 ± 5.8

59.7 ± 10.2

1.9 ± 0.5

974.6 ± 161.7

Control

Study 2: severe Infarct 10.000 eMSC

0.027 ± 0.015

-1051 ± 82.8

55.0 ± 4.0

72.4 ± 13.2

5.2 ± 1.0

1417.1 ± 113.2

20.000 eMSC

0.0168 ± 0.028

-634.5 ± 94.3

67.9 ± 6.8

56.7 ± 7.8

6.5 ± 1.3

1071.4 ± 135.3

0.02 ± 0.003

-851.0 ± 66.8

47.7 ± 2.1

51.2± 17.7

5.2 ± 1.7

1226.4 ± 136.2

40.000 eMSC

4.8 ± 1.8

5.0 ± 3.3

837.3 ± 137.5

0.03

P-value

0.004

Supplemental Table SIII represents some admittance derived parameters. End-systolic pressure volume relationship (ESPVR) is significantly better in all animals indicating a preserved contractility. PRSW indicates; pre-recruitable stroke work. EDPVR: end-diastolic pressure volume relationship.

Tau 8 week FU

Diastolic Parameters

measure

Systolic Parameters

Parameter

Supplemental Table III. PV-loop parameters

3.7 ± 0.8

7.2 ± 2.1

15.4 ± 0.6

951.3 ± 181.5

Supplemental Table III. Overview of histological analysis. NS indicates not significant. eMSC: encapsulated mesenchymal stem cells

% apoptotis/view

Border

Cardiomyocyte apoptosis

capillaries/mm2

Border

Capillary density

arterioles/mm2

Infarct

Arteriole density

Supplemental Table II. Continued

PART THREE CHAPTER 11

ENCAPSULATED MSC PRESERVE LV FUNCTION AFTER AMI

REFERENCES 1. Stone GW, Brodie BR, Griffin JJ, Morice MC, Costantini C, St. Goar FG, Overlie PA, Popma JJ, McDonnell J, Jones D, O’Neill WW, Grines CL. Prospective, Multicenter Study of the Safety and Feasibility of Primary Stenting in Acute Myocardial Infarction: In-Hospital and 30-Day Results of the PAMI Stent Pilot Trial. J Am Coll Cardiol. 1998;31:23–30. 2. Folland ED, Parisi a. F, Moynihan PF, Jones DR, Feldman CL, Tow DE. Assessment of left ventricular ejection fraction and volumes by real- time, twodimensional echocardiography. A comparison of cineangiographic and radionuclide techniques. Circulation. 1979;60:760–766. 3. Soliman OII, Krenning BJ, Geleijnse ML, Nemes A, van Geuns R-J, Baks T, Anwar AM, Galema TW, Vletter WB, ten Cate FJ. A comparison between QLAB and TomTec full volume reconstruction for real time three-dimensional echocardiographic quantification of left ventricular volumes. Echocardiography. 2007;24:967–74. 4. Hout G van, Jong R De. Admittance‐based pressure–volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;l:1–11. 5. Porterfield JE, Pearce JA. The admittance technique: an improvement over conductance catheter technology. Comput Eng. 78712. 6. Kottam A, Dubois J, Mcelligott A, Henderson KK. ovel Approach to Admittance to Volume Conversion for Ventricular Volume Measurement. In Vivo (Brooklyn). C:4–7. 7. Houtgraaf JH, de Jong R, Kazemi K, de Groot D, van der Spoel TIG, Arslan F, Hoefer I, Pasterkamp G, Itescu S, Zijlstra F, Geleijnse ML, Serruys PW, Duckers HJ. Intracoronary infusion of allogeneic mesenchymal precursor cells directly after experimental acute myocardial infarction reduces infarct size, abrogates adverse remodeling, and improves cardiac function. Circ Res. 2013;113:153– 66.

231

part four

GENERAL DISCUSSION AND DUTCH SUMMARY

chapter 12

General Discussion

PART FOUR CHAPTER 12

Coronary artery disease is the single most frequent cause of death worldwide, accounting for 12.8% of all deaths1,2. For the Netherlands this means that approximately 18 patients die every day from MI3. The mortality after MI is declining, most likely due to optimized revascularization therapies and improved medical care for patients after MI. Nevertheless, mortality after acute MI still remains considerable, with reports of mortality rates of 12% after 6 months1. These numbers justify the search for both interventional and pharmacological therapies after MI to further optimize post-MI care. MI also directly increases the risk of HF, since irreparable cardiac damage decreases cardiac function4. Due to the inability of the myocardium to replace damaged cells by new cardiomyocytes, the necrotic tissue is replaced by a collagen-based scar that influences cardiac physiology5. To this extent, the inflammatory response that is triggered after cardiac ischemia-reperfusion is essential. After MI, the myocardium releases danger molecules that activate circulating leukocytes6. These inflammatory cells migrate into the damaged myocardium to induce clearance of necrotic tissue. Besides exerting beneficial effects, leukocytes also cause additional damage to viable tissue, thereby directly contributing to myocardial ischemia-reperfusion injury and infarct expansion7. Infarct expansion, together with rapid clearance of damaged tissue, results in myocardial wall thinning. Due to the existing contractile forces on the scar tissue, infarct expansion and wall thinning enable cardiac dilatation. This process of adverse cardiac remodeling eventually culminates in HF8. HF is a chronic disorder that has considerable health care costs4. Moreover, HF directly decreases the quality of life and has a very poor prognosis9. To further optimize post-MI patient care and confine cardiac damage, novel therapies are needed10. Before such therapies can be tested in MI patients, thorough assessment of the efficacy of such therapies is mandatory. Although mechanistic insights are most likely to be obtained by performing in vitro experiments and transgenic mouse studies, these experiments do not represent clinically relevant treatment protocols. To translate newly discovered therapeutic targets to clinical applications, highly translational animal models are therefore needed11,12. These animal models enable a detailed assessment of cardiac function after administration of clinically feasible dosing of new therapies in a controlled setting and enable clinically feasible treatment regimens. Nevertheless, the step from positive large animal studies to clinical implementation has been proven to be very difficult, since large clinical trials with experimental cardioprotective compounds have shown very limited beneficial effects13â&#x20AC;&#x201C;16. This translational failure is based on discrepancies between large animal MI studies and clinical trials. The quest for novel therapies in patients after MI should therefore not only focus on the discovery of new therapies, but also at improving the translational value of large animal MI models. The aim of this thesis therefore was to 1) optimize large animal model study protocols, 2) validate new measurements that reflect cardiac function and 3) continue the search for therapeutic strategies that attenuate cardiac injury after MI. This thesis was divided into three parts, based on these aims.

236

GENERAL DISCUSSIONâ&#x20AC;&#x192;

PART ONE Optimization of large animal MI models In patients, MI occurs due to rupture of an atherosclerotic plaque, leading to thrombus formation and occlusion of a coronary artery17. The development of atherosclerosis and occurrence of MI is associated with many risk factors such as age, gender, hypercholesterolemia, hypertension, obesity, diabetes and smoking18. These factors undoubtedly influence the extent of cardiac ischemia-reperfusion injury and adverse pathological remodeling of the heart after coronary occlusion. The absence of these risk factors in large animals could possibly be a major confounder of outcome and therefore decrease the translational value of large animal MI studies. Although it is possible to expose animals to these risk factors by e.g. artificially inducing diabetes19, disease initiation is very costly, time consuming and disease progression has not yet reached similarities with the human situation. For this reason this specific limitation is currently hard to overcome in large animal research. However, many other limiting factors of preclinical large animal studies can and should be optimized. These experimental MI studies should mimic clinical treatment regimens, delivery routes and follow-up as much as possible20. Moreover, extensive knowledge about these animal models is required to determine follow-up duration, timing of therapy and the overall similarity of the particular model to the clinical situation. To this extent we investigated the effect of the occlusion site on infarct size and cardiac function in chapter 2. Here we carefully assessed cardiac function at different points in time in a severe MI model, in which we occluded the left anterior descending artery (LAD) and a moderate MI model that underwent left circumflex (LCx) occlusion. The LAD model showed a severely reduced ejection fraction (EF). Moreover, this model showed progressive left ventricular dilatation, indicating the occurrence of adverse cardiac remodeling and progression into HF. On the other hand, LCx occlusion only showed a moderate decay of cardiac function with stabilization over time and no signs of progressing into HF. In this chapter we conclude that for testing therapies targeted at the prevention of adverse remodeling that use left ventricular geometrics as a primary read-out, one should rather occlude the LAD after the first diagonal artery, while both models are suitable for studies that target ischemiareperfusion injury and use infarct size as primary endpoint. Many different methods have been used to artificially induce MI in large animal models. Although the actual occlusion method and duration vary among different models, generally two surgical approaches are used in the field to induce MI21. The first approach requires invasive surgery (sternotomy) to reach the coronary arteries and perform an external ligation. The second method follows an intravascular approach to occlude the coronary artery and only requires minimally invasive surgery. In chapter 3 we investigated the effect of extensive surgery on infarct size in a porcine model of MI. To this extent we subjected pigs to transluminal balloon occlusion with or without preceding medial sternotomy. Although we did not fully elucidate the mechanisms in chapter 3, we observed a marked infarct size reduction and preservation of cardiac function in the open chest versus the closed chest MI model after 75 minutes or LAD occlusion. These findings are of importance,

237

PART FOUR CHAPTER 12

since infarct size reduction by preceding injury could be mediated through pathways related to preconditioning or inflammation22,23. Sternotomy could therefore obscure effects of therapies targeting these pathways and should in our opinion be avoided when possible. Apart from the surgical approach and the choice which coronary artery to occlude, many other aspects of preclinical study design can influence study outcome and effect size. We investigated which factors could influence outcome in large animal MI studies that assess the effects of anti-inflammatory compounds. The inflammatory response after MI has long since known to play an essential role in the decay of cardiac function after myocardial infarction24. Many different anti-inflammatory compounds, comprising the whole range of anti-inflammatory mechanisms, have therefore been tested in large animal models. Although many positive effects have been observed, anti-inflammatory therapies are not implemented in post-MI clinical care25. Since this discrepancy suggests translational failure, we systematically analyzed the overall effect of anti-inflammatory therapies in chapter 4 and 5 by performing a meta-analysis. In chapter 4 we discuss our study design of the meta-analysis. To identify study parameters that are of significant influence on effect size in these studies we performed meta-regression. In chapter 5 we provide the results of our meta-analysis and reveal that the effect of anti-inflammatory compounds depends on the duration of follow-up. We also show that blinding of the investigator decreases the reported efficacy of anti-inflammatory compounds. This indicates that the effect of novel compounds is higher when clinically irrelevant study protocols are used, leading to overestimation of the effect of anti-inflammatory compounds in large animal MI studies. Additionally, chapter 5 also reveals that a pronounced difference exists between different anti-inflammatory drug groups, based on working mechanism. This allowed identification of promising compounds that could be interesting for future clinical testing. PART TWO Measurements and markers for cardiac function Large animal models enable extensive and thorough assessment of cardiac function to determine therapy efficacy. Real-time invasive pressure volume measurements (PV-loops) provide detailed information on systolic and diastolic left ventricular function combined with cardiac (patho)physiology26. PV-loop measurements require placement of a conductance catheter in the left ventricle. The blood conductance measured by this catheter first has to be separated from parallel (myocardial) conductance before it can be converted into volume27. Recently, a novel admittance-based PV-system has been developed. This system allows for rapid separation of blood conductance and parallel conduction by using admittance28. Moreover, this admittance based PV-system does not assume a linear relation between conductance and volume and allows constant measurements of parallel conduction during the cardiac cycle29,30. In chapter 6, this PV-system was compared with a conventional conductance based PVsystem and 3D-echocardiography (3DE). In the healthy in vivo pig heart, the admittancebased system did not show any difference with 3D-echocardiography regarding end diastolic volume (EDV), end systolic volume (ESV) or EF. Although the admittance based

238

GENERAL DISCUSSION

system was able to reveal a difference in cardiac function between infarcted and healthy porcine hearts, it underestimated EF and overestimated cardiac volumes in pigs after MI in chapter 6 compared to 3DE. Since 3DE is known to underestimate cardiac volumes, especially in large damaged hearts after MI31,32, we compared the admittance PV system with the gold standard cardiac magnetic resonance imaging (CMR) in chapter 7. This chapter shows that indeed 3DE underestimates EDV and ESV in pigs subjected to MI after a follow-up duration of 8 weeks, justifying comparison of the PV-system with CMR. This comparison revealed that the admittance based PV-system also overestimates cardiac volumes when compared to CMR in the infarcted porcine heart. Nevertheless, cardiac function and volumes correlated significantly between the PV measurements and CMR indicating that the PV system, although with more variation, reliably measures cardiac function after MI. Since invasive and detailed measurements of cardiac function can be time-consuming and burdening for MI patients, serological markers that are easily and inexpensively measurable and possess diagnostic or prognostic information could be of value in this setting. The inflammatory response plays an essential role in acute MI and neutrophils are among the first responders after cardiac injury33. Neutrophil numbers have shown to have great prognostic value in clinical MI34. Instead of looking at neutrophil quantity, in chapter 8, we investigated whether neutrophil size reflected the extent of cardiac damage after MI. Mean neutrophil volume (MNV) showed to be elevated in ST-elevation MI patients and correlated with CK-levels. In a more controlled setting where pigs were subjected to 75 minutes of myocardial ischemia-reperfusion injury, early changes in MNV correlated with cardiac function, infarct size and troponin measurements. Since the mechanisms behind MNV elevation were unknown, in chapter 8 we elucidated that higher MNV levels were observed due to the migration of banded neutrophils into the systemic circulation in combination with a direct activation of circulating neutrophils. Future studies should focus on the validation of our findings in large patient cohorts. PART THREE Novel therapeutic strategies In 1994, a paradigm shift took place in the way investigators looked at the inflammatory response after MI. The theory that the immune systems discriminates between ‘self’ and ‘non-self’ no longer covered new findings and as a result Matzinger was the first to propose the danger model, in which pathogen associated molecular patterns (PAMPs) or danger associated molecular patterns (DAMPs) serve as ligands for pattern recognition receptors (PRRs)35,36. When binding to their membrane bound PRRs, DAMPs can induce the synthesis and secretion of pro-inflammatory cytokines. Inhibition of this signaling cascade therefore reduces the inflammatory response after MI and these DAMPs and PRRs could hence serve as therapeutic targets. In chapter 9 we provide a concise overview of the DAMPs and PRRs that are known to play a role in MI. We conclude that interference with toll-like receptor (TLR)2, TLR4 and NLRP3-inflammasome mediated signaling are promising interventions to confine cardiac damage after MI by reducing the immune reaction. Since clinically

239

PART FOUR CHAPTER 12

applicable pharmacological agents have been developed, translation of mechanistic rodent studies is now possible. In chapter 10 we performed such a study with the selective small molecule inhibitor MCC950. This compound interferes with NLRP3-inflammasome mediated signaling, thereby decreasing the secretion of interleukin (IL)-1Î˛ and IL-18. In chapter 10 we reveal that continuous inhibition of the NLRP3-inflammasome is feasible by daily intravenous administration of MCC950 in a pig MI model. This inhibition results in a dose-dependent reduction of infarct size and a pronounced difference in left ventricular function after 7 days follow-up in favor of animals treated with MCC950. We conclude that targeting the NLRP3-inflammasome is a very promising pharmacological intervention for patients after myocardial infarction. Chapter 11 covers experiments that were performed to test the efficacy of intracoronary infusion of encapsulated mesenchymal stem cells (eMSCs). The encapsulation consists of alginate, which allows for nutrients to reach the eMSCs and secreted molecules by the MSCs to leave the encapsulation. At the same time the hostâ&#x20AC;&#x2122;s immune system is unable to attack these cells. These eMSCs have been genetically programmed to produce the cardioprotective glucagon-like peptide-1 (GLP-1) and could therefore serve as paracrine factories that provide long-term cardioprotection during cardiac wound healing after MI. In this chapter we have studied intracoronary delivery of these eMSCs in a pig model in which we occluded the LCx and in a model that underwent LAD occlusion. We show that administration of 20.000 beads containing MSCs preserves cardiac function and prevents adverse remodeling by reducing apoptosis and promoting vessel growth in the LAD model. However, the LCx model did not show a difference in left ventricular function. Although efficacious, the therapeutic dose for these eMSCs is very narrow since a high dose could cause microvascular obstruction and a low dose fails to show any effect, making the clinical applicability of this intervention through intracoronary delivery limited at this moment.

FUTURE PERSPECTIVES The current thesis aimed to increase the translational value of large animal MI models and translate novel treatment strategies that protect the heart from damage after MI from mechanistic studies to possible clinical applications. The ultimate goal in this research field is to develop a clinically applicable method that protects the heart and leads to a preservation of cardiac function and less adverse events in patients suffering from MI10,37. However, large randomized clinical trials have revealed that promising experimental compounds were not beneficial in MI patients. This translational failure can be attributed to 1) clinically irrelevant preclinical study design, 2) selection of inappropriate animal models, 3) poor clinical study design and 4) a possible difference in mechanisms of ischemia-reperfusion injury and adverse remodeling in humans versus other animals. To enable good study design and select the appropriate animal model, extensive knowledge about the development of cardiac dysfunction in these large animal models is

240

GENERAL DISCUSSIONâ&#x20AC;&#x192;

mandatory. Importantly, preclinical study design should always resemble the clinical situation as closely as possible, and methodologically unreliable large animal studies with e.g. pretreatment protocols, obscure the true effect of certain therapies and promote translational failure. Apart from improving methodology, future research should aim at optimizing the resemblance of large animal models with MI patients. The most important hurdle that has to be taken is the introduction of risk factors in large animal MI models. Since risk factors can influence disease progression and therapy efficacy, introducing these risk factors in the experimental-setup of large animal MI studies will increase translational value of these models. Although this step will take an incredible amount of time and money, creating these animal models will add to our knowledge on clinically relevant ischemiareperfusion injury and the post-MI inflammatory response. Moreover, large animal MI models will better predict clinical efficacy by exposing animals to risk factors for MI and in the long term hopefully less animals are needed to predict the efficacy of new compounds. The research field of cardioprotection has experienced many setbacks in the past decades. Therapeutics that showed a positive effect in large animal studies were not beneficial in MI patients13â&#x20AC;&#x201C;16,38,39, raising the question whether the mechanisms that have been observed in animal models, are also occurring in MI patients. Although at this moment this is still not completely clear, MI and HF have considerable mortality and morbidity rates and future studies should therefore keep focusing on the development of interventions that preserve cardiac function in post-MI patients, despite the low success rate. To this extent, many newly discovered mechanisms have been proven to play an essential role in infarct expansion and impairment of cardiac function after MI. Especially the interaction of DAMPs and PRRs along with their downstream signal transducers form interesting targets for future experimental and clinical studies. The first large randomized clinical trials targeting these pathways are now recruiting and pilot results look promising40,41. In the near future we will hopefully be able to determine if selective inhibition of these pathways is effective in patients and whether pharmacological intervention in patients after MI to reduce infarct size, prevent adverse remodeling and preserve cardiac function makes it into daily clinical practice.

241

PART FOUR CHAPTER 12

REFERENCES 1. Steg PG, James SK, Atar D, Badano LP, Lundqvist CB, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, Van’T Hof A, Widimsky P, Zahger D, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, FunckBrentano C, Hasdai D, Hoes A, Kirchhof P, Kolh P, McDonagh T, Moulin C, Popescu B a., Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Astin F, Åström-Olsson K, Budaj A, Clemmensen P, Collet JP, Fox K a., Fuat A, Gustiene O, Hamm CW, Kala P, Lancellotti P, Maggioni A Pietro, Merkely B, Neumann FJ, Piepoli MF, Van De Werf F, Verheugt F, Wallentin L, Blömstrom-Lundqvist C, Borger M a., Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, van ’t Hof A, Widimsky P, Zahger D. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 2012;33:2569– 2619. 2. Who. WHO | The Atlas of Heart Disease and Stroke. 3. Hartstichting. Feiten en Cijfers Hartinfarct. 2008; 4. Mcmurray JJ V, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, Falk V, Filippatos G, Fonseca C, Gomez-Sanchez MA, Jaarsma T, Kober L, Lip GYH, Maggioni A Pietro, Parkhomenko A, Pieske BM, Popescu B a., Ronnevik PK, Rutten FH, Schwitter J, Seferovic P, Stepinska J, Trindade PT, Voors A a., Zannad F, Zeiher A, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Kirchhof P, Knuuti J, Kolh P, Mcdonagh T, Moulin C, Reiner Ž, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Bonet LA, Avraamides P, Ben Lamin H a., Brignole M, Coca A, Cowburn P, Dargie H, Elliott P, Flachskampf FA, Guida GF, Hardman S, Iung B, Merkely B, Mueller C, Nanas JN, Nielsen OW, Orn S, Parissis JT, Ponikowski P. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012. Eur J Heart Fail. 2012;14:803–869. 5. Frangogiannis NG. The Immune System and the Remodeling Infarcted Heart. J Cardiovasc Pharmacol. 2014;63:185–195. 6. Frangogiannis NG. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 2014;11:255–65. 7. Arslan F, de Kleijn DP, Pasterkamp G. Innate immune signaling in cardiac ischemia. Nat Rev Cardiol. 2011;8:292–300. 8. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981–2988.

242

9. Hartstichting. Feiten en Cijfers Hartfalen. 2008; 10. Hausenloy DJ, Yellon DM. Myocardial ischemiareperfusion injury: a neglected therapeutic target. J Clin Invest. 2013;123:92–100. 11. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411. 12. Hausenloy DJ, Baxter G, Bell R, Bøtker HE, Davidson SM, Downey J, Heusch G, Kitakaze M, Lecour S, Mentzer R, Mocanu MM, Ovize M, Schulz R, Shannon R, Walker M, Walkinshaw G, Yellon DM. Translating novel strategies for cardioprotection: The Hatter Workshop Recommendations. Basic Res Cardiol. 2010;105:677–686. 13. Armstrong PW, Granger CB, Adams PX, Hamm C, Holmes D, O’Neill WW, Todaro TG, Vahanian A, Van de Werf F. Pexelizumab for acute ST-elevation myocardial infarction in patients undergoing primary percutaneous coronary intervention: a randomized controlled trial. JAMA. 2007;297:43– 51. 14. Baran KW, Nguyen M, McKendall GR, Lambrew CT, Dykstra G, Palmeri ST, Gibbons RJ, Borzak S, Sobel BE, Gourlay SG, Rundle a C, Gibson CM, Barron H V. Double-blind, randomized trial of an anti-CD18 antibody in conjunction with recombinant tissue plasminogen activator for acute myocardial infarction: limitation of myocardial infarction following thrombolysis in acute myocardial infarction (LIMIT AMI) stu. Circulation. 2001;104:2778–2783. 15. Arai M, Lefer DJ, So T, DiPaula A, Aversano T, Becker LC. An anti-CD18 antibody limits infarct size and preserves left ventricular function in dogs with ischemia and 48-hour reperfusion. J Am Coll Cardiol. 1996;27:1278–1285. 16. Granger CB, Mahaffey KW, Weaver WD, Theroux P, Hochman JS, Filloon TG, Rollins S, Todaro TG, Nicolau JC, Ruzyllo W, Armstrong PW. Pexelizumab, an anti-C5 complement antibody, as adjunctive therapy to primary percutaneous coronary intervention in acute myocardial infarction: The COMplement inhibition in Myocardial infarction treated with Angioplasty (COMMA) trial. Circulation. 2003;108:1184–1190. 17. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, De Lemos J a., Ettinger SM, Fang JC, Fesmire FM, Franklin B a., Granger CB, Krumholz HM, Linderbaum J a., Morrow D a., Newby LK, Ornato JP, Ou N, Radford MJ, Tamis-Holland JE, Tommaso CL, Tracy CM, Woo YJ, Zhao DX. 2013 ACCF/AHA guideline for the management of stelevation myocardial infarction: A report of the American college of cardiology foundation/ american heart association task force on practice

GENERAL DISCUSSION

guidelines. J Am Coll Cardiol. 2013;61:e78–140. 18. Yeh RW, Sidney S, Chandra M, Sorel M, Selby J V, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med. 2010;362:2155–2165. 19. Koopmans SJ, Schuurman T. Considerations on pig models for appetite, metabolic syndrome and obese type 2 diabetes: From food intake to metabolic disease. Eur J Pharmacol. 2015; 20. Heide RS Vander, Steenbergen C. Cardioprotection and myocardial reperfusion: Pitfalls to clinical application. Circ Res. 2013;113:464–477. 21. Lukács E, Magyari B, Tóth L, Petrási Z, Repa I, Koller A, Horváth I. Overview of large animal myocardial infarction models (review). Acta Physiol Hung. 2012;99:365–81. 22. Jones WK, Fan GC, Liao S, Zhang JM, Wang Y, Weintraub NL, Kranias EG, Schultz J El, Lorenz J, Ren X. Peripheral nociception associated with surgical incision elicits remote nonischemic cardioprotection via neurogenic activation of protein kinase C signaling. Circulation. 2009;120:S1–9. 23. Ren X, Wang Y, Jones WK. TNF-alpha is required for late ischemic preconditioning but not for remote preconditioning of trauma. J Surg Res. 2004;121:120–129. 24. Christia P, Frangogiannis NG. Targeting inflammatory pathways in myocardial infarction. Eur J Clin Invest. 2013;43:986–995. 25. Kristensen SD, Laut KG, Fajadet J, Kaifoszova Z, Kala P, Di Mario C, Wijns W, Clemmensen P, Agladze V, Antoniades L, Alhabib KF, De Boer MJ, Claeys MJ, Deleanu D, Dudek D, Erglis A, Gilard M, Goktekin O, Guagliumi G, Gudnason T, Hansen KW, Huber K, James S, Janota T, Jennings S, Kajander O, Kanakakis J, Karamfiloff KK, Kedev S, Kornowski R, Ludman PF, Merkely B, Milicic D, Najafov R, Nicolini F a., Noč M, Ostojic M, Pereira H, Radovanovic D, Sabaté M, Sobhy M, Sokolov M, Studencan M, Terzic I, Wahler S, Widimsky P. Reperfusion therapy for ST elevation acute myocardial infarction 2010/2011: Current status in 37 ESC countries. Eur Heart J. 2014;35:1957–1970. 26. Krenz M. Conductance, admittance, and hypertonic saline: should we take ventricular volume measurements with a grain of salt? J Appl Physiol. 2009;107:1683–1684. 27. Baan J, van der Velde ET, de Bruin HG, Smeenk GJ, Koops J, van Dijk a D, Temmerman D, Senden J, Buis B. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation. 1984;70:812– 823. 28. Porterfield JE, Pearce J a. The admittance technique: an improvement over conductance catheter technology. Comput Eng. :78712. 29. Porterfield JE, Kottam ATG, Raghavan K, Escobedo D, Jenkins JT, Larson ER, Treviño RJ, Valvano JW, Pearce J a, Feldman MD. Dynamic correction for

parallel conductance, GP, and gain factor, alpha, in invasive murine left ventricular volume measurements. J Appl Physiol. 2009;107:1693– 1703. 30. Wei CL, Valvano JW, Feldman MD, Pearce J a. Nonlinear conductance-volume relationship for murine conductance catheter measurement system. IEEE Trans Biomed Eng. 2005;52:1654–1661. 31. Greupner J, Zimmermann E, Grohmann A, Dübel HP, Althoff T, Borges AC, Rutsch W, Schlattmann P, Hamm B, Dewey M. Head-to-head comparison of left ventricular function assessment with 64-row computed tomography, biplane left cineventriculography, and both 2- and 3-dimensional transthoracic echocardiography: Comparison with magnetic resonance imaging as the reference s. J Am Coll Cardiol. 2012;59:1897–1907. 32. Dorosz JL, Lezotte DC, Weitzenkamp D a., Allen L a., Salcedo EE. Performance of 3-dimensional echocardiography in measuring left ventricular volumes and ejection fraction: A systematic review and meta-analysis. J Am Coll Cardiol. 2012;59:1799–1808. 33. Vinten-Johansen J. Involvement of neutrophils in the pathogenesis of lethal myocardial reperfusion injury. Cardiovasc Res. 2004;61:481–497. 34. Horne BD, Anderson JL, John JM, Weaver A, Bair TL, Jensen KR, Renlund DG, Muhlestein JB. Which white blood cell subtypes predict increased cardiovascular risk? J Am Coll Cardiol. 2005;45:1638–1643. 35. Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol. 1994;12:991–1045. 36. Matzinger P. The danger model: a renewed sense of self. Science. 2002;296:301–305. 37. Heusch G. Cardioprotection: Chances and challenges of its translation to the clinic. Lancet. 2013;381:166–175. 38. Mahaffey KW, Granger CB, Nicolau JC, Ruzyllo W, Weaver WD, Theroux P, Hochman JS, Filloon TG, Mojcik CF, Todaro TG, Armstrong PW. Effect of pexelizumab, an anti-C5 complement antibody, as adjunctive therapy to fibrinolysis in acute myocardial infarction: The COMPlement inhibition in myocardial infarction treated with thromboLYtics (COMPLY) trial. Circulation. 2003;108:1176–1183. 39. Rusnak JM, Kopecky SL, Clements IP, Gibbons RJ, Holland a E, Peterman HS, Martin JS, Saoud JB, Feldman RL, Breisblatt WM, Simons M, Gessler CJ, Yu a S. An anti-CD11/CD18 monoclonal antibody in patients with acute myocardial infarction having percutaneous transluminal coronary angioplasty (the FESTIVAL study). Am J Cardiol. 2001;88:482– 487. 40. Sonnino C, Christopher S, Oddi C, Toldo S, Falcao RA, Melchior RD, Mueller GH, Abouzaki NA, Varma A, Gambill ML, Van Tassell BW, Dinarello CA, Abbate A. Leukocyte activity in patients with STsegment elevation acute myocardial infarction treated with anakinra. Mol Med. 2014;20:486–9.

243

PART FOUR CHAPTER 12

41. Abbate A, Kontos MC, Abouzaki NA, Melchior RD, Thomas C, Van Tassell BW, Oddi C, Carbone S, Trankle CR, Roberts CS, Mueller GH, Gambill ML, Christopher S, Markley R, Vetrovec GW, Dinarello CA, Biondi-Zoccai G. Comparative Safety of Interleukin-1 Blockade With Anakinra in Patients With ST-Segment Elevation Acute Myocardial Infarction (from the VCU-ART and VCU-ART2 Pilot Studies). Am J Cardiol. 2015;115:288â&#x20AC;&#x201C;292.

244

245

chapter 13

Nederlandse Samenvatting

PART FOUR CHAPTER 13

Ziekten van de kransslagaders zijn de meest frequente doodsoorzaak wereldwijd, hetgeen neerkomt op 12.8% van alle doden1. Een hartinfarct, of myocardinfarct (MI), is een van de belangrijkste onderdelen van deze ziektegroep en ontstaat wanneer een kransslagader acuut afgesloten raakt door een afgescheurde atherosclerotische plaque (aderverkalking). Hierdoor raakt de hartspier permanent beschadigd en functioneert deze minder goed. Dagelijks worden er in Nederland 78 personen opgenomen met een MI en sterven er gemiddeld 18 personen per dag aan een MI2. De laatste jaren is de medische zorg rondom een MI sterk verbeterd en ondergaan de meeste patiënten een percutane coronaire interventie (PCI), waardoor het aantal patiënten dat uiteindelijk sterft als direct gevolg ervan gestaag afneemt. Echter, de mortaliteit is nog steeds aanzienlijk waardoor nieuwe therapieën nodig zijn die patiënten beschermen na een acuut MI. Patiënten die een MI overleven lopen het risico dat zij hartfalen ontwikkelen doordat het hart ernstig beschadigd is geraakt en niet meer voldoende bloed kan rondpompen. Hartfalen is een chronische ziekte die ervoor zorgt dat de kwaliteit van leven snel achteruit gaat en enorme zorgkosten veroorzaakt. Bovendien is 37% van de patiënten die de diagnose hartfalen krijgen na een jaar overleden3. Om het aantal patiënten dat hartfalen ontwikkelt na een MI terug te dringen is de ontwikkeling van nieuwe medicijnen een vereiste. Deze medicijnen dienen de schade na een MI te reduceren. Naast de permanente, irreversibele schade die wordt veroorzaakt door langdurig zuurstofgebrek aan het hart, is er na het verrichten van een PCI namelijk ook sprake van reversibele schade. Deze schade wordt veroorzaakt doordat het herstel van de perfusie (reperfusie) ertoe leidt dat er additionele schade aan het hart ontstaat (reperfusieschade)4. Naast andere complexe moleculaire processen, speelt de ontstekingsreactie een essentiële rol in reperfusieschade. Na een hartinfarct worden circulerende leukocyten geactiveerd en migreren deze cellen naar het beschadigde myocard (hartspierweefsel). Deze ontstekingscellen spelen een belangrijke rol bij het opruimen van het permanent beschadigde weefsel, echter ze veroorzaken ook additionele schade aan de hartspier, hetgeen op termijn kan leiden tot een verslechtering van de hartspierfunctie, pathologisch remodeleren en hartfalen5. Voordat therapieën die op deze ontstekingsreactie of andere processen ingrijpen, getest kunnen worden in patiënten, is nauwkeurig en grondig testen van de efficiëntie en veiligheid van dergelijke behandelingen vereist. Hoewel mechanistische in vitro studies en studies in kleine proefdieren inzicht kunnen verschaffen in de moleculaire werking van bepaalde behandelingen, zijn deze experimenten niet representatief voor de klinische behandelprotocollen van patiënten met een hartinfarct. Om recentelijk ontdekte veelbelovende therapieën te vertalen naar klinische behandelingen zijn daarom translationele grote proefdier modellen vereist6. Deze proefdier modellen hebben grote gelijkenissen met mensen met betrekking tot hemodynamiek, cardiale anatomie en (coronair) farmacokinetiek en –dynamiek. Een voordeel van studies in grote diermodellen is ook dat er in een gecontroleerde setting een zeer gedetailleerde inschatting van de hartspierfunctie kan worden gedaan, die

248

NEDERLANDSE SAMENVATTING

onderzoekers meer informatie over de effectiviteit van een therapie kan verschaffen. Ondanks deze toegevoegde waarde van grote proefdieren zijn er veel behandelingen positief getest in deze experimenten, terwijl klinisch vervolgonderzoek negatief bleek te zijn. Dit suggereert dat een optimalisatie van translationele protocollen nodig is om de voorspellende waarde van grote proefdierstudies te vergroten De queeste naar nieuwe therapieën voor patiënten na een acuut hartinfarct dient zich dus niet alleen te richten op het valideren van deze experimentele behandelingen in grote proefdier modellen, maar ook op de verdere optimalisatie van translationele protocollen en meetmethodes voor hartfunctie. Het doel van het huidige proefschrift is daarom om 1) translationele grote proefdierstudies te optimaliseren, 2) nieuwe functie metingen en surrogaat markers van hartfunctie te onderzoeken en 3) de zoektocht naar veelbelovende therapeutische strategieën voor patiënten na MI te continueren. Het huidige proefschrift is opgedeeld in drie delen waarin hoofdstukken aan de orde komen die deze vraagstukken behandelen. DEEL ÉÉN Optimalisatie van grote proefdierstudies In patiënten ontstaat een MI in verreweg de meerderheid van de gevallen door het afscheuren van een atherosclerotische plaque. De vorming van atherosclerose is sterk geassocieerd met een aantal risicofactoren zoals roken, diabetes, hypertensie, obesitas, leeftijd en geslacht. Het is zeer aannemelijk dat deze risicofactoren niet alleen effect hebben op de vorming van atherosclerose, maar ook de processen van reperfusieschade en ontsteking na een MI beïnvloeden. De afwezigheid van deze risicofactoren in grote diermodellen zou dus een mogelijke reden kunnen zijn van het verschil in effectiviteit van bepaalde behandelingen in grote proefdier modellen enerzijds en grote klinische patiënten studies anderzijds. Hoewel het mogelijk is om een aantal van deze risicofactoren te introduceren in groot proefdieronderzoek zijn deze processen erg tijdrovend en kosten veel geld. Bovendien zijn de gelijkenissen met humane situatie niet altijd evident. Om deze reden is deze limitatie van groot proefdier onderzoek op dit moment moeilijk te verhelpen. Andere limiterende factoren van preklinische grote proefdierstudies kunnen wel geoptimaliseerd worden. Het ontwerp van deze dierstudies dient zoveel mogelijk te zijn gebaseerd op bestaande klinische behandelprotocollen, toedieningswegen en duur van follow-up van patiënten in klinische studies en reguliere patiëntenzorg. Om te bepalen of het diermodel gelijkend is met de klinische situatie en het studiedesign van grote proefdierstudies te optimaliseren, is een gedetailleerde kennis van de representatieve diermodellen vereist. In dit perspectief hebben we in hoofdstuk 2 onderzocht hoe de hartfunctie zich over de tijd ontwikkelt in twee MI varkensmodellen. In het ene model hebben we de linker circumflex arterie (LCx) en in het andere model de linker anterior descending arterie (LAD) geoccludeerd voor 120 minuten. Het LCx model liet een matige afname van de hartfunctie zien die stabiel was over de tijd, terwijl het LAD model een vrij

249

PART FOUR CHAPTER 13

ernstige afnamen van de linker ventrikelfunctie liet zien met progressieve dilatatie van het linker ventrikel. In dit hoofdstuk hebben we geconcludeerd dat therapeutica die inwerken op reperfusieschade getest kunnen worden in beide modellen terwijl behandelingen die focussen op pathologisch remodeleren van hart beter gebruik kunnen maken van het LAD model. Om kunstmatig een MI te induceren bij grote proefdier modellen zijn er veel verschillende methodes gebruikt. Hoewel de occlusie-methode verschilt tussen deze verschillende modellen, worden er grofweg twee verschillende chirurgische benaderingen gehanteerd voor MI inductie. De eerste methode vereist invasieve chirurgie waarbij the thorax middels een sternotomie wordt geopend om vanuit een externe benadering de coronairarterie the occluderen. De tweede methode maakt gebruik van minimaal invasieve chirurgie om een vasculaire toegang te verschaffen om op deze manier een coronair occlusie the induceren. In hoofdstuk 3 hebben we onderzocht wat het effect van invasieve chirurgie is op de infarct grootte in een MI varkensmodel. We hebben hiertoe varkens onderworpen aan een transluminale ballonocclusie waarbij de helft van de varkens voorafgaand aan deze procedure onderworpen werd aan een mediale sternotomie. Alhoewel de exacte moleculaire mechanisme voor het geobserveerde effect niet geïdentificeerd zijn in hoofdstuk 3, concluderen we wel dat invasieve chirurgie tot een uitgesproken reductie van de infarctgrootte en een behouden hartfunctie leidt na 75 minuten ischemie-reperfusie. Omdat dit effect mogelijk gemedieerd wordt door bepaalde signaaltransductie mechanismen die ook een rol spelen bij de bescherming van het hart na MI kan mediale sternotomie een mogelijke confounder zijn wanneer bepaalde therapieën in grote proefdier MI modellen getest worden. Naast de chirurgische benadering en de keuze welke coronairarterie te occluderen, zijn er veel andere aspecten van preklinisch studiedesign die de studie-uitkomst kunnen beïnvloeden. In een systematische review hebben we daarom onderzocht welke factoren van invloed zijn op de infarct grootte in grote proefdierstudies die onderzochten wat het effect van een anti-inflammatoire behandeling na MI was. Het is algemeen geaccepteerd dat de inflammatoire response een essentiële rol speelt in de achteruitgang van de hartfunctie na MI. Veel verschillende anti-inflammatoire behandelingen zijn daarom getest in grote proefdier modellen die op een scala van verschillende mechanismen ingrijpen. Hoewel er een heel aantal positieve dierstudies zijn geweest, wordt er geen antiinflammatoire therapie toegepast bij patiënten na een MI. Dit suggereert dat er spraken is van een discrepantie tussen groot proefdieronderzoek enerzijds en patiënten studies anderzijds. In hoofdstuk 4 en 5 hebben we daarom onderzocht wat het algehele effect is van anti-inflammatoire therapie in grote proefdier MI studies door middel van metaanalyse. Hoofdstuk 4 beschrijft het studieprotocol van deze meta-analyse en in hoofdstuk 5 bepalen we met behulp van meta-regressie welke factoren van invloed zijn op de infarct grootte na MI. Dit hoofdstuk laat zien dat het effect van anti-inflammatoire therapieën onder ander afhangt van het geslacht van het proefdier en van de follow-up duur van de studie. Daarnaast heeft ook de blindering van de onderzoek een significant effect op de infarct grootte. De conclusie van dit hoofdstuk is dan ook dat klinisch irrelevante

250

NEDERLANDSE SAMENVATTING

studieprotocollen leiden tot een overschatting van anti-inflammatoire behandelingen in grote proefdier modellen. Daarnaast laten we in hoofdstuk 5 ook zien dat er een significant verschil bestaat in de effectiviteit van verschillende anti-inflammatoire behandelmethode. Op deze manier kunnen de meest effectieve medicamenten geïdentificeerd worden die mogelijk in patiënten getest kunnen worden. DEEL TWEE Metingen en markers voor hartfunctie In grote proefdier modellen is het mogelijk om een zeer nauwkeurige en grondige inventarisatie van de hartfunctie te bewerkstelligen om op deze manier de effectiviteit van een therapie in kaart te brengen. Invasieve druk-volume metingen (PV-Loops) verschaffen gedetailleerde informatie over de systolische en diastolische ventriculaire functie. Om PV-loop metingen te verrichten is de plaatsing van een conductantie katheter in het linker ventrikel noodzakelijk. Om deze conductantie te kunnen converteren naar linker ventrikel volumes, moet de bloed conductantie eerst gescheiden worden van de parallelle conductantie. Recentelijk is er een nieuw PV-loop systeem ontwikkeld dat de bloed conductantie snel en gemakkelijk kan isoleren op basis van admittantie. Bovendien heeft dit nieuwe PV-Loop systeem de aanname dat er een niet-lineaire relatie bestaat tussen de conductantie van het bloed en het linker ventrikel volume. Daarnaast is het systeem in staat om de contributie van de parallelle conductantie continue tijdens de cardiale cyclus te meten. Theoretisch gezien zouden volumina met dit systeem daardoor nauwkeuriger geschat moeten worden. In hoofdstuk 6 hebben we dit PV-systeem vergeleken met een conventioneel, op conductantie gebaseerd, PV-systeem en echocardiografie. In het gezonde in vivo varkenshart liet het admittantie-systeem (AS) geen significante verschillen zien ten opzichte van echocardiografie voor eind diastolisch volume (EDV), eind systolisch volume (ESV) of ejectie fractie (EF). Hoewel het AS ook een verschil aantoonde in hartfunctie tussen gezonde varkens en varkens onderworpen aan een hartinfarct, werd dit systeem wel onnauwkeuriger in het zieke hart. EF werd door het AS onderschat, terwijl EDV en ESV overschat werden. Omdat echocardiografie linker ventrikel volumina doorgaans onderschat in vergelijking met de gouden standaard ‘cardiac magnetic resonance imaging’ (CMR), hebben we in hoofdstuk 7 het AS vergeleken met echocardiografie en CMR. Dit hoofdstuk liet inderdaad een onderschatting van EDV en ESV zien voor echocardiografie ten opzichte van CMR, hetgeen een vergelijking tussen het AS en CMR rechtvaardigt. De bevindingen in dit hoofdstuk duiden erop dat ook ten opzichte van CMR het AS zowel EDV als ESV overschat. Echter, deze volumina correleren (deels) wel met CMR, wat erop wijst dat de metingen met het AS – zij het met meer variatie –betrouwbaar de hartfunctie meten. Aangezien invasieve en gedetailleerde meting van de hartfunctie veel tijd kan kosten en tevens belastbaar kan zijn voor patiënten na een MI, kan de ontwikkelingen van goedkope en snel meetbare circulerende markers die diagnostische of prognostische waarde bezitten, in deze setting van toegevoegde waarde zijn. De ontstekingsreactie speelt een belangrijke rol in een acuut MI en neutrofiele granulocyten zijn de eerste circulerende cellen die

251

PART FOUR CHAPTER 13

worden geactiveerd na cardiale schade. Neutrofiel aantallen hebben bewezen grote prognostische waarde te bezitten in patiënten met een MI. In plaats van neutrofiel kwantiteit te onderzoeken, hebben we in hoofdstuk 8 onderzocht wat het effect van MI is op morfologische veranderingen van de gemiddelde neutrofiel. In het bijzonder hebben we bepaald hoe het gemiddelde volume van een neutrofiel (MNV) verandert na MI. Dit hoofdstuk laat zien dat MNV significant hoger is in patiënten na een MI in vergelijking tot patiënten met stabiele angineuze klachten. Daarnaast correleren deze MNV waardes ook met creatine kinase waardes in patiënten. In een MI varkensmodel van 75 minuten ischemiereperfusie laten we vervolgens zien dat MNV veranderingen ook correleren met infarctgrootte, Troponine waardes en met achteruitgang van de EF. Daarnaast laten we in hoofdstuk 8 ook zien dat de oorzaak van de veranderingen van MNV gelegen zijn in een veranderde populatie van circulerende neutrofiele granulocyten alsmede de lokale activatie van deze circulerende neutrofielen. DEEL DRIE Nieuwe therapeutische strategieën Sinds ongeveer 20 jaar wordt de inflammatoire response na MI anders bekeken door onderzoekers. Hier ligt de theorie aan ten grondslag dat het immuun systeem geen onderscheid maakt tussen pathogene substanties en endogene substanties, zoals eerder werd gedacht, maar dat, het immuun systeem bepaalde moleculen als gevaren signalen waarneemt ongeacht de bron. Wanneer deze moleculen aan bepaalde receptoren binden ontstaat er een signaaltransductie die direct leidt tot een ontstekingsreactie. In hoofdstuk 9 geven we een overzicht van de meest onderzochte gevaar-moleculen die een belangrijke rol spelen na MI. We beschrijven daarnaast aan welke receptoren deze moleculen kunnen binden en in wat voor soort ontstekingsreactie dit resulteert. We concluderen in dit overzicht dat blokkade van bepaalde interacties tussen deze gevaar-moleculen en receptoren zou kunnen dienen als mogelijke toekomstige therapie na een hartinfarct. In hoofdstuk 10 hebben we de effectiviteit van één van deze veelbelovende therapieën getest in een varkensmodel van MI. De stof MCC950 werkt in op het NLRP3-inflammasoom. Het NLRP3-inflammasoom is een intracellulaire molecuulstructuur die geactiveerd wordt na MI. Na activatie is dit eiwitcomplex in staat om interleukiene (IL)-1β en IL-18 tot hun actieve vorm om te zetten, waarna deze moleculen de cel uit gesecreteerd worden. IL-1β en IL-18 zijn twee zeer potente stimulatoren van de ontstekingsreactie na een MI. In hoofdstuk 10 tonen we aan dat het remmen van deze moleculaire reactie leidt tot een verminderde intrede van circulerende neutrofielen in het myocard en een verminderde systemische ontstekingsreactie. Deze inhibitie zorgt ook voor een dosis-afhankelijke reductie van de infarct grootte na MI, alsmede een sterk behouden hartfunctie in de varkens die behandeld zijn met MCC950. In deze studie concluderen we dan ook dat inhibitie van het NLRP3-inflammasoom een zeer veelbelovende therapie is voor patiënten na een MI. Vervolg studies zullen moeten uitwijzen of deze therapie mogelijk in de toekomst in patiënten getest kan gaan worden.

252

NEDERLANDSE SAMENVATTING

Hoofdstuk 11 behelst experimenten die zijn uitgevoerd om de effectiviteit te bepalen van mesenchymale stamcellen die omgeven zijn van een beschermlaag van alginaat. Dit alginaat zorgt ervoor dat voedingstoffen de cellen kunnen bereiken en geproduceerde stoffen door de stamcellen naar het omliggende weefsel kunnen defunderen, terwijl het immuunsysteem niet direct de cellen kan aanvallen. Deze stamcellen zijn genetische geprogrammeerd om het cardioprotectieve GLP-1 te produceren en zouden daarom dus gedurende lange tijd bescherming kunnen geven aan het hart na een MI. In dit hoofdstuk hebben we onderzocht wat het effect is van intra-coronaire toediening van de alginaat kraaltjes die deze stamcellen bevatten in een varkensmodel waarin we de LCx hebben geoccludeerd en een model waarin we de LAD hebben geoccludeerd voor 90 minuten. In deze studie observeren we dat toediening van 20.000 van deze kraaltjes leidt tot een behoud van de linker ventrikel functie en een reductie van het pathologisch remodeleren van het linker ventrikel door het verminderen van apoptose en het stimuleren van vaatnieuwvorming in het LAD model. Hoewel de therapie effectief lijkt te zijn in deze specifieke groep, laten de andere groepen geen verschil zien. Daarnaast is geen van de doseringen die we in het LCx model getest hebben effectief. De klinische toepasbaarheid van deze therapie lijkt daarom gering te zijn op dit moment.

TOEKOMST PERSPECTIEF Het huidige proefschrift richtte zich op het verhogen van de translationele waarde van grote proefdier MI studies en het vertalen van veelbelovende nieuwe therapeutische strategieën tot klinisch toepasbare behandelingen. Het ultieme doel in dit specifieke onderzoeksveld is het ontwikkelen van een klinisch toepasbare methode om het hart te beschermen direct na MI. Deze bescherming dient te leiden tot een behoud van de hartspierfunctie, een betere kwaliteit van leven, minder aandoeningen zoals hartfalen en mogelijk een verbeterde overleving van patiënten. Ondanks vele positieve dierstudies in het verleden, hebben grote klinische studies keer op keer bewezen dat deze bevindingen niet gereproduceerd konden worden in patiënten. Het falen van translatie van deze therapieën naar de kliniek kan worden toegedicht aan 1) klinisch irrelevant preklinisch studieontwerp, 2) het gebruiken van niet-representatieve diermodellen, 3) suboptimaal ontwerp van klinische studies en 4) een mogelijk verschil in mechanismen in patiënten in vergelijking met proefdieren. Om toekomstig goed translationeel studieontwerp en adequate keuze van diermodellen te garanderen, is gedetailleerde kennis van de ontwikkeling van cardiale dysfunctie in deze grote proefdier modellen noodzakelijk. Daarnaast is het van essentieel belang dat preklinische dierstudies zoveel mogelijk worden ontworpen op basis van bestaande klinische behandelprotocollen aangezien het effect van een behandeling anders erg moeilijk te interpreteren is. Naast de verbetering van deze methodiek dient toekomstig onderzoek zich te richten op het optimaliseren van de gelijkenis van grote proefdier modellen met MI patiënten. Het belangrijkste struikelblok dat hierbij overkomen dient te

253

PART FOUR CHAPTER 13

worden is de introductie van risicofactoren in het hedendaagse grote proefdier onderzoek, aangezien deze factoren ongetwijfeld van invloed kunnen zijn op het effect van nieuwe behandelingen na MI. Mede door deze stap zal de translationele waarde van grote proefdieren mogelijkerwijs vergroot kunnen worden waardoor deze studies de effectiviteit van een therapie in patiënten beter kunnen voorspellen. Daarnaast zijn er door deze toegevoegde waarde op termijn mogelijk ook minder proefdieren nodig om dergelijk onderzoek uit te voeren. Het onderzoeksveld van cardioprotectie heeft veel tegenslagen te verwerken gehad de laatste decennia. Door de vele tegenvallende klinische studies is de vraag ontstaan of de mechanismen die worden waargenomen in proefdieren wel van toepassing zijn op patiënten. Hoewel hier op dit moment nog geen duidelijkheid over is, brengen MI en HF nog steeds aanzienlijke morbiditeit en mortaliteit met zich mee. Om deze reden is continuerend onderzoek op dit gebied vooralsnog gerechtvaardigd, ondanks de relatief lage kans van slagen. Recentelijk zijn er een aantal veelbelovende mechanismen ontdekt die een essentiële rol in de ontstekingsreactie na MI lijken te spelen. In proefdierstudies is onomstotelijk bewezen dat inhibitie van deze mechanismen leidt tot een reductie in infarct grootte, een behouden hartfunctie en een preventie van het pathologisch remodeleren van het hart. Grote klinische studies zijn al van start gegaan die het effect van deze behandelingen onderzoeken en de eerste resultaten lijken veelbelovend. In de nabije toekomst is het hopelijk mogelijk om vast te stellen of selectieve inhibitie van deze mechanismen effectief is in klinische studies en of een farmacologische interventie om infarct grootte te reduceren, een goede hartfunctie te behouden en pathologisch remodeleren te voorkomen, het haalt tot een klinisch toepasbare behandeling in patiënten.

254

NEDERLANDSE SAMENVATTING

REFERENTIES 1. 2. 3. 4.

Who. WHO | The Atlas of Heart Disease and Stroke. Hartstichting. Feiten en Cijfers Hartinfarct. 2008; Hartstichting. Feiten en Cijfers Hartfalen. 2008; Hausenloy DJ, Yellon DM. Myocardial ischemiareperfusion injury: a neglected therapeutic target. J Clin Invest. 2013;123:92–100. 5. Frangogiannis NG. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 2014;11:255–65. 6. Lecour S, Botker HE, Condorelli G, Davidson SM, Garcia-Dorado D, Engel FB, Ferdinandy P, Heusch G, Madonna R, Ovize M, Ruiz-Meana M, Schulz R, Sluijter JPG, Van Laake LW, Yellon DM, Hausenloy DJ. ESC Working Group Cellular Biology of the Heart: Position Paper: improving the preclinical assessment of novel cardioprotective therapies. Cardiovasc Res. 2014;104:399–411.

255

part five

ADDENDUM

Review committee Dankwoord / Acknowledgements List of publications Curriculum Vitae

PART FIVE ADDENDUM

REVIEW COMMITTEE Prof. dr. T. Leiner Prof. dr. L. Koenderman Prof. dr. R.M. Schiffelers Prof. dr. N. van Royen Prof. dr. P. Ferdinandy

260

DANKWOORD / ACKNOWLEDGEMENTS

DANKWOORD / ACKNOWLEDGEMENTS Een woord van dank voor de velen die mij hebben geholpen bij het tot stand komen van het huidige proefschrift is zeker op zijn plaats. Of de bijdrage nu bestond uit wetenschappelijke input, het aanleren van bepaalde handelingen, chirurgische verrichtingen, laboratorium werk of emotionele ondersteuning, er zijn veel mensen die een belangrijke bijdrage hebben geleverd aan dit proefschrift. Daarnaast is samenwerken voor mij altijd van grote toegevoegde waarde geweest omdat het eindresultaat binnen de verschillende projecten vaak groter is dan de simpele som der delen. Er is geen moment geweest gedurende mijn promotietraject dat ik het gevoel had dat ik er alleen voor stond en daarvoor ben ik de mensen om mijn heen erg dankbaar. Zonder anderen tekort te doen zou ik graag een aantal van deze mensen persoonlijk willen bedanken. Dr. Hoefer, beste Imo, in de eerste fase van mijn promotie benaderde ik jou, mijn copromotor, nogal formeel. Ik was gewend aan een klinische setting maar ontdekte al snel dat het in de onderzoekswereld toch iets anders werkt. De afgelopen jaren heb ik erg veel van je geleerd over het schrijven van papers, experimentele dierstudies, cardiale fysiologie, Duits bier en circulerende cellen. Ik heb je altijd om alles kunnen vragen en mede door jou hebben we de hoeveelheid werk kunnen verzetten die gebundeld is in dit proefschrift. Ik ben je erg dankbaar om jouw bijdrage aan alle projecten die we samen hebben gedaan en hoop dat de toekomst nog veel moois voor je gaat brengen. Prof. Dr. Pasterkamp, beste Gerard, vanaf de dag dat ik mijn sollicitatiegesprek had bij jou had ik een prettig gevoel bij het feit dat je mijn promotor zou worden. Je stimuleert mensen constant door de juiste vragen te stellen. Hoewel je jouw bijdrage aan de verschillende projecten vaak bagatelliseert, ben je op deze manier belangrijker geweest voor mijn experimenten dan je jezelf misschien realiseert. Je gaf me het gevoel dat je mij het beste gunt en ik vond het mooi om te zien dat onze gedeelde competitiviteit pas echt naar boven komt bij een potje pingpong. Ik ben van mening dat je jouw lab erg goed managet en promoveren in deze setting heeft me in staat gesteld een zelfstandige wetenschapper te worden en met mijn eigen ideeĂŤn te komen. Mede daarvoor ben ik jou ontzettend dankbaar. Prof. Dr. Van Solinge, beste Wouter, Hoewel we in de laatste fase van mijn promotietraject wat minder frequent contact hadden ben jij toch van wezenlijk belang geweest als promotor. Onze meetings waren altijd erg vruchtbaar en aan het eind had ik vaak een helder beeld van welke kant ik op wilde met mijn moeilijkste projecten. Daarnaast heb je ervoor gezorgd dat ik moeiteloos mijn weg heb kunnen vinden op het klinische chemie lab. Ik wens je veel succes met het managen van deze afdeling in de toekomst. Graag zou ik via deze weg graag alle co-auteurs van mijn verschillende manuscripten willen bedanken voor hun bijdrage. In het bijzonder Dr. Steven Chamuleau,

261

PART FIVE ADDENDUM

Prof. Pieter Doevendans en Prof. Leo Koenderman, voor de input die ervoor gezorgd heeft dat de kwaliteit van een aantal manuscripten aanzienlijk verbeterd is. Gedurende mijn promotie heb ik ongeveer evenveel mede-promovendi zien komen als gaan binnen het lab van de experimentele cardiologie. Met allemaal heb ik in meer of mindere mate samengewerkt en daarvoor wil ik graag iedereen bedanken. Specifiek zijn er nog een aantal collega’s met wie ik intensiever heb samengewerkt. Allereerst Lena, nadat jij al eerder kort op het lab aan een project had gewerkt, kwam je in je laatste jaar als SUMMA student bij mij je masterstage doen. Al vrij snel had ik door dat je een erg goede, zelfstandige wetenschapper bent die bovendien bijzonder fijn is in de omgang. Binnen de kortste keer ontwikkelde wij een efficiënte manier van samenwerken waarbij we elkaar steeds op nieuwe ideeën brachten. Gedurende de eerste fase van jouw promotie hebben we die samenwerking doorgezet met een erg mooi project als gevolg. Ik denk dat je in de toekomst een geweldig proefschrift gaat afleveren en wie weet komen we elkaar daarna weer tegen! Jelte en Crystel, mijn roomies van het laatste jaar. Promoveren betekent ook het delen van je frustraties met elkaar. Ik denk dat we qua arbeidsethos, humor en biertijd over het algemeen op dezelfde lijn zitten. We hebben elkaar daar erg goed in gevonden en samen met Jonne wil ik jullie bedanken voor de mooie tijd in onze kamer en ons appartement in Barcelona. Marten, Sanne, Ellen en Pleunie, mijn roomies van de eerste jaren. Ook jullie wil ik natuurlijk bedanken voor de gezellige tijd die we hadden ‘’back in the days’’. Zoals gezegd, verslindt het experimentele cardiologie lab aanzienlijk wat promovendi de laatste jaren. Daarom zijn er heel wat (oud)collega’s die ik graag zou willen bedanken. Allereerst de collega’s in de ‘’toren’’ in willekeurige volgorde: Vince, Sander, Saskia, Aisha, Bas, Dave, Amir, Quirina, Ingrid en Joyce. Daarnaast mag ik natuurlijk niet de echte lab-rats vergeten van zowel de experimentele cardiologie, traumatologie als klinische chemie: Judith, Erik, Petra, Peter-Paul, Janine, Frederieke, Alain, Krijn, Dries, Marcel, Steven, Michel, Marjolein, Pieter en alle andere collega’s, bedankt en heel veel succes! Ook zou ik mijn dank willen uitspreken aan alle senior onderzoekers die over de jaren input hebben geleverd aan mijn projecten. Saskia, Dennie, Hester, Mirjam, Hamid, Fatih, Leo en Joost bedankt voor jullie bijdrage. Experimenteren is hard werken. Experimenten met grote proefdieren kun je bovendien onmogelijk alleen uitvoeren. Soms kwam het voor dat er gedurende een experiment wel 8 mensen tegelijk aan het rennen waren. Noortje, Sander, Loes, Elske en Julie, jullie wil ik ontzettend bedanken voor de vele lab-experimenten die jullie voor mij gedaan hebben tijdens mijn operaties. Petra, ook jou wil ik graag bedanken voor het optimaliseren en uitvoeren van de meeste kleuringen in het huidige proefschrift! Corina en Esther, bedankt voor het bieden van hulp als ik even vast zat tijdens experimenteren op het lab. Arjan, als baas van het lab moet je de touwtjes strak aangetrokken houden. Volgens mij doe jij dat

262

DANKWOORD / ACKNOWLEDGEMENTS

erg goed, maar er schuilt achter jouw alles verwoestende blik een ongelofelijke behulpzaamheid, dank daarvoor! De kern van mijn proefschrift wordt gevormd door onderzoeksdata die verworven zijn op de grote proefdier operatiekamers van het gemeenschappelijk dierlaboratorium. Daar heb ik vele uren doorgebracht met de meest belangrijke personen voor de totstandkoming van dit proefschrift. Evelyn, Marlijn, Joyce, Grace, Martijn, Merel, Danny en Cees, een paar woorden in dit boek zijn misschien niet genoeg, maar ik wil jullie toch bedanken voor de onnoemelijke hoeveelheid werk die jullie voor mij verzet hebben. Zeker gedurende mijn laatste studie opereerde we als een goed geoliede machine. Ik heb van jullie erg veel mogen leren op de OK en jullie zijn onmisbaar voor het groot proefdier onderzoek binnen het UMC Utrecht. Daarnaast wil ik ook graag Maike en Ben bedanken voor de gezellige tijd op het GDL. Marjolein en later ook Irene, het lijkt me niet makkelijk om een chaotisch en dynamisch secretariaat te runnen als het onze. Veel dank daarvoor en heel veel succes in de toekomst. Ineke, jij doet er altijd alles aan om de (financiële) zaken van het lab op orde te hebben. Ik heb altijd het gevoel gehad dat ik bij je terecht kon en daar wil ik je graag voor bedanken. Jij hebt ongelofelijk veel loyaliteit naar het lab toe en daar heb ik veel bewondering voor. Naast mijn collega’s bij de experimentele cardiologie wil ik ook graag alle arts-onderzoekers van de klinische cardiologie bedanken. Sanne, jou wil ik natuurlijk in het bijzonder bedanken. We hebben voor de totstandkoming van onze boekjes maar liefst 3 gedeelde eerste-auteurschappen en wie weet worden het er in de toekomst nog meer! Ik vond samenwerken met jou altijd erg prettig, je hebt een sterke eigen mening maar luistert tegelijkertijd ook erg goed naar wat de ander te zeggen heeft. Ik heb ongelofelijk veel respect en bewondering voor hoe jij je werkzaamheden hebt opgepakt na 4 maanden afwezigheid. Heel veel succes in de kliniek in Zwolle en bedankt voor de fijne samenwerking! Natuurlijk wil ik ook alle overige collega arts-onderzoekers bedanken voor de fijne tijd. In het bijzonder mijn fietsmaatjes Willemien en Sofieke, maar natuurlijk ook Ing Han, Frebus, René, Remco, Thomas, Manon, Tycho, Stefan, Cheyenne, Iris, Mira, Freek, Martine, Rosemarijn, Marloes, Anneline, Cas en alle andere (oud)collega’s, bedankt voor de gezellige tijd! Ook mijn mede GDL maatjes hebben een belangrijke bijdrage geleverd aan mijn tijd als arts-onderzoeker. Ten eerste Renate, doordat ik kon meewerken aan jouw projecten ben ik in een zeer kort tijdsbestek kundig geworden in de grote proefdierstudies. Naast het feit dat onze samenwerking heel erg soepel en goed verliep, heb ik ontzettend veel van je geleerd gedurende mijn eerste jaren en jij hebt daardoor een erg grote bijdrage geleverd aan mijn proefschrift. David, dagen lang hebben wij naast elkaar gezwoegd op OK 4 en 5 van het GDL. De vele ritten die we tussen Amsterdam en Utrecht in de vroege morgen

263

PART FIVE ADDENDUM

en late avond hebben gemaakt met jouw (vriendin haar) auto zorgde altijd weer voor veel vermaak en het is mooi dat we uiteindelijk een week na elkaar promoveren! Hanna, ook jij weet wat het is grote proefdier experimenten te verrichten. Door jou heb ik ook de nodige experimentele ervaring opgedaan met schapen, iets dat ik niet had willen missen. Bedankt voor de gezelligheid in het GDL en succes in de kliniek. Marjon & Ron, gedurende mijn promotietraject hebben jullie meer dan 1000 metingen gedaan op de Sapphire geloof ik. Ik wil jullie bedanken voor het doen van de meeste metingen en voor het feit dat ik jullie altijd laagdrempelig heb kunnen benaderen als er problemen waren met de metingen of het extraheren ervan! Lisanne, zo vroeg in je studie geneeskunde al mee publiceren is uniek, maar je hebt het helemaal verdiend. Dank voor het harde werk dat je hebt geleverd voor de verschillende projecten! Dear Prof. Matthew Cooper and Dr. Avril Robertson, I would sincerely like to thank you for the collaboration and all the effort you put into developing MCC950. It turned out to be one of the best projects of my thesis. Dear Prof. Malcolm Macleod, Prof. David Howells, Dr. Emily Sena and Dr. Kim Wever, thank you for helping us out with performing the meta-analysis and raising the methodological quality of our manuscript. De leden van de leescomissie, prof. dr. T. Leiner, prof. dr. L. Koenderman, prof. dr. R. Schiffelers, prof dr. N. van Royen en prof. dr. P. Fernandy. Veel dank voor uw bereidheid zitting te nemen in de beoordelingscommissie van mijn proefschrift. Mijn paranimfen, na schoolbanken gedeeld te hebben in het Eindhoven, gingen we alle drie geneeskunde studeren in Maastricht. Sjoerd, ik ben er van overtuigd dat jij de hoofdoorzaak van mijn interesse in de cardiologie bent. Ik kijk meer tegen je op (letterlijk) dan ik durf toe te geven. Je bent een begenadigd medicus en zal een hele goede cardioloog worden. Jij staat voor betrouwbaarheid en loyaliteit, 2 dingen die in de collegebanken iets heel anders kunnen betekenen dan in de kroeg en ik ben blij dat ik beiden door de jaren heen heb kunnen meemaken. Ik ben ontzettend trots dat jij mijn paranimf wil zijn en weet dat ik zowel op persoonlijk als professioneel vlak altijd bij jou terecht kan. Mark, als vanzelfsprekend ben ik ook ontzettend trots dat jij naast mij staat! Van jou vermoed ik juist dat ik 1 van de redenen ben waarom jij uiteindelijk cardiologie bent gaan doen (al zal je dat zelf niet zo snel toegeven). De laatste jaren heb ik jou van ondeugende betweter zien veranderen in een verantwoordelijke vader van 2 kinderen. Ik heb er gruwelijk veel respect voor hoe jij tussen het aankleden, boterhammen smeren en naar de speeltuin gaan door, een hoogwaardig promotieonderzoek aflevert. Net als

264

DANKWOORD / ACKNOWLEDGEMENTS

met mijn andere paranimf deel ik met jou zowel mijn middelbare schooltijd, studentenleven en professionele toekomst, een unieke band waar ik de rest van mijn leven van zal blijven genieten! Jac, Pieter en Tineke. Samen zijn we zorgeloos opgegroeid in het zuidoosten van Brabant. Ook al hebben we alle vier een verschillend karakter en een uitgesproken mening, ik denk dat we over het algemeen toch heel erg op dezelfde lijn zitten. Het is eigenlijk bizar dat we alle vier al enkele jaren op nog geen 10 minuten fietsen van elkaar af wonen in Amsterdam en dat we dat de normaalste zaak van de wereld vinden. Het feit dat onze vriendengroepen verweven zijn, ons ouderlijk huis bomvol zit met carnaval en dat alle afspraken en etentjes vanzelf gaan, zegt genoeg. Toch zou ik mijn dank nog even naar jullie willen uitspreken. Ik weet dat ik altijd op jullie terug kan vallen en dat geeft een erg veilig gevoel. Lieve, lieve papa en mama, de fundering voor mijn proefschrift ligt ongetwijfeld bij jullie. Ik ben ontzettend trots op waar ik vandaan kom en het gevoel dat ons gezellige en warme gezin bij me oproept is onbetaalbaar. Het is altijd weer een drukte van jewelste als 4 kinderen, inclusief aanhang, bij jullie langskomen. Desondanks weten jullie je aandacht perfect te verdelen en blijven jullie, ondanks de roerige tijden in onze familie, altijd veel tijd en aandacht houden voor jullie kinderen. Heel veel dank voor alle steun die jullie aan me gegeven hebben. Allerliefste Marjet, ik was nog maar net begonnen aan mijn promotie-onderzoek toen we elkaar leerden kennen. Inmiddels wonen we al ruim 2 jaar samen en ben jij mijn steun en toeverlaat geworden. Nog nooit heb ik me bij iemand zo op mijn plaats gevoeld als bij jou. Hoewel onze dagelijkse activiteiten compleet van elkaar verschillen, ben je altijd een luisterend oor voor me geweest. Je hebt ontzettend veel geduld met me en bent bovenal heel erg lief en leuk. Ik heb heel erg veel zin in de toekomst die wij samen tegemoet gaan!

265

PART FIVE ADDENDUM

LIST OF PUBLICATIONS Published manuscripts Van Hout GPJ, Arslan F, Pasterkamp G, Hoefer IE. Targeting danger associated molecular patterns after myocardial infarction. Expert opinion on Therapeutic Targets - Accepted August 2015 Van Hout GPJ, Teuben MPJ, Heeres M, De Maat S, De Jong R, Maas C, Kouwenberg LHJA, Koenderman L, van Solinge WW, de Jager SCA, Pasterkamp G, Hoefer IE. Invasive surgery reduces infarct size and preserves cardiac function in a porcine model of myocardial infarction. Journal of Cellular and Molecular Medicine. Accepted June 2015 Zwetsloot PPM, Jansen of Lorkeers SJ, Vegh AMD, Van Hout GPJ, Currie GL, Goumans MJ, Chamuleau SAJ, Sluijter JPG. Cardiac stem cell treatment in myocardial infarction; protocol for a systematic review and meta-analysis of preclinical studies. Evidence based preclinical medicine. Accepted may 2015. De Jong R, Van Hout GPJ, Houtgraaf JH, Takashima S, Pasterkamp G, Hoefer IE, Duckers HJ. Cardiac Function in a Long-Term Follow-Up Study of Moderate and Severe Porcine Model of Chronic Myocardial Infarction. Biomed Res Int. Epub 2015 Feb 23. Van Hout GPJ*, Jansen of Lorkeers SJ*, Wever KE, Sena ES, van Solinge WW, Doevendans PA, Pasterkamp G, Chamuleau, Hoefer IE. Anti-inflammatory compounds to reduce infarct size in large-animal models of myocardial infarction: A meta-analysis. Evidence based preclinical medicine. 2014 Dec: 1(1): 4-10. *Contributed equally De Jong R, Van Hout GPJ*, Houtgraaf JH*, Kazemi K, Wallrapp C, Lewis A, Hoefer IE, Pasterkamp G, Duckers HJ. Intracoronary infusion of encapsulated GLP-1 eluting mesenchymal stem cells (CellBeads™) improves left ventricular function in a porcine model of acute myocardial infarction. Circ Cardiovasc Interv. 2014 Oct;7(5):673-83. *Contributed equally Van Oorschot JWM, Gho JMIH, Van Hout GPJ, Froeling M, Jansen of Lorkeers SJ, Hoefer IE, Doevendans PA, Luijten PR, Chamuleau SAJ, Zwanenburg JJM. MRI of cardiac fibrosis – beyond late gadolinium enhancement. J Magn Reson Imaging. 2014 May;41(5):1181-9. Koudstaal S, Janssen of Lorkeers SJ, Gho JMIH, van Hout GPJ, Jansen M, Gründeman PF, Pasterkamp G, Doevendans PA, Hoefer IE, Chamuleau SAJ. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014 Apr 25;(86). doi: 10.3791/51269.

266

LIST OF PUBLICATIONS

Van Hout GPJ*, Jansen of Lorkeers SJ*, Gho JMIH, Doevendans PA, van Solinge WW, Pasterkamp G, Chamuleau SAJ, Hoefer IE. Admittance based pressure volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014 Apr 22;2(4):e00287. *Contributed equally Van Hout GPJ*, de Jong R*, Vrijenhoek JEP, Timmers L, Duckers HJ, Hoefer IE. Admittancebased pressure-volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. Nov 2013;98(11):1565-75. *Contributed equally

Submitted manuscripts Van Hout GPJ, Solinge WW, Gijsberts CM, Teuben MPJ, Leliefeld PHC, Heeres M, Nijhoff F, De Jong S, Bosch L, De Jager SCA, Huisman A, Stella PR, Pasterkamp G, Koenderman LJ, Hoefer IE. Elevated mean neutrophil volume represents altered neutrophil composition and reflects damage after myocardial infarction. In revision - Basic Research in Cardiology Van Hout GPJ*, Jansen of Lorkeers SJ*, Wever KE, Sena EM, Kouwenberg LHJA, van Solinge WW, Malceod MR, Doevendans PA, , Pasterkamp G, Chamuleau SAJ, Hoefer IE Anti-inflammatory compounds reduce infarct size after myocardial infarction: A metaanalysis of large-animal models. *Contributed equally. In revision - Cardiovascular Research Van Hout GPJ, Bosch L, Ellenbroek GHJM, De Haan JJ, Van Solinge WW, Cooper MA, Arslan F, De Jager SCA, Robertson AAB, Pasterkamp G, Hoefer IE. The selective NLRP3inflammasome inhibitor MCC950 reduces infarct size and preserves cardiac function in a pig model of myocardial infarction. In revision - European Heart Journal Jansen of Lorkeers SJ, Gho JMIH, Koudstaal S, Van Hout GPJ, Zwetsloot PPM, Van Oorschot JWM, Van Eeuwijk ECM, Leiner T, Hoefer IE, Goumans MJ, Doevendans PA, Sluijter JPG, Chamuleau SAJ. Xenotransplantation of human cardiomyocyte progenitor cells does not improve cardiac function in a porcine model of chronic ischemic heart failure. In revision - Plos One

267

PART FIVE ADDENDUM

CURRICULUM VITAE Gerardus (Geert) Petrus Johannes van Hout werd geboren op 15 november 1986 te Geldrop. Na de eerste jaren doorgebracht te hebben in Mierlo, verhuisde het gezin Van Hout in 1991 naar Nuenen, waar Geert is opgegroeid in een warm gezin met een broer, een broertje en een zusje. In 2004 behaalde Geert zijn VWO diploma aan het Lorentz Casimir Lyceum te Eindhoven. In deze periode ontwikkelde hij interesse in biologische en fysiologische processen. In datzelfde jaar ging hij geneeskunde studeren aan de Universiteit Maastricht waar zijn voorliefde voor de fysiologie al snel kon worden toegepast op de menselijke circulatie. In 2006-2007 heeft Geert zijn opleiding voor een jaar stil gelegd om het Zuid-Amerikaanse werelddeel te verkennen. Gedurende zijn geneeskunde opleiding heeft hij zijn semi-arts stage gevolgd onder supervisie van Dr. C.J. Begeman en Dr. N.H.K.M. Lencer op de afdeling cardiologie van het Maastricht UMC, alwaar zijn interesse in de klinische cardiologie definitief vorm kreeg. Zijn passie voor de biologie nam wetenschappelijk gestalte aan door een stage aan het Cardiovasculair Research Instituut Maastricht onder supervisie van Dr. M.M. Gladka en Prof. Dr. L.J. de Windt, waar hij de post-transcriptionele regulatie van een specifieke transcriptie factor onderzocht. Gedurende de laatste fase van zijn studie geneeskunde kreeg Geert eind 2011 de kans om zijn keuze co-schap te volgen op de afdeling Experimentele Cardiologie van het UMC Utrecht, gevolgd door een 3Â˝-jarig promotietraject als arts-onderzoeker op diezelfde afdeling. Gedurende deze periode heeft Geert onderzoek gedaan naar het ontwikkelen van nieuwe therapieĂŤn die het hart beschermen na een acuut hartinfarct onder de supervisie van Dr. I.E. Hoefer en Prof. Dr. G. Pasterkamp. De resultaten van dit promotietraject zijn gebundeld in het huidige proefschrift. Inmiddels is Geert werkzaam als arts-assistent op de afdeling Cardiologie van het UMC Utrecht onder de supervisie van Dr. J.H. Kirkels. Per 1 november 2015 zal hij aanvangen met zijn opleiding tot cardioloog in het UMC Utrecht, door te beginnen met de vooropleiding interne geneeskunde onder supervisie van Dr. Y.F.C. Smets in het OLVG te Amsterdam.

268