Excessive degradation of type II collagen in articular cartilage in equine osteochondrosis

ELSEVIER

Journal of Orthopaedic Research

Journal of Orthopaedic Research 20 (2002) 1282-1289

www.elsevier.com/locate/orthres

Excessive degradation of type I1 collagen in articular cartilage in equine osteochondrosis S. Laverty

‘,

S. O’Kouneff ‘, M. Ionescu b, A. Reiner I. Pidoux Y. Rossier a, R.C. Billinghurst A.R. Poole

‘>’,

b,

C. Webber ’,

Faculk de M4decine Ve!t&inaiw, Dipartrmeni tie Sciences Cliiziques, Unii;ersitt dr Moniiicd, CP 5000, Suint HJvxinilie. QuC, Canada J2S 7C6 Dicisiori vf’ Surgical Rescwch, Departments vf Surgrry and Mdicine, .Joint Diseases Luhorutory, S h r i n m Hospital.7 for Cliilclren. MrGill Uniuersity, Montreul, Q u ~ . Canada , H3G IA6 ”

Received 9 March 2002; accepted 14 March 2002

Abstract Articular osteochondrosis (OCD) occurs in both man and animals. The etiology remains to be determined. Studies of OCD lesions in animals may provide clues as to its pathogenesis. The aim of our study was to determine whether there was evidence for increased degradation namely proteoglycan (PG) release and type I1 collagen cleavage in articular cartilage harvested from O C D lesions. We examined ex vivo explants at post-mortem from equine OCD lesions and macroscopically normal site and age matched cartilage. These were cultured over a 10 day period in serum-free medium. Type I1 collagen cleavage was measured in articular cartilage and media using an Elisa assay to detect the coL2-3/4c,h,r, epitope, which is generated on cleavage of the triple helix of type I1 collagen by collagenases. PC release was measured by a dye-binding assay. Cumulative release of P C and COL2-3/4CShor, and their contents in cartilage at the end of the culture period were determined. In O C D lesions there was a significant increase in type I1 collagen cleavage by collagenase but no evidence for increase of PG degradation. These findings point to a selective increase in type 11 collagen cleavage by collagenases, in O C D lesions of the kind observed in osteoarthritis. Further work is needed to determine whether changes represent primary or secondary events in the pathogenesis of OCD. 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved. Ke)word,r: Cartilage; Osteochondrosis; Horse; Equine; Collagen; Joint disease

Introduction

Articular osteochondrosis (OCD) is characterized by a focal failure of endochondral ossification in the growing epiphysis. It affects both humans [40,43] and animals [10,19,29]. A cartilaginous flap may develop and detach creating “loose bodies” in the affected joint [24,33]. OCD may also result in incongruency of the articular surface which may eventually lead to degenerative arthritis [29,40]. The overall pathology of OCD is strikingly similar in all 6 species where it’s presence has been reported [30]. * Corresponding author. Tel.: +I-450-778-8100; fax: +1-450-7788102. E-niuil uddrcss: sheila.laverty@umontreal,ca ( S . Laverty). Present address: Department of Clinical Sciences, Colorado State University, Fort, Collins, Co, USA.

’

There are many similarities between OCD in humans and animals: these include its occurrence during growth, the bilateral nature of the condition and it’s capacity to involve multiple sites [29,41,43]. Pathologic material for study from OCD lesions in humans has been scarce [7,41] and has been studied only in the chronic stages of disease. Studies of lesions at an earlier stage of disease in animals may provide insight into the etiology andlor pathophysiology of OCD. Among all species, lesions have been found to be remarkably similar. This strongly suggests that the pathophysiology of this condition is similar in different species [10,30]. Because of the similarities between OCD in horses and humans, and the availability of equine tissue for analysis, pathophysiological studies of naturally occurring OCD in horses are therefore relevant to the further understanding of the disease in man. OCD occurs most commonly in the knees of adolescents and

0736-0266/021$ - see front matter 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved PII: S 0 7 3 6 - 0 2 6 6 ( 02)000 5 3 - 0

S. Luverty et al. I Journal of' Ortlzopuedic Reseurch 20 (2002) 1282 1289

young adults [31,33] and lesions are frequently observed on the trochleas of the femur, the site evaluated in this study, in young growing horses [24]. Although the exact pathogenesis of OCD remains to be elucidated, trauma, focally deficient blood supply, genetic, endocrinologic and nutritional factors have all been proposed as possible causes of OCD in humans [31,33,43,46] and horses [8, 14,18,19,32,36,42]. In animals OCD is defined as a disturbance of cell differentiation in joint cartilage leading to altered endochondral ossification [19,29,43]. Key cellular events in the normal developing epiphysis include chondrocyte proliferation, the creation of an extensive extracellular matrix, chondrocyte differentiation (hypertrophy), vascular invasion and calcification of the matrix. The early lesion of OCD develops as a small retained core of cartilage extending into the subchondral bone. There is a disruption in the normal sequential transition of the chondrocytes through proliferation and maturation resulting in the accumulation of large numbers of small rounded chondrocytes apparently arrested in the prehypertrophic stage within the retained cartilage [ 191. Matrix calcification or vascular penetration of the affected cartilage does not occur [ 10,301. Secondary events such as necrosis of thickened cartilage give rise to fissures and flaps and clinical disease. Primary molecular changes in the pericellular matrix of cartilage have been hypothesized as a cause of OCD in both humans [43] and animals [19]. Cartilage extracellular matrix alterations have been identified in naturally occurring OCD in the horse [22]. These included a loss of glycosaminoglycans, including chondroitin sulfate in cartilage from OCD lesions when compared to normal [22]. Proteoglycan (PG) production was decreased and there was an increased turnover of newly synthesized PGs in explant cultures of cartilage from equine OCD lesions in a similar age group as our study [471. Type I1 collagen alterations in OCD cartilage have also been reported. In studies of a copper deficient model of OCD in horses [6,17] type I1 collagen solubility was increased [6] and the number of collagen cross links in the articular cartilage of OCD affected foals was reduced [17]. A decrease in the total type 11 collagen content and reduced levels of type I1 collagen cross links were noted in the articular cartilage from naturally occurring OCD lesions in pigs [48]. Recently we reported an increase, in the synovial fluid, of CPII, a specific marker of articular cartilage type I1 collagen synthesis [28] in young horses with naturally occurring tarsocrural OCD [21]. Together these observations point to an increase in type I1 collagen turnover in this disease. Studies of cartilage type I1 collagen degradation in OCD lesions are lacking. Using recently established methods to measure type I1 degradation [3,15,16], we investigated cartilage type I1 collagen breakdown ex vivo

I283

in articular cartilage explants from lesions of OCD and from normal cartilage of young horses, We show that unlike in the case of P G (principally aggrecan), type I1 collagen undergoes increased cleavage by collagenases, in lesions, suggesting that this increased proteolysis accounts in part for some of the molecular changes that have been observed in this molecule in the development of OCD.

Methods Anirnuls

Young horses, 4 females and 2 males, with a radiographic diagnosis of osteochondritis dissecans (OCD) of the femoropatellar joint had a mean age of 8.9 months (range 7-12 months). All horses included in the study had grade 3 O C D lesions (>4 cm length on radiographs) [12]. A representative radiograph of a lesion is shown in Fig. 1. A large defect is present in the ridge of the trochlea of the femur, containing islands of mineralized cartilage. Three horses, 2 males and I female, euthanised for reasons unrelated to ~nusculoskeletaldisease and with macroscopically normal femoropatellar joints, were also studied. They were 4 and 8 months (2 horses) old. Articulur curtilage Femoral trochlear ridge full depth articular cartilage slices were harvested, within 2 h of euthanasia. Only one joint per animal was examined. A schematic drawing of each joint was made indicating the site of the lesion and where the articular cartilage was harvested. In horses with OCD, cartilage was harvested from the OCD lesion on the lateral trochlear ridge (lesion) and the opposing proximal medial trochlear ridge wherever this appeared macroscopically normal (nonlesion). Samples were prepared from similar sites on both trochlear ridges in the normal horses (controls) (Fig. 2). H&E stained sections were prepared of osteochondral samples and a representative sample is shown in Fig. 3. Curiiluge explant preparation Cartilage was harvested aseptically and transported to the laboratory under sterile conditions in Dulbecco's modified eagle's medium (DMEM) (Cibco BRL, Life Technologies, Grand Island, NY). The cartilage was cut into 2 x 2 mm pieces and placed in culture wells (24well falcon 3047 plate; Becton Dickenson, Mountain View Ca). Before culture, the cartilage samples were washed for 10 min in D M E M with 2.5 mg/ml amphotericin B (fungizone) followed by amphotericin B plus 10 times the normal concentrations of penicillin and streptomycin (see below), before being washed with standard serum-free culture medium. The culture medium was DMEM with penicillin (100 U/ml), streptomycin (100 pglml), 50 mM hepes buffer, pH 7.4, bovine serum albumin (100 &nil), fresh ascorbic acid (50 pgi ml) and 5 pglml insulin, 5 pglml transferrin, and 5 n d m l sodium selenite (ITS Supplement, Boehringer Mannheim, Indianapolis, IN). They were then cultured in the presence of 5% of COz in air at 37 "C. The media were changed every 2 days up to and including 10 days and stored at -20 "C until assayed. The mean total tissue wet weight per well at the time of harvest of'the cartilage was recorded after removal of excess fluid and the cartilage was stored at -70 "C until further analyses. Quadruplicate wells were ordinarily analyzed for each experimental point. Axsay for content of collagenase-cleaved type I I cotlugen in nzerlin

The media were concentrated approximately threefold by lyophilization and then resuspended in H 2 0 and analyzed by immunoassay for the COL2-314C,,,,,, [3,4]. Cumulative release was expressed as the

I284

S. Lucerty et ul I Joumul of Orthopuetlic Reseurclz 20 (2002) 1282-1289

Normal

OCD

Fig. I . Lateroinedial radiographic views of equine femoropatellar joints. The normal radiographic view (left) illustrates the smooth contour of the ridge of the lateral trochlea of the femur (white arrows). An O C D lesion is evident o n the radiograph on the right (between white arrows). A defect is present in the ridge of the trochlea, containing islands of mineralized cartilage (white arrow head), causing joint incongruency. Large radiolucent areas are seen within the subchondral bone (black arrows).

Drlerminution oj protcwglyctni i'onient in mediu und curtiluge The culture media and both the r-chymotrypsin and proteinase K cartilage digests were analyzed for the PG content (predominantly aggrecan), using a modification of the colorimetric 1.9-dimethylmethylene bluc (DMMB) dye assay [Ill, released into the culture inedinm and present in the cartilage. PG content was also expressed as pdpg DNA of cartilage. Determination of D N A contcnt in the curtiluge DNA content of the articular cartilage was measured fluorimetrically in the proteinase K digests as described previously [20]. The DNA content was expressed as pg DNAlmg wet weight. Stutisticul ~ i n u l y ~ ~ ~ . ~

Fig. 2. Schematic representation of the femur of 3 horse with O C D demonstrating the SI site on the lateral trochlear ridge where the O C D lesions were sampled in all horses in this study. The S2 site is on the medial trochlear ridge of the femur and was classified as a remote normal cartilage sample. Similar sites were sampled in unaffected normal horses.

Box plots were prepared to show analyses where the upper and lower margins represent 75% and 25% percentiles respectively. The upper and lower whiskers represent 90'% and 10% limits respectively: outliers are shown as symbols. Medians are indicated within the boxes. A paired t test was performed between the lesion cartilage and normal appearing cartilage remote from the lesion (non-lesion) to determine differences. A p value of p < 0.05 was considered significant.

Results Histology

total of all culture media measurements at each time point. The contents of COL2-3/4C,,,,,, epitope were expressed as pmollpg DNA. As.rci~.(fcurtiluge / O r c~~llugctiii.rr-i~ki~ureu' epitoipe ( COL2-314C,,,,,,) rlrnuturrd type I1 collugm epitope (COL2-3/4,,,) unil totcil typi' I1 c,olltrgm content The methods used to extract the COL2-314C,ho,, [3,4]and COL2-314, [I 51 epitopes from cartilage were described previously. Briefly, the cartilage was digested with 9-chymotrypsin overnight at 37 "C and extracts were analyzed for both epitopes. The undigested cartilage was then incubated overnight at 56 "C with proteinase K. The contents of each epitope were expressed as pinole of peptidelpg D N A of cartilage. Total type 11 collagen content was determined from the content of COL2-314,,, epitope in both the chyniotrypsin and proteinase K digests [15].

Histological analyses were performed to assess histopathology. Characteristic features included the presence of cartilaginous flaps and retained cartilage in the underlying bone causing disruption of normal architecture. A representative section is shown in Fig. 3. This shows a cartilage flap, which has detached from the subchondral bone. The cartilage is generally intact except for some erosion at the cartilage bone interface. The subchondral bone exhibits considerable pathological change compared to the adjacent normal bone under

S. Luoerfy et ul. I Jouinnl of' Orthopuedic Rescurch 20 (2002) 1282-1289

1255

Fig. 3. Sections of cartilage-bone from the ridge of the lateral trochlea of the femur of horses from the study. The section on the left demonstrates normal articular cartilage, which is attached to the underlying subchondral bone. The H&E section on the right demonstrates a focal osteochondritis dissecans lesion. There is an area of architecturally normal cartilage (NC) and bone (B). Adjacent to this a cartilaginous flap (CF) is detached from the underlying bone. Focal areas of retained, degenerate cartilage (RC) are evident within a large lesion in the subchondral bone.

the structurally normal looking cartilage which is still attached. m

D N A content of the articular cartilage n

As in earlier work [47] there were no significant differences between the DNA content of lesion cartilage when compared to non-lesion cartilage within the same joint (Fig. 4). For this reason we expressed the data per DNA. Similar results were obtained when the data were expressed per wet weight of cartilage (data not shown). Collagenase-cleaved type II collagen release into culture media The amount of the COL2-3/4C,,,c,,tepitope released over time varied considerably within samples removed from a lesion (comparison of different wells) and between lesions. The amount released from lesions was

1

0.75

E .-m m

T

0.50

0.00

'

si Lesion

52 Non-lesion

U OCD Joints

-

s1 Control

s2 Control

g

40-

0 0

30 20-

O

10-

0

E

-

s1 Lesion

s2

Non-lesion

OCD Joints

* s1 Control

s2 Control

u

Control Joints

Fig. 5. Cumulative release of cleaved type I1 collagen into the media following 10 days of culture. The amount of COL2-3/4C,h,,, is expressed as pmol/pg DNA. Cartilage explants were harvested from the O C D lesion (SI), from a remote normal site within the O C D affected joint (S2) and from joints of age matched horses without O C D (normal S1 and S2 sites). A paired / test was performed on results from media analyses from the OCD lesion and remote normal cartilage explants.

however significantly greater than that released from non-lesion and control cartilages (Fig. 5). Release of C0L2-3/4Csh,,, into the media progressively increased over time in OCD cartilage cultures but not in non-lesion or control cartilage (not shown). The release of COL2-3/4C,ho,l epitope from non-lesion cartilage and cartilage from horses with normal joints (controls) were similar in amount. They were not compared statistically because of small numbers of horses available in the control group.

Control Joints

Fig. 4. D N A per wet weight in cultured cartilage. Results are expressed as pg DNA/mg wet weight of cartilage. CartIlage was sampled from the O C D lesion (SI), from a remote normal site within the O C D affected joint (S2) and from joints of age matched horses without O C D (normal S1 and S2 sites). A paired t test was performed on results from the OCD lesion and remote normal cartilage.

Cleavage in cartilage There was greater variation in the content of the COL2-3/4Csh0,.,epitope present in OCD lesions cartilage compared to no11 lesion and normal (controls) cartilages. There were however, no significant differences in

S. Laoerty et al. I Jourrial of Orthopaedic Research 20 (2002) 1282-1289

1286

1000

-

90 5j!

750-

5m a

=8

-

500-

b

250-

m

a

0' s1 Lesion

s2

Non-lesion

u OCD Joints

s1 Control

s2

I,~,,, Fig. 6. The amount of cleaved type I 1 collagen ( C O L ~ - ~ / ~ C ,assay) in cartilage is represented as pmollDNA. Cartilage was obtained from the OCD lesion (S1) from a remote normal site within the OCD affected joint (S2) and from joints of age matched horses without OCD (normal SI and S 2 sites). A paired f test was performed on results of cartilage analyses from the OCD lesion and remote normal cartilage explants.

content unlike the differences in release (Fig. 6). A higher proportion of specimens from lesions indicated increased collagenase activity when compared to nonlesion.

Proteoglycaii release into media PG release into the media was initially high at 2 days from lesion, non-lesion and control cartilage but decreased in the majority of wells at day 4 and remained below the levels released at day 2 in all cartilage samples (data not shown). The cumulative release of PG into the media after 10 days of culture was significantly greater (p = 0.04) from the non-lesion cartilage when compared to cartilage harvested from the OCD lesion in the same joint (Fig. 8). Overall, the cumulative GAG release from the cartilage of normal horses was similar to that seen with the lesion, though not compared statistically because of small numbers of horses in the group. Proteoglycan in cartilage There was no significant difference in the GAG content between the cartilage from the OCD lesion and the

s1

Control

s2 Control

U Control Joints

Fig. 7. The total type 11 collagen was calculated from the assays of COL2-3/4,,, and COL2-3/4C,1,,,,. Cartilage explants were harvested from the OCD lesion (Sl) froin a remote normal site within the OCD affected joint (S2) and from joints of age matched horses without OCD (normal S1 and S2 sites). A paired t test was performed on results of cartilage analyses from the OCD lesion and remote normal cartilage explants.

300 1

$

200-

m

J

$ 100SL

0'

Total type I1 collugen content in cartilage There were no significant differences in the total amount of type I1 collagen in OCD cartilage or compared to non OCD cartilages. The degree of variation observed was also similar in both (Fig. 7).

s2

Non-lesion

OCD Joints

n

Type I1 collagen denaturation in cartiluge There was no significant difference in the content of COL2-3/4, epitope in cartilage cultured from the OCD lesions and non-lesion cartilage. The amount of variation in content was also similar for both (not shown).

s1 Lesion

u

Control

U Control Joints

T 4T T T

i4 €5 'is1

Lesion

52 Non-lesion

U OCD Joints

s1 Control

s2 Control

U Control Joints

Fig. 8. Cumulative release of GAG into the media during 10 days of culture. Results are expressed as pg GAGlpg DNA. Cartilage was sampled from the OCD lesion (SI) from a remote normal site within the OCD affected joint (S2) and from joints of age matched horses without OCD (normal S1 and S 2 sites). A paired t test was performed on results from media analyses from the OCD lesion and remote normal cartilage explants.

remote normal lesion within the same joint. There was a greater overall variation in the GAG content from the OCD lesion cartilage (Fig. 9).

Discussion

The etiology and pathophysiology of OCD which is associated with growth in both humans and animals, remains to be clearly elucidated. Earlier studies have provided indirect evidence for abnormalities in the turnover of type I1 collagen [6,17,21]. In this, the first

120,

,- ,

S Lurrstj) et al I Journal of Orthopedic Resenrch 20 (2002) 1282-1289

~=0.0411

1004

401

z! 20

@ s1 Lesion

s2 Non-lesion

u OCD Joints

-

s1 Control

52 Control

Control Joints

Fig. 9. PG content in cartilage. Results are expressed as pg GAGlpg DNA. Cartilage was sampled from the OCD lesion (Sl) from a remote normal site within the OCD affected joint (S2) and from joints of age matched horses without OCD (normal SI and S2 sites). A paired / test was performed on results of cartilage analyses from the OCD lesion and remote normal cartilage explants.

study, to our knowledge, of type I1 collagen degradation in OCD we clearly show that there is an increase in the cleavage of this molecule by collagenases in OCD lesions. This cleavage of collagen and anticipated associated synthesis, in view of the lack of loss of type I1 collagen would explain the earlier observations of a reduction in collagen cross-links [ 171, the formation of which are age dependant and the increase solubility of the collagens [6],observations which point to the active synthesis of new collagen replacing that which is lost. Collagenases cleave the intact triple helix of type I1 collagen into 3/4 and 114 fragments, leading to degeneration of the articular cartilage. The antibody COL2-3/ 4Cshortthat we used here only recognizes the carboxy terminal neoepitope of this collagenase-specific cleavage site and permits the detection of this specific cleavage by collagenases [3]. Our results also indicate that there is no evidence for an increase in PG degradation whereas we observed an increase in type I1 collagen cleavage by collagenases in articular cartilage from OCD lesions in young growing horses when compared to normal cartilage. In skeletal tissue, collagenases play an important role in the growth cartilage of the normal developing epiphysis [2,44,50] and also in articular cartilage degeneration in osteoarthritis [9]. In human OA there is an increase in collagenase mRNA in OA chondrocytes [13,25,39], together with increased type I1 collagen degradation [ 15,161 around human chondrocytes which is caused by collagenases IS]. In cultured explants of OA cartilage [3,9] there is a selective increase in cleavage reflected by an increase in the cleavage neoepitope [9]. Again as observed here and as observed by others [47] in OCD, we observed no increase in OA cartilage PG release.

1287

Although it is possible, though not considered likely, that PG release was maximal in all the OCD and human samples studied. Thus our findings reveal that increased collagenase activity is a feature of OCD lesions that is shared with degenerate articular cartilage in advanced human osteoarthritis studied at arthroplasty. Thus as in osteoarthritis, collagenases may play a rolc in the development of pathology of OCD lesions and may also be part of a final common pathway in the pathology of OCD and OA. Comparisons were not made in the horse between OCD and OA lesions since equine OA lesions are usually encountered on the femoral condyles, as in human disease. In the horse it is on the trochlear ridges that we commonly see evidence of OCD. Tissue would havc therefore been from different sites. Moreover we had considerable difficulty in obtaining material for this study. It would have been extremely difficult to obtain sufficient OA material at this same site if not impossible. An integral part of the process of normal endochondral ossification involves rapid selective degradation of type I1 collagen and resorption and remodelling of the extracellular matrix in the growth plate [27]. Type I1 collagen is extensively cleaved and denatured and lost in the hypertrophic zone [ 1,491 as a result of collagenase 3 activity (MMP-13) [26]. Any alteration in type I1 collagen loss can inhibit normal chondrocyte maturation [49] and this loss of type I1 collagen in development is also seen in OA cartilages [15] again in association with excessive collagenase activity [3]. In contrast to human OA and normal endochondral remodelling in the growth plate, there is no loss of type I1 collagen in OCD lesions associated with increased cleavage. This may relate to the active synthesis of the new type I1 collagen molecules associated with growth. Although type I1 collagen synthesis is upregulated in the growth plate it is reduced during hypertrophy. Also in human OA synthesis is upregulated [28] but never to the levels seen in skeletal growth. Our previous work [21] has provided evidence for an increase in type I1 collagen synthesis in OCD over that of normal growth, based on analyses in synovial fluid of the propeptide of this molecule. There was no evidence of an increase in PG degradation, unlike collagen, degradation in OCD lesions. In fact, the increase in PG release seen in non-lesion cartilage (compared to OCD lesions and control cartilage) suggests that within the OCD joint there are other differences in PG turnover. In our previous study of synovial fluids we noted a decrease in the 846 epitope which is ordinarily associated with growth and is increased in human OA cartilage [37] and joint fluid [23]. This is clearly related to the differences in PG turnover indicated by thesc observations. For some reasons unknown, there is an upregulation of collagenase activity in OCD cartilage which may hinder the normal progression of chondrocyte

1288

S. Lmerty et ul. I Journal of Orthopuedic Reseurch 20 (2002) 1282-1289

maturation in the developing epiphysis. Trauma could cause such an upregulation of collagenolytic activity. Following ACL rupture in humans [35] and transection of the ACL in rats [45] there is an early degradation of type I1 collagen in cartilage probably due to subsequent altered biomechanical loading of the articular cartilage created by joint instability. It is difficult to explain the consistently focal presence of lesions of OCD at specific sites in joints across the species. An internal trauma either isolated or repetitive, as proposed in horses [34] and humans [38] at a certain stage of development could in some way contribute to focal enzyme up regulation and altered metabolic activity focally in the cartilage leading to the development of OCD lesions. Studies of this kind are difficult, as it is not easy, with current surgical management, to obtain specimens in sufficient number and amount which limits our opportunity to obtain larger sample sizes and our ability to draw firm conclusions. In addition we observed heterogeneity of samples particularly among the OCD samples as evidenced by the large standard deviations. However, this and earlier studies clearly point to marked differences in the turnover of type I1 collagen and as we show here, specifically it’s cleavage, in OCD.

Acknowledgements

The authors would like to thank Guylaine Bedard for assistance with illustrations. Sheila Laverty is funded by the Groupe de Recherche en MCdecine Equine du Quebec (GREMEQ), Facultt de Medecine VCtCrinaire, Universitk de MontrCal and The Canadian Arthritis Network. Robin Poole is funded by Shriners Hospitals for Children, National Institute for Ageing, National Institute of Health, Canadian Institute for Health Research and the Canadian Arthritis Network. References Alini M, Matsui Y , Dodge GR, Poole AR. The extracellular matrix of cartilage in the growth plate before and during calcification; changes in composition and degradation of type I1 collagen. Calcif Tissue Int 1992;50:327-35. Alvarez J, Balbin M , Saiitos F, Fernandez M, Ferrando S, Lopez JM. Different bone growth rates are associated with changes in the expression pattern of types I1 and X collagens and collagenase in proximal growth plates of the rat tibia. J Bone Min Res 2000;15:82-94. Billinghurst RC, Dahlberg L, Ionescu M, Reiner A, Bourne R, Rorabeck C, et al. Enhanced cleavage of type 11 colagen by collagenases in osteoarthritic articular cartilage. J Clin Invest 1997:99:1534 45. Billinghurst RC, Ionescu M, Poole AR. Immunoassay for collagenase-mediated cleavage of types I and I1 collagens. Method Mol Biol 2001;151:457-72. Billinghurst RC, Wu W, Ionescu M, Reiner A, Dahlberg L, Chen J, et al. Comparison of the degradation of type I1 collagen and

proteoglycan in nasal and articular cartilages induced by interleukin-I and the selective inhibition of type I1 collagen cleavage by collagenase. Arth Rheum 2000;43(3):664-72. Bridges CH, Harris ED. Experimentally induced cartilaginous fractures (osteochondritis dissecans) in foals fed low-copper diets. J Am Vet Med Assoc 1988;193:215-21. Campbell CJ, Ranawat CS. Osteochondritis dissecans: the question of etiology. J Trauma 1966;6(2):201-21. Carlson CS, Cullins LD. Meuten DJ. Osteochondrosis of the articular-epiphyseal cartilage complex in young horses: evidence for a defect in cartilage canal blood supply. Vet Pathol 1995; 32:641-7. Dahlberg L, Billinghurst RC, Manner P, Nelson F, Webb G, Ionescu M, et al. Selective enhancement of collagenase-mediated cleavage of resident type 11 collagen in cultured osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collagenase 1 (matrix metalloproteinase 1). Arth Rheum 2000;43(3): 673-83. Ekman S, Carlson CS. The pathophysiology of osteochondrosis. Vet Clin North Am: Small Animal Practice 1998;28:17 32. Farndale R, Buttle DJ, Barret AJ. Improved quantitation and discrimination of sulphated glycosaminoglycans by use of dimethylmethlene blue. Biochem Biophys Acta 1986;883:173~7. Foland J , McIlwraith C, Trotter G. Arthroscopic surgery for osteochondritis dissecans of the femoropatellar joint of the horse. Equine Vet J 1992;24(6):419-23. Freemont AJ, Hampson V, Tilman R, Goupille P, Taiwo Y, Hoyland JA. Gene expression of matrix metalloproteinases I, 3, and 9 by chondrocytes in osteoarthritic human knee articular cartilage is zone and grade specific. Ann Rheum Dis 1997;56: 542-9. Glade MJ, Belling TH. A dietary etiology for osteochondrotic cartilage. J Eq Vet Sci 1986;6(3):151-5. Hollander AP, Heathfield TF, Webber C , Iwata Y . Bourne R, Rorabeck C, et al. Increased damage to type I1 collagen in osteoarthritic articular cartilage detected by a new immunoassay. J Clin Invest 1994;93:1722-32. Hollander AP, Pidoux I, Reiner A, Rorabeck C, Poole AR. Damage to type I1 collagen in aging and osteoarthritis starts at the articular surface, originates around chondrocytes, and extends into the cartilage with progressive degeneration. J Clin Invest 1995;96:2859-69. Hurtig M, Green SL, Dobson H, Mikuni-Takagaki Y. Choi J . Correlative study of defective cartilage and bone growth in foals fed a low-copper diet. Equine Vet J Suppl 1993;16:66 73. Hurtig MB, Pool RR. Pathogenesis of equine osteochondrosis. In: Mc Ilwraith WC, Trotter GW, editors. Joint disease in the horse. Philadelphia: W.B. Saunders Company; 1996. p. 335-58. Jeffcott LB, Henson FMD. Studies on growth cartilage in the horse and their application to the aetiopathogenesis of dyschondroplasia (osteochondrosis). Vet J 1998;156:177-92. Labarca C, Paigen K. A simple, rapid, and sensitive DNA assay procedure. Anal Biochem 1980:102:34451. Laverty S, Ionescu M, Marcoux M, Boure L, Doize B, Poole AR. Alterations in cartilage type41 procollagen and aggrecan contents in synovial fluid in equine osteochondrosis. J Orthop Res 2000;18(3):399405. Lillich JD, Bertone AL, Malemud CJ, Weisbrode SE, Ruggles AJ, Stevenson SS. Biochemical, histochemical and immunochemical characterization of distal tibia1 osteochondrosis in horses. Am J Vet Res 1997;58:89-98. Lohmander L, Ionescu M, Jugessur H, Poole A. Changes in joint cartilage aggrecan after knee injury and in osteoarthritis. Arth Rheum 1999;42:53444. Mc Ilwraith WC. Clinical aspects of osteochondritis dissecans. In: Mc Ilwraith WC, Trotter GW, editors. Joint disease in the horse. Philadelphia: W.B. Saunders Company; 1996. p. 362-83.

S Luvercy et al I Joirrnul of Orthopuedrr K e ~ ~ a r t20 l i (2002) 1282-1289

[25] Mitchell PG, Magnd HA, Reeves LM, Loprcsti-Morrow LL, Yocum SA, Rosner PJ, et al. Cloning. expression and type TI collagenolytic activity of matrix metalloproteinasc-13from human osteoarthritic cartilage. J Clin Invest 1996;97(3):761-8. [26] Mwale F, Billinghurst C, Wu W, Alini M, Webber C, Reiner A, et al. Selectivc assembly and remodelling of collagens I1 and IX associated with expression of the chondrocyte hypertrophic phenotype. Dev Dyn 2000;218(4):648-62. [27] Mwale F, Tchetiiia E, Wu WC, Poole AR. The assembly and remodeling of the extracellular matrix in the growth in relationship to mineral deposition and cellular hypertrophy: an in situ study of collagens I1 and IX and Proteoglycan. J Bone Jt Surg 2002; 17(2):275-83. [28] Nelson F, Dahlberg L, Laverty S, Reiner A, Pidoux I, Ionescu M, et al. Evidence for altered synthesis of type 11 colhgcn in patients with osteoarthritis. J Clin Invest 1998;102(12):2115-25. [29] Olsson SE. Introduction. Acta Radiol 1978;I(Suppl 358):7-14. [30] Olsson SE, Reiland S. The nature of osteochondrosis in animals. Acta Radiol 1978;(Suppl 358):299-306. [31] Pappas AM. Osteochondrosis dissecans. Clin Orthop Re1 Res 1981;158:59-69. [32] Philipsson J, Andreasson E, Sandgren B, Carlsten J. Osteochondrosis in the tarsocrural joint and osteochondral fragments in the fetlock joints in Standardbred trotters. 11 Heritability. Equine Vet J Suppl 1993;16:3841. 1331 Pinals RS. Traumatic arthritis and allied conditions. In: Koopman W, editor. Arthritis and allied conditions, Vol. 2. Philadelphia: Lippincott Williams & Wilkins; 2000. p. 1971-86. [34] Pool RR. Difficulties in definition of equine osteochondrosis; differentiation of developmental and acquired lesions. Equine Vet J Suppl 1993;16:5-12. [35] Price JS, Till SH, Bickerstaff DR, Bayliss MT, Hollander AP. Degradation of cartilage type I1 collagen precedes the onset of osteoarthritis following anterior cruciate ligament rupture. Arth Rheum 1999;42(I1):2390-8. [36] Ralston S. Hyperglycemia/hyperinsulinemia after feeding a meal of grain to young horses with osteochondritis dissecans (OCD) lesions. Pferdeheilkunde 1996;3:320-2. [37] Rizkalla G, Reiner A, Bogoch E, Poole AR. Studies of the articular cartilage proteoglycan aggrecan in health and osteoarthritis. Evidence for molecular heterogeneity and extensive molecular changes in disease. J Clin Invest 1992;90:2268-77.

1289

[38] Schcnk RC, Athanasiou KA, Constantinideb G, Gomez E. A biomechanical analysis of articular cartilage of the human elbow and a potential relationship to osteochondritis dissecdns. Clin Orthop Re1 Res 1994299305- 12. [39] Shlopov BV, Lie W-R, Mainardi CL, Cole AA, Chubinskaya S. Hasty KA. Osteoarthritic lesions. Involvement of three diffcrcnt collagenases. Arth Rheum 1997;40(11):2065-74. [40] Siffert RS. Classification of the osleochondroses. C h i Orthop Re1 Res 1981;158:10-8. 1411 Siffert RS. The Osteochondroses. Clin Orthop Re1 Res 1981;158:2-3. [42] Sloet van Oldruitenborgh-oosterbaan MM, Mol JA, Barneveld A. Hormones, growth factors and other plasma variables in relation to osteochondrosis. Equine Vet J 1999;31:45-54. [43] Sokoloff L. Acquired chondronecrosis. Ann Rheum Dis 1990:49: 2624. 1441 Stahle-Backdahl M, Sandstedt B, Bruce K, Lindahl A. Jiininez MG, Vega JA, et al. Collagenase-3 (MMP-13) is expressed during human fetal ossification and re-expressed in postnatal bone remodeling and in Rheumatoid Arthritis. Lab Invest 1997; 76(5):717-28. 1451 Stoop R, van der Kraan PM, Buma P, Hollander AP, Poole AR, van den Berg WB. Denaturation of type I1 collagen in articular cartilage in experimental murine arthritis. Evidence for collagen degradation in both reversible and irreversible cartilage damage. J Path 1999;188:329-37. [46] Stougdard J. The hereditary factor in osteochondritis dissecans. J Bone Jt Surg 1961;43B(2):256-8. [47] van den Hoogen BM, van de Lest CHA, van Weeren PR, van Golde LMG, Barneveld A. Changes in proteoglycan metabolism in osteochondrotic articular cartilage of growing foals. Equine Vet J 1999;(Suppl 3 1):3844. 1481 Wardale RJ, Duance VC. Characterisation of articular and growth plate cartilage collagens in porcine osteochondrosis. J Cell Sci 1994;107:47-59. [49] Wu CW, Tchetina EV, Mwale F, Hasty K, Pidoux I, Reiner A, et al. Proteolysis involving Matrix Metalloproteases 13 (Collagenase-3) is required for chondrocyte differentiation that is associated with matrix mineralization. J Bone Min Res 2002; 17(4):639-51. [50] Wu W, Mwale F, Tchetina E. Kojima T, Yasuda T. Poole AR. Cartilage matrix resorption in skeletogenesis. Novart Found Symp 2001;232:158-66.