Distribution of myofibroblasts, smooth muscle– like cells, macrophages, and mast cells in mitral valve leaflets of dogs with myxomatous mitral valve disease Richard I. Han, MScMed; Alexander Black, MSci; Geoff J. Culshaw, BVMS; Anne T. French, MVB; Roderick W. Else, BVSc, PhD; Brendan M. Corcoran, MVB, PhD

Objective—To map the cellular distribution and phenotypic alteration of the predominant stromal cell population throughout the entire valve length of dogs with myxomatous mitral valve disease (MMVD). Sample Population—31 mitral valve complexes (ie, mitral valve leaflets) collected from 4 clinically normal dogs and 27 dogs with MMVD of varying severity. Procedures—A combination of standard histologic and immunohistochemical techniques was used to identify pathologic changes, the presence of mast cells, and the density and distribution of cells expressing vimentin, desmin, α-smooth muscle actin (α-SMA), smooth muscle myosin, and the macrophage marker MAC387. Results—Vimentin-positive cells predominated in the mitral valve leaflets from clinically normal dogs and were located throughout the leaflet, but cell density was appreciably decreased with disease progression, and minimal cell numbers were found in distinct myxomatous areas. Cells that were positive for α-SMA were uncommon in the mitral valve leaflets from clinically normal dogs and only seen in appreciable numbers in mitral valves of dogs with severe late-stage disease, in which cells were typically located close to the ventricularis valve surface. A slight increase in mast cell numbers was observed in the distal zone of affected leaflets. Conclusions and Clinical Relevance—Activated-myofibroblasts (α-SMA–positive cells) were increased and inactive-myofibroblasts (vimentin-positive cells) were reduced in mitral valve leaflets of dogs with MMVD, compared with that of clinically normal dogs. Impact on Human Medicine—This is the first description of spatial and temporal alterations in mitral valve cells of any species with MMVD and has clinical importance in the understanding of disease development in dogs and humans. (Am J Vet Res 2008;69:763–769)

M

yxomatous mitral valve disease is the single most common acquired cardiovascular disease of dogs and is a major cause of morbidity and early death in this species.1,2 The disease is of substantial veterinary importance but is also of comparative interest as it bears close similarity to mitral valve prolapse in humans.3,4 The disease is particularly prevalent in certain breeds, has a strong age association, and is slowly progressive over a number of years.2,5 The disease pathologically is manifested as myxomatous degeneration of the valve leaflets. This is typified by destruction of the fibrosa layer of the leaflets, expansion of the loose connective tissue Received July 23, 2007. Accepted October 21, 2007. From the Hospital for Small Animals (Han, Culshaw, French, Corcoran) and Section of Pathology (Else), Division of Veterinary Clinical Sciences, the Royal (Dick) School of Veterinary Studies, University of Edinburgh, Roslin, Midlothian EH25 9RG, Scotland; and Department of Anatomy, National University of Ireland (Galway), Galway, Republic of Ireland (Black). Mr. Han is supported by a joint PhD studentship funded by the Cavalier King Charles Spaniel Club of England and the Kennel Club Charitable Trust, London. Address correspondence to Dr. Corcoran.

α-SMA MMVD

Abbreviations

α-Smooth muscle actin Myxomatous mitral valve disease

spongiosa layer, and excessive accumulation of acidic mucopolysaccharides.6 The localization of changes is mainly toward the free edge of the leaflets (distal zone) and can involve the midzone, but the base of the leaflet is typically disease free. Small nodules form at the valve edge, and as the disease progresses, these lesions coalesce, resulting in mechanically distorted leaflets that fail to coapt during ventricular systole. The underlying cause of this degenerative process is unknown but appears to be associated with phenotypic alteration in the valvular interstitial cell population.7–10 Evidence from studies11,12 in dogs suggests that this phenotypic alteration is crucial to the development of myxomatous degeneration, and coupled with this is alteration to the valve endocardial endothelium (dysfunction, degeneration, denuding, and detachment). With respect to the interstitial cell population, transformation from an inactive fibroblast phenotype to a more active myofibroblast type has been documented for

AJVR, Vol 69, No. 6, June 2008

763

dogs10 and humans.7 This is suspected to result in a derangement of dynamic valve matrix remodeling, which may be the result of overexpression of a range of matrix metalloproteinases and other catabolic enzymes.7,9 However, it may be that the activated myofibroblast is the predominant cell of unaffected mitral valves, particularly in neonates, or under the normal circumstances of tissue remodeling in response to changes in hemodynamics.8 In dogs with MMVD, evidence of phenotypic alteration from a fibroblast-like through an activated myofibroblast to a smooth muscle–like cell phenotype exists, and we have previously suggested that the latter cell type represents a final phenotypic cellular stage in dogs with MMVD.10 To date, the only data illustrating interstitial cell phenotypic alteration in dogs with MMVD have been at the ultrastructural level and this has been found only in small localized, obviously diseased areas of affected valve leaflets.10 This is also the case in humans with MMVD, where no data exist on regional or age-related development of pathologic changes. The purpose of the study reported here therefore was to map the cellular distribution and phenotypic alteration of the predominant stromal cell population throughout the entire valve length of dogs with MMVD. This was achieved by use of histologic and immunocytochemical techniques to identify cell phenotype on the basis of expression of standard cell markers. Furthermore, the relationship between disease severity (which has a close association with age in dogs), cell distribution, and cell immunophenotype was also investigated. Materials and Methods Tissue collection—Mitral valve leaflets were obtained from 4 clinically normal dogs and 27 dogs with MMVD euthanatized at the Hospital for Small Animals at the University of Edinburgh. All samples were obtained with full owner consent. Valves were collected from dogs with proven mitral valve disease (n = 13), clinically normal dogs that were euthanatized for noncardiac reasons (4), and dogs similarly euthanatized for noncardiac reasons and subsequently found to have mitral valve lesions typical of the disease (14). Dogs were euthanatized via IV administration of pentobarbital sodium.a The institutional animal care and use committee approved procedures. All valves were obtained and processed within 15 minutes after euthanasia. Gross and histologic examination of mitral valve leaflets—The presence or absence of mitral valve lesions was determined on gross examination of excised valves and graded 1 to 4 according to the Whitney criteria.13 A grade for each leaflet was determined by at least 2 authors (RIH, GJC, and BMC). No disagreement occurred between observers on grading. Disease was confirmed on histologic examination of H&E-stained sections and collagen distribution identified by use of Masson trichrome stain and van Gieson combined with Weigert’s method (Weigert-van Gieson) stain. Anterior (septal) and posterior (mural) mitral valve leaflets were separated to enable comparison between leaflets and immersion fixed in 4% paraformaldehyde for 24 hours and then wax-embedded for microtome 764



sectioning. Samples for sectioning were consistently obtained from the midpoint of each leaflet, avoiding the adjacent areas of chordal attachment. All sections were dewaxed in xylene and rehydrated through serial descending concentrations of ethanol. For routine histologic examination, 10-µm-thick sections were obtained longitudinally and so included the entire valve profile from the annulus to the valve leaflet tip. Sections were stained with H&E for routine assessment and Masson trichrome and Weigert-van Gieson stains for identification of connective tissue elements. For immunostaining, sections were cut at a thickness of 5 µm. To improve antigen retrieval, sections were microwaved in citrate buffer (0.01M; pH, 6.0) prior to application of antibodies. A variety of standard antibodies was used for the identification of cell types. A standard peroxidase methodb with diaminobenzidine as the chromagen was used for antibody detection. Sections were pretreated with 1% hydrogen peroxide in PBS solution to inhibit endogenous peroxidase activities. Monoclonal antibodies against vimentinc (80 mL/slide), desmind (1:40 dilution), α-SMAd (1:400 dilution), smooth muscle myosind (1:400 dilution), and the macrophage marker MAC387e (1:200 dilution) were applied to sections for immunohistochemical identification of cells. All sections were counterstained with hematoxylin. For the identification of mast cells, 5-µmthick sections were stained with acidic toluidine blue. For all types of staining, a single section was assessed. Cell density in this part of the study was subjectively (qualitatively) assessed in the basal, mid-, and distal zones of the leaflet (each zone approximating a third of the leaflet length) and in the atrialis, spongiosa, and fibrosa ventricularis layers, and compared between mitral valve leaflets from clinically normal dogs and from dogs with MMVD and among various grades of disease. A small number of cells stained positive for MAC387 or by use of toluidine blue. This allowed total cell numbers to be accurately counted; comparisons were made between grades, and the total cell numbers for both leaflets were combined. For further comparison

Figure 1—Photomicrographs of sequential sections of a mitral valve leaflet from a dog with grade 4 MMVD (A, H&E stain; B, Masson trichrome stain). Notice the marked myxomatous change in the distal third of the leaflet with loss of the fibrosa layer and connective tissue remnants (green). Bars = 1 mm. AJVR, Vol 69, No. 6, June 2008

of immunostaining results for vimentin and α-SMA, a semiquantitative scoring scheme of the cell density in a particular zone or layer was used: 0 = no cells stained positive, 1 = < 33% of cells stained positive, 2 = 33% to 67% of cells stained positive, and 3 = > 67% of cells stained positive. Because each zone and each layer could have a maximal score of 3, the combined score for a layer or zone could have a maximal score of 9. Results Gross and histologic findings—Anterior and posterior mitral valve leaflets were obtained from 31 dogs of 12 breeds and of varying ages. Only 1 dog was in heart failure prior to euthanasia and tissue specimen

collection. On gross examination, the valve leaflets were graded as follows: normal (n = 4) and grade 1 (6), grade 2 (8), grade 3 (8), and grade 4 (5) disease. When grouped according to age, disease severity (extent of visible lesions) increased with age. Looking at the age profile of the total cohort, all dogs ≥ 9 years old had evidence of disease and no dogs < 9 years old had grade 3 or grade 4 MMVD. For mitral valves from dogs with MMVD, pathologic changes were consistent with myxomatous degeneration in all leaflets examined. Changes were typically localized toward the distal zone of the valve leaflet, with expansion of the spongiosa layer and disruption of the fibrosa layer. The normal arrangement of fibrosa and spongiosa layers was difficult to recognize in the distal part of mitral valve leaflets from dogs with MMVD, particularly in mitral valves with grade 3 or 4 MMVD. Remnants of the fibrosa (positive Masson trichrome and Weigerts-van Gieson staining results for collagen) could be seen extending into the myxomatous areas of all mitral valve leaflets from dogs with MMVD, but the distribution was sparse in the mitral valves with more severe disease grades (Figure 1).

Figure 2—Photomicrographs of sections of a mitral valve leaflet from a dog with MMVD following immunostaining for vimentin in areas of unaffected (A) and affected (B) spongiosa layer. Notice the loss of immunostaining for vimentin in the myxomatous area. Bars = 10 µm.

Figure 3—Photomicrographs of sections of a mitral valve leaflet from a dog with MMVD following immunostaining for desmin in the unaffected spongiosa (A) and affected fibrosa (B) layers. Notice the grouping of chondrocyte-like cells in the affected fibrosa layer of the mitral valve leaflet. Bars = 10 µm.

Figure 4—Photomicrographs of sections of a mitral valve leaflet from a dog with MMVD following immunostaining for α-SMA in the unaffected midzone of the atrialis layer (A) and affected tissue adjacent to the distal zone of the ventricularis layer (B). Notice the differential expression of α-SMA toward the edge of the mitral valve leaflet. Bars = 10 µm (A) and 50 µm (B).

Cell identification and localization in mitral valve leaflets—Cells were identified in mitral valve leaflets from clinically normal dogs and from dogs with MMVD by evaluation of the expression of the antigens vimentin, desmin, α-SMA, smooth muscle myosin, and MAC387 on immunostaining and by staining with toluidine blue (mast cells). Comparing the anterior and posterior leaflets, no distinct differences were found in cell density and distribution, and the qualitative descriptions applied to both leaflets. Mitral valves of clinically normal dogs—Immunostaining for vimentin was extensive, and positive cells were visible in the fibrosa and spongiosa layers, being distributed evenly throughout the leaflets, although the greatest density was in the midzone, with somewhat fewer positive cells in the distal zone. At the free edge of the leaflet, fewer cells were visible, compared with other areas. Cells were visible between muscle bundles at the valve base but were somewhat sparse. Vimentin-positive cells were in close alignment with the direction of the collagen bundles in the fibrosa layer and were elongated and spindle shaped, whereas vimentin-negative cells were more rounded. The density of positive cells increased in the spongiosa and ventricularis layers toward the mid- and distal zones, whereas the proportion of positive cells in the atrialis layer remained the same for each zone.

AJVR, Vol 69, No. 6, June 2008

765

Immunostaining for desmin in mitral valve leaflets from clinically normal dogs was predominantly associated with cardiac myocytes in the basal zone, with negative immunostaining results for desmin in the distal zone and ventricularis layer. In the midzone, some immunostaining for desmin was observed, but desminpositive cells were sparse. Immunostaining for α-SMA was minimal in mitral valve leaflets from clinically normal dogs. Only occa-

sional α-SMA-positive cells were seen in the atrialis layer of the midzone. Smooth muscle myosin–positive cells were sparse in mitral valve leaflets from clinically normal dogs, with few cells identified. Smooth muscle myosin–positive cells were mainly seen in the mid- and distal zones and the atrialis and spongiosa layers but not in the ventricularis layer. Macrophage (MAC387-positive cells) numbers were low with only a mean of 6 cells/dog and were predominantly found in the spongiosa layer of the basal zone. Mast cell numbers were also low (mean of 5 cells/dog) and had a similar distribution to macrophages.

Mitral values of dogs with MMVD— Immunostaining for vimentin was still strong in mitral valve leaflets of dogs with MMVD, but a recognizable decrease in vimentin-positive cell density was found with increased disease severity. This reduction was most noticeable for all layers in all zones of grade 4 disease. Figure 5—Photomicrographs of sections of a mitral valve leaflet from a dog with In contrast to mitral valve leaflets from MMVD following immunostaining for MAC387 in the affected midzone of the spon- clinically normal dogs, vimentin-negagiosa layer (A) and toluidine blue staining for mast cells in the affected basal zone of the spongiosa layer (B). Notice the presence of macrophages and mast cells (at low tive cells with an interstitial cell spinnumbers) in mitral valve leaflets associated with MMVD. Bars = 10 µm. dle-shape were observed and these were typically identified in myxomatous areas, while in adjacent unaffected areas, the cells were positive for vimentin (Figure 2). Cells in myxomatous areas were often orientated in a wavy or circular pattern (whorl-like), but in the most severe form of the disease (grade 4), little evidence of vimentin-positive cells was observed in myxomatous areas. Overall, little observable variation was found between individuals within each grade of disease in vimentin-positive cell density, suggesting changes were consistent with disease severity and progression. For all grades of disease, a slight increase was found in the number of desmin-positive cells, with the appearance of immunostaining in the distal zone (Figure 3). However, desmin-positive cells were sparse in the distal zone and were only occasionally observed in myxomatous areas; overall, any changes were trivial. The number of α-SMA–positive cells increased with an increase in disease severity. α-Smooth muscle actin was differentially expressed in cardiomyocytes at the valve base. In early stage disease, immunostaining for α-SMA was mainly seen in the atrialis layer. In grade 3 disease, a distinct increase was found in the density of α-SMA–positive cells in the distal zone with the greatest increase occurring in the ventricularis layer, whereas in grade 4 disease, this increase was equally obvious in the midzone. However, α-SMA–positive cells were not distributed evenly throughout the tissue but formed agFigure 6—Semiquantitative (positive cells as a percentage of togregations as clusters and layers, typically located near tal cell numbers) comparison of changes in vimentin-positive (A) to the ventricularis side of the valve surface. This was and α-SMA–positive cells (B) in the 3 zones and 3 layers of mitral particularly so in the distal zone (Figure 4). Positive valve leaflets from clinically normal dogs (normal) and of mitral valve leaflets of dogs with grade 4 MMVD (grade 4). Scoring cells were mainly associated with dense connective tisscheme: 0 = no cells stained positive, 1 = < 33% of cells stained sue and were absent from the distinct myxomatous arpositive, 2 = 33% to 67% of cells stained positive, and 3 = > 67% eas. No appreciable change was found in expression of of cells stained positive. Because each zone and each layer could have a maximal score of 3, the combined score for a layer or zone smooth muscle myosin in mitral valve leaflets of dogs could have a maximal score of 9. Compared with normal values, with MMVD. notice the decrease in vimentin-positive cell density in all layers Macrophages were predominantly found in the and all zones for grade 4 disease and an increase in α-SMA–posisame locations in mitral valve leaflets from dogs with tive cells in all zones and all layers for grade 4 disease. 766



AJVR, Vol 69, No. 6, June 2008

MMVD as in mitral valve leaflets from clinically normal dogs. An increase in mean macrophage numbers was found from 6 cells in mitral valve leaflets from clinically normal dogs to 26 cells in mitral valve leaflets of dogs with grade 3 and 4 MMVD. However, this change was confined to the base of the valve. Mast cell numbers increased slightly from a mean of 5 cells in mitral valve leaflets from clinically normal dogs to 8 cells in mitral valve leaflets of dogs with grade 4 MMVD; although the change was small, the increase was mostly in the distal zone (Figure 5). Comparing changes in vimentin-positive and αSMA–positive cell numbers semiquantitatively, a decrease was found in vimentin-positive cells as a percentage of the total cell population with progression of disease, whereas the opposite was found with α-SMA– positive cells (Figure 6). Discussion In this study, we mapped the stromal cellular profile of the mitral valve of dogs, qualitatively illustrated the changes that occur with MMVD, and semiquantitatively determined changes in vimentin-positive and αSMA–positive cells. Furthermore, this is the first study in which the regional distribution of cell types in the mitral valve leaflet of dogs and how it might relate to disease progression has been examined. To date, studies in dogs10 and humans8 with forms of myxomatous degeneration have described cellular changes in localized areas of disease and have not addressed the issue of spatial organization of this change or change with disease development. Rabkin et al7 did describe, in a limited fashion, regional distribution of cells in myxomatous valve leaflets of humans, with α-SMA–positive cells localizing mainly to the base and midzones and the atrialis layer. We believe the description of the localization of cell types is important in improving our understanding of the pathologic process of this disease, as are changes in cell type, density, and distribution with age-related disease progression. In addition, results of this study suggest a potential role for mast cells in the development of mitral valve disease, as evidenced by increased cell numbers in the distal zone, but that macrophages are probably not implicated as any increase is restricted to the disease-free basal zone. The valvular interstitial cell population is recognized as having phenotypic plasticity, in that the cell phenotype can change from a quiescent fibroblast to an activated myofibroblast and eventually to a more smooth muscle–like cell phenotype.8,10,14 This cell plasticity seems to be crucial to heart valve development, remodeling, and repair and in fetal development, and in diseased states, the myofibroblast cell type predominates.8,14 Vimentin is widely used for the purpose of immunophenotyping interstitial cells, and it is expressed predominantly in mature cells. In mitral valves of dogs with MMVD, a mixed population of vimentin-positive and vimentin-negative cells was found, with a reduction in positive cell numbers in the most severe form of the disease in all zones and layers. This reduction in vimentin-positive cells agrees with previous reports7,10 for the human disease, but the overall reduction in cell density in myxomatous areas does not. The difficulty in

comparing our results with those of other studies7,10,15 is knowing the amount of the valve leaflet that was assessed, and either this has not been stated or the amount was small. In the current study, we have assessed the entire valve length for the first time. The nature of MMVD in dogs is that the pathologic change is focal in distribution, and varying grades of pathologic severity of the disease are described.13 A pathologic grading system is lacking for the human form of the disease, so it is difficult to interpret what level of disease has been described in previous human reports. Nevertheless, our own impression in dogs is that cellular changes in visibly pathologic areas of mitral valves are reasonably consistent irrespective of severity but a variation in change is found across the entire valve length and width. We have not been able to identify the immunophenotype of the vimentin-negative cells in myxomatous areas. These cells were also α-SMA negative, suggesting they are not the typical activated myofibroblasts seen with myxomatous degeneration.7,8,16 They do have an elongated spindle interstitial cell shape that contrasted with the more rounded vimentin-negative cells seen in the spongiosa layer of unaffected mitral valves. Recently, Barth et al15 have identified the predominant cells in unaffected and affected mitral valves of humans to be CD34+ fibrocytes. These cells are known to have a functional myofibroblast phenotype producing matrix metalloproteinase-9 and collagen I and III. The increase in α-SMA–positive cells in this study was expected as this has been reported in studies7,9 from human patients. α-Smooth muscle actin is a cytoskeletal microfilament and can be used to distinguish smooth muscle cells from fibroblasts, but has been mainly used in valve studies to identify activated-myofibroblasts. Findings in an electron microscopy study10 in dogs with MMVD revealed a population of smooth muscle–like cells and an increase in these cells in the most severe forms of the disease.10 It has been postulated the phenotypic change from a predominantly fibroblast to myofibroblast phenotype cell population and eventually to a more smooth muscle phenotype might represent the final phenotypic alteration as the disease enters the terminal stage in dogs.10 What is interesting in dogs is the localization of α-SMA–positive cells to the edge of the valve, with few cells visible in the overtly myxomatous areas, and this contrasts with that reported for human myxomatous degeneration.7 The consensus is that α-SMA–positive cells represent an activated myofibroblast population that are involved in extracellular matrix remodeling, but some disagreement exists as to whether such a phenotypic expression is associated only with disease.7,8,17,18 It would be reasonable to presume activated myofibroblasts contribute to normal tissue remodeling and also to damage. Our data from dogs would suggest a more quiescent vimentin-positive fibroblast population predominates in the mitral valve of clinically normal dogs and only in disease is there an obvious increase in the activated myofibroblast-smooth muscle–like cell population. This agrees with reports of human disease8 and animal models19,20 of valve damage repair, but it is the distinct localization of the α-SMA– positive cells in dogs that is different. The transition from fibroblast to myofibroblast would appear to be un-

AJVR, Vol 69, No. 6, June 2008

767

der the control of tumor growth factor-β1, and in cell culture systems at least, this results in the development of α-SMA–positive cells with enhanced contractility.20 Results of a recent study21 on genomic expression in dogs with MMVD has revealed, among numerous other changes, upregulation of the 5-hydroxytryptamine2-B receptor gene, and the authors speculated that a serotonin-transforming growth factor-β mediated activation of myofibroblasts might be implicated in disease pathogenesis. The serotonin signal could come from platelets adhering to damaged endothelium, and endothelial denuding has been known to occur in dogs with MMVD, and it can be further speculated that the localization of α-SMA–positive cells to the subendothelial layers may be in response to such signaling.10,11 Indeed, it has long been recognized that in in vitro models of valve injury interstitial cells proliferate and migrate toward the valve surface, and these cells express α-SMA.19,20 Furthermore, in such models, there appears to be a preferential accumulation of α-SMA–positive cells on the ventricularis side of the leaflet.19 The change in distribution of α-SMA– positive cells in our study was consistent and related to disease severity, but what is surprising is that we did not see an increase in α-SMA–positive cells in the overtly myxomatous areas. This increase in αSMA–positive cells roughly matched the decrease in vimentin-positive cells observed with increased disease severity but not at the same sites. This has also been reported21 for myxomatous mitral valves of humans where CD34 has been used as a marker of myofibroblasts, although exact comparison of complementary sites was not reported. The importance of the initial increase in α-SMA–positive cells adjacent to the atrialis is unknown. The clustering of cells and their organization close to the valve edge in the distal zone may reflect an attempt to compensate for the reduced mechanical integrity of the valve. Walker et al17 suggest this augmented contractile myofibroblast response may be a precursor to the alteration in valve matrix seen with myxomatous degeneration, but data from the current study and previous electron microscopy data from dogs suggest it might be a later consequence of the disease process.10 The role of other cell types in the etiopathogenesis of this disease is worth considering. No evidence exists that MMVD involves an inflammatory process, and this would appear also to be the case in human myxomatous degeneration.7 Our study provided evidence of a possible role of mast cells in the development of MMVD, but the cellular response is small and could be argued to be trivial. The number of mast cells found in mitral valves of humans is also small, but because an association between mast cells, mast cell chymases, matrix metalloproteinase activity, and extracellular matrix remodeling is recognized in the myocardium, the existence of similar mechanisms in MMVD is worth considering.22 In contrast, no evidence exists of direct involvement of the resident macrophage population in the development of MMVD, at least at the sites where disease develops, and this agrees with findings in human mitral valve disease.7 Data presented here are a comprehensive description of the cellular changes in mitral valves of dogs with 768



MMVD, and although we have found changes in cell populations in affected valves, it is the information on temporal and spatial development of these changes that may be the most important in the understanding of the disease in dogs and humans. It is possible that the increase in activated myofibroblast numbers in MMVD is a reaction to the disease process rather than a cause, and indeed, in humans with MMVD, variation in the reports7,9 of the degree of matrix metalloproteinase expression is found, with some human patients having no expression. a. b. c. d. e.

Euthatal, Merial Animal Health, Harlow, England. Vector ABC Elite Kit, Vector Laboratories Inc, Burlington, Calif. Euro/DPC Ltd, Gwynedd, Wales. Sigma-Aldrich Co Ltd, Poole, England. Dako Cytomation Denmark A/S, Glostrup, Denmark.

References 1. 2. 3. 4. 5. 6. 7.

8.

9. 10.

11.

12. 13. 14.

15. 16. 17.

Darke PG. Valvular incompetence in cavalier King Charles spaniels. Vet Rec 1987;120:365–366. Beardow AW, Buchanan JW. Chronic mitral valve disease in cavalier King Charles spaniels: 95 cases (1987–1991). J Am Vet Med Assoc 1993;203:1023–1029. Pomerance A, Whitney JC. Heart valve changes common to man and dog: a comparative study. Cardiovasc Res 1970;4:61–66. Pedersen HD, Haggstrom J. Mitral valve prolapse in the dog: a model of mitral valve prolapse in man. Cardiovasc Res 2000;47:234–243. Thrusfield MV, Aitken CGG, Darke PGG. Observations on breed and sex in relation to canine heart valve incompetence. J Small Anim Pract 1985;26:709–717. Buchanan JW. Chronic valvular disease (endocardiosis) in dogs. Adv Vet Sci Comp Med 1977;21:75–106. Rabkin E, Aikawa M, Stone JR, et al. Activated interstitial myofibroblasts express catabolic enzymes and mediate matrix remodeling in myxomatous heart valves. Circulation 2001;104:2525– 2532. Rabkin-Aikawa E, Farber M, Aikawa M, et al. Dynamic and reversible changes of interstitial cell phenotype during remodeling of cardiac valves. J Heart Valve Dis 2004;13:841– 847. Soini Y, Satta J, Maatta M, et al. Expression of MMP2, MMP9, MT1-MMP, TIMP1, and TIMP2 mRNA in valvular lesions of the heart. J Pathol 2001;194:225–231. Black A, French AT, Dukes-McEwan J, et al. Ultrastructural morphologic evaluation of the phenotype of valvular interstitial cells in dogs with myxomatous degeneration of the mitral valve. Am J Vet Res 2005;66:1408–1414. Corcoran BM, Black A, Anderson H, et al. Identification of surface morphologic changes in the mitral valve leaflets and chordae tendineae of dogs with myxomatous degeneration. Am J Vet Res 2004;65:198–206. Olsen LH, Mortensen K, Martinussen T, et al. Increased NADPHdiaphorase activity in canine myxomatous mitral valve leaflets. J Comp Pathol 2003;129:120–130. Whitney JC. Observations on the effect of age on the severity of heart valve lesions in the dog. J Small Anim Pract 1974;15:511– 522. Merryman WD, Youn I, Lukoff HD, et al. Correlation between heart valve interstitial cell stiffness and transvalvular pressure: implications for collagen biosynthesis. Am J Physiol Heart Circ Physiol 2005;290:H224–H231. Barth PJ, Koster H, Moosdorf R. CD34+ fibrocytes in normal mitral valves and myxomatous mitral valve degeneration. Pathol Res Pract 2005;201:301–304. Rabkin-Aikawa E, Aikawa M, Farber M, et al. Clinical pulmonary autograft valves: pathologic evidence of adaptive remodeling in the aortic site. J Thorac Cardiovasc Surg 2004;128:552–561. Walker GA, Masters KS, Shah DN, et al. Valvular myofibroblast activation by transforming growth factor-beta: implications for AJVR, Vol 69, No. 6, June 2008

pathological extracellular matrix remodeling in heart valve disease. Circ Res 2004;95:253–260. 18. Taylor PM, Batten P, Brand NJ, et al. The cardiac valve interstitial cell. Int J Biochem Cell Biol 2003;35:113–118. 19. Lester WM, Damji AA, Gedeon I, et al. Interstitial cells from the atrial and ventricular sides of the bovine mitral valve respond differently to denuding endocardial injury. In Vitro Cell Dev Biol 1993;29A:41–50. 20. Lester WM, Damji AA, Tanaka M, et al. Bovine mitral valve or-

gan culture: role of interstitial cells in repair of valvular injury. J Mol Cell Cardiol 1992;24:43–53. 21. Oyama MA, Chittur SV. Genomic expression patterns of mitral valve tissues from dogs with degenerative mitral valve disease. Am J Vet Res 2006;67:1307–1318. 22. Veinot JP, Prichett-Pejic W, Song J, et al. CD117-positive cells and mast cells in adult human cardiac valves—observations and implications for the creation of bioengineered grafts. Cardiovasc Pathol 2006;15:36–40.

AJVR, Vol 69, No. 6, June 2008

769