ARTHRITIS & RHEUMATISMVol. 46, No. 8, August 2002, pp 2087–2094DOI 10.1002/art.10428© 2002, American College of Rheumatology

Sites of Collagenase Cleavage and Denaturation ofType II Collagen in Aging and Osteoarthritic ArticularCartilage and Their Relationship to the Distribution of

Matrix Metalloproteinase 1 and Matrix Metalloproteinase 13

William Wu,1 R. Clark Billinghurst,1 Isabelle Pidoux,1 John Antoniou,2 David Zukor,2

Michael Tanzer,3 and A. Robin Poole4

Objective. To determine the sites of cleavage anddenaturation of type II collagen (CII) by collagenase(s)in healthy and osteoarthritic (OA) human articularcartilage and their relationship to the distribution ofmatrix metalloproteinase 1 (MMP-1) and MMP-13.

Methods. Single (per subject) full-depth speci-mens from femoral condylar cartilage were isolatedfrom articulating surfaces at autopsy from 8 subjectswithout arthritis and during arthroplasty from 10 pa-tients with OA. Fixed frozen sections of cartilage wereexamined by immunoperoxidase localization, using an-tibodies to the collagenase-generated cleavage site inCII, to an intrachain epitope recognized only in dena-tured CII, and to MMP-1 and MMP-13 (proenzyme,activated enzyme, or enzyme/inhibitor complex).

Results. Staining for collagen cleavage, denatur-ation, and both MMPs was weak to moderate and wasfrequently observed in pericellular sites in cartilagefrom younger, nonarthritic subjects. In specimens from

older subjects, this staining was often more widespreadand of greater intensity. Similar staining was usually,but not always, seen for all antibodies. In OA cartilage,staining was often stronger and more intense than thatin normal cartilage from older subjects, and the distri-bution of staining was often similar for the differentantibodies. Pericellular staining in the deep zone wasfrequently more pronounced in arthritic cartilage andextended to territorial and sometimes interterritorialsites. In very degenerate specimens, staining was dis-tributed throughout most of the cartilage matrix.

Conclusion. These observations provide evidencefor the presence of limited cleavage and denaturation ofCII restricted to mainly pericellular and superficialsites in cartilage from younger, healthy subjects, whereMMP-1 and MMP-13 are also selectively localized.Collagen degradation is more extensive and often morepronounced in cartilage from older, nonarthritic sub-jects. Characteristic changes in early OA are similar tothose seen with aging in cartilage from older, healthysubjects, with collagen damage and collagenases con-centrated closer to the articular surface. There wasusually a close correspondence between the cleavageand denaturation of CII and the sites at which thesecollagenases were detected, suggesting that both MMPsare involved in the physiology and pathology. There wasno evidence that the damage to CII is ordinarily initi-ated in sites other than at and near the articular surfaceand around chondrocytes.

Degradation and loss of articular cartilage arefundamental features of osteoarthritis (OA). The pro-cess ordinarily is thought to be very slow, and pathologymay not be clinically apparent for up to 20–30 years.Type II collagen (CII), the major component of the

Dr. Poole’s work was supported by the Shriners Hospitals forChildren, the National Institute on Aging, the NIH, the CanadianInstitute of Health Research, and the Canadian Arthritis Network. Dr.Billinghurst was a recipient of a fellowship from the Fonds de laRecherche et Sante du Quebec.

1William Wu, BSc, PhD (current address: MassachusettsGeneral Hospital, Boston), R. Clark Billinghurst, DVM, MSc, PhD(current address: Colorado State University, Fort Collins), IsabellePidoux, RT: Shriners Hospitals for Children, Montreal, Quebec,Canada; 2John Antoniou, MD, PhD, David Zukor, MD: McGillUniversity and Jewish General Hospital, Montreal, Quebec, Canada;3Michael Tanzer, MD: McGill University, Montreal, Quebec, Canada;4A. Robin Poole, PhD, DSc: Shriners Hospitals for Children, andMcGill University, Montreal, Quebec, Canada.

Address correspondence and reprint requests to A. RobinPoole, PhD, DSc, Joint Diseases Laboratory, Shriners Hospital forChildren, 1529 Cedar Avenue, Montreal, Quebec H3G 1A6, Canada.E-mail: rpoole@shriners.mcgill.ca.

Submitted for publication October 18, 2001; accepted inrevised form April 16, 2002.

2087

extracellular matrix of cartilage and the collagen fibril,provides this tissue with its tensile properties (1,2). Inthe setting of OA, denaturation of this molecule inarticular cartilage is excessive (3) and is accompanied byincreased cleavage by collagenases (4). The net result isa loss of tensile properties and CII (5,6), associated withdamage to this molecule. Recent studies using inhibitorsof collagenases have provided evidence that matrixmetalloproteinase 1 (MMP-1; collagenase 1) and espe-cially MMP-13 (collagenase 3) are involved in thiscleavage (7,8). Excessive activity of MMP-13 can gener-ate the type of pathology observed in OA (9).

We previously demonstrated that in OA, earlydamage to articular cartilage is usually characterized bydenaturation of CII at and close to the articular surface,extending progressively into the underlying cartilage,first in pericellular sites and then in territorial and moreremote interterritorial sites (10). There have also been

reports of more pronounced staining for MMPs, such asstromelysin 1 (an activator of proMMPs such as procolla-genase) in sites similar to those where some investigatorshave observed damage (11–13). Other authors havereported the presence of MMP-13 in the deep layer ofarticular cartilage during early cartilage degeneration inOA (14).

To our knowledge, there have been no humanstudies in which damage to matrix molecules such as CIIhas been compared with localization of the proteinases(e.g., collagenases) that may be involved in cleavage ofthese molecules. Moreover, there is also a lack ofdefinitive studies of collagenase cleavage of CII inhuman articular cartilage, although recent studies re-vealed a close association between lesion formation inmice, rats, and guinea pigs and cleavage of this molecule(15–17). The present study examines in detail the sites ofcleavage and denaturation of CII by collagenases in OAand nonarthritic human articular cartilage and theirrelationship to the presence of the collagenases MMP-1and MMP-13.

MATERIALS AND METHODS

Human cartilage. Full-depth specimens of femoralcondylar cartilage (0.5–1.0 cm2 surface area) were removed

Figure 1. Normal articular femoral condylar cartilage specimens re-moved at autopsy from 4 nonarthritic subjects (ages 20–35 years)(a–d), showing type II collagen denaturation (Col2-3/4m antibody),cleavage by collagenase (Col2-3/4Cshort antibody), matrix metallopro-teinase 1 (MMP-1), and MMP-13 by immunolocalization. Specimensare positioned so that the articular surface is at the top of eachphotomicrograph. The superficial zone (SZ), mid zone (MZ), anddeep zone (DZ) of samples are designated in the top row (a). Bar �500 �m.

Figure 2. High-magnification views of nonarthritic articular cartilagefrom a 32-year-old subject (a) and osteoarthritic (OA) cartilage froma 78-year-old subject (b), showing relative immunolocalization of typeII collagen denaturation, cleavage by collagenase, matrix metallopro-teinase 1 (MMP-1), and MMP-13. Pericellular staining can be clearlyseen. Specimens are positioned so that the articular surface is at thetop of each photomicrograph. Bar � 200 �m.

2088 WU ET AL

with a sharp scalpel from the anterior articulating regions ofthe knee joints of 8 adult autopsy subjects of various ages andsexes, within 15 hours of death (Figures 1, 2a, and 3); thesesubjects had no observable arthritic joint abnormalities andhad not recently (within 2–3 months) undergone chemo-therapy. Corresponding site–matched full-depth specimens offemoral condylar cartilage from 10 patients (Figures 2b, 4–6)who underwent total knee arthroplasty for OA, diagnosedaccording to the criteria of the American College of Rheuma-tology (formerly, the American Rheumatism Association)(18), were removed to the laboratory within an hour ofsurgery. Osteophytic cartilage was excluded from this study.

Antibodies. The antibodies used in this study havebeen previously described and include a mouse monoclonalantibody to an intrachain epitope in CII, which is used todetect denaturation of CII (i.e., the antibody does not reactwith triple helical CII unless the CII has been denatured[proteolytically cleaved by a collagenase]) (3); a rabbit poly-clonal antibody to a neoepitope generated by cleavage of CIIby a collagenase that resides at the carboxy terminus of thisprimary cleavage site (named Col2-3/4Cshort) (4); a rabbitantipeptide antibody to a human MMP-13 peptide sequence(9,19) that recognizes the sequence PNPKHPKTPEK corre-sponding to amino acids 275–285; and a rabbit antipeptideantibody to human MMP-1 (19). These antibodies, recognizingthe peptide sequence AEKAFQLWSNV corresponding toamino acids 133–143, exhibited no cross-reactivity or activitywith any other known protein.

Immunolocalization. The methods used were generallysimilar to those previously described (10). Briefly, frozensections were cut and fixed in paraformaldehyde. After rinsingwith phosphate buffered saline, sections were always treatedwith chondroitinase ABC (10) to deplete chondroitin sulfate.Depletion was confirmed by a lack of Safranin O staining,which was performed as previously described (10). Localiza-tion by immunoperoxidase staining involved use of primaryantibody binding of diluted antiserum or monoclonal antibody,which is detected by secondary anti-mouse or anti-rabbitantibodies labeled with biotin. Primary antibody binding isfinally detected with streptavidin–peroxidase. The reagentswere obtained from Dako Cedarlane (Hornby, Ontario, Can-ada). Binding of these antibodies in 6-�m–thick frozen sec-tions was examined in triplicate on at least 2 different occa-sions to ensure reproducibility. Antibody dilutions and colordevelopment were arranged to provide maximum stainingintensity. Duplicate controls were prepared by prior absorp-

Figure 3. Normal articular cartilage obtained at autopsy from 4nonarthritic subjects (a–d), showing type II collagen denaturation(Col2-3/4m antibody), cleavage by collagenase (Col2-3/4Cshort anti-body), matrix metalloproteinase 1 (MMP-1), and MMP-13 by immu-nolocalization. Specimens are positioned so that the articular surface isat the top of each photomicrograph. Bar � 500 �m.

Figure 4. Demonstration of all 4 antibody specificities (a–d) in acartilage specimen obtained from a 78-year-old woman with OA. Forcontrol, each antibody was preabsorbed with 100 �g/ml of epitopecontaining peptide for 1 hour at 37°C and overnight at 4°C prior to use.The test sample was incubated under the same conditions, without thepeptide. The lack of staining in control specimens is clearly apparent.Specimens are positioned so that the articular surface is at the top ofeach photomicrograph. Bar � 500 �m. See Figure 2 for definitions.

COLLAGEN DEGRADATION IN OSTEOARTHRITIS 2089

tion of the primary antibody, using the peptides listed aboveand in the manner described (10).

RESULTS

Nonarthritic cartilage. Figures 1 and 3 showspecimens obtained from 8 autopsy subjects. In speci-mens from the 4 younger subjects (ages 20–35 years),staining patterns for CII denaturation, cleavage,MMP-1, and MMP-13 were similar (Figure 1). Stainingwas generally weak to moderate and was restrictedmainly to pericellular sites throughout the full depth ofthe articular cartilage (Figures 1 and 2a). Sometimespericellular staining was more pronounced in the mid todeep zones. In cartilage from 2 subjects, staining wasweak to moderately diffuse immediately under the artic-ular surface in more superficial sites. Staining was alsoobserved at the articular surface (Figure 1).

In cartilage from older subjects (�52 years), thepresence of pericellular staining and more intense stain-ing was often also observed more remote from the cell interritorial sites, particularly in the lower mid and deepzones (Figures 3b and d). A band of strong, diffusestaining for all epitopes throughout the matrix in thesuperficial and mainly mid zones (Figure 3b) and some-times extending further into the deep zone, wherepericellular staining was less pronounced than inyounger individuals (compare Figure 3b with Figure 1),was often observed in cartilage obtained from oldersubjects. Frequently, this strong staining ceased abruptlyin the mid-deep zone, as shown in Figure 3b.

In cartilage from older subjects, evidence ofcollagen denaturation was observed throughout the car-tilage matrix and the depth of the cartilage and was mostpronounced in pericellular sites (Figure 3d). In somecases staining for both denaturation and cleavage of

Figure 5. Osteoarthritic articular cartilage specimens obtained from 5patients (a–e) at knee arthroplasty, showing type II collagen denatur-ation (Col2-3/4m antibody), cleavage by collagenase (Col2-3/4Cshort

antibody), matrix metalloproteinase 1 (MMP-1), and MMP-13 byimmunolocalization. Specimens are positioned so that the articularsurface is at the top of each photomicrograph. Bar � 500 �m.

Figure 6. Osteoarthritic articular cartilage specimens obtained from 5patients (a–e) at knee arthroplasty, showing type II collagen denatur-ation (Col2-3/4m antibody), cleavage by collagenase (Col2-3/4Cshort

antibody), matrix metalloproteinase 1 (MMP-1), and MMP-13 byimmunolocalization. Specimens are positioned so that the articularsurface is at the top of each photomicrograph. Bar � 500 �m.

2090 WU ET AL

collagen was similar (Figures 1–3), but in other casesstaining for cleavage was either more or less pronouncedthan that for denaturation (Figures 3c and d).

In some samples, the intensity and distribution ofstaining patterns for MMP-1 and MMP-13 were similar(Figures 1, 2a, and 3c), but in others the intensity forMMP-13 was clearly more pronounced (Figures 3a, b,and d). In most specimens, the distribution and relativeintensity for cleavage of collagen were comparable tothat for MMP-13. In one specimen, however (Figure 3c),staining for cleavage was selectively observed in thesuperficial zone, while staining for MMP-13 was weak inthis site and elsewhere.

Specificity of staining. The specificity of stainingfor all antibodies was clearly demonstrated by an almostcomplete inhibition of antibody/Fab binding followingprior absorption with the specific collagen or MMPpeptide against which the antibody had been prepared.Figure 4 shows an example of OA cartilage in whichstaining for denaturation and cleavage of CII, as well asMMP-1 and MMP-13, was largely prevented followingthis pretreatment. The only staining observed was weak,diffuse, and of an intensity that was generally muchlower than that observed with antibody alone. Becausesome superficial staining persisted in these OA cartilagespecimens, the specificity is not proven. Control sampleswith intact articular surfaces (prepared from cartilageobtained from nonarthritic subjects) did not reveal anysurface staining after absorption of the peptide, pointingto its specificity (data not shown).

OA cartilage. A variety of staining patterns wereobserved in OA cartilage, several of which were similarto those observed in nonarthritic cartilage. Figures 5 and6 show results in cartilage specimens obtained from 10different patients during knee arthroplasty.

Pericellular staining similar to that seen in carti-lage from younger (ages 20–30 years), nonarthritic per-sons (Figure 5a) was observed occasionally, althoughsuperficial fibrillation was clearly detectable. Pericellularcleavage was again seen in the absence of significantdenaturation (see Figure 3d) and corresponded more tothe presence and intensity of staining for MMP-13 thanto that for MMP-1, as observed in nonarthritic cartilage(Figures 3a and d).

What was most striking in these OA cartilagespecimens was the frequent similarity of staining for allantibodies in articular cartilage that exhibited earlysuperficial fibrillation. In 6 of the 8 specimens in thiscategory (early degeneration) that showed staining otherthan pericellular (Figures 5b–e and Figures 6b and c),staining was generally present throughout the matrix in

the superficial and mid zones, and was usually mostintense in pericellular sites. It was in these pericellularsites, in the deep zone, that weaker staining in nonar-thritic cartilage was most commonly seen. In all 6 ofthese cases, staining for collagen denaturation and cleav-age was generally seen in the same sites as was stainingfor MMP-1 and MMP-13. As was sometimes observed innonarthritic cartilage (Figure 3b), OA cartilage oftendisplayed a sharp boundary between the zone of almosttotal cartilage matrix involvement and the zone wherestaining was more confined to pericellular sites (Figures5b–d and Figure 6b). OA cartilage with an intact artic-ular surface often demonstrated more evidence of col-lagen damage, associated with the presence of MMPs, inthe deep zone. In 3 OA specimens showing intactarticular surfaces, collagen degradation was seen pri-marily in pericellular sites (Figures 5e and 6b and c). Asimilar observation was made in 1 specimen of normalcartilage (Figure 3d).

Only occasionally did we observe discontinuousdistribution of staining, where collagen damage was seenin more than 1 separate site. In 1 specimen (Figure 6a),bands of cleavage were shown in the superficial-mid anddeep zones, with an isolated pocket of activity in the midzone that closely corresponded to the distribution ofstaining for both MMP-1 and MMP-13. This cleavagewas not accompanied by collagen denaturation in thissample or in the sample shown in Figure 4a, unlike in themajority of other specimens (8 of 10). In the setting ofadvanced degeneration (Figure 6e), collagen denatur-ation and cleavage were observed throughout the carti-lage in sites similar to those where MMP-13 and MMP-1were localized. The specimen shown in Figure 6e wastypical of very degenerate, advanced lesions (data notshown).

In essentially all specimens from patients withOA, we observed a codistribution of MMP-1 andMMP-13 in sites where cleavage by collagenase wasobserved, as was seen in the majority of specimens fromnonarthritic subjects. In both normal and OA cartilage,collagen denaturation was sometimes seen in interterri-torial sites in the deep zone, remote from chondrocytes,in the absence of staining for collagen cleavage and forMMP-1 and MMP-13 (Figures 3c, 5b–e, and 6c).

DISCUSSION

Immunoanalyses of the cleavage (4) and denatur-ation (3) (which follows cleavage) of the triple helix ofCII have clearly revealed a significant increase in theseparameters in human OA cartilage recovered at arthro-

COLLAGEN DEGRADATION IN OSTEOARTHRITIS 2091

plasty. It is apparent that immunolocalization can beused to distinguish the excessive damage to CII in OAcartilage. Immunoperoxidase localization shows the in-creased damage to CII and the increased presence ofMMP-1 and MMP-13 in pericellular and territorial sitesin the deep zone of less degenerate OA cartilage. Itdemonstrates the widespread distribution of collagendamage and the presence of these collagenases through-out very degenerate OA cartilage. It also shows wheresuch damage occurs in both aging and OA cartilage,regardless of whether sites of denaturation correspondto sites of cleavage (and vice versa), and whether theseevents are related to the sites in which collagenases arelocalized in these cartilages. However, because primaryantibody dilution and color development were arrangedto provide maximum intensity of staining, it is notpossible in this study to compare the relative concentra-tions of the epitopes against which the antibodies weredirected. Therefore, we make no attempt to comparerelative amounts of collagenases using these methods.

What is very clear from the present study is thatthe cleavage of CII by collagenases, as revealed byantibodies to this primary cleavage site in the triple helixof CII, frequently (but not always) corresponds to thesites where MMP-1 and MMP-13 are mainly detected, asobserved by others (17). Such an association was re-cently observed for collagen cleavage and MMP-13 in atransgenic mouse overexpressing active human MMP-13in articular cartilage, resulting in development of degen-erative focal lesions resembling OA (9). At our labora-tory, recent studies of human articular cartilage haveprovided evidence for a role for increased activities ofthese collagenases in the cleavage of both residentmolecules (MMP-13) and those that are newly synthe-sized (MMP-1) (7).

In the present study, we cannot discriminatebetween these collagen molecules in terms of when theywere synthesized, nor can we ascribe activity to thesecollagenases, because our antibodies detect proenzyme,activated enzyme, and tissue inhibitor ofmetalloproteinases–enzyme complex (Wu W, Poole AR:unpublished observations). However, there was oftenclose correspondence in terms of staining location be-tween cleavage of CII by collagenases and the presenceof MMP-1 and MMP-13. The concentration of collage-nase-cleaved CII mainly in pericellular sites in cartilageobtained from the 4 younger subjects (20–35 years ofage) no doubt reflects a normal turnover process that isa feature of healthy cartilage in the young adult. The factthat this concentration is seen predominantly in pericel-lular sites indicates that chondrocytes release the colla-

genases that produce these cleavages. In such cases, weusually found the MMPs in these same pericellular sites,suggesting that they also originated from chondrocytes.These proteinases are probably present in extracellularsites because collagenases can bind to collagen fibrils bytheir hemopexin domains (12), providing a means ofretaining them in the extracellular matrix, bound tocollagen fibrils at sites of activity. Involvement in thisdegradative process of collagen molecules more remotefrom the chondrocytes in territorial and interterritorialsites, which was seen in older individuals and patientswith OA, indicates cleavage of collagen fibrils in siteswhere this is not usually observed in younger adults.Thus, some of the changes observed in nonarthriticcartilage may represent preclinical changes that couldlead to OA.

The predominance of collagen damage andMMPs in sites closer to the articular surface (with lesspronounced evidence of this in the deep zone) suggests,as in earlier studies of denaturation alone (10), that thisdegradative process begins in the more superficial car-tilage and proceeds progressively into the underlyingarticular cartilage of the mid and deep zones. Theoften-present line of demarcation between extensivematrix staining and the adjacent deeper, predominantlypericellular staining suggests that the deeper cells maybe activated by cytokines acting in a paracrine manner,which are generated by the adjacent, more superficialcells. This suggests a creeping substitution of good tissuewith damaged cartilage as part of the pathologic process,which we believe takes place over many years.

This observation is in contrast to a report thatMMP-13 is found predominantly in the deep zone indeveloping human OA (14). In studies of the naturaldevelopment of OA in mice (16), guinea pigs (17), andrats (15), CII cleavage was usually observed in moresuperficial sites and progressed deeper into the articularcartilage as lesions developed. In a study of cartilage-specific expression of the transgene for active MMP-13in the mouse, which causes development of a lesionclosely resembling human OA (9), MMP-13 and CIIcleavage were also detected in pericellular sites in themore superficial cartilage prior to fibrillation, althoughin the present study, the opposite was sometimes seen innonarthritic cartilage. This suggests that in OA it isordinarily these more superficial cells from which thesecollagenases are initially up-regulated, leading to theexcessive cleavage and denaturation of CII seen inexperimental OA and in the present investigation.

The striking demarcation between sites in carti-lage where CII was cleaved by collagenase and dena-

2092 WU ET AL

tured serves to demonstrate the strict localization of thiscleavage with respect to the cell and, on a larger scale,with respect to the articular surfaces. Remarkably, thesecollagenases usually exhibited the same distribution ofstaining, suggesting that the activity of at least some ofthe MMP localized by the antibodies accounted for thiscleavage.

In specimens in which denaturation was detectedin the absence of detectable cleavage, it is likely that theintrachain epitope reflecting denaturation may havepersisted from earlier cleavage or damage to the colla-gen molecule. A cleavage epitope may no longer beobserved, because it can be subsequently removed fromcartilage, as we have shown in culture (7,8).

Together, these studies represent a comprehen-sive investigation of the changes and interrelationshipsbetween CII cleavage and denaturation in human OAand nonarthritic cartilage and show how these eventsrelate to the presence of collagenases that have thepotential to produce these events. They provide evi-dence for the often close relationship of these collage-nases to development of damage to CII in aging and OA.The results demonstrate that collagen cleavage anddenaturation, matrix degeneration, and the presence ofcollagenases in human articular cartilage in healthy andOA cartilage are usually closely interrelated. They donot provide further evidence supporting the activity ofone collagenase over another, because only the localiza-tion of these MMPs, not the concentrations, was exam-ined (conditions of staining did not permit this compar-ison). Neither do these studies provide direct evidenceto indicate that these proteases are in fact involved in thedegradative process. The activation of these collage-nases is as important as their localization.

These reagents and methods, when used in themanner we have described, do not offer a definitivemeans of distinguishing between OA cartilage and nor-mal cartilage from older persons. The staining patternsin cartilage from older subjects are often similar to thoseseen in OA. Using immunoassay, however, we previouslyshowed that cleavage and denaturation of CII are in-creased in OA cartilage at arthroplasty (3,4). Thepresent results do reveal that changes that occur withaging are often similar to those seen in OA specimensrecovered at arthroplasty. Whether the age-relatedchanges are fully representative of earlier degenerativechanges that occur in patients with OA remains to beestablished. Therefore, these immunohistochemicalstudies indicate that development of OA may sometimesbe part of the natural aging process.

REFERENCES

1. Kempson GE. Mechanical properties of articular cartilage. In:Freeman MAR, editor. Adult articular cartilage. 2nd ed. Turn-bridge Wells (UK): Pitman Medical; 1979. p. 333–414.

2. Akizuki S, Mow VC, Muller F, Pita JC, Howell DS. Tensileproperties of human knee joint cartilage. I. Influence of ionicconditions, weight bearing, and fibrillation on tensile modulus.J Orthop Res 1986;4:379–92.

3. Hollander AP, Heathfield TF, Webber C, Iwata Y, Rorabeck C,Bourne R, et al. Increased damage to type II collagen in osteoar-thritic cartilage detected by a new immunoassay. J Clin Invest1994;93:1722–32.

4. Billinghurst RC, Dahlberg L, Ionescu M, Reiner A, Bourne R,Rorabeck C, et al. Enhanced cleavage of type II collagen bycollagenases in osteoarthritic articular cartilage. J Clin Invest1997;99:1534–45.

5. Kempson GE, Muir H, Pollard C, Tuke M. The tensile propertiesof the cartilage of human femoral condyles related to the contentof collagen and glycosaminoglycans. Biochem Biophys Acta 1973;297:465–72.

6. Akizuki S, Mow VC, Muller F, Pita JC, Howell DS. Tensileproperties of human knee joint cartilage. II. Correlations betweenweight bearing and tissue pathology and the kinetics of swelling.J Orthop Res 1987;5:173–86.

7. Dahlberg L, Billinghurst C, Manner P, Ionescu M, Reiner A,Tanzer M, et al. Selective enhancement of collagenase-mediatedcleavage of resident type II collagen in cultured osteoarthriticcartilage and arrest with a synthetic inhibitor that spares collage-nase 1 (matrix metalloproteinase 1). Arthritis Rheum 2000;43:673–82.

8. Billinghurst RC, Wu W, Ionescu M, Reiner A, Dahlberg L, ChenJ, et al. Comparison of the degradation of type II collagen andproteoglycan in nasal and articular cartilages induced by interleu-kin-1 and the selective inhibition of type II collagen cleavage bycollagenase. Arthritis Rheum 2000;43:664–72.

9. Neuhold LA, Killar L, Zhao W, Sung ML, Warner A, Kulik L, etal. Postnatal expression in hyaline cartilage of constitutively activehuman collagenase-3 (MMP-13) induces osteoarthritis in mice.J Clin Invest 2001;107:35–44.

10. Hollander AP, Pidoux I, Reiner A, Rorabeck C, Bourne R, PooleAR. Damage to type II collagen in aging and osteoarthritis startsat the articular surface, originates around chondrocytes, andextends into the cartilage with progressive degeneration. J ClinInvest 1995;96:2859–69.

11. Poole AR, Alini M, Hollander AP. Cellular biology of cartilagedegradation. In: Henderson B, Edwards JCW, Pettipher ER,editors. Mechanisms and models in rheumatoid arthritis. London:Academic Press; 1995. p. 163–204.

12. Poole AR. Cartilage in health and disease. In: Koopman WJ,editor. Arthritis and allied conditions. 14th ed. Philadelphia (PA):Lippincott, Williams & Wilkins; 2001. p. 226–84.

13. Poole AR, Howell DS. Etiopathogenesis of osteoarthritis. In:Moskowitz RW, Howell DS, Altman RD, Buckwalter JA, Gold-berg VM, editors. Osteoarthritis diagnosis and management. 3rded. Philadelphia (PA): WB Saunders; 2001. p. 29–47.

14. Moldovan F, Pelletier JP, Hambor J, Cloutier JM, Martel-PelletierJ. Collagenase-3 (matrix metalloproteinase 13) is preferentiallylocalized in the deep layer of human arthritic cartilage in situ: invitro mimicking effect by transforming growth factor �. ArthritisRheum 1997;40:1653–61.

15. Stoop R, Buma P, van der Kraan PM, Hollander AP, BillinghurstRC, Poole AR, et al. Differences in type II collagen degradationbetween peripheral and central cartilage of rat stifle joints aftercranial cruciate ligament transection. Arthritis Rheum 2000;43:2121–31.

16. Stoop R, van der Kraan PM, Buma P, Hollander AP, Billinghurst

COLLAGEN DEGRADATION IN OSTEOARTHRITIS 2093

RC, Poole AR, et al. Type II collagen degradation in spontaneousosteoarthritis in C57Bl/6 and BALB/c mice. Arthritis Rheum1999;42:2381–9.

17. Huebner JL, Otterness IG, Freund EM, Caterson B, Kraus VB.Collagenase 1 and collagenase 3 expression in a guinea pig modelof osteoarthritis. Arthritis Rheum 1998;41:877–90.

18. Altman R, Asch E, Block D, Bole G, Borenstein D, Brandt K, etal, and the Diagnostic and Therapeutic Criteria Committee of the

American Rheumatism Association. Development of criteria forthe classification and reporting of osteoarthritis: classification ofosteoarthritis of the knee. Arthritis Rheum 1986;29:1039–49.

19. Wu CW, Tchetina EV, Mwale F, Hasty K, Pidoux I, Reiner A, etal. Proteolysis involving matrix metalloproteinase 13 (collage-nase-3) is required for chondrocyte differentiation that is associ-ated with matrix mineralization. J Bone Miner Res 2002;17:639–51.

2094 WU ET AL