Osteoarthritis and Cartilage 20 (2012) 940e948

Transcription

1 Osteoarthritis and Cartilage 20 (2012) 940e948 The impact of forced joint exercise on lubricin biosynthesis from articular cartilage following ACL transection and intra-articular lubricin s effect in exercised joints following ACL transection K.A. Elsaid y, L. Zhang z, K. Waller x, J. Tofte z, E. Teeple k, B.C. Fleming xk, G.D. Jay zx y Department of Pharmaceutical Sciences, Massachusetts College of Pharmacy and Health Sciences, Boston, MA 02115, USA z Emergency Medicine, Rhode Island Hospital, Providence, RI 02903, USA x Department of Engineering, Brown University, Providence, RI 02903, USA k Department of Orthopaedics, Rhode Island Hospital, Providence, RI 02903, USA article info summary Article history: Received 27 July 2011 Accepted 28 April 2012 Keywords: Lubricin Cartilage degeneration ACL Forced exercise Objective: To evaluate the impact of forced joint exercise following acute knee injury on lubricin metabolism and its relationship to cartilage degeneration and to assess chondroprotection of a singledose purified human lubricin injection in exercised injured joints. Methods: Anterior cruciate ligament transection (ACLT) was performed in rats with six experimental groups; 3-week post-aclt, 3-week post-aclt þ exercise, 5-week post-aclt, 5-week post- ACLT þ exercise, and 5-week post-aclt þ exercise treated with intra-articular phosphate buffered saline (PBS) or lubricin. Joint exercise was achieved using a rotating cylinder at a speed of 6 rpm for 30 min daily, 5 days a week starting 1 week following surgery. Cartilage lubricin expression in injured joints was determined. Histological analyses included Safranin O/Fast Green, activated caspase-3, and lubricin mrna in-situ hybridization. Assessment of cartilage damage was performed by osteoarthritis research society international (OARSI) modified Mankin scoring and urinary CTXII (uctxii) levels. Results: At 3 weeks, lubricin expression in exercised ACLT joints was significantly (P < 0.001) lower compared to ACLT joints. The OARSI scores were significantly (P < 0.001) higher in the ACLT þ exercise animals compared to ACLT animals at 5 weeks. Compared to 3-week ACLT, 3-week ACLT þ exercise cartilage showed increased caspase-3 staining. Compared to ACLT þ exercise and PBS-treated ACLT þ exercise, lubricin intra-articular treatment resulted in a significant increase (P < 0.001) in cartilage lubricin gene expression and a reduction (P < 0.05) in uctxii levels. Conclusion: Joint exercise resulted in decreased lubricin cartilage expression, increased cartilage degeneration and reduced superficial zone chondrocyte viability in the ACLT joint. Intra-articular lubricin administration ameliorated cartilage damage due to exercise and preserved superficial zone chondrocytes viability. Ó 2012 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Introduction Acute knee injuries involving cruciate ligament tears, is a significant risk factor for development of post-traumatic osteoarthritis (OA) 1e3. The mechanism of cartilage degeneration following an acute injury is multifaceted and may involve joint instability, inflammation and possibly loss of joint lubrication 4,5. Address correspondence and reprint requests to: K.A. Elsaid, Department of Pharmaceutical Sciences, Massachusetts College of Pharmacy and Health Sciences, Boston, MA 02115, USA. Tel: address: khaled.elsaid@mcphs.edu (K.A. Elsaid). Joint lubrication is provided in part by lubricin/proteoglycan 4 (PRG4), a glycoprotein secreted from synoviocytes 6 and superficial zone chondrocytes 7. In addition to its boundary lubricating properties 8,9, lubricin prevents synoviocyte overgrowth, protects cartilage surfaces and prevents cartilage wear 10,11. Patients with anterior cruciate ligament (ACL) injuries are typically prescribed an exercise regimen as part of their rehabilitation process. However, the impact of joint exercise following ACL injury on long-term cartilage integrity, superficial chondrocyte viability and joint lubrication mediated by lubricin is not understood. In this investigation, we utilized the rat ACL injury model 12e14 to evaluate the impact of an exercise regimen on lubricin cartilage biosynthesis and its relationship to superficial zone /$ e see front matter Ó 2012 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved. doi: /j.joca

2 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e chondrocytes viability and cartilage degeneration. We hypothesized that joint exercise following injury will modulate lubricin metabolism leading to accelerating cartilage degeneration. We also posit that lubricin expression by chondrocytes will be enhanced through the intra-articular lubricin supplementation which will serve to prevent cartilage degeneration compared to non-lubricintreated counterparts. Methods ACL transection (ACLT) in the rat ACLT was performed in 8e10 weeks old male Lewis rats (n ¼ 144) which were randomly assigned to six experimental groups; (1) 3-week post-aclt (n ¼ 24), (2) 3-week post- ACLT þ exercise (n ¼ 24), (3) 5-week post-aclt (n ¼ 24), (4) 5-week post-aclt þ exercise group (n ¼ 24), (5) 5-week post- ACLT þ exercise treated with one intra-articular sham injection of phosphate buffered saline (PBS) (n ¼ 24) and (6) 5-week post- ACLT þ exercise treated with one intra-articular injection of purified human lubricin (n ¼ 24). Following anesthesia with intraperitoneal Ketamine and Dexmedetomidine, the right knee joint skin was shaved, cleaned with a topical antiseptic, and a lateral skin incision was performed to gain access to the joint capsule. After the joint capsule was opened, the ACL was severed using a surgical scalpel. A positive anterior draw test confirmed the transection of the ACL. Closure of the joint capsule was performed with biodegradable sutures and the animal skin was finally closed using surgical staples. In all animals, the right knee joint was the ACLT joint. All surgeries were performed by KE and all required approvals were obtained from MCPHS IACUC committee prior to the commencement of the study. Control animals (n ¼ 5) were age and sex-matched to the operated animals and were used for histological comparisons. Forced joint exercise Consistent doses of joint exercise were achieved in the animals assigned to the exercise groups by placing each rat on a custom rotating cylinder apparatus that rotated toward the animal, at a speed of 6 rpm for 30 min daily, 5 days a week starting 1 week following surgery. Animals were trained for 1 week on the rotating cylinder prior to ACLT. This exercise regimen was used to simulate repeated joint flexion and extension. Intra-articular lubricin and PBS (sham) injections One week following ACLT, animals were anesthetized with isoflurane and received intra-articular injections of purified human lubricin, obtained from patients undergoing knee and hip joint replacement as previously described 6, at a dose of 40 mg in50ml or PBS. Animals were injected through the patellar tendon of the operated knee joint. Quantitative lubricin expression Immediately following joint harvest, tibial plateau cartilage was carefully extracted from operated and contralateral joints of ACLT animals at 3 and 5 weeks following surgery, with or without exercise and from animals that received PBS or lubricin (n ¼ 9in each group). Tibial plateau cartilage was immediately snap-frozen and stored at 80 C until RNA isolation. Tibial cartilage tissue was pulverized using a mortar and pestle. Total RNA was isolated from synovial tissue, using a Ribopure kit (Ambion, Austin, TX, USA). Contaminating genomic DNA was removed by DNase I treatment of isolated RNA. Messenger RNA was converted to cdna using a Cells-to-cDNA II kit (Ambion). Quantitative polymerase chain reaction (PCR) assays were performed as previously described 15 using lubricin 5 0 -GGAACCGATCTCTTGGTTGA-3 0 (forward), 5 0 -ATCCACTGGCTTACCATTGC-3 0 (reverse) and GAPDH 5 0 -CAGTGCCAGCCTCGTCTCAT-3 0 (forward) and 5 0 -AGGGGCCATC CACAGTCTTC-3 0 (reverse) primers. In-situ lubricin hybridization Representative articular cartilage specimens (n ¼ 9 in each group) were sectioned and probed using a 373-bp fragment corresponding to nucleotides 1283e1655 of the rat PRG4 NCBI RGD: gene. Random-primed digoxigenin (DIG)-labeled probe was used and the hybridized probes were immunodetected with anti-dig-fluorescein Fab fragments (Roche Applied Science) as previously described 16e18. Fluorescence microscopy was utilized to image the cartilage specimens using a Nikon E800 microscope using a 40 Plan Fluor objective and a Spot II digital camera (Diagnostic Instruments). All images were acquired with the same camera settings. Image processing and analysis were performed using ivision (Biovision, Exton, PA, USA) image analysis software. Positive staining was defined through intensity thresholding and integrated optical density (IOD) was calculated by examining the thresholded area multiplied by the mean 13. Lubricin concentrations in synovial fluid (SF) lavages Lavaging of SF from ACLT joints at 3 or 5 weeks ACLT with or without exercise as well as 5-week ACLT joints that received intraarticular PBS or lubricin injections was performed by injecting 100 ml of normal saline in the joint capsule followed by flexing and extending the articular joint for 10 times. A total of 30 mloffluid was aspirated. Lubricin SF lavage concentrations were determined using a sandwich enzyme-linked immunosorbent assay (ELISA) using peanut agglutinin (PNA; SigmaeAldrich) and monoclonal antibody 9G3 as previously described 12. Histological analysis and immunostaining Paraffin-embedded coronal sections were taken from weightbearing areas of the articular cartilage of ACL transected joints of each animal (n ¼ 6 in each group). Microtomed sections were collected every 250 mm to find a representative area showing both femoral condyles, tibial plateaus, and the menisci. Adjacent sections were collected through this region and stained with Safranin O/Fast green for histological scoring and assessment of sulfated glycosaminoglycans (sgag) content, immunoprobed with antibody 9G3, a monoclonal antibody specific for lubricin (provided by M. Warman) to evaluate the presence of lubricin, and rabbit polyclonal antibody against active caspase-3 (Abcam, Cambridge, MA, USA) to assess chondrocyte apoptosis. mab 9G3 was incubated with the histological specimens at 1:200 dilutions followed by biotinylated anti-mouse IgG at 1:500 dilution and detected using the Vectastain ABC kit (VECTOR Laboratories, Burlingame, CA, USA). Activated caspase-3 antibody was incubated with the histological specimens at 1:50 dilution followed by biotinylated anti-rabbit IgG at 1:200 dilution and detected using the Vectastain ABC kit. Histological scoring The osteoarthritis research society international (OARSI) modified Mankin scoring was utilized to assess cartilage degeneration in the different groups 19. We did not examine osteophyte formation as part of reported scores. Scoring was performed on the medial and

3 942 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e948 lateral tibial plateau cartilage (n ¼ 6 in each group). At least three blinded investigators scored Safranin-O stained cartilage specimens and a consensus score was reported. Histological cartilage surface roughness analysis Digital images of the lateral and medial tibial plateaus were acquired at 20x magnification from Safranin-O stained rat knee joint sections and processed using Photoshop (Adobe, San Jose, CA, USA). A line image describing the cartilage surface was obtained and further processed using Matlab (Mathworks, Natick, MA, USA) to yield a single-pixel-thick, contiguous curve. A third degree polynomial approximation was fitted to the curve. Head-to-tail piecewise vectors were fitted to the original curve and the polynomial approximation. The magnitude of the cross product of each corresponding pair of vectors was calculated. This quantity was multiplied by the absolute value of the vertical distance between the midpoints of each vector pair. The root mean square (RMS) of this unitless value was calculated for all pairs of vectors. Sections of the cartilage boundary that had greater slopes than and/or deviated vertically from the polynomial approximation thus contributed to an increase in the RMS roughness value for the tibial plateau overall. Determination of urinary CTXII (uctxii) levels in urine specimens from ACLT animals for different groups On days 20 and 34 following ACLT of exercised and nonexercised animals and on day 34 following ACLT and intraarticular injection of lubricin or PBS, animals were housed in metabolic cages, and 24-h urine collections were performed. Urine samples were subsequently centrifuged at 3,000 rpm for 20 min and stored at 20 C. uctxii concentrations were determined using the Pre-clinical Urine CartiLaps ELISA (Immunodiagnostic Systems). Urinary creatinine was determined using the QuantiChrom creatinine assay (BioAssay Systems). uctxii levels were normalized to urinary creatinine. Statistical analyses Relative lubricin expression at 3 and 5 weeks following ACLT, lubricin SF lavage concentrations, fluorescence intensity of lubricin fluorescence in-situ hybridization (FISH), and OARSI modified Mankin score in ACLT and ACLT þ exercise, uctxii in ACLT and ACLT þ exercise as well as the PBS-treated and lubricin-treated joints were initially tested for equal variance and normality. Variables that passed the equal variance and normality tests were compared using one-way analysis of variance (ANOVA) with Fisher s least significant difference (LSD) test for pairwise comparisons. Analyses that included data with non-normally distributed residuals were performed using KruskaleWallis ANOVA on the ranks with Tukey s test for pairwise comparisons. Statistical significance was set at a ¼ 0.05 a priori. Unless otherwise reported, mean is presented as mean standard deviation (SD). All statistical tests were performed using Sigma Plot (version 11.0; San Jose, CA, USA). When appropriate, data are presented as scatter plots with the median value highlighted. Results Effects of time and exercise following ACLT Lubricin gene expression in ACLT joints at 3 and 5 weeks following ACLT was lower than lubricin expression in contralateral joints as determined by an average relative lubricin expression ratio of 0.59 and 0.28, respectively [Fig. 1(A)]. In exercised ACLT animals, lubricin cartilage expression in ACLT joints was lower than lubricin expression in contralateral joints at 3 and 5 weeks as determined by an average lubricin expression ratio of 0.34 and 0.28, respectively. At 3 weeks, relative lubricin expression in exercised ACLT joints was significantly (P < 0.001) lower than relative lubricin expression in ACLT joints. There was no significant difference (P ¼ 0.35) between relative lubricin expression in 5-week exercised ACLT joints and relative lubricin expression in 5-week ACLT joints. The SF lavage lubricin concentrations in the 3-week ACLT joints were significantly (P ¼ 0.017) higher compared to 3-week ACLT þ exercise joints, 5-week ACLT (P ¼ 0.002) and 5-week ACLT þ exercise (P ¼ 0.003) [Fig. 1(B)]. Lubricin immunostaining and lubricin in-situ hybridization of representative cartilage specimens is presented in Fig. 1(C) top and bottom panels, respectively. Control cartilage showed lubricin deposition on the surface of articular cartilage and in the superficial zone cartilage (A; top panel). Compared to control, cartilage from ACLT joints at 3 and 5 weeks following ACLT showed minimal lubricin deposition on the surface of articular cartilage and lack of lubricin in the superficial zone [Fig. 1(B and D; top panel)]. The exercised ACLT joints resulted in greater reduction in lubricin staining intensity on the surface of articular cartilage at 3 weeks [Fig. 1(C; top panel)] compared to ACLT joints at 3 weeks (B; top panel). At 5 weeks following ACLT and with exercise, no lubricin staining was detected on the surface of articular cartilage (E; top panel). The mean IOD of lubricin in-situ hybridization in control cartilage specimens (n ¼ 5) was calculated and compared to mean IOD in the 3 and 5-week ACLT with or without exercise groups (n ¼ 9 in each group), respectively. The mean IOD was calculated using four randomly selected fields from each cartilage specimen, and was averaged across specimens. The mean IOD in control animals was The mean IOD in ACLT joints at 3 weeks was , while the mean IOD in the ACLT joints with exercise at 3 weeks was The mean IOD in ACLT joints at 5 weeks was , while the mean IOD in the ACLT joints with exercise at 5 weeks was The mean IOD values in ACLT animals with or without exercise were significantly (P < 0.001) lower than the mean IOD value of control animals, indicating decreased lubricin mrna levels in the superficial zone of articular cartilage as a result of injury. The mean IOD value of 3-week ACLT þ exercise animals was significantly (P < 0.001) lower than mean IOD value of 3-week ACLT. There was no significant difference (P ¼ 0.39) between the mean IOD of 5-week ACLT and 5-week ACLT þ exercise animals. The OARSI modified Mankin scores of animals that underwent ACLT with or without exercise at 5 weeks is presented in Fig. 2(A). The OARSI modified Mankin scores (n ¼ 6 in each group) were significantly (P < 0.001) higher in the ACLT þ exercise group compared to the ACLT group. There was no significant difference (P ¼ 0.24) in 24 h uctxii levels in the 3-week ACLT þ exercise group and the 3-week ACLT group. The uctxii levels in the 5-week ACLT þ exercise group were significantly (P < 0.001) higher than the uctxii levels in the 5-week ACLT group (n ¼ 15 in each group). Representative cartilage specimens of caspase-3 immunostaining and Safranin O/fast Green from control, 3-week ACLT, 3-week ACLT þ exercise, 5-week ACLT and 5-week ACLT þ exercise are presented in Fig. 2(C). Control cartilage demonstrates lack of activated caspase-3 staining, normal cartilage surface integrity and strong proteoglycans staining. Compared to 3-week ACLT, 3-week ACLT þ exercise cartilage shows a greater number of caspase-3 positive chondrocytes and loss of proteoglycans staining in the superficial zone, extending into the middle zone of articular cartilage. Compared to 3-week ACLT, 5-week ACLT showed increased activated caspase-3 staining and increased cartilage degeneration,

4 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e Fig. 1. Quantitative lubricin expression of articular cartilage, lubricin SF lavage concentrations, lubricin immunostaining and in-situ hybridization of lubricin mrna in joints that underwent ACLT with or without exercise. (A) Relative lubricin cartilage expression in ACLT joints compared to contralateral joints normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in animals that underwent ACLT with or without exercise (n ¼ 9 in each group). Indicates that lubricin cartilage expression in 3-week ACLT joints was significantly (P < 0.001) higher compared to lubricin cartilage expression in 3-week ACLT þ exercise, 5-week ACLT and 5-week ACLT þ exercise joints. (B) SF lavage lubricin concentrations in ACLT joints at 3 and 5 weeks with or without exercise (n ¼ 9 in each group). Indicates that lubricin concentrations in the 3-week ACLT joints were significantly (P ¼ 0.017) higher compared to 3-week ACLT þ exercise joints, 5-week ACLT (P ¼ 0.002) and 5-week ACLT þ exercise (P ¼ 0.003). (C) mab 9G3 immunostaining for lubricin (top panel) and in-situ hybridization of lubricin mrna (bottom panel) of representative articular cartilage from control (a), week-3 ACLT (b), week-3 ACLT þ exercise (c), week-5 ACLT (d) and week-5 ACLT þ exercise (e). Lubricin surface accumulation was detected in control, week-3 and week-5 ACLT and in the superficial zone of control cartilage. Lubricin surface accumulation was absent in cartilage of animals that underwent exercise following ACLT. In-situ hybridization showed lubricin expression in superficial zone chondrocytes of control, week-3 ACLT and to a lesser extent in week-3 ACLT þ exercise cartilage and lack of lubricin expression in week-5 ACLT and week-5 ACLT þ exercise cartilage. exemplified by loss of proteoglycans staining in the superficial and middle cartilage layers. The 5-week ACLT þ exercise cartilage demonstrated hypocellularity in the different cartilage layers and virtually all chondrocytes stained positive for activated caspase-3. The Safranin O/Fast Green staining demonstrates loss of proteoglycans in the superficial and middle zone with clefts extending into the middle layers. The histological analysis of cartilage roughness is presented in Fig. 3. The RMS roughness values significantly (P < 0.05) greater in the 3-week ACLT þ exercise group than in the 3-week ACLT group (n ¼ 4 in each group). Comparison of these two groups with the control (n ¼ 2) did not reveal significant differences (P ¼ 0.26) between groups. The RMS roughness value was also significantly (P < 0.05) higher in the 5-week ACLT þ exercise group than in the 5-week ACLT group (n ¼ 4 for ACLT and n ¼ 6 for ACLT þ exercise). PBS vs lubricin injection The impact of intra-articular lubricin injection on lubricin gene expression, SF lubricin concentration, cartilage degeneration and activated caspase-3 positivity in ACLT and exercised joints are presented in Fig. 4. Compared to ACLT þ exercised animals and PBSinjected ACLT þ exercised animals, intra-articular lubricin injection resulted in a significant increase (P < 0.001) in lubricin cartilage gene expression [Fig. 4(A)]. PBS injection did not result in a change in cartilage lubricin expression, indicating no placebo effect on lubricin expression. SF lavage lubricin concentrations in the lubricin group were significantly higher than lubricin concentrations in the PBS group (P ¼ 0.012) and the ACLT group (P < 0.001)[Fig. 4(B)]. The OARSI modified Mankin scores were significantly lower (P < 0.001) in the PBS and lubricin-treated groups (n ¼ 6 in each group). There was no significant difference (P ¼ 0.13) between the OARSI scores of the PBS and lubricin groups [Fig. 4(C)]. The uctxii levels in lubricintreated ACLT þ exercise animals were significantly lower (P < 0.05) than uctxii levels in the PBS-treated or untreated animals [Fig. 4(D)]. The mean IOD value of 5-week ACLT þ exercise joints was The mean IOD value of PBS-treated 5-week ACLT þ exercise joints was and the mean IOD value of lubricin-treated 5-week ACLT þ exercise joints was [Fig. 4(E)]. The mean IOD value of lubricin-treated joints was significantly higher (P < 0.001) than PBS-treated or untreated joints (n ¼ 9 in each group). The mean IOD value in the PBS-treated group was significantly (P ¼ 0.03) higher than the untreated group. The

5 944 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e948 Fig. 2. Cartilage degeneration and activated caspase-3 cartilage immunostaining from animals that underwent ACLT with or without exercise. (A) OARSI modified Mankin scores of articular cartilage from animals that underwent ACLT followed by exercise or ACLT alone at week 5 following ACLT (n ¼ 6 in each group). Indicates that cartilage degeneration scores were significantly (P < 0.001) higher in the ACLT þ exercise group compared to ACLT. (B) uctxii levels normalized to urinary creatinine collected over 24 h prior to joint harvest of animals that underwent ACLT with or without exercise and harvested at 3 and 5 weeks (n ¼ 15 in each group). Indicates that uctxii levels from week-5 ACLT þ exercise animals were significantly (P < 0.001) higher than uctxii levels from week-5 ACLT, week-3 ACLT þ exercise and week-3 ACLT. (C) Activated caspase-3 immunostaining (top panel) and Safranin O/Fast Green (bottom panel) of representative articular cartilage from control (a), week-3 ACLT (b), week-3 ACLT þ exercise (c), week-5 ACLT (d) and week-5 ACLT þ exercise (e). Control cartilage shows uniform Safranin O staining and lack of activated caspase-3 staining. Week-3 ACLT þ exercise and week-5 ACLT þ exercise show stronger activated caspase-3 staining and greater loss of proteoglycans staining compared to week-3 ACLT and week-5 ACLT, respectively. activated caspase-3 immunostaining of representative control, 5-week ACLT þ exercise, lubricin-treated 5-week ACLT þ exercise and PBS-treated 5-week ACLT þ exercise cartilage is presented in Fig. 4(E). Control cartilage showed no activated caspase-3 staining. ACLT þ exercise cartilage showed intense caspase-3 staining. Similarly, PBS-treated ACLT þ exercise joints demonstrated activated caspase-3 staining. Lubricin-treated joints showed no activated caspase-3 staining in articular chondrocytes. The mean cartilage RMS roughness value of PBS-treated 5-week ACLT þ exercise animals (n ¼ 6) was compared to in lubricin-treated 5-week ACLT þ exercise (n ¼ 6) and in untreated 5-week ACLT þ exercise animals (n ¼ 6). RMS roughness values were not significantly higher (P ¼ 0.09) in the PBS-treated compared to the lubricin-treated joints. Discussion Acute joint injury models have demonstrated an association between decreased lubricin expression and its accumulation on the surface of articular cartilage and cartilage degeneration 4,20e22. Lubricin is down-regulated by inflammatory cytokines, e.g., interleukin-1 alpha (IL-1a) and tumor necrosis factor alpha (TNF-a) and is upregulated by transforming growth factor-beta (TGF-b) 23,24. Lubricin is also upregulated by mechanical compression and shear under homeostatic conditions in cartilage explant cultures 25,26 and in whole joints 27. In this study, exercise of ACL transected joints conducted in vivo resulted in down-regulation of lubricin cartilage expression at an early stage following the injury. The reduction in lubricin expression was mirrored with a lowering of lubricin SF concentrations as well as its deposition on the cartilage surface, as evidenced by ELISA and immunohistochemical analysis, respectively. The reduction in lubricin expression following exercise was particularly evident in the superficial zone chondrocytes by low levels of lubricin mrna detected by in-situ hybridization. Joint exercise following ACL injury resulted in an increase in the extent of cartilage damage as determined by an increase in OARSI histological scores and elevation of uctxii levels at 5 weeks following injury. The impact of exercise on cartilage health depends on the extent of injury, the method of exercise and the intensity of the exercise regimen. While a running wheel exercise protocol was found to protect articular cartilage and stimulate cartilage matrix synthesis in an uninjured hamster model 28, a forced joint exercise regimen similar to the one employed in this study accelerated cartilage degeneration in a rat meniscectomy model 29. Strenuous joint exercise has resulted in cartilage degeneration similar to early OA changes in normal rats 30, while a mild exercise regimen following ACLT provided a protective effect on articular cartilage which was lost with intense exercise 31. In our model, a time-

6 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e A B a b c d e Control 3 Week ACLT 3 Week ACLT + Exercise 5 Week ACLT 5 Week ACLT+ Exercise RMS Value Fig. 3. Cartilage surface roughness of control, ACLT and ACLT þ exercise animals. (A) RMS values of cartilage roughness (mean and standard error) (top panel) of control, 3-week ACLT, 5-week ACLT, 3-week ACLT þ exercise and 5-week ACLT þ exercise animals. Indicates that 5-week ACLT cartilage was significantly (P < 0.05) rougher than 3-week ACLT cartilage and week-5 ACLT þ exercise cartilage was significantly (P < 0.05) rougher than week-3 ACLT þ exercise. Indicates that a one-way ANOVA comparing control, week-5 ACLT, and week-5 ACLT þ exercise cartilage showed significant (P < 0.01) differences among groups. Control, 3-week ACLT, and 3-week ACLT þ exercise cartilage did not show significant differences. (B) Safranin O staining of representative articular cartilage (bottom panel) with extracted cartilage boundary curve (right). (a) Control, (b) 3-week ACLT, (c) 3-week ACLT þ exercise, (d) 5-week ACLT and (e) 5-week ACLT þ exercise. dependent effect of joint exercise on cartilage degeneration is exemplified by increased uctxii release, OARSI scores and cartilage surface roughness. The increased cartilage degeneration at 5 weeks following injury and exercise was preceded by activation of caspase-3 in chondrocytes of superficial and middle cartilage zones. Caspase-3 is an effector intracellular enzyme in cellular apoptotic pathways. Apoptosis plays a significant role in chondrocyte cell death during the pathogenesis of OA 32e34, and is considered a key early event of cartilage pathogenesis following acute injury 35. Furthermore, inhibition of cartilage caspase activity in experimental OA models was shown to reduce the extent of cartilage degradation 36. The early increase in caspase-3 activity following exercise indicates that superficial zone and middle zone chondrocytes are actively undergoing apoptosis and explains the observed reduction of lubricin biosynthesis and SF concentrations. Additionally, this process contributes to the increased cartilage degradation by interfering with repair processes mediated by chondrocytes. In our model, increased apoptosis of chondrocytes as a response to exercise following the ACL injury can be explained by a combination of shear strain and compression 37e39. Following acute injury, lack of boundary lubrication by lubricin was evident in this study and others. In this environment, loss of lubricin s ability to reduce shearing between cartilage surfaces, coupled with joint exercise, leads to increasing the extent of chondrocyte apoptosis in the superficial and middle cartilage zones resulting in increased cartilage degradation. Lubricin intra-articular supplementation was shown to provide disease-modifying anti-osteoarthritis effect 13,14,40. Lubricin was shown to preserve superficial zone chondrocytes synthetic and repair capability and stimulate endogenous lubricin biosynthesis 13. We have observed an additional chondroprotective effect in ACLT joints undergoing strenuous exercise following a singledose lubricin intra-articular injection. The single-dose lubricin used in our study represents a supra-physiologic level of SF lubricin, aimed to create a lubricin depot for a sustained chondroprotective effect. While we were not able to observe a significant difference in OARSI scores between PBS and lubricin treatments, only lubricin treatment resulted in restoring endogenous lubricin biosynthesis, increased SF endogenous lubricin levels and prevented caspase-3 activation. The efficacy of a single intra-articular injection has clinical translational value, especially in situations where protecting against increased cartilage damage is required for patients undergoing post-injury rehabilitation or who wish to return quickly to pre-injury levels of chondroprotection. Lubricin protected the surface of articular cartilage in exercised animals and stimulated cartilage lubricin biosynthesis compared to placebo. Chondrocytes in the superficial and middle zones of cartilage treated with lubricin lacked caspase-3 activation and was associated with reduced uctxii levels and increased lubricin SF concentrations. The endogenous lubricin expression in the superficial zone chondrocytes as a result of lubricin injection points to the ability of lubricin to preserve the viability of superficial zone chondrocytes, possibly by lowering shear strain upon the surface of cartilage. Protecting the viability of superficial zone chondrocytes is a potentially useful early intervention to prevent chondrocyte apoptosis and halt the progression of early cartilage damage. Tribosupplementation with lubricin is emerging as a promising intervention to preserve the cartilage superficial zone integrity following injury. One limitation is that the impact of this joint exercise regimen on cartilage integrity of normal joints was not evaluated in this study. Using contralateral joints of ACLT joints as controls, we did not detect an impact of joint exercise on lubricin mrna levels as determined by quantitative PCR compared to non-exercised normal joints. Additionally, we have used lubricin mab 9G3 to quantify rat SF lubricin concentrations 4 weeks following intraarticular administration of human lubricin. The antibody detects human and rat lubricin. Although we have not formally determined the half-life of full-length lubricin following intra-articular administration, recent evidence suggests that less than 10% of a recombinant truncated lubricin will remain in the joint at 24 h following a single intra-articular administration 41. We have confidence that the predominant form of lubricin measured in our ELISA assay is the endogenously-produced rat lubricin with minimal, if any, interference of the exogenously administered human lubricin. In summary, we have described the deleterious effect of forced joint exercise following ACL injury on lubricin cartilage metabolism. The enhanced loss of boundary lubrication provided by lubricin following exercise in the setting of an ACL injury may explain the increased cartilage degradation with exercise. Intraarticular lubricin administration appears to slow the increased

7 946 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e948 A 0.8 Relative Lubricin Expression Raw Data Median C OARSI Modified Mankin Score Raw Data Median B Lubricin SF Lavage Concentration ( g/ml) E ACLT 60 Raw Data Median ACLT+PBS ACLT+Lubricin ACLT PBS Lubricin a b c d D Urinary CTXII (pg per mg creatinine per 24 hours) 3 ACLT PBS Lubricin 180 Raw Data 160 Median ACLT PBS Lubricin 50 µm Fig. 4. Effect of lubricin tribosupplementation on lubricin expression, SF lavage lubricin concentrations and cartilage degeneration following ACLT and exercise. Lubricin (800 mg/ml; 50 ml) intra-articular injection was performed at day 7 following ACLT. PBS (50 ml) intra-articular injection was performed at day 7 following ACLT. (A) Relative lubricin cartilage expression in ACLT joints at week 5 compared to contralateral joints normalized to GAPDH in animals that were exercised (ACLT group; n ¼ 9), exercised and received PBS (PBS group; n ¼ 9) and exercised and received lubricin (lubricin group; n ¼ 9). Indicates that lubricin cartilage expression in lubricin-treated group is significantly higher (P < 0.001) than lubricin cartilage expression in PBS-treated or untreated animals. (B) SF lavage lubricin concentrations in transected joints of ACLT, PBS and lubricin groups (n ¼ 9 in each group). Indicates that SF lavage lubricin concentrations in the lubricin group are significantly higher than lubricin concentrations in the PBS group (P ¼ 0.012) and the ACLT group (P < 0.001). There was no significant difference (P ¼ 0.295) between the lubricin SF concentrations in the PBS and ACLT groups. (C) OARSI modified Mankin scores of articular cartilage from week-5 animals that underwent ACLT and exercise (ACLT group; n ¼ 6), underwent ACLT and exercise and received PBS (PBS group; n ¼ 6) or underwent ACLT and exercise and received lubricin (lubricin group; n ¼ 6). Indicates that OARSI modified Mankin scores in the PBS and lubricin groups were significantly lower (P ¼ 0.005) and (P < 0.001) than untreated group, respectively. There was no significant difference (P ¼ 0.099) between the OARSI scores of the PBS and lubricin groups. (D) uctxii levels normalized to urinary creatinine collected over 24 h prior to joint harvest of week-5 animals that underwent ACLT and exercise (ACLT group; n ¼ 15), underwent ACLT and exercise and received PBS (PBS group; n ¼ 15). Indicates that uctxii levels from animals in the lubricin-treated group were significantly (P < 0.05) lower than uctxii levels from animals in the PBStreated and untreated groups. (E) In-situ hybridization of lubricin mrna (top panel) and activated caspase-3 immunostaining (bottom panel) of representative articular cartilage from control (a), week-5 ACLT þ exercise (b), week-5 ACLT þ exercise treated with lubricin (c) and week-5 ACLT þ exercise treated with PBS (d). Lubricin tribosupplementation restored lubricin expression in superficial zone chondrocytes and reduced cartilage activated caspase-3 levels compared to untreated and PBS-treated ACLT joints. cartilage degradation with joint exercise via preserving superficial zone chondrocytes viability. Author contributions All authors were involved in drafting the article or revising it critically for content, and all authors approved the final version to be published. Dr. Elsaid had full access to all of the data in the study and takes responsibility for the data integrity and data analysis accuracy. Study conception and design: Elsaid, Jay, Fleming. Data acquisition: Elsaid, Tofte, Zhang. Data analysis and interpretation: Elsaid, Jay, Fleming, Waller, Teeple, Tofte, Zhang. Role of funding source Funding source had no role in the design, collection, and interpretation of the data or the decision to submit for publication.

8 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e Conflict of interest Author GJ holds patents on the use of recombinant lubricin. All other authors have no potential conflict to declare related to the content of the manuscript. Acknowledgments The authors would like to thank Dr Matt Warman for providing Mab 9G3. This work was funded by a faculty development grant to KE. References 1. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Joint injury in young adults and risk for subsequent knee and hip osteoarthritis. Ann Intern Med 2000;133:321e8. 2. Nelson F, Billinghurst RC, Pidoux I, Reiner A, Langworthy M, McDermott M, et al. Early post-traumatic osteoarthritis-like changes in human articular cartilage following rupture of the anterior cruciate ligament. Osteoarthritis Cartilage 2006;14:114e9. 3. Roos E, Adalberth T, Dahlberg L, Lohmander LS. Osteoarthritis of the knee after injury to the anterior cruciate ligament or meniscus: the influence of time and age. Osteoarthritis Cartilage 1995;3:261e7. 4. Elsaid KA, Fleming BC, Oksendahl HL, Machan JT, Fadale PD, Hulstyn MJ, et al. Decreased lubricin concentrations and markers of joint inflammation in the synovial fluid of patients with anterior cruciate ligament injury. Arthritis Rheum 2008;58:1707e Anderson DD, Chubinskaya S, Guilak F, Martin JA, Oegema TR, Olson SA, et al. Post-traumatic osteoarthritis: improved understanding and opportunities for early intervention. J Orthop Res 2011;29:802e9. 6. Jay GD, Britt DE, Cha CJ. Lubricin is a product of megakaryocyte stimulating factor gene expression by human synovial fibroblasts. J Rheumatol 2000;27:594e Flannery CR, Hughes CE, Schumacher BL, Tudor D, Aydelotte MB, Kuettner KE, et al. Articular cartilage superficial zone protein (SZP) is homologous to megakaryocyte stimulating factor precursor and is a multifunctional proteoglycan with potential growth-promoting, cytoprotective, and lubricating properties in cartilage metabolism. Biochem Biophys Res Commun 1999;254:535e Jay GD. Characterization of a bovine synovial fluid lubricating factor. I. Chemical, surface activity and lubricating properties. Connect Tissue Res 1992;28:71e Gleghorn JP, Jones AR, Flannery CR, Bonassar LJ. Boundary mode lubrication of articular cartilage by recombinant human lubricin. J Orthop Res 2009;27:771e Rhee DK, Marcelino J, Brown M, Gong Y, Smits P, Lefebvre V, et al. The secreted protein lubricin protects cartilage surfaces and inhibits synovial cell adhesion. J Clin Invest 2005;115:622e Jay GD, Torres JR, Rhee DK, Helminen HJ, Cha CJ, Elsaid KA, et al. Association between friction and wear in diarthrodial joints lacking lubricin. Arthritis Rheum 2007;56:3662e Elsaid KA, Machan JT, Waller K, Fleming BC, Jay GD. The impact of anterior cruciate ligament injury on lubricin metabolism and the effect of inhibiting tumor necrosis factor alpha on chondroprotection in an animal model. Arthritis Rheum 2009;60:2997e Jay GD, Fleming BC, Watkins BA, McHugh KA, Anderson SC, Zhang LX, et al. Prevention of cartilage degeneration and restoration of chondroprotection by lubricin tribosupplementation in the rat following anterior cruciate ligament transection. Arthritis Rheum 2010;62:2382e Teeple E, Elsaid KA, Jay GD, Zhang L, Badger GJ, Akelman M, et al. Effects of supplemental intra-articular lubricin and hyaluronic acid on the progression of posttraumatic arthritis in the anterior cruciate ligament-deficient knee. Am J Sports Med 2011;39:164e Elsaid KA, Jay GD, Chichester CO. Reduced expression and proteolytic susceptibility of lubricin/superficial zone protein may explain early elevation in the coefficient of friction in the joints of rats with antigen-induced arthritis. Arthritis Rheum 2007;56:108e Roche PJ. Preparation of template DNA and labeling techniques. In: Darby IA, Hewitson TD, Eds. In situ Hybridization Protocols. 3rd edn. Totowa (NJ): Humana Press; 2006:9e Tesch GH, Lan HY, Nikolic-Paterson DJ. Treatment of tissue sections for in situ hybridization. In: Darby IA, Hewitson TD, Eds. In situ Hybridization Protocols. 3rd edn. Totowa (NJ): Humana Press; 2006:1e Ge NL, Kocan KM, Murphy GL, Blouin EF. Detection of Anaplasma marginale DNA in bovine erythrocytes by slot-blot and in situ hybridization with a PCR-mediated digoxigenin-labeled DNA probe. J Vet Diagn Invest 1995;7:465e Pritzker KP, Gay S, Jimenez SA, Ostergaard K, Pelletier JP, Revell PA, et al. Osteoarthritis cartilage histopathology: grading and staging. Osteoarthritis Cartilage 2006;14:13e Young AA, McLennan S, Smith SM, Cake MA, Read RA, Melrose J, et al. Proteoglycan 4 downregulation in a sheep meniscectomy model of early osteoarthritis. Arthritis Res Ther 2006;8:R Elsaid KA, Jay GD, Warman ML, Rhee DK, Chichester CO. Association of articular cartilage degeneration and loss of boundary-lubricating ability of synovial fluid following injury and inflammatory arthritis. Arthritis Rheum 2005;52:1746e Jones AR, Flannery CR. Bioregulation of lubricin expression by growth factors and cytokines. Eur Cell Mater 2007;13:40e Niikura T, Reddi AH. Differential regulation of lubricin/superficial zone protein by transforming growth factor beta/bone morphogenetic protein superfamily members in articular chondrocytes and synoviocytes. Arthritis Rheum 2007;56:2312e Jones AR, Chen S, Chai DH, Stevens AL, Gleghorn JP, Bonassar LJ, et al. Modulation of lubricin biosynthesis and tissue surface properties following cartilage mechanical injury. Arthritis Rheum 2009;60:133e Nugent GE, Aneloski NM, Schmidt TA, Schumacher BL, Voegtline MS, Sah R. Dynamic shear stimulation of bovine cartilage biosynthesis of proteoglycans 4. Arthritis Rheum 2006;54:1888e Nugent GE, Schmidt TA, Schumacher BL, Voegtline MS, Bae WC, Jardin KD, et al. Static and dynamic compression regulate cartilage metabolism of proteoglycans 4 (PRG4). Biorheology 2006;43:191e Nugent-Derfus GE, Takara T, O Neil JK, Cahill SB, Görtz S, Pong T, et al. Continuous passive motion applied to whole joints stimulates chondrocyte biosynthesis of PRG4. Osteoarthritis Cartilage 2007;15:566e Otterness IG, Eskra JD, Bliven ML, Shay AK, Pelletier JP, Milici AJ. Exercise protects against articular cartilage degeneration. Arthritis Rheum 1998;41:2068e Appelton CT, McErlain DD, Pitelka V, Schwartz N, Bernier SM, Henry JL, et al. Forced mobilization accelerates pathogenesis: characterization of a preclinical surgical model of osteoarthritis. Arthritis Res Ther 2007;9:R Pap G, Eberhardt R, Sturmer I, Machner A, Schwarzberg H, Roessner A, et al. Development of osteoarthritis in the knee

9 948 K.A. Elsaid et al. / Osteoarthritis and Cartilage 20 (2012) 940e948 joints of Wistar rats after strenuous running exercise in a running wheel by intracranial self-stimulation. Pathol Res Pract 1998;194:41e Galois L, Etienne S, Grossin L, Watrin-Pinzano A, Cournil- Henrionnet C, Loeuille D, et al. Doseeresponse relationship for exercise on severity of experimental osteoarthritis in rats; a pilot study. Osteoarthritis Cartilage 2004;12:779e Almonte-Becerril M, Navarro-Garcia F, Gonzalez-Robles A, Vega-Lopez MA, Lavalle C, Kouri JB. Cell death of chondrocytes is a combination between apoptosis and autophagy during the pathogenesis of osteoarthritis within an experimental model. Apoptosis 2010;15:631e Thomas CM, Fuller CJ, Whittles CE, Sharif M. Chondrocyte death by apoptosis is associated with the initiation and severity of articular cartilage degradation. Int J Rheum Dis 2011;14:191e Musumeci G, Loreto C, Carnazza ML, Martinez G. Characterization of apoptosis in articular cartilage derived from the knee joints of patients with osteoarthritis. Knee Surg Sports Traumatol Arthrosc 2011;19:307e Colwell CW, D Lima DD, Hoenecke HR, Fronek J, Pulido P, Morris BA, et al. In vivo changes after mechanical injury. Clin Orthop Relat Res 2001;391:S116e D Lima D, Hermida J, Hashimoto S, Colwell C, Lotz M. Caspase inhibitors reduce severity of cartilage lesions in experimental osteoarthritis. Arthritis Rheum 2006;54:1814e Zhu F, Wang P, Lee NH, Goldring MB, Konstantopoulos K. Prolonged application of high fluid shear to chondrocytes recapitulates gene expression profiles associated with osteoarthritis. PLoS One 2010;5:e Hashimoto S, Nishiyama T, Hayashi S, Fukishiro T, Takebe K, Kanzaki N, et al. Role of p53 in human chondrocyte apoptosis in response to shear strain. Arthritis Rheum 2009;60:2340e D lima DD, Hashimoto S, Chen PC, Colwell CW, Lotz MK. Human chondrocyte apoptosis in response to mechanical injury. Osteoarthritis Cartilage 2001;9:712e Flannery CR, Zollner R, Corcoran C, Jones AR, Root A, Rivera-Bermúdez MA, et al. Prevention of cartilage degeneration in a rat model of osteoarthritis by intraarticular treatment with recombinant lubricin. Arthritis Rheum 2009;60:840e Vugmeyester Y, Wang Q, Xu X, Harrold J, Daugusta D, Li J, et al. Disposition of human recombinant lubricin in naïve rats and in a rat model of post-traumatic arthritis after intra-articular or intravenous administration. AAPS J 2012;14:97e104.