Osteoarthritis and Cartilage 21 (2013) 1950e1957

Loss of extracellular matrix from articular cartilage is mediated by thesynovium and ligament after anterior cruciate ligament injury

C.M. Haslauer y *, K.A. Elsaid z, B.C. Fleming x, B.L. Proffen y, V.M. Johnson k, M.M. Murray yy Department of Orthopaedic Surgery, Boston Children’s Hospital, Boston, MA, USAz Department of Pharmaceutical Sciences, Massachusetts College of Pharmacy and Health Sciences University, Boston, MA, USAx Department of Orthopaedics, Warren Alpert Medical School of Brown University/Rhode Island Hospital, Providence, RI, USAk Department of Anesthesiology, Boston Children’s Hospital, Boston, MA, USA

a r t i c l e i n f o

Article history:Received 15 February 2013Accepted 4 September 2013

Keywords:CartilageMetalloproteinasesAggrecanaseACLJoint tissues

* Address correspondence and reprint requests to:Orthopaedic Surgery, Boston Children’s Hospital,Avenue, Boston, MA 02115, USA.

E-mail address: Carla.Haslauer@childrens.harvard.

1063-4584/$ e see front matter � 2013 Osteoarthritihttp://dx.doi.org/10.1016/j.joca.2013.09.003

s u m m a r y

Objective: Post-traumatic osteoarthritis (PTOA) occurs after anterior cruciate ligament (ACL) injury. PTOAmay be initiated by early expression of proteolytic enzymes capable of causing degradation of thearticular cartilage at time of injury. This study investigated the production of three of these key proteasesin multiple joint tissues after ACL injury and subsequent markers of cartilage turnover.Methods: ACL transection was performed in adolescent minipigs. Collagenase (MMP-1 and MMP-13) andaggrecanase (ADAMTS-4) gene expression changes were quantified in the articular cartilage, synovium,injured ligament, and the provisional scaffold at days 1, 5, 9, and 14 post-injury. Markers of collagendegradation (C2C), synthesis (CPII) and aggrecan synthesis (CS 846) were quantified in the serum andsynovial fluid. Histologic assessment of the cartilage integrity (OARSI scoring) was also performed.Results: MMP-1 gene expression was upregulated in the articular cartilage, synovium and ligament afterACL injury. MMP-13 expression was suppressed in the articular cartilage, but upregulated 100-fold in thesynovium and ligament. ADAMTS-4 was upregulated in the synovium and ligament but not in thearticular cartilage. The concentration of collagen degradation fragments (C2C) in the synovial joint fluidnearly doubled in the first five days after injury.Conclusion: We conclude that upregulation of genes coding for proteins capable of degrading cartilageECM is seen within the first few days after ACL injury, and this response is seen not only in chondrocytes,but also in cells in the synovium, ligament and provisional scaffold.

� 2013 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Introduction

Post-traumatic osteoarthritis (PTOA) develops in more than 50%of individuals who tear their anterior cruciate ligament (ACL)within 10e20 years1. Recent work has shown that patients begin tolose the proteoglycan and collagen molecules from cartilage in thefirst few weeks after injury, losses that are typically irrecover-able2,3. In this study, we employed a large animal model to studychanges in gene and protein expression within the cartilage andsurrounding tissues as well as cartilage degradationwithin the first2 weeks following an induced ACL injury. There aremultiple tissueswhich are in contact with the articular cartilage via the synovialfluid and thus could influence the response of the cartilage to

C.M. Haslauer, Department ofEnders 270, 300 Longwood

edu (C.M. Haslauer).

s Research Society International. P

injury. These tissues include the synovium, the ACL, and the pro-visional scaffold.

Organized connective tissueswhichhave a net anabolic responseto injury and go on to heal do so in an orderly and predictablefashion. The first major step in that process is the formation of aprovisional scaffoldwithin thewound site that fills the gap betweenthe two damaged ends of the ligament. The trigger for scaffold for-mation is thought to be the contact of blood with the exposedcollagen of the wound edges. This contact starts a cascade of pro-cesses which result in the release of anabolic cytokines, includingfibroblast growth factor 2 (FGF-2), vascular endothelial growthfactor (VEGF), platelet-derived growth factor (PDGF)-A and -B andothers, and wound healing progresses in an orderly and reliablefashion4. However, after the ACL tears, it does not heal thewayotherligaments do, even with suture repair. Recently, a mechanism hasbeenproposed for this failure, namely thatwhen theACL tears insidethe knee joint, the contact of bloodwith the newly exposed collagenresults in formation of a provisional scaffold onlyover the torn edgesof the tissue, while the joint fluid prevents formation of a bridging

ublished by Elsevier Ltd. All rights reserved.

Table ISequences of porcine-specific qPCR primers

Gene Forward primer Reverse primer

GAPDH19 GGG CAT GAA CCA TGA GAA GT GTC TTC TGG GTG GCA GTG ATMMP-1 CCA GAG AAG ATG TGG ACC

GTG CCCCT GGG CCT GGC TGA AAAGCA

MMP-13 CAC GCC TGA TTT GAC TCA TT CAT CAA AAT GGG CAT CTC CTADAMTS-420 CAG GGT CCC ATG TGC AAC GT CAT CTG CCA CCACCA GGG TCT

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e1957 1951

scaffold4,5. In this study, we wished to determine if this dysfunc-tional provisional scaffold had any catabolic effect on the sur-rounding cartilage of the knee joint after an ACL transection.

After injury, not only does the ACL fail to heal, but also the injuryleads to the development of PTOA of the knee joint in many pa-tients. Matrix metalloproteinase-1 (MMP-1) and MMP-13 havebeen found to play a role in the development of osteoarthritis(OA)6,7. MMP-1 is responsible for breaking down interstitial colla-gens, including types I, II, and III. MMP-13 is five to ten times moreactive than MMP-1 in degrading collagen type II and is capable ofcontributing to aggrecan degradation7,8. Therefore, it has beensuggested that MMP-13 may play an important role in articularcartilage turnover and the pathophysiology associated with PTOA8.The characteristic fragments generated from this collagenase ac-tivity on type II collagen are quantifiable through ELISA (enzyme-linked immunosorbent assay). The C2Cmarker identifies fragmentsfollowing type II collagen cleavage by collagenases, includingMMP-1 and MMP-139. The CPII marker, in contrast, is a marker of type IIprocollagen synthesis10. Cartilage turnover is thus largely depen-dent on the balance between collagen and aggrecan synthesis anddegradation.

In addition, aggrecanases are also thought to play a role in thedevelopment of PTOA. A disintegrin and metalloproteinase withthrombospondin motifs (ADAMTS-4) efficiently degrades aggre-can11. This aggrecanase activity is balanced by aggrecan synthesis,which also increases during the course of OA12. The chondroitinsulfate 846 epitope (CS 846) has been examined as a marker ofaggrecan synthesis, with the greatest concentrations observed inthe synovial fluid of OA patients with the longest disease durationand greatest cartilage loss12,13.

Kraus et al. suggest that the early phase of acute injury providesa window of opportunity to decrease the destructive joint pro-cesses that may ultimately lead to the development of PTOA14. Inthis study, we hypothesized that an isolated ACL injury prompts aresponse in multiple tissues as early as 1 day after injury, whichresults in changes in cytokine and protein expression over time.Wehypothesized that even in the absence of impact trauma to thecartilage, ACL injury would produce a catabolic response in severaljoint tissues (i.e., ligament, synovium, cartilage, provisional scaf-fold). We further hypothesized that this upregulation of catabolicgenes would result in aggrecan and collagen destructionwithin thecartilage within the first 2 weeks of injury.

Materials and methods

Thirty adolescent Yucatan minipigs (Coyote CCI, Douglas, MA),aged12e15months,were obtained for use in this study. Allminipigswere handled according to approved Institutional Animal Care andUse Committee (IACUC) protocols at Animal Resources at Children’sHospital (ARCH, Boston, MA). Minipigs were assigned to euthanasiaat one of four time points (day 1, 5, 9 or 14; n¼ 6 per group). Four ofthe groupswere subjected to unilateral ACL transection, followed bytissue harvest at the designated time point. A fifth group of six non-operated minipigs served as ACL-intact controls.

Surgical procedure

Twenty-four minipigs underwent unilateral ACL transectionas previously described5,15. Briefly, the ACL of anesthetized ani-mals were exposed by performing a medial arthrotomy andpartial resection of the fat pad. The ACL was then cut using ascalpel blade at the junction of the proximal and middle thirds.Functional loss of the ACL was verified using the Lachman ma-neuver, a clinical exam used to assess the integrity of the ACL16.The knee was closed in layers. Animals were allowed normal

nutrition and ad lib activity following surgery throughout theexperimental period.

Tissue collection

At time of euthanasia, samples of cartilage, synovium, ACL andthe provisional scaffold (i.e., ACL scar tissue located between thetorn ends of the ligament) were harvested. Cartilage was harvestedfrom the weight-bearing surface of the femoral condyle and asynovium sample was taken from the medial aspect of the joint at alocation remote from the arthrotomy site. Each tissue specimenwassubmerged in a cryovial containing RNAlater� stabilization solution(Ambion, Austin, TX, USA), then flash frozen in liquid nitrogen andstored at �80�C until gene expression analysis. A second portion ofthe cartilage tissue was embedded within Optimal Cutting Tem-perature (OCT) medium (Sakura Finetek, CA, USA), frozen, andstored at �80�C for histological analysis. Systemic blood of controlanimals was clotted to serve as a provisional scaffold control for theintact group. Cartilage, intact ACL tissue, and synovium sampleswere also harvested from the six unoperated animals.

The six minipigs euthanized at day 14 were also subjected toserial serumand synovialfluiddraws. Serumand synovialfluidweresampledpre-transection, then at 5 and14days post-injurywhile thesubjects were under anesthesia. Blood was collected in serumseparator tubes, allowed to clot at room temperature, centrifuged at1,000� g for 10min, and the serum aliquoted in 500 mL aliquots andstored at �80�C. Synovial fluid was centrifuged at 3,000�g for10 min to remove any cells. The supernatant was removed andstored in 120 mL aliquots in cryovials at �80�C, with approximately240e500 mL of synovial fluid recovered at each time point.

Quantitative real-time PCR

The cartilage, ligament, synovium, and provisional scaffoldsamples were examined for mRNA expression of several genesusing real-time reverse transcriptase polymerase chain reaction(qPCR) run in duplicate. Briefly, total RNA was extracted from thefrozen tissue using the PureLink RNA Mini Kit (Ambion, Austin, TX,USA), treated with DNAse I (PureLink DNase, Invitrogen, LifeTechnologies, NY, USA) according to the manufacturer’s protocoland quantified. Total RNAwas reverse transcribed to generate cDNAusing the RETROscript kit (Ambion, Austin, TX, USA). Previouslyreported primers were validated by sequencing the PCR productand performing a basic local alignment search tool (BLAST) searchwith these results. Primers are summarized in Table I. Sybr GreenPCR Mastermix (Applied Biosystems, Foster City, CA, USA) (10 mL),nuclease-free water, forward and reverse primer (2 mL each), and3 mL (cartilage) or 0.5 mL (ligament, synovium, provisional scaffold)of the 1 mg cDNAweremixed and quantified in a reaction volume of10 ml. No template controls were included to indicate contaminantsor non-specific amplification. An Applied Biosystems 7900HT(Applied Biosystems, Foster City, CA, USA) was used for amplifica-tion and detection. The PCR profile conditions were 50�C for 2 min,95�C for 10min, followed by 40 cycles of 95�C for 15 s, 50�C for 45 s,and 60�C for 45 s. Levels of gene expressionwere normalized to thehousekeeping gene, glyceraldehyde 3-phosphate dehydrogenase

Fig. 1. Median fold change in MMP-1 (A), MMP-13 (B), and ADAMTS-4 (C) geneexpression levels in cartilage, synovium, ligament, and provisional scaffold in theinitial 2 weeks following ACL injuries compared to uninjured controls (Day 0). Note thelog 10 scale on the y-axis. Asterisks indicate significant differences at P < 0.05 frombaseline (tissue from uninjured controls) within each tissue type for the six minipigs.

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e19571952

(GAPDH). Relative changes in gene expression were calculated us-ing the 2�DDCt method17.

Collagen and aggrecan fragment detection by ELISA

Synovial fluid and serum C2C, CPII, CS 846 levels were deter-mined using commercially-available ELISA kits (60-1001-001, 60-1004, and 60-1003-01; IBEX Pharmaceuticals Inc, Montreal, Can-ada) as previously described for porcine tissue18e20. The IBEX C2Csequence and porcine proteome have a 100% maximum identitybetween the assay and porcine collagen alpha-1 (II) sequence.Spiking studies were performed to document adequate antigenretrieval for the porcine synovial fluid. Synovial fluid samples weredigested using 75 units/mL recombinant Streptomyces hyaluroni-dase (Sigma Aldrich, USA) for 10 min at 37�C prior to ELISA testing.

Safranin-O, active caspase 3, and chondroitin-6-sulfateimmunostaining in porcine cartilage specimens

Frozen section slides of porcine cartilage specimens from control(non-injured) joints (n¼ 2) and day 14 post-ACL transection (n¼ 5)knees were stained with Safranin-O and Fast Green. All sectionswere evaluated for histologic signs of cartilage degeneration using amodified Osteoarthritis Research Society International (OARSI)scoring system inwhich two investigators blinded to the treatmentindependently graded the best stained sections and arrived at aconsensus score as previously described21,22. Active caspase 3staining was performed using rabbit polyclonal primary antibody(Cat#ab13847, Abcam, Cambridge,MA) at 1:50 dilution overnight at4�C and a goat anti-rabbit secondary antibody (Life Technologies,Molecular Probes) at 1:50 dilution. Vectashield mounting mediumwith DAPI (Vector Laboratories Inc., Burlingame, CA) was used.Frozen sections were incubated in 1% hydrogen peroxide in meth-anol for 20min to block endogenous peroxidase activity. Slideswerewashed in PBS then digested with 0.2 units/ml chondroitinase ABC(Sigma Aldrich, St. Louis, MO) for 2 h at room temperature. Slideswere blocked with 3% normal horse serum for 60 min at roomtemperature and primary antibody incubation with chondroitin-6-sulfate (MK 302, abcam, Cambridge, MA) was performed over-night at 4�C at a 1:100 dilution. The specimenswere then incubatedwith biotinylated goat anti-mouse IgG in horse serum for 90 minfollowed by avidin-horse radish peroxidase complex for 60min andthen 3,30-diaminobenzidine (DAB) for 5 min. Slides were counter-stained with methyl green, dehydrated and mounted.

Statistical analyses

Gene expression was summarized as median (25th percentile,75th percentile). Overall differences in gene expression levels wereassessed using KruskaleWallis H Test followed by post-hoc ManneWhitney U Tests (SAS Version 9.2, SAS Institute Inc., Cary, NC). Sy-novial fluid and serum levels of cartilage biomarkers were reportedas median (25th percentile, 75th percentile). Comparisons weredetermined using a Wilcoxon signed-rank test (R Statistical Soft-ware; http://www.r-project.org/)23. P < 0.05 was considered sta-tistically significant for both gene expression and cartilagebiomarker analyses.

Results

MMP-1 gene expression changes in cartilage, synovium, ligament,and provisional scaffold over 14 days following ACL injury

MMP-1 gene expression in the articular cartilage increasedsignificantly by 16-fold compared to baseline levels by day 1

(P¼ 0.0162) and remained moderately elevated at day 14, althoughthis difference was not significant (P ¼ 0.0751) [Fig. 1(A)]. MMP-1gene expression was significantly upregulated by at least 9-fold atday 14 in the synovium (P ¼ 0.0123) and 3-fold in the ligament(P ¼ 0.0162), compared to baseline levels within each tissue. MMP-1 gene expression was at least 9-fold higher than baseline at alltime points in the provisional scaffold; however, this change was

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e1957 1953

not statistically significant (P ¼ 0.0544, 0.1134, and 0.0544,respectively).

MMP-13 gene expression changes in cartilage, synovium, ligament,and provisional scaffold over 14 days following ACL injury

MMP-13 gene expression in the articular cartilage was found tobe suppressed by 16-fold at day 1, increasing to over 200-fold byday 5, a change that was statistically significant [P ¼ 0.0312;Fig. 1(B)]. In contrast, in the synovium, a 350-fold increase in MMP-13 gene expression was seen by day 5, and remained that elevatedat day 14 (P ¼ 0.0123 for comparison to baseline for both timepoints). Changes in the ligament were similar to those seen in thesynovium, with a 500-fold increase in MMP-13 gene expressionrelative to baseline by five days after injury (P ¼ 0.0123), andremained upregulated by at least 1,000-fold at days 9 and 14(P ¼ 0.0123 and 0.0162, respectively). In the provisional scaffold, asignificant 80-fold increase in MMP-13 expressionwas seen at bothdays 9 and 14 (P ¼ 0.0342 and 0.0255, respectively).

ADAMTS-4 gene expression changes in the cartilage, synovium,ligament, and provisional scaffold over 14 days following injury

ADAMTS-4 gene expression in the articular cartilage did notchange significantly over the 2-week experiment [Fig. 1(C)]. In thesynovium, a 3-fold increase in ADAMTS-4 gene expression wasobserved from days 1 to 14 (P ¼ 0.0162, 0.0123, 0.0123, and 0.0123,respectively). In the ligament, a significant 2-fold upregulation ofgene expression was seen by day 5, and persisted throughout theremainder of the experiment (P ¼ 0.0123, 0.0123, 0.0162, respec-tively). In the provisional scaffold, a significant upregulation of 40-fold was seen on the first day after injury, and persisted throughoutthe entire experiment (P ¼ 0.0255, 0.0481, 0.0342, 0.0255,respectively).

Quantification of aggrecan and collagen fragments in synovial fluidand serum

The synovial fluid C2C levels increased by a factor of 1.5 in thefirst 5 days after injury and remained elevated at 14 days afterinjury, changes that were significant at both time points (P ¼ 0.041and 0.024, respectively; Fig. 2). CPII levels in the synovial fluid didnot change over time (P ¼ 0.32 and 0.079); nor did the C2C:CPIIratios (P ¼ 0.54 and 0.25). The synovial fluid CS 846 levels were3,600 ng/mL at baseline and over 10,000 ng/mL at day 14; however,this change was not statistically significant (P ¼ 0.10). Spikingstudies for C2C, CPII and CS846 resulted in a 97%, 93% and 86%recovery of the protein respectively from the synovial fluid. Thisrecovery rate is consistent with prior reports of the use of the IBEXELISA kit for analysis of synovial fluid24e26.

The C2C, CPII and CS 846 levels in porcine sera are presented inFig. 3. There was no significant increase in any of the serum bio-markers in the first 2 weeks following ACL injury.

Histological assessment of Safranin-O, apoptosis, and chondroitin-6-sulfate

The articular cartilage in the uninjured control knees appearedsmooth [OARSI score of 1.0, Fig. 4(A)], increasing to an averageOARSI score of 1.8 [Fig. 4(B)], primarily due to an increase in surfaceroughness of the articular cartilage rather than observation ofcellular changes or glycosaminoglycan loss. There was no observedchange in C-6-S, a marker for chondroitin sulfate, seen withimmunohistochemistry at the 2-week time point. While there wasno caspase 3 detected in the cartilage from intact knees [Fig. 4(E)],

active caspase 3 was detected in the superficial zone chondrocytesat day 14 [Fig. 4(F)].

Discussion

Results from this study convey the potential importance of anearly intervention strategy following ACL injury to minimizecartilage damage. MMP-1, MMP-13 and ADAMTS-4 gene expressionwere upregulated in the joint as early as 1 day following isolatedACL injury, and the amount of type II collagen fragments in thesynovial fluid nearly doubled within 5 days. While MMP-13 andADAMTS-4 were not significantly upregulated in the articularcartilage, these genes were upregulated in the adjacent synovium,transected ACL, and provisional scaffold formed at the injury site;tissues which had the potential to influence cartilage degradationby releasing these degradative enzymes into the synovial fluid. Thesignificant increase in collagen degradation fragment concentra-tions in the synovial fluid within the first 2 weeks after injurysuggest that this acute phase may be suited for interventions toprevent cartilage loss after ACL injury.

Our results indicate that all of the joint tissues in this study(articular cartilage, synovium, ligament and provisional scaffold)are capable of upregulating MMP-1 expression following ACLinjury. This observed increased expression of MMP-1 in the artic-ular cartilage is consistent with prior reports of MMP-1 geneexpression in cartilage of white rabbits subjected to ACL transec-tion27 and in humans with OA28. In addition, the synovium and ACLwere found to upregulate their gene expression of MMP-13. This isconsistent with prior reports of a 28-fold increase in MMP-13 geneexpression in the injured ligament after 1 week in a rabbit ACLtransection model and an upregulation in both the ACL and syno-vium in a rat ACL injurymodel after 1, 2, and 3 days29,30. In addition,the lack of upregulation of MMP-13 in the articular cartilage is alsoconsistent with prior reports of a lack of MMP-13 upregulation inthe articular cartilage in the first 7 days following destabilization ofthe medial meniscus in a murine model31.

The upregulation of bothMMP-1 andMMP-13 expression by theintra-articular tissues may be responsible for the increasing C2Cconcentration in the synovial fluid seen in the first few days afterACL injury. The increases in C2C concentration reported here aresimilar to the increased levels previously reported at day 5 afterinjury to the human knee3. While MMP-1 regulated cartilagedegradation may be due to the actions of chondrocytes as well assynoviocytes and fibroblasts, the MMP-13 regulated cartilagedegradation appears to be mediated by the synovium and injuredligament, rather than the articular cartilage. As MMP-13 is thoughtto be five to ten times more active in cleaving type II collagen thanMMP-1, it is likely that the synovium and ligament MMP-13 pro-duction plays a key role in the early cartilage degradation seen afterACL injury.

Aggrecan has been noted to be one of the first cartilage extra-cellular matrix proteins to undergo measurable loss after jointinjury, and thus is thought to be one of the early measures ofPTOA11,32. In this study, while we did not see increases in the geneexpression for ADAMTS-4 in the articular cartilage, we did see a 10-and 3-fold increase in gene expression in the synovium and ACL,respectively. These findings are consistent with those observedafter destabilization of the medial meniscus in the murine kneejoint, where ADAMTS-4 gene expression was noted to be elevatedat time points as early as 6 h after injury in whole joint prepara-tions, following removal of the skin and muscle31. In addition to theupregulation of ADAMTS-4, we also found a relative increase inaggrecan synthesis fragments (CS 846) within the synovial fluid atday 14 post-injury, though this increase was not statistically sig-nificant (P ¼ 0.15 and 0.10 for day 5 and 14, respectively). An

Fig. 2. Synovial fluid concentrations of C2C, CPII, C2C:CPII, and CS 846. The synovial fluid aspirated from unoperated knees at Day 0 serves as the baseline control. Sample sizeindicated in parentheses and asterisks indicate significant differences from control at P < 0.05.

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e19571954

increase in aggrecan synthesis fragments would be consistent withprior reports of a similar increase found in human patients withindays after injury, peaking at 1 week32,33. The increased geneexpression for ADAMTS-4 observed in the synovium and transectedACL and the associated production of aggrecanase synthesis frag-ments suggests that the initiation of proteoglycan loss in thearticular cartilage after injury may be due to actions within tissuesother than the articular cartilage.

The provisional scaffold is thought to lead the repair of extra-synovial ligaments, including the medial collateral ligament35.However, in this study, the provisional scaffold that forms after anintra-synovial ACL injury was found to serve a catabolic function aswell, with significant upregulation of MMP-13 and ADAMTS-4. Thiswould suggest it is a catabolic mediator of similar or greater sig-nificance than the synovium and ligament. This is the first report ofthe catabolic components of the response of the native provisionalscaffold within the joint. This finding is consistent with what isnoted clinically, where patients with recurrent hemarthroses arenoted to prematurely develop cartilage loss36,37. Increased expres-sions of MMPs have been observed in patients diagnosed withpigmented villonodular synovitis, which has been suggested toplay a critical role in the cartilage destruction observed clinically37.Osteochondral damage has also been detected via magnetic reso-nance imaging in hemophiliac patients experiencing bleedingwithin the joint36. This study suggests the provisional scaffoldformed after ACL injury may play a similar role in joint destruction,ultimately leading to cartilage degradation.

In prior studies of PTOA after ACL injury, there has been muchdebate over the root causeof this disease.While the impact damage istypically in the lateral compartment, the resultingOA typically occursin themedial compartment, suggesting it is a “whole joint” response,

rather than a site-specific response to local cartilage injury38. Cellularapoptosis has previously been observed at both the impaction site aswell as adjacent, non-impacted areas34. Apoptosis was accompaniedby progressive cartilage degeneration, even at these non-impactedsites. Additionally, delivery of a surfactant has been shown to delaythe expansion of cellular apoptosis, thereby protecting cartilageintegrity to a higher extent than inhibition of caspase 3 alone34. Re-sults of this study suggest a post-traumatic hemarthrosis (whichroutinely occurs after ACL injury) may thus be one of the factors thatinitiate the catabolic reaction of the joint and the journey to PTOA.Further studies to examine the role of the coagulation proteins andblood degradation products on cartilage metabolism may shed newlight on the mechanism of PTOA development.

There are several limitations for this study. First, an animalmodel was required in order to harvest the tissue and to evaluatethe changes over time. The porcinemodel was chosen in part due toits large size, which is larger and easier to access surgically (i.e.,more typical to that seen for humans) in comparison to smalleranimal models. The porcine model was also chosen because ofsimilarities in gait biomechanics and dependence on the ACL39e42.However, it is possible that the timing of cytokine release andprotease activity would be different in humans. Secondly, this studywas performed with only a 2-week follow-up period. While thiswas long enough to accomplish the initial goals of the study e

namely to determine the acute changes in the joint tissues after ACLinjury e it was not long enough to see macroscopic cartilagebreakdown. Finally, the joint is a complex structure comprised ofmultiple tissue types. Though we have performed qPCR analysis onseveral of these tissues, it is likely that each secretes proteins andother signaling molecules into the synovial fluid, which subse-quently interact with the other tissues.

Fig. 3. Serum concentrations of C2C, CPII, C2C:CPII, and CS 846. The synovial fluid aspirated from unoperated knees at Day 0 serves as the baseline control. No statistical differenceswere found across days (n ¼ 6 per analysis).

Fig. 4. Representative Safranin-O/Fast Green (A, B) and active caspase 3 (E,F) immu-nostaining in porcine cartilage from control (day 0 e A, C, E) and from day 14 (B, D, F)post-ACL transection. Day 14 cartilage specimens exhibited slightly rougher cartilagesurface (shown by arrows). Active caspase 3 staining (red, shown by arrows) wasdetected in cartilage specimens at days 0 and 14.

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e1957 1955

Patients who have an ACL tear are at increased risk of PTOA,whether or not they have an ACL reconstruction43. In this study, wedetermined that gene expression of a keymatrix metalloproteinaseand aggrecanase were upregulated in the adjacent joint synovium,transected ACL, and provisional scaffold, but not in the articularcartilage. Collagen fragments in the synovial fluid were increased atonly 5 days after injury, but the histologic changes appear to lagbehind, with only minor changes in surface roughness seen even at2 weeks after injury. These finding suggest that cartilage changesseen after ACL injury, even without impact trauma, may be medi-ated not just by chondrocytes, but also by other tissues of the joint,and thus the joint should be considered a complex organ with anorchestrated response to injury, rather than a set of individualtissues. These various joint tissues share the same synovial fluidenvironment, which makes determining which tissue is the keyplayer in cartilage degradation a challenge. However, the upregu-lation of catabolic factors seen in the torn ligament, synovium,provisional scaffold and articular cartilage after ACL injury suggeststhe response of these tissues should be considered when studyingmechanisms of PTOA development after an ACL injury.

Author contributions

Conception and design of the study: MMM, BCF, KAE, and CMH.Sample and data collection: BLP, CMH, and KAE.Statistical analysis: VMJ.Drafting of manuscript: MMM, BCF, KAE, VMJ, BLP, and CMH.Critical revision of manuscript: MMM, BCF, KAE, VMJ, BLP,

and CMH.Approval of final manuscript version: MMM, BCF, KAE, VMJ, BLP,

and CMH.

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e19571956

Funding sourcesThis investigation was supported by National Institutes of Healthunder NIAMS AR054099 (MMM), AR056834 (BCF/MMM), and theRuth L. Kirschstein National Research Service Award (F32AR061186) (CMH). Dr. Haslauer’s salary was also supported by Dr.Matthew Warman and the Children’s Orthopaedic Surgery Foun-dation. The content is solely the responsibility of the authors anddoes not necessarily represent the official views of the NationalInstitutes of Health.

Competing interestsThe authors do not have any conflicts of interest to report.

Acknowledgments

The authors thank the ARCH staff, Dr. Arthur Nedder, KathrynMullen, Dana Bolgen, and Courtney White, for their assistanceand care in handling the minipigs. We would also like to thankDrs. Adele Hill, Sebastian Kalamajski, and Justin Allen for theirassistance in qPCR design and Elise Magarian, Ryu Yoshida, andDr. Patrick Vavken for their assistance in surgery and tissuecollection.

References

1. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-termconsequence of anterior cruciate ligament and meniscus in-juries: osteoarthritis. Am J Sports Med 2007;35(10):1756e69.

2. Fleming BC, Oksendahl HL, Mehan WA, Portnoy R, Fadale PD,Hulstyn MJ, et al. Delayed Gadolinium-Enhanced MR Imagingof Cartilage (dGEMRIC) following ACL injury. OsteoarthritisCartilage 2010;18(5):662e7.

3. Catterall JB, Stabler TV, Flannery CR, Kraus VB. Changes inserum and synovial fluid biomarkers after acute injury(NCT00332254). Arthritis Res Ther 2010;12(6):R229.

4. Murray MM, Spindler KP, Devin C, Snyder BS, Muller J,Takahashi M, et al. Use of a collagen-platelet rich plasmascaffold to stimulate healing of a central defect in the canineACL. J Orthop Res 2006;24(4):820e30.

5. Murray MM, Magarian E, Harrison SL, Mastrangelo A,Zurakowski D, Fleming BC. The effect of skeletal maturity onfunctional healing of the anterior cruciate ligament. J BoneJoint Surg 2010;92(11):2039e49.

6. Tchetverikov I, Lohmander LS, Verzijl N, Huizinga TW,TeKoppele JM, Hanemaaijer R, et al. MMP protein and activitylevels in synovial fluid from patients with joint injury, in-flammatory arthritis, and osteoarthritis. Ann Rheum Dis2005;64(5):694e8.

7. 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(7):1534e45.

8. Burrage PS, Mix KS, Brinckerhoff CE. Matrix metal-loproteinases: role in arthritis. Front Biosci A J Virtual Libr2006;11:529e43.

9. Poole AR, Ionescu M, Fitzcharles MA, Billinghurst RC. Theassessment of cartilage degradation in vivo: development ofan immunoassay for the measurement in body fluids of type IIcollagen cleaved by collagenases. J Immunological Methods2004;294(1e2):145e53.

10. Nelson F, Dahlberg L, Laverty S, Reiner A, Pidoux I, Ionescu M,et al. Evidence for altered synthesis of type II collagen in pa-tients with osteoarthritis. J Clin Invest 1998;102(12):2115e25.

11. Karsdal MA, Madsen SH, Christiansen C, Henriksen K,Fosang AJ, Sondergaard BC. Cartilage degradation is fully

reversible in the presence of aggrecanase but not matrixmetalloproteinase activity. Arthritis Res Ther 2008;10(3):R63.

12. Poole AR, Ionescu M, Swan A, Dieppe PA. Changes in cartilagemetabolism in arthritis are reflected by altered serum andsynovial fluid levels of the cartilage proteoglycan aggrecan.Implications for pathogenesis. J Clin Invest 1994;94(1):25e33.

13. Lohmander LS, Ionescu M, Jugessur H, Poole AR. Changes injoint cartilage aggrecan after knee injury and in osteoarthritis.Arthritis Rheum 1999;42(3):534e44.

14. Kraus VB, Birmingham J, Stabler TV, Feng S, Taylor DC,Moorman 3rd CT, et al. Effects of intraarticular IL1-Ra for acuteanterior cruciate ligament knee injury: a randomizedcontrolled pilot trial (NCT00332254). Osteoarthritis Cartilage2012;20(4):271e8.

15. Murray MM, Spindler KP, Abreu E, Muller JA, Nedder A,Kelly M, et al. Collagen-platelet rich plasma hydrogel enhancesprimary repair of the porcine anterior cruciate ligament.J Orthop Res 2007;25(1):81e91.

16. Magarian EM, Fleming BC, Harrison SL, Mastrangelo AN,Badger GJ, Murray MM. Delay of 2 or 6 weeks adversely affectsthe functional outcome of augmented primary repair of theporcine anterior cruciate ligament. Am J Sports Med2010;38(12):2528e34.

17. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by thecomparative C(T) method. Nat Protoc 2008;3(6):1101e8.

18. Frantz NZ, Friesen KG, Andrews GA, Tokach MD, Yamka RM,Loughin TL, et al. Use of serum biomarkers to predict thedevelopment and severity of osteochondrosis lesions in thedistal portion of the femur in pigs. Am J Vet Res 2010;71(8):946e52.

19. Hembry RM, Dyce J, Driesang I, Hunziker EB, Fosang AJ,Tyler JA, et al. Immunolocalization of matrix metal-loproteinases in partial-thickness defects in pig articularcartilage. A preliminary report. J Bone Jt Surg Am Volume2001;83-A(6):826e38.

20. Fosang AJ, Last K, Stanton H, Weeks DB, Campbell IK,Hardingham TE, et al. Generation and novel distribution ofmatrix metalloproteinase-derived aggrecan fragments inporcine cartilage explants. J Biol Chem2000;275(42):33027e37.

21. Pritzker KP, Gay S, Jimenez SA, Ostergaard K, Pelletier JP,Revell PA, et al. Osteoarthritis cartilage histopathology:grading and staging. Osteoarthritis Cartilage 2006;14(1):13e29.

22. Jay GD, Fleming BC, Watkins BA, McHugh KA, Anderson SC,Zhang LX, et al. Prevention of cartilage degeneration andrestoration of chondroprotection by lubricin tribosupple-mentation in the rat following anterior cruciate ligamenttransection. Arthritis Rheum 2010;62(8):2382e91.

23. Hornik K. The R FAQ, http://CRANR-projectorg/doc/FAQ/R-FAQhtml. 2012; ISBN: 3-900051-08-9.

24. Matyas JR, Atley L, Ionescu M, Eyre DR, Poole AR. Analysis ofcartilage biomarkers in the early phases of canine experi-mental osteoarthritis. Arthritis Rheum 2004;50(2):543e52.

25. Chu Q, Lopez M, Hayashi K, Ionescu M, Billinghurst RC,Johnson KA, et al. Elevation of a collagenase generated type IIcollagen neoepitope and proteoglycan epitopes in synovialfluid following induction of joint instability in the dog. Oste-oarthritis Cartilage 2002;10(8):662e9.

26. Kojima T, Mwale F, Yasuda T, Girard C, Poole AR, Laverty S.Early degradation of type IX and type II collagen with theonset of experimental inflammatory arthritis. Arthritis Rheum2001;44(1):120e7.

27. BluteauG, Conrozier T,Mathieu P, Vignon E, HerbageD,Mallein-Gerin F. Matrix metalloproteinase-1, -3, -13 and aggrecanase-1

C.M. Haslauer et al. / Osteoarthritis and Cartilage 21 (2013) 1950e1957 1957

and -2 are differentially expressed in experimental osteoar-thritis. Biochim biophys acta 2001;1526(2):147e58.

28. Wu W, Billinghurst RC, Pidoux I, Antoniou J, Zukor D,Tanzer M, et al. Sites of collagenase cleavage and denaturationof type II collagen in aging and osteoarthritic articular cartilageand their relationship to the distribution of matrix metal-loproteinase 1 and matrix metalloproteinase 13. ArthritisRheum 2002;46(8):2087e94.

29. Attia E, Brown H, Henshaw R, George S, Hannafin JA. Patternsof gene expression in a rabbit partial anterior cruciate liga-ment transection model: the potential role of mechanicalforces. Am J Sports Med 2010;38(2):348e56.

30. Tang Z, Yang L, Zhang J, Xue R, Wang Y, Chen PC, et al. Coor-dinated expression of MMPs and TIMPs in rat knee intra-articular tissues after ACL injury. Connect Tissue Res2009;50(5):315e22.

31. Burleigh A, Chanalaris A, Gardiner MD, Driscoll C, Boruc O,Saklatvala J, et al. Joint immobilization prevents murine oste-oarthritis and reveals the highly mechanosensitive nature ofprotease expression in vivo. Arthritis Rheum 2012;64(7):2278e88.

32. Sward P, Frobell R, Englund M, Roos H, Struglics A. Cartilageand bone markers and inflammatory cytokines are increasedin synovial fluid in the acute phase of knee injury (hemarth-rosis) e a cross-sectional analysis. Osteoarthritis Cartilage2012;20(11):1302e8.

33. Lohmander LS, Atley LM, Pietka TA, Eyre DR. The release ofcrosslinked peptides from type II collagen into human synovialfluid is increased soon after joint injury and in osteoarthritis.Arthritis Rheum 2003;48(11):3130e9.

34. Anderson DD, Chubinskaya S, Guilak F, Martin JA, Oegema TR,Olson SA, et al. Post-traumatic osteoarthritis: improved un-derstanding and opportunities for early intervention.J Orthopaedic Res Official Publ Orthopaedic Res Soc2011;29(6):802e9.

35. Chamberlain CS, Crowley EM, Kobayashi H, Eliceiri KW,Vanderby R. Quantification of collagen organization andextracellular matrix factors within the healing ligament.Microsc Microanal 2011;17(5):779e87.

36. Kraft J, Blanchette V, Babyn P, Feldman B, Cloutier S,Israels S, et al. Magnetic resonance imaging and joint out-comes in boys with severe hemophilia A treated withtailored primary prophylaxis in Canada. J Thromb Haemost2012;10(12):2494e502.

37. Uchibori M, Nishida Y, Tabata I, Sugiura H, Nakashima H,Yamada Y, et al. Expression of matrix metalloproteinases andtissue inhibitors of metalloproteinases in pigmented villo-nodular synovitis suggests their potential role for jointdestruction. J Rheumatol 2004;31(1):110e9.

38. Potter HG, Jain SK, Ma Y, Black BR, Fung S, Lyman S. Cartilageinjury after acute, isolated anterior cruciate ligament tear:immediate and longitudinal effect with clinical/MRI follow-up.Am J Sports Med 2012;40(2):276e85.

39. Andriacchi TP, Briant PL, Bevill SL, Koo S. Rotational changes atthe knee after ACL injury cause cartilage thinning. Clin OrthopRelat Res 2006;442:39e44.

40. Levy AS, Meier SW. Approach to cartilage injury in the anteriorcruciate ligament-deficient knee. Orthop Clin North Am2003;34(1):149e67.

41. Sharma L, Lou C, Felson DT, Dunlop DD, Kirwan-Mellis G,Hayes KW, et al. Laxity in healthy and osteoarthritic knees.Arthritis Rheum 1999;42(5):861e70.

42. Seitz H, Hausner T, Schlenz I, Lang S, Eschberger J. Vascularanatomy of the ovine anterior cruciate ligament. A macro-scopic, histological and radiographic study. Arch OrthopTrauma Surg 1997;116(1e2):19e21.

43. Oiestad BE, Holm I, Engebretsen L, Aune AK, Gunderson R,Risberg MA. The prevalence of patellofemoral osteoarthritis 12years after anterior cruciate ligament reconstruction. KneeSurg Sports Traumatol Arthrosc 2013;21(4):942e90.