Issuu on Google+



2:52 PM

Page 67

LYMPHATIC RESEARCH AND BIOLOGY Volume 4, Number 2, 2006 Š Mary Ann Liebert, Inc.

Measurement of Skin Desmosine as an Indicator of Altered Cutaneous Elastin in Draft Horses With Chronic Progressive Lymphedema HILDE E.V. DE COCK, D.V.M., Ph.D.,1 VERENA K. AFFOLTER, D.V.M., Ph.D.,1 THOMAS B. FARVER, Ph.D.,2 LEEN VAN BRANTEGEM, D.V.M.,3 BRAD SCHEUCH, D.V.M.,4 and GREGORY L. FERRARO, D.V.M.5

ABSTRACT Background: Chronic progressive lymphedema in Clydesdale and Shire draft horses causes severe disability of the limbs which leads to premature death of these horses. Since appropriate function of lymph vessels is dependent on the presence of viable elastin fibers, the goal of this study was to document differences in skin elastin fibers in affected horse breeds, compared to a nonaffected draft horse breed. Methods and Results: Biochemical analysis of cutaneous desmosine, a cross-linking amino acid found only in elastin, was used to measure elastin in the skin from 110 draft horses. This included 7 normal, 38 mildly affected, 30 moderately, and 15 severely affected horses, and 20 horses of a nonaffected draft breed. Desmosine concentrations in neck, considered a nonaffected skin region, and left forelimb, an affected skin region, were compared between the groups. A significantly lower desmosine concentration was found in the skin of the neck and limb of clinically normal animals of affected draft breeds compared to a nonaffected draft horse breed. During the progression of the disease in the affected breeds, cutaneous desmosine concentrations most prominently increased in the skin of the distal limbs. Conclusions: Chronic progressive lymphedema in draft horses was associated with an initially systemic lower cutaneous elastin level and a deposition of elastin during the progression of the disease. A failure of elastic fibers to appropriately support the skin and its lymphatics is proposed as a possible contributing factor for chronic progressive lymphedema in Shires and Clydesdales. chronic progressive disease starts at an early age, progresses throughout the life of the horse, and often ends in disfigurement and disability of the limbs which inevitably leads to the horse’s premature death.1 The clinical signs and pathologic changes closely resemble a con-




hyperkeratosis, and fibrosis of distal limbs has been recognized in Shires, Clydesdales and Belgian Draft horses.1,2 This

Departments of 1Pathology, Microbiology, and Immunology; 2Population Health and Reproduction; and 4Equine Medicine, Veterinary Medical Teaching Hospital; and 5Center for Equine Health, University of California, Davis, California; and 3Department of Pathology, Bacteriology and Poultry Diseases, Ghent University, Merelbeke, Belgium. This research was supported by The Center for Equine Health (UC Davis) and the Marcia MacDonald Rivas Fund. The authors wish to thank Barry Starcher, Department of Biomedical Research, The University of Texas Health Center at Tyler, Texas, for the biochemical analysis of the skin.




2:52 PM

Page 68



dition known in humans as chronic lymphedema or elephantiasis nostras verrucosa.1,3–6 In a previous study, fragmentation and disorganization of elastic fibers supporting the dermal lymphatics in the extremities has been observed in these horses.1 Anchoring filaments composed of microfibrils and elastin are very important for the function of skin lymphatics. The network of elastic fibers that surrounds the dermal lymphatics is believed to constitute an elastic device aiding propulsion of the lymph.7 Damaged or congenital abnormalities of elastic fibers might interfere with normal lymphatic function and therefore result in chronic lymphedema. Quantitative evaluation of the elastin content of the skin can be accomplished by measuring desmosine, a unique crosslink of elastin by means of biochemical analyses.8,9 In this study we present measurements of desmosine levels in skin samples of affected Shires and Clydesdales and correlate it with the severity of the clinical disease. These desmosine levels are compared to the levels in the study controls, that is, in normal horses of affected breeds and in horses of a nonaffected draft horse breed. This study indicates a possible association between altered desmosine levels and associated elastin levels with the condition of chronic progressive lymphedema in Shires and Clydesdales.

MATERIAL AND METHODS Animals One hundred and ten draft horses from three different breeds (Clydesdale, Shire horse, and Percheron) were selected for this study. The control group of a normal, nonaffected draft horse breed consisted of 20 Percherons. The affected breeds consisted of 49 Clydesdales and 41 Shire horses. The animals are owned by private breeders. All control animals were in good health. Animals in the affected breeds were in good health except for the animals that had skin lesions on the distal limbs associated with chronic lymphedema. The skin lesions on the limbs in the affected breeds were clinically evaluated by the same person (GLF) and scored

on a 3-point scale as follows: 0  normal; 1  mild lesions composed of slight swelling of the limb, and 1 or 2 small skin folds in the plantar pastern region; 2  moderate lesions composed of distinct swelling of the limb, several thick skin folds, and occasionally small hard nodules in the plantar pastern region; 3  severe lesions composed of severe swelling of the limb, several thick skin folds, and hard nodules in the plantar pastern region. The affected animals were divided in four groups according to the score of their lesions: group 1 or normal, and group 2, group 3, and group 4, respectively, for mild, moderate, and severe lesions. The animals of the nonaffected breed were grouped separately in group 5. Tissues Two 6-mm punch biopsies were taken, one from the latero-palmar pastern region of the left forelimb (clinically involved skin) and one biopsy from the left caudo-dorsal neck region (clinically uninvolved skin). The skin samples were fixed in 10% buffered formalin and embedded in paraffin. Quantitation of desmosine Two 50 m sections were cut lengthwise, including the total depth of the skin biopsy, for biochemical analysis as described by Starcher (1977).8 In brief, each 50 m section was placed in a microfuge tube, the paraffin removed with xylene, and the sample hydrolyzed in 6 N HCl for 24 h at 100°C. The HCl was evaporated and the residue redissolved in 500 l water. Desmosine was determined by RIA, and total protein in the original section by a ninhydrin method.10,11 Total protein contents was used as denominator for desmosine for comparisons between samples. Statistics SPSS for Windows (SPSS Inc., Chicago, USA) was used for the statistical analysis of the data. One factor considered was “clinical group” using the clinical groups 1 to 5 as described above. In addition, the animals were categorized into three age groups: group A contained the young animals, ages 0 to 4; group B con-


2:52 PM

Page 69


tained the middle-aged animals, ages 5 to 9; and group C contained the older animals ages 10 and older. A general linear model with and without repeated measures was used to compare desmosine values between the different clinical groups, between normal animals of affected breeds and an unaffected breed, and between the two skin regions, neck and pastern region (a repeated measures factor). The results were adjusted for the influence of age and gender.

RESULTS In total, 90 animals of an affected breed and 20 control animals (nonaffected draft horse breed) were biopsied. In the control animals (Percheron, group 5) 11 were mares, 8 were geldings, and 1 was a stallion. Their mean age was 5.6 years (SD  5.2; min 1, max 17). In the affected breeds (group 1 to 4) there were 51 mares, 28 geldings, and 11 stallions. Seven animals were clinically normal (group 1), 38 were given score 1 (mild lesions, group 2), 30 were given score 2 (moderate lesions, group 3), and 15 were given score 3 (severe lesions, group 4). The mean age of all animals was 7.5 years


(SD  4.8; min 1, max 20). The mean age for the normal animals (group 1) in the affected group was 3.3 years (SD  4.0; min 1, max 12), for the mildly affected animals (group 2) 6.2 years (SD  4.0; min 2, max 20), the moderately affected animals (group 3) 9.5 years (SD  4.6; min 1, max 20) and for the severely affected animals (group 4) 9.9 years (SD  3.4; min 4, max 17). Mean desmosine concentrations in the skin are shown in Figure 1 and summarized in Table 1. Preliminary analysis showed there to be a highly significant (p  0.006) interaction between location (neck vs. left forelimb) effects and clinical group effects. This interaction indicates that the magnitude of the difference in mean desmosine concentration for the forelimb and the neck region is not the same for each clinical group. Table 1 shows that the desmosine concentration in the distal limb was higher than in the neck region for all clinical groups except the group with severe lesions. Subsequent statistical analysis indicated that the most significant difference (p  0.0005) was that observed for the group with mild lesions. Other differences that were statistically significant were those observed for the normal affected animals (p  0.014) and for the group

700 Desmosine concentration pmD/mgP


Neck Fore limb

600 500 400 300 200 100 0 1


3 Group



FIG. 1. Desmosine concentration in the skin of neck and left forelimb of the different groups. 1  normal animals; 2  mildly affected animals; 3  moderately affected animals; 4  severely affected animals in the affected breeds; 5  control animals from a nonaffected draft horse breed.



2:52 PM

Page 70





Groupยง 1 2 3 4 5

Number of horses

Neck region desmosine

7 38 30 15 20

191 286 266 319 281

( ( ( ( (

Forelimb desmosine

94) 106) 80) 151) 99)

269 406 323 291 305

( ( ( ( (

51) 167) 125) 167) 81)


concentration in pmol/mg protein ( SD). 1  normal animals; group 2  mild lesions; group 3  moderate lesions; group 4  severe lesions in affected breeds; group 5  normal control nonaffected breed. ยงGroup

with moderate lesions (p  0.021). The differences observed for the group with severe lesions and the control group of a nonaffected breed were not statistically significant (p  0.35). When these clinical group-specific location effects were adjusted for age and gender, the results noted above remained the same with the single exception that the difference observed for the group with moderate lesions was no longer statistically significant (p  0.27). The interaction between location (neck vs. left forelimb) effects and clinical group effects also indicates that the difference in mean desmosine among the clinical groups is not the same for the forelimb and neck regions. Subsequent statistical analysis indicated there to be no statistically significant difference (p  0.15) among the five groups in mean desmosine concentration of the neck region both before and after adjusting for the effects of age and gender. In contrast, a highly significant difference (p  0.004) was observed among the five groups in mean desmosine concentration of the forelimb. Multiple pairwise comparisons of the means indicated that the mean desmosine concentration of group with mild lesions was significantly higher than the mean desmosine concentrations of group with severe lesions and the control group of a nonaffected breed; no other comparisons were statistically significant with a level of significance of 5% over all comparisons. The clinical group effect in the forelimb was no longer statistically significant (p  0.23) when adjusted for age and gender. When the desmosine concentration of the neck and limb was statistically compared for

the control group of a nonaffected breed (group 5) and the normal animals of affected breeds (group 1), a borderline nonsignificantly lower desmosine concentration was found in the skin from the normal affected breed (p  0.053). This difference became significant (p  0.045) when adjusted for age and gender.

CONCLUSIONS Lymph capillaries differ from blood capillaries in certain structural features. One of them is the presence of a small network of elastic fibers surrounding lymphatics.12 This network is believed to constitute an elastic device aiding propulsion of the lymph. Defective function or destruction of the elastic fibers prevents the transmission of tissue traction to the lymph capillary wall, reducing the reabsorption and propulsion of lymph and cells along the lymphatic. A consequence of this is edema and impaired immune response.7 In this study, differences in dermal elastin were noticed in the skin of draft horses affected with the condition known as chronic progressive lymphedema when compared to a nonaffected draft horse breed. Normal animals of the affected draft horse breeds had lower elastin concentrations in their skin when compared to a nonaffected draft horse breed. Interestingly, these changes were similar in the skin of the pastern and neck region, previously considered as clinically normal skin. This might suggest that the condition is a generalized skin disease rather than a dis-



2:52 PM

Page 71


ease affecting the skin of the distal limbs as previously thought. An inadequate amount of elastin support of the lymphatics and subsequent impaired function might explain the increased tendency to develop lymphedema. However, only the total dermal elastin was measured, which includes dermal and perilymphatic elastin. Therefore the elastin in the dermal, perilymphatic, or both regions might be affected. A possible genetic defect in skin elastic fibers has to be considered. An alternative etiology for the condition is a systemic factor causing early degradation of skin elastin. With development of the clinical signs of lymphedema, the Shires and Clydesdales had a significant increase in elastin levels within the skin of affected limbs which seemed to be correlated with age. This is not unexpected. Lymphedema is known to be a progressive disease.6 Every event that evokes local inflammation such as trauma, insect bites, and skin irritation induces local tissue edema and damages lymphatics and worsens clinical signs.6,14 These conditions commonly affect horse limbs. However, since it takes some time to develop the typical lesions, older animals are more likely to show worse clinical signs compared to young animals. The increase in dermal elastin is however in the form of disorganized elastin and therefore most likely nonfunctional, as was previously shown.1 Again, the skin elastin in the neck region increased when clinical signs worsened. The onset was slower and the increase less severe and not statistically significant. This is not unexpected in view of the higher stress on lymphatics in distal limbs compared to the neck region. This is especially true in draft horses because of their large size. This observation again might be indicative of a more generalized condition of the skin in these horses. The cause for the seemingly contradictory observation of decrease in skin elastin in draft horses with end stage lymphedema (group 4) can be explained by the marked increase in dermal collagen which causes a relative decrease in dermal elastin.1,2 This massive increase of fibrous tissue and marked deposition of collagen has also been documented in hu-


mans with late stage of chronic extremity lymphedema.6,15,16 Although this study shows some differences in skin elastin in the affected horse breeds, it is still unknown if this is a primary or secondary change. A genetic defect of elastic fibers, an early degradation of skin elastin or a combination thereof have to be considered as possible pathology in these horses. More research is definitively warranted to further characterize the nature of the elastin changes in CPL.

REFERENCES 1. De Cock HEV, Affolter VK, Wisner ER, Ferraro GL, MacLachlan NJ. Progressive swelling, hyperkeratosis, and fibrosis of distal limbs in Clydesdales, shires, and Belgian draft horses, suggestive of primary lymphedema. Lymph Res Biol 2003;3:191–199. 2. Ferraro G. Chronic progressive lymphedema in draft horses. Eq Vet Sci 2003;23:189–190. 3. Fields GS, Howard I, Stuart M. Elephantiasis nostras. A case report and review of the subject. J Am Podiatry Assoc 1976;66:28–31. 4. Richards RN. Verrucous and elephantoid lymphedema: morphologic spectrum and terminology. Int J Dermatol 1981;20:177–187. 5. Harwood CA, Mortimer PS. Causes and clinical manifestations of lymphatic failure. Clin Dermatol 1995;13:459–472. 6. Rockson SG. Lymphedema. Am J Med 2001;110:288– 295. 7. Gerli R, Ibba L, Fruschelli C. A fibrillar elastic apparatus around human lymph capillaries. Anat Embryol 1990;181:281–286. 8. Starcher BC. Determination of the elastin content of tissues by measuring desmosine and isodesmosine. Anal Biochem 1977;79:11–15. 9. Uitto J, Santa Cruz DJ, Starcher BC, Whyte MP, Murphy WA. Biochemical and ultrastructural demonstration of elastin accumulation in the skin lesions of the Buschke–Ollendorff syndrome. J Invest Dermatol 1981; 76:284–287. 10. Starcher BC, Mecham RP. Desmosine radioimmunoassay as a means of studying elastogenesis in cell cultures. Connec Tissue Res 1981;8:255–258. 11. Starcher B. A ninhydrin-based assay to quantitate the total protein content of tissue samples. Anal Biochem 2001;292:125–129. 12. Skobe M, Detmar M. Structure, function, and molecular control of the skin lymphatic system. J Invest Dermatol Symp Proc 2000;5:14–19. 13. Davidson JM, Zang MC, Zoia O, Giro G. Regulation of elastin synthesis in pathological states. The molecular biology and pathology of elastic tissues. Wiley,



2:52 PM

Page 72

72 Chichester (Ciba foundation Symposium 192) 1995; 85–99. 14. MacLaren JA. Skin changes in lymphoedema: pathophysiology and management options. Int J Palliat Nurs 2001;7:381–388. 15. Daroczy J. Pathology of lymphedema. Clin Dermatol 1995;13:433–444. 16. Vaccaro M, Borgia F, Guarneri F, Cannavo SP. Elephantiasis nostras verrucosa. Int J Dermatol 2000;39: 760–773.


Address reprint requests to: Dr. Hilde De Cock Department of Pathology, Microbiology, and Immunology University of California One Shields Ave Davis, CA 95616, USA E-mail:

This article has been cited by: 1. Francine Blei . 2006. Literature Watch. Lymphatic Research and Biology 4:3, 167-176. [Citation] [PDF] [PDF Plus]

Draft Horses With Chronic Progressive Lymphedema