Effect of muscle weakness and joint inflammation on the onset and progression of osteoarthritis in the rabbit knee

Osteoarthritis and Cartilage 22 (2014) 1886e1893

Effect of muscle weakness and joint inflammation on the onset and progression of osteoarthritis in the rabbit knee C. Egloff y z, A. Sawatsky y, T. Leonard y, D.A. Hart x, V. Valderrabano z, W. Herzog y * y Human Performance Laboratory, Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada z Orthopaedic Department, University Hospital, University of Basel, Basel, Switzerland x McCaig Institute for Bone & Joint Health, University of Calgary, Calgary, Alberta, Canada

a r t i c l e i n f o

s u m m a r y

Article history: Received 11 April 2014 Accepted 29 July 2014

Objectives: Interactions between mechanical and non-mechanical independent risk factors for the onset and progression of Osteoarthritis (OA) are poorly understood. Therefore, the goal of the present study was to investigate the in vivo effects of muscle weakness, joint inflammation and the combination on the onset and progression of OA in a rabbit knee joint model. Materials and methods: Thirty 1-year-old female New Zealand White rabbits (average 5.7 kg, range 4.8e6.6 kg) were divided into four groups with one limb randomly assigned to be the experimental side: (1) surgical denervation of the vastus lateralis (VL) muscle; (2) muscle weakness induced by intramuscular injection of Botulinum toxin A (BTX-A); (3) intraarticular injection with Carrageenan to induce a transient inflammatory reaction; (4) combination of Carrageenan and BTX-A injection. After 90 days, cartilage histology of the articular surfaces were microscopically analyzed using the Osteoarthritis Research Society International (OARSI) histology scoring system. Results: VL denervation resulted in significantly higher OARSI scores in the patellofemoral joint (group 1). BTX-A administration resulted in significant cartilage damage in all four compartments of the knee (group 2). Carrageenan did not cause significant cartilage damage. BTX-A combined with Carrageenan lead to severe cartilage damage in all four compartments. Conclusion: Muscle weakness lead to significant OA in the rabbit knee. A transient local inflammatory stimulus did not promote cartilage degradation nor did it enhance OA progression when combined with muscle weakness. These results are surprising and add to the literature the conclusion that acute inflammation is probably not an independent risk factor for OA in this rabbit model. © 2014 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Keywords: Osteoarthritis Rabbit model Muscle weakness Joint inflammation BTX-A Carrageenan

Introduction Estimated costs attributable to arthritis and other rheumatic conditions (AORC) in the United states in 2003 were approximately $128 billion1. The Arthritis Research Center of Canada calculated the total cost of treating OA in Canada to be 1.8 billion dollars in 2010 and it is expected to quadruple over the next two decades to reach $8.1 billion by the year 20312. Despite this substantial socioeconomic impact, the onset, natural history, or progression of OA is still poorly understood. Scientific research over the last decade

* Address correspondence and reprint requests to: W. Herzog, Human Performance Laboratory, Kinesiology, University of Calgary, Calgary, Alberta, Canada. E-mail addresses: [email protected] (C. Egloff), [email protected] (A. Sawatsky), [email protected] (T. Leonard), [email protected] (D.A. Hart), [email protected] (V. Valderrabano), [email protected], ch.egloff@ gmail.com, [email protected] (W. Herzog).

has increasingly described OA as a whole joint disease and the result of a complex interplay between several causes or risk factors. Risk factors initiating or advancing OA can be grouped into mechanical and non-mechanical factors. Mechanical alterations, such as anterior cruciate ligament ruptures, meniscal tears, muscle weakness and malalignment, especially in joints of the lower limb, have been widely studied in animal models in vitro, and in human studies3e8. Muscle weakness and atrophy is one of the earliest signs and always accompanies OA9. Muscle weakness has also been shown to be an independent risk factor for OA10 leading to increased loads in the lower limb, whereas neuromuscular training protocols seem to have a protective effect on the onset and progression of OA11,12. Intramuscular administration of Botulinum toxin A (BTX-A) has been used in the past to induce controlled and reversible muscle weakness and destabilization of the knee13, and has caused weakness and significant cartilage degradation and promoted OA8,14. It is generally believed that abnormal mechanical

http://dx.doi.org/10.1016/j.joca.2014.07.026 1063-4584/© 2014 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

forces are almost always present in OA, but the question remains if they act as independent risk factors. Non-mechanical risk factors are more difficult to define but include genetics, obesity, age, sex and inflammation. Especially Inflammation and subsequent synovitis have become an intensively studied area in the pathogenesis of OA. Although OA is defined usually as a non-inflammatory disease, there is evidence that inflammatory reactions mediated by the release of proinflammatory cytokines result in cartilage degradation, impaired tissue repair, and act as a trigger for symptomatic OA15e18. Recently, Erhart-Hledik and colleagues linked mechanical stimuli and changes in serum cartilage oligomeric matrix protein (COMP) to cartilage thinning within 5 years indicating that proinflammatory cytokines contribute to levels of soluble COMP that may be released through a trigger event, and subsequently have a detectable impact on the long term progression of OA19. However, to date there is no clear evidence if joint inflammation represents a primary cause for the onset of OA, or if inflammation is merely a secondary synovitis resulting from mechanical alterations that is associated, but not the direct cause of OA. Carrageenan (a marine polysaccharide) has been used as an inflammatory stimulus in a number of experimental models (e.g., mice, rats, rabbits) for half a century. In rabbits, injection of carrageenan into the knee has been used to induce local and acute inflammation, and the temporal aspects of these inflammatory episodes have been carefully assessed using a variety of techniques20e23. The aim of this study was to elucidate the effects of mechanical and inflammatory interventions, and the combination of the two, on the onset and development of OA in the rabbit knee in vivo. Mechanical interventions included selective and general quadriceps muscle weakness, and inflammation was introduced using intra-articular injections of carrageenan. We hypothesized that muscle weakness and inflammation were independent risk factors for the development of OA, and that a combination of the two factors would accelerate osteoarthritic changes in the knee.

1887

equally among three lines of the thigh e medial, central, and lateral e corresponding to the vastus medialis, rectus femoris, and VL muscle. The contralateral side received the same amount of sterile saline. Group 3: Animals (n ¼ 8) received an intra-articular injection of commercially available sterile Carrageenan (Sigma Chemical Co; St. Louis, MO) into the experimental knee joint using 0.5 ml of a 2% solution21 on day 0 and day 30. This regime was chosen to make sure that the inflammatory reaction induced by the carrageenan had been resolved by the end of the study. The contralateral side received 0.5 ml of sterile saline. Group 4: Animals (n ¼ 8) received a combination of intramuscular BTX-A injection monthly and intra-articular injection of Carrageenan on days 0 and 30 of the experimental period. The contralateral side received an intra-articular and intra-muscular injection of sterile saline. At the end of the experimental period (day 90), rabbits were sacrificed by an intravenous injection of Euthanyl (Bimeda-MTC). Muscle mass assessment The heads of the quadriceps muscle group (rectus femoris, vastus medialis, VL, and vastus intermedius muscle) were isolated and dissected individually and their wet mass was measured using a commercial balance with a measuring accuracy of 0.001 g (Mettler-Toledo, Switzerland). Joint width assessment On the day of sacrifice, the width of the experimental and contralateral knee was measured in all animals using a commercially available digital caliper. Prior to the measurement, hind limbs were shaved, the joint space between femur and tibia was identified and the largest medial-lateral diameter was marked. Measurements were repeated three times and the average value was recorded.

Methods Cartilage histology Experimental design Thirty 1-year-old skeletally mature female New Zealand White rabbits (weight: average 5.7 kg, range 4.8e6.6 kg; Riemen's Furriers, St. Agatha, Ontario, Canada) were used in accordance with an experimental protocol approved by the University of Calgary Animal Care Committee. Animals were housed in pairs in floor pens to permit normal activity. Rabbits received a standard diet and water ad libitum. Animals were divided into four experimental groups, and one hind limb was randomly assigned for intervention (experimental hind limb) while the other hind limb served as a contralateral non-intervened control (control hind limb). For all interventions, rabbits were anesthetized using a 3.5% isoflurane (Benson Medical) to oxygen mixture. Group 1: Animals (n ¼ 6) underwent surgical transection of the nerve branch arising from the common femoral nerve leading to the vastus lateralis (VL) muscle, thereby preventing activation of VL. The VL was chosen for nerve transection because it is the biggest contributor to the extensor force in rabbits and has a laterally oriented fiber insertion into the patella, and therefore, when transected it was thought to result in considerable weakness of the knee extensor group24, and imbalance in the knee, specifically the patellar tracking in the femoral groove. Group 2: Animals (n ¼ 8) received an injection of BTX-A (BOTOX, Allergan Inc., Toronto, Ontario, Canada) into the experimental hind limb. Injections of 3.5 U/kg of BTX-A were given once each month (day 0, 30, 60). Each injection of 3.5 U/kg was divided

The femoral condyle, tibial plateau, and patella were harvested, photographed with a high resolution digital camera and fixed in a 10% neutral buffered formalin solution for 10 days (Fisher Scientific, Wohlen, Switzerland) and then decalcified with Cal-Ex II decalcifying solution (10% formic acid solution in 4% formaldehyde, Fisher Scientific) and kept at room temperature. The solution was changed daily. After 2 weeks, joints were opened and returned to a fresh decalcifying solution for an additional 2e3 weeks. Decalcification was finished when joints could be cut smoothly and without any grittiness. The joints were then washed thoroughly in running tap water for 2 h. After decalcification, the femur was cut into three pieces (femoral groove, medial and lateral femoral condyle) and the tibia into two pieces (medial and lateral plateau). Samples were then processed in an automatic paraffin processor (Leica TP 1020 Leica Microsystems, Heerbrugg, Switzerland) where they were dehydrated in a graded series of alcohols ranging between 80% and 100%, cleared in xylene and infiltrated in a mixed Paraplast-plus/ extra wax (Fisher Scientific). The individual bones of the knee (tibia, femur, femoral groove, and patella) were then embedded in paraffin molds and stored at room temperature until they were sectioned. Serial sections of the knee articular cartilage were cut at 12e14 mm thickness using a Leica RM 2165 microtome. Every fifth section was collected, adhered to Superfrost Plus slides (Fisher Scientific) and allowed to dry at 40 C for 7 days. All slides were then stained with hematoxylin, Safranin-O, and fast green (Fisher Scientific). Sections were then dehydrated in ethanol, cleared in

1888

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

their diet to a high fat diet, both animals recovered quickly and regained their bodyweight at the time of sacrifice (6.3%, 5.6%, respectively). All analyses were performed by two independent raters. Interobserver reliability was assessed using the ICC. Cronbach's alpha resulted in 0.85, which was considered as good to very good reliability. Quadriceps muscle mass

Fig. 1. Mean (95% CI) of total wet muscle mass (VL, Vastus medialis, Rectus femoris and Vastus intermedius) for the experimental and control limbs for each intervention group after 12 weeks. Bars marked with * show significant differences in the total muscle mass between experimental and control limb.

xylene, and finally cover-slipped with Polymount mounting media (Fisher Scientific) before they were allowed to dry at room temperature for several days. Sections were viewed on a light microscope (Axiostar plus, Carl Zeiss, Jena, Germany) by two examiners who were blinded to the study design. Examinations were made with a 25 objective for cartilage scoring and a 200 objective to confirm cellular changes. Two independent graders assessed all sections in a blinded manner using the OARSI histology scoring system25. A piece of synovium from the infrapatellar fat pad of all 30 rabbit knees (experimental and contralateral side) was also embedded in paraffin and sectioned. These sections were stained with Hematoxylin & Eosin and cut into 8 mm sections for assessment by image analysis using the OARSI scoring system25. Statistical analysis For analysis of the histological data (OARSI), a non-parametric Wilcoxon's signed rank test was used to compare differences between experimental and contralateral control limbs. Differences between groups were analyzed using a KruskaleWallis followed by a ManneWhitney U test. SPSS software (version 21.0, SPSS Science Inc., Chicago, Ill, USA) was used for all analysis. Statistical significance was set at P < 0.05. Two independent observers scored all histological data separately. For inter-examiner reliability testing, the internal consistency coefficient (ICC) was calculated using Cronbach's alpha.

Surgical transection of the femoral branch leading to the VL resulted in a significant loss in wet muscle mass of the VL muscle compared to the contralateral side (13.1 g vs 21.8 g, respectively; P ¼ 0.002). Interestingly, total muscle mass of the quadriceps group was not different between experimental and contralateral hind limb, because vastus medialis, intermedius and rectus femoris had increased masses on the experimental side (38.3 g vs 33.0 g, respectively; P ¼ 0.06), presumably in response to the loss of force in the VL following denervation. The isolated administration of BTX-A led to a significant loss in total mass of the quadriceps muscles, as well as all individual heads of the quadriceps group (48%, P < 0.001). A similar loss in muscle mass was also found in the combined carrageenan/BTX-A group (47%, P < 0.001). There were no changes in total quadriceps muscle mass in the carrageenan group 3 animals (2%, P ¼ 0.6). Figure 1 summarizes the changes in quadriceps muscle mass (Fig. 1, Table II). Joint diameter Knee joint diameter of the experimental limb was significantly increased compared to contralateral control knees in the carrageenan, group 3, animals (means 22.9 mm vs 21.1 mm, resp.; P ¼ 0.002) and the carrageenan/BTX-A, group 4, animals (means 21.8 mm vs 20.8 mm, resp.; P ¼ 0.01). Joint diameters in the VL denervation, group 1, (means 20.7 mm vs 21.2 mm, resp. P ¼ 0.17) and BTX-A injected, group 2, animals, were not different (means 21.1 mm vs 21.1 mm, resp. P ¼ 0.96) (Fig. 2, Table I). Cartilage histology

Animals

Microscopic analysis showed gross visible changes in the tibiofemoral and the patello-femoral articulations (Fig. 6). Histologic assessment of the different regions of the knee joint demonstrated characteristic side-to-side changes in each of the groups. Scores are shown in Table I as means and corresponding 95% Confidence Interval (CI).

Throughout the course of the study, there were no surgical complications, infections and no animals died. Two animals in the Carrageenan/BTX-A group lost temporarily about 20% of their bodyweight, probably due to the combined systemic effect of BTX-A and Carrageenan (20.4%, 21.6%, respectively). After changing

VL transection (group 1 animals) The femoral groove and the backside of the patella showed significantly higher OARSI scores than their contralateral counterpart (patella: 9.5 vs 6.0, P ¼ 0.03; groove: 8.7 vs 5.8, P ¼ 0.03). The tibial surface and the femoral condyles showed slightly higher

Results

Table I Joint diameter and total muscle mass for each intervention group after 12 weeks. Values presented are the mean with 95% Confidence Interval (CI, lower and upper bound). Significant effect of VL transection, Botox, Carrageenan and Botox plus Carrageenan were determined using a repeated-measures analysis of variance (ANOVA). Bonferroni correction was used to determine differences within groups Location and measurement Joint diameter (mm) Total muscle mass (g)

VL transsection (n ¼ 6)

BTX-A (n ¼ 8)

Carrageenan (n ¼ 8)

BTX-A* carrageenan (n ¼ 8)

Experimental

Control

Experimental

Control

Experimental

Control

Experimental

Control

20.7 (19.8, 21.5)

21.3 (20.8, 21.7)

21.1 (20.2, 22)

21.1 (20, 22.1)

22.9 (22.1,23.7)

21.1 (20.5, 21.8)

21.8 (21.3,22.3)

20.8 (20.2, 21.3)

33.0 (26.4, 37.3)

38.9 (36.4, 41.3)

20.0 (16.1, 23.9)

38.3 (35.3, 41.2)

44.3 (40.4, 48.1)

45.1 (40.8, 49.4)

21.4 (18.6, 24.2)

40.1 (38.2, 42)

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

1889

Table II Mean OARSI Scores for the experimental limbs and for all four intervention groups at 12 weeks (95% CI, lower and upper bound). Significant differences between experimental and contralateral side were determined using the Wilcoxon signed rank test. Numbers marked with * show significant increases in OA in the experimental knee across the intervention groups using a KruskaleWallis test, followed by a ManneWhitney U test. Bonferroni correction was used for posthoc analysis. NS ¼ not significant Intervention

VL trans Botox Carrageenan B&C

Patella

Groove

Femur

Tibia

Synovium

OARSI score (CI 95%, LI, UI)

P-value

OARSI score (CI 95%, LI, UI)

P-value

OARSI score (CI 95%, LI, UI)

P-value

OARSI score (CI 95%, LI, UI)

P-value

OARSI score (CI 95%, LI, UI)

P-value

9.5 10.6 8.6 11.0

0.03* 0.04* 0.13 (ns) 0.017*

8.7 11.4 8.6 11.2

0.03 0.03 0.28 (ns) 0.012

8.3 10.3 8.4 15.4

0.24 (ns) 0.012* 0.07 (ns) 0.011*

6.2 9.3 7.8 10.9

0.23 (ns) 0.011 0.48 (ns) 0.012

2.8 3.8 10.6 8.9

0.29 (ns) 0.13 0.012 0.012

(8, 10.9) (9.2, 11.7) (7.4, 9.7) (10.1, 12)

(8.1, 9.8) (10.8, 12.1) (7.7, 9.6) (12, 10.3)

scores on the experimental sides without reaching statistical significance (tibia: 8.3 vs 6.9; femur 8.8 vs 6.1, respectively). Cartilage damage occurred primarily on the medial side of the patella and femoral groove. Cartilage damage in the tibio-femoral joints occurred exclusively in the medial compartment. BTX-A (group 2 animals) Administration of BTX-A resulted in severe cartilage damage in all four compartments of the knee with significant increases in OARSI scores: patella: 10.6 vs 6.9, P ¼ 0.04; groove: 12.4 vs 7.6, P ¼ 0.03; tibia: 8.3 vs 4.0, P ¼ 0.01; femur: 10.0 vs 5.3, P ¼ 0.012. Cartilage damage in the tibio-femoral joints was located in the medial compartment, whereas cartilage damage occurred primarily on the lateral side of the patello-femoral joint. Carrageenan (group 3 animals) Intraarticular injection of carrageenan did not result in consistent differences between experimental and control knees. Experimental knees showed a trend towards marginally higher scores, but did not reach statistical significance: patella: 8.6 vs 7.6, P ¼ 0.13; groove: 8.6 vs 6.8, P ¼ 0.28; tibia: 8.5 vs 7.0, P ¼ 0.46; femur: 9.0 vs 7.1, P ¼ 0.07. Cartilage damage in the patello-femoral joint occurred about equally on the medial and lateral side. Cartilage damage in the tibio-femoral joint occurred predominantly on the medial side.

(6.1, 10.8) (8.5, 11.9) (6.9,9.9) (13.5, 17)

(5.1, 7.1) (6.9, 11.6) (6, 9.5) (7.9, 13.8)

(2.4, (3.4, (7.1, (8.3,

3.3) 4.4) 13.2) 9.5)

Cartilage subanalysis The overall analysis between interventions showed a trend towards the lowest OARSI scores in the VL denervation group 1 animals and the highest scores for the carrageenan/BTX-A group 4 animals (P ¼ 0.065). Analysis across knee joint compartments (patella, groove, tibia, femur) revealed that the Carrageenan group 3 animals showed significantly lower OARSI scores in the femoral groove (P ¼ 0.039), the femur (P ¼ 0.027) and the tibia (P ¼ 0.042) than the carrageenan/BTX-A group. The VL denervation group 1 rabbits had lower scores in the femoral groove (P ¼ 0.037), tibia (P ¼ 0.05) and femur (P < 0.01) compared to the carrageenan/BTX-A group animals (Table II). Further analysis of the OARSI score revealed specific differences between intervention groups: For the patella, the carrageenan/BTXA group 2 animals had significantly more cellular cluster formations than the VL denervation group 1 animals (P ¼ 0.027) and the carrageenan group 3 animals (P ¼ 0.01). For the femoral groove, the carrageenan/BTX-A group 4 animals had more cellular clustering than the Carrageenan group 3 animals, and significant more loss of staining than the VL denervation group 1 animals. For the femoral condyles and the tibial plateau, the Botox and Carrageenan (B&C) group 4 animals had significantly more structural deficits than the VL denervation group 1 animals (P ¼ 0.017, P ¼ 0.042; respectively). Synovial analysis

Carrageenan/BTX-A (group 4 animals) The combination of BTX-A induced muscle weakness and carrageenan induced joint inflammation resulted in severe cartilage damage in the patella (11.0 vs 6.1; P ¼ 0.017), the groove (8.6 vs 6.8; P ¼ 0.012), the tibia (11.3 vs 5.8; P ¼ 0.012) and the femur (14.4 vs 6.0; P ¼ 0.011). Cartilage damage occurred primarily on the medial side of the tibio-femoral joint and the lateral side of the patello-femoral joint.

The histological analysis according to the OARSI score revealed significant higher scores in the Carrageenan (10.6 vs 3.9; P ¼ 0.01) and carrageenan/BTX-A groups (8.9 vs 2.9; P ¼ 0.01) (Fig. 4, Table II). The synovium of the VL transection (2.8 vs 2; P ¼ 0.29) and the isolated Botox group (3.8 vs 2.8; P ¼ 0.13) were unaffected (Fig. 5). The synovial subanalysis across intervention groups showed significant higher scores for proliferation and hypertrophy (P < 0.001), lymphoplasmatic infiltrate (P < 0.001) and angiogenesis (P ¼ 0.006) in both groups with injected carrageenan (group 2, 4) compared to groups one and three. Discussion

Fig. 2. Mean (95% CI) of joint diameter (mm) between experimental and control limbs across intervention groups. Bars marked with * show significant differences in the joint diameter between experimental and control limbs.

Over the past decade, it has become accepted that abnormal joint mechanics are one of the greatest risk factors for the onset and increased rate of OA severity26. Also, it has been found that in the advanced stages of OA, an inflammatory reaction is almost always present. In the present study, we aimed to determine the effects of isolated muscle weakness and isolated joint inflammation, as well as the superposition of these two factors, on the onset and severity of OA. Our results showed that severe muscle weakness, produced by intramuscular administration of BTX-A, resulted in significant cartilage degradation in all compartments of the knee when compared to the contralateral, non-injected side. Also, a select muscle imbalance caused by a denervation of the VL muscle, was

1890

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

Fig. 3. Mean (95% CI) OARSI scores for the experimental and control limbs for each intervention group after 12 weeks according to the anatomical region. Bars marked with * show significant differences in the OARSI score between experimental and control limb.

associated with a significant loss of cartilage in the patello-femoral joint, indicating a disturbance of normal patella-femoral joint mechanics. In contrast, selective administration of a transient inflammatory stimulus through intra-articular injection of carrageenan did not result in a significant increase in the progression of OA over 3 months. However, it did lead to a severe inflammatory reaction in synovial tissue as seen in Figs. 4 and 5. Our hypothesis that an inflammatory stimulus in an otherwise healthy knee joint acts as an independent risk factor for OA progression was not supported by our results. However, when muscle weakness was combined with a local inflammation there was a trend towards an increased rate of cartilage damage compared to muscle weakness alone (Fig. 3). Therefore, either the carrageenaninduced inflammatory reaction was not sufficiently strong to

Fig. 4. Mean (95% CI) of OARSI scores of the synovium for the experimental and control limbs for each intervention group after 12 weeks. Bars marked with * show significant differences in the OARSI score between experimental and control limbs.

enhance the impact of the muscle weakness at the doses used, or perhaps being an acute stimulus, it did not lead to a chronic inflammation which may be required for OA progression. The results of this study suggest that alterations in joint mechanics through muscle weakness are an important and powerful risk factor for OA, and an accompanying joint inflammation might be a contributing factor in the pathogenesis of OA. OA is often described as a multi-factorial disease, and thus, in the human disease, it is often hard to study and virtually impossible to disentangle the various contributing factors. In order to prevent or stop the progression of OA, it may be useful to evaluate to what extent different causes or risk factors of OA contribute to the disease. Major risk factors for knee OA are age, female gender, obesity, traumatic injuries, pathological biomechanics and chronic excessive overuse26e28. Muscle weakness has been linked to an increased rate of radiographic joint space narrowing9 and to an acceleration of OA development in elderly community-dwelling women29. Moreover, decreased isokinetic quadriceps muscle strength in women has been found to lead to increased lower limb loading during walking, and has been associated with both the initiation and progression of OA11. Furthermore, it has been demonstrated that muscle weakness can result in OA, and our results agree with those described in the literature6,8,14. In BTX-A group animals (group 2), we found significant cartilage degradation not only in the patello-femoral but also in the tibio-femoral joints suggesting that the severe muscle weakness caused by the toxin acted as a major contributor to OA progression in the rabbit model. Our results showed more severe cartilage damage than comparable previous studies. One reason for these differences may be that this experiment was conducted over a longer period of time, e.g., 3 months instead of 4 weeks8. The denervation of the VL resulted in cartilage degradation concentrated on the medial side of the patello-femoral joint, but did not produce differences in cartilage damage between

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

1891

Fig. 5. Microscopic analysis of synovium sample (20 magnification). A: Synovial probe of group 1 animals (VL transection) with OARSI score 2. B: Synovial example of group 2 (BTX-A injection) showing mild synovial reaction with an OARSI score of 4. C: Example of group 4 (BTX-A/Carrageenan) showing severe synovial inflammatory reaction with an OARSI score of 11. D: Synovial example of group 3 (Carrageenan) with severe inflammatory reaction with an OARSI score of 16.

experimental and control knees in the tibio-femoral joint. This observation confirmed our hypothesis that an asymmetrical quadriceps weakness has a greater effect on the mechanics of the patello-femoral than the tibio-femoral compartments. A limitation of this study is that we did not measure absolute muscle forces or patello-femoral kinematics or contact pressure distributions to substantiate this hypothesis. However, previous studies showed the effect of selective muscle weakness on muscle strength and patellofemoral joint kinematics30. Therefore, we assume that the denervation of the VL, and associated loss of force, led to a force overload on the medial side of the patello-femoral joint, a corresponding alteration in patello-femoral joint kinematics, and finally, to the localized cartilage loss on the medial side of the patella and femoral groove. This result emphasizes the importance of balanced muscle strength on knee joint mechanics, especially for the patello-femoral compartment. Another limitation of this study is that we used the contralateral joint as the control and did not use a sham control group. It is possible that the impact of the experimental side could have influenced the control side and would have derogated the results. However, we observed significant changes in histology in the experimental knees of BTX-A group animals compared to the corresponding contralateral control knee, while the carrageenan group animals did not show consistent changes between experimental and control knees. The OARSI scores of these joints were comparable with values from non-experimental control animals observed in other studies using sham controls, suggesting that our observations were similar to sham control knees8. Acute inflammatory reactions, with an up-regulation of proinflammatory cytokines, are commonly observed after traumatic

intra-articular injuries15,31. Recent studies suggest that a stimulusresponse system between mechanical and biochemical factors could play a role in the incidence and progression of OA with crucial long-term effects19. In this study, we did not find a significant effect of a transient inflammatory stimulus on articular cartilage damage of the knee. However, we observed an increase in the OARSI histology score when the inflammatory reaction was superposed to the muscle weakness compared to when muscle weakness was induced in isolation. A similar trend of increased cartilage damaged was seen when we compared the carrageenan-injected with the corresponding control limbs. One explanation for our findings could be that mechanical factors have a much greater effect than transient inflammatory reactions. If this is indeed the case, one might speculate that inflammation plays only a secondary, accompanying role in the onset and progression of OA, which then raises the question of the effect of treatments of acute inflammation in OA. In this study, the healthy contralateral knee was used as a control for the experimental limb, thereby accounting for inter-animal differences in the natural progression of OA with aging. Using this approach, we found distinct and highly significant differences in OARSI scores between experimental and contralateral control knees in the BTX-A and the VL group animals, but not the inflammation group animals. Another explanation for our findings could be that the inflammatory stimulus administered by the intra-articular injection of carrageenan was too small to result in a relevant inflammatory response. Due to the experimental setup we could not quantify the inflammatory reactions with synovial fluid or blood serum analyses. However, we observed significant soft tissue and joint swelling of the knee in both carrageenan groups. Also we did find a severe synovial

1892

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

Fig. 6. Microscopic images of three different animals in 5 magnification (A, C, E) and 20 magnification (B, D, F). All histological slides are examples from the femoral condyle. A, B: contralateral control side from a specimen of the BTX-group showing an OARSI score of 3 (normal). C, D: experimental side from a specimen of the BTX/carrageenan-group showing an OARSI score of 12 (moderate OA) E, F: experimental side from a specimen of the BTX-group showing an OARSI score of 18 (severe OA).

reaction in both groups treated with carrageenan suggesting that a true inflammatory reaction had taken place in those experimental groups. Furthermore, other studies indicate that a single carrageenan injection causes a severe but transient inflammatory reaction and tissue response, which returns to baseline after about 28 days20,21. Thus, injections on days 0 and 30 as done here, would produce an acute phase of inflammation that would most likely be resolved by the end of the 3 months experimental period. However, we focused on investigating the progression of OA in the presence of a local inflammatory stimulus and/or specific muscle weakness, rather than on the acute effects of inflammation. Conclusion Quadriceps weakness and imbalance over a period of time leads to the onset and progression of OA in the rabbit knee. A transient local inflammatory stimulus did not promote cartilage degradation, nor did it significantly increase the severity of OA when combined with muscle weakness. This result is somewhat surprising and adds to the literature indicating that acute inflammation may not be a strong independent risk factor for OA in this model system.

Author contributions Study concept and design: Egloff, Herzog. Acquisition of data: Egloff, Sawatsky. Analysis and interpretation of data: Egloff, Sawatsky, Hart, Herzog. Drafting of the manuscript: Egloff. Critical revision of the manuscript for important intellectual content: Egloff, Hart, Herzog. Obtained funding: Egloff, Valderrabano, Hart, Herzog. Administrative, technical, or material support: Sawatsky, Leonard. Study supervision and take responsibility for the integrity of the work as a whole: Herzog.

Conflicts of interest statement Each author certifies that he or she has no commercial associations (e.g., consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article.

C. Egloff et al. / Osteoarthritis and Cartilage 22 (2014) 1886e1893

We are confirming that all authors were fully involved in the study and preparation of the manuscript, and that the material within has not been and will not be submitted for publication elsewhere.

15.

Acknowledgments Funding from the Canadian Institutes of Health Research, the AIHS Team Grant on Osteoarthritis (200700596), the Canada Research Chair Program ((1) 950-207002), the Killam Foundation and the technical expertise of Mrs Ruth Seerattan are gratefully acknowledged. References 1. Centers for Disease C, Prevention. National and state medical expenditures and lost earnings attributable to arthritis and other rheumatic conditionseUnited States, 2003. MMWR Morb Mortal Wkly Rep 2007;56(1):4e7. s P, et al. The 2. Sharif B, KJ, Rahman M, Bansback N, Sayre E, Fine cost of treating osteoarthritis in Canada expected to quadruple. In: American College of Rheumatology Annual Meeting 2012,. Washington, D.C., USA 2012. 3. Roos H, 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(4):261e7. 4. Englund M, Guermazi A, Roemer FW, Aliabadi P, Yang M, Lewis CE, et al. Meniscal tear in knees without surgery and the development of radiographic osteoarthritis among middleaged and elderly persons: the Multicenter Osteoarthritis Study. Arthritis Rheum 2009;60(3):831e9. 5. Sharma L, Chmiel JS, Almagor O, Felson D, Guermazi A, Roemer F, et al. The role of varus and valgus alignment in the initial development of knee cartilage damage by MRI: the MOST study. Ann Rheum Dis 2013;72(2). 6. Herzog W, Longino D. The role of muscles in joint degeneration and osteoarthritis. J Biomech 2007;40(Suppl 1):S54e63. 7. Hasler EM, Herzog W, Leonard TR, Stano A, Nguyen H. In vivo knee joint loading and kinematics before and after ACL transection in an animal model. J Biomech 1998;31(3):253e62. 8. Rehan Youssef A, Longino D, Seerattan R, Leonard T, Herzog W. Muscle weakness causes joint degeneration in rabbits. Osteoarthritis Cartilage 2009;17(9):1228e35. 9. Palmieri-Smith RM, Thomas AC, Karvonen-Gutierrez C, Sowers MF. Isometric quadriceps strength in women with mild, moderate, and severe knee osteoarthritis. Am J Phys Med Rehabil 2010;89(7):541e8. 10. Roos EM, Herzog W, Block JA, Bennell KL. Muscle weakness, afferent sensory dysfunction and exercise in knee osteoarthritis. Nat Rev Rheumatol 2011;7(1):57e63. 11. Mikesky AE, Meyer A, Thompson KL. Relationship between quadriceps strength and rate of loading during gait in women. J Orthop Res 2000;18(2):171e5. 12. Neuman P, Englund M, Kostogiannis I, Friden T, Roos H, Dahlberg LE. Prevalence of tibiofemoral osteoarthritis 15 years after nonoperative treatment of anterior cruciate ligament injury: a prospective cohort study. Am J Sports Med 2008;36(9):1717e25. 13. Longino D, Frank C, Herzog W. Acute botulinum toxin-induced muscle weakness in the anterior cruciate ligament-deficient rabbit. J Orthop Res 2005;23(6):1404e10. 14. Leumann A, Longino D, Fortuna R, Leonard T, Vaz MA, Hart DA, et al. Altered cell metabolism in tissues of the knee joint in a

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

1893

rabbit model of Botulinum toxin A-induced quadriceps muscle weakness. Scand J Med Sci Sports 2012;22(6):776e82. Ayral X, Pickering EH, Woodworth TG, Mackillop N, Dougados M. Synovitis: a potential predictive factor of structural progression of medial tibiofemoral knee osteoarthritis e results of a 1 year longitudinal arthroscopic study in 422 patients. Osteoarthritis Cartilage 2005;13(5):361e7. Echtermeyer F, Bertrand J, Dreier R, Meinecke I, Neugebauer K, Fuerst M, et al. Syndecan-4 regulates ADAMTS-5 activation and cartilage breakdown in osteoarthritis. Nat Med 2009;15(9): 1072e6. Polur I, Lee PL, Servais JM, Xu L, Li Y. Role of HTRA1, a serine protease, in the progression of articular cartilage degeneration. Histol Histopathol 2010;25(5):599e608. Vlad SC, Neogi T, Aliabadi P, Fontes JD, Felson DT. No association between markers of inflammation and osteoarthritis of the hands and knees. J Rheumatol 2011;38(8):1665e70. Erhart-Hledik JC, Favre J, Asay JL, Smith RL, Giori NJ, Mundermann A, et al. A relationship between mechanicallyinduced changes in serum cartilage oligomeric matrix protein (COMP) and changes in cartilage thickness after 5 years. Osteoarthritis Cartilage 2012;20(11):1309e15. Byers S, Handley CJ, Lowther DA, Sriratana A. Carrageenininduced arthritis. VI. Alterations in amino acid transport by articular cartilage in acute inflammatory arthritis. Ann Rheum Dis 1985;44(7):477e84. Najafipour H, Ferrell WR. Nitric oxide modulates sympathetic vasoconstriction and basal blood flow in normal and acutely inflamed rabbit knee joints. Exp Physiol 1993;78(5):615e24. Watrin-Pinzano A, Loeuille D, Goebel JC, Lapicque F, Walter F, Robert P, et al. Quantitative dynamic contrast enhanced MRI of experimental synovitis in the rabbit knee: comparison of macromolecular blood pool agents vs. Gadolinium-DOTA. Biomed Mater Eng 2008;18(4e5):261e72. Achari Y, Reno CR, Frank CB, Hart DA. Carrageenan-induced transient inflammation in a rabbit knee model: molecular changes consistent with an early osteoarthritis phenotype. Inflamm Res 2012;61(8):907e14. Hart HF, Ackland DC, Pandy MG, Crossley KM. Quadriceps volumes are reduced in people with patellofemoral joint osteoarthritis. Osteoarthritis Cartilage 2012;20(8):863e8. Laverty S, Girard CA, Williams JM, Hunziker EB, Pritzker KP. The OARSI histopathology initiative e recommendations for histological assessments of osteoarthritis in the rabbit. Osteoarthritis Cartilage 2010;18(Suppl 3):S53e65. Felson DT. Osteoarthritis as a disease of mechanics. Osteoarthritis Cartilage 2013;21(1):10e5, http://dx.doi.org/10.1016/ j.joca.2012.09.012. Epub 2012 Oct 4. Blagojevic M, Jinks C, Jeffery A, Jordan KP. Risk factors for onset of osteoarthritis of the knee in older adults: a systematic review and meta-analysis. Osteoarthritis Cartilage 2010;18(1):24e33. Egloff C, Hugle T, Valderrabano V. Biomechanics and pathomechanisms of osteoarthritis. Swiss Med Wkly 2012;142: w13583. Slemenda C, Heilman DK, Brandt KD, Katz BP, Mazzuca SA, Braunstein EM, et al. Reduced quadriceps strength relative to body weight: a risk factor for knee osteoarthritis in women? Arthritis Rheum 1998;41(11):1951e9. Sawatsky A, Bourne D, Horisberger M, Jinha A, Herzog W. Changes in patellofemoral joint contact pressures caused by vastus medialis muscle weakness. Clin Biomech (Bristol, Avon) 2012;27(6):595e601. Benito MJ, Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B. Synovial tissue inflammation in early and late osteoarthritis. Ann Rheum Dis 2005;64(9):1263e7.