K. Coeleveld

A single intra‐articular injection with IL‐4 plus IL‐10 ameliorates blood‐induced cartilage degeneration in haemophilic mice

The combination of interleukin (IL)‐4 and IL‐10 protects against blood‐induced cartilage damage in vitro. It has been hypothesized that the combination of these cytokines is effective if applied early in the process of cartilage damage. The present study investigated whether a single intra‐articular injection of IL‐4 plus IL‐10 immediately after a joint bleed limits cartilage damage in an in vivo haemophilia mouse model of blood‐induced joint damage. Factor VIII knockout mice with severe haemophilia A were punctured once with a needle below the patella to induce a joint haemorrhage. Subsequently IL‐4 plus IL‐10 (n = 24) or vehicle (n = 24) was injected intra‐articularly. After 35 days, the time needed for development of detectable joint degeneration, knee joints were examined for cartilage damage by macroscopic and microscopic evaluation. A single intra‐articular injection of IL‐4 plus IL‐10 ameliorated progression of cartilage degeneration caused by a single joint bleed to a certain extent. No effect on inflammation was observed at this time point. A single intra‐articular injection of IL‐4 plus Il‐10 directly after a single joint bleed limits progression of cartilage degeneration over time. Improved bioavailability (half‐life) of both cytokines might improve their protective ability in the development of cartilage degeneration, and probably also inflammation.

IL-1β, in contrast to TNFα, is pivotal in blood-induced cartilage damage and is a potential target for therapy.

Joint bleeding after (sports) trauma, after major joint surgery, or as seen in hemophilia in general leads to arthropathy. Joint degeneration is considered to result from the direct effects of blood components on cartilage and indirectly from synovial inflammation. Blood-provided proinflammatory cytokines trigger chondrocytes and induce the production of cartilage-degrading proteases. In the presence of erythrocyte-derived iron, cytokines stimulate radical formation in the vicinity of chondrocytes inducing apoptosis. To unravel the role of interleukin (IL) 1β and tumor necrosis factor (TNF) α in the pathogenesis of this blood-induced cartilage damage, the effect of antagonizing these cytokines was examined in human in vitro cultures. Addition of recombinant human IL-1β monoclonal antibody or IL-1 receptor antagonist resulted in a dose- and time-dependent protection of cartilage from blood-induced damage. In higher concentrations, almost complete normalization of cartilage matrix proteoglycan turnover was achieved. This was accompanied by a reduction in IL-1β and IL-6 production in whole blood cultures, whereas TNFα production remained unaffected. Interestingly, addition of a TNFα monoclonal antibody, although demonstrated to inhibit the direct (transient) effects of TNFα on cartilage, exhibited no effect on blood-induced (prolonged) cartilage damage. It is demonstrated that IL-1β is crucial in the development of blood-induced joint damage, whereas TNFα is not. This hierarchical position of IL-1β in blood-induced joint damage warrants studies on targeting IL-1β to potentially prevent joint degeneration after a joint bleed.

Hemarthrosis in hemophilic mice results in alterations in M1-M2 monocyte/macrophage polarization

INTRODUCTION Joint bleedings result in iron-mediated synovitis and cartilage destruction. Monocyte/macrophage polarization affects their role in iron homeostasis. This study evaluates the effects of hemarthrosis on monocyte/macrophage polarization. MATERIALS AND METHODS Using a murine hemophilia model of acute joint bleeding and flow cytometry, we evaluated monocyte/macrophage polarization in blood, spleen, synovium, and knee lavage at day 1, 2, and 7 following the induction of hemarthrosis. RESULTS Induction of hemarthrosis resulted in a transient shift of blood monocytes towards a M1 type (control 13 vs. 1847 counted cells at day 1; p<0.01), a temporary decrease of spleen M1 monocytes (control 2841 vs. 1086 counted cells at day 1; p=0.02), and a sustained decrease of spleen M2 red pulp macrophages (control 1853 vs. 673 counted cells at day 7; p=0.01). In addition, an increase in M1 (control 119 vs. 592 counted cells at day 1; p=0.04) and M2 (control 247 vs. 650 counted cells at day 1; p=0.02) synovial macrophages was noted. In the joint lavage, a temporary increase in M1 monocytes (control 20 vs. 125 counted cells at day 1; p=0.04) and a more sustained increase in M2 monocytes (control 73 vs. 186 counted cells at day 2; p<0.01) was observed. CONCLUSIONS This study demonstrates alterations in monocyte/macrophage polarization following hemarthrosis resulting in a blood monocyte M1 phenotype and a combined M1-M2 monocyte/macrophage phenotype in the joint. Based on the different capabilities of M1 and M2 cells, modulating polarization of distinct monocyte/macrophage populations might represent interesting prophylactic or therapeutic approaches for joint bleedings.

Haemarthrosis stimulates the synovial fibrinolytic system in haemophilic mice

Recurrent joint bleeding is the most common manifestation of haemophilia resulting in haemophilic arthropathy (HA). The exact pathophysiology is unknown, but it is suggested that arthropathy is stimulated by liberation of fibrinolytic activators from the synovium during haemarthrosis. The aim of this study was to test the hypothesis that haemarthrosis activates the local synovial fibrinolytic system in a murine haemophilia model. The right knees of haemophilic and control mice were punctured to induce haemarthrosis. The left knees served as internal control joints. Synovial levels of urokinase-type plasminogen activator (uPA), plasminogen activator inhibitor 1 (PAI-1), plasmin, and alpha-2-antiplasmin (A2AP) were compared between the punctured and control knees. In haemophilic mice, an increase in synovial cells expressing urokinase-type plasminogen activator (uPA) in the right punctured knee versus the left unaffected knee was observed: (47% vs 43%) (p=0.03). Additionally, in haemophilic mice, haemarthrosis induced an increase in uPA (0.016 ng/ml vs 0.01 ng/ml) (p=0.03) and plasmin (0.53 μg/ml vs 0.46 μg/ml) (p=0.01) as promoters of fibrinolysis. Synovial levels of PAI-1 (0.32 ng/ml vs 0.17 ng/ml) (p<0.01) was also increased, whereas synovial levels of A2AP were unchanged: (0.021 μg/ml vs 0.021 μg/ml) (p=0.15). Enhanced uPA production was confirmed in human stimulated synovial fibroblast cultures and elevated levels of plasmin were confirmed harmful to human cartilage tissue explants. In this study we demonstrate that haemarthrosis in haemophilic mice induces synovial uPA expression and results in an increase in synovial plasmin levels, making the joint more vulnerable to prolonged and subsequent bleedings, and adding directly to cartilage damage.

Antiplasmin, but not amiloride, prevents synovitis and cartilage damage following hemarthrosis in hemophilic mice.

Blood‐induced joint damage is characterized by synovitis and cartilage damage. Recently, we demonstrated that joint bleeding in hemophilic mice results in elevated synovial levels of urokinase plasminogen activator (u‐PA) and plasmin, and in plasmin‐mediated cartilage damage.

Deferasirox limits cartilage damage following haemarthrosis in haemophilic mice

Joint bleeds in haemophilia result in iron-mediated synovitis and cartilage damage. It was evaluated whether deferasirox, an iron chelator, was able to limit the development of haemophilic synovitis and cartilage damage. Haemophilic mice were randomly assigned to oral treatment with deferasirox (30 mg/kg) or its vehicle (control) (30 mg/kg). Eight weeks after start of treatment, haemarthrosis was induced. After another five weeks of treatment, blood-induced synovitis and cartilage damage were determined. Treatment with deferasirox resulted in a statistically significant (p< 0.01) decrease in plasma ferritin levels as compared to the control group (823 ng/ml ± 56 and 1220 ng/ml ±114, respectively). Signs of haemophilic synovitis, as assessed by the Valentino score [range 0 (normal) - 10 (most affected)], were not different (p=0.52) when comparing the control group with the deferasirox group. However, deferasirox treatment resulted in a statistically significant (p< 0.01) reduction in cartilage damage, as assessed by the loss in Safranin O staining [range 0 (normal) - 6 (most affected)], when comparing the deferasirox group with the control group: score 2 (65.4 % vs 4.2 %), score 3 (26.9 % vs 4.2 %), score 4 (7.7 % vs 20.8 %), score 5 (0 % vs 54.2 %), and score 6 (0 % vs 16.7 %). Treatment with deferasirox limits cartilage damage following the induction of a haemarthrosis in haemophilic mice. This study demonstrates the role of iron in blood-induced cartilage damage. Moreover, these data indicate that iron chelation may be a potential prevention option to limit the development of haemophilic arthropathy.

The groove model of tibia-femoral osteoarthritis in the rat.

Several experimental models of osteoarthritis in rats are used to study the pathophysiology of osteoarthritis. Many mechanically induced models have the limitation that permanent joint instability is induced by, for example, ligament transection or meniscal damage. This permanent instability will counteract the potential beneficial effects of therapy. The groove model of osteoarthritis uses a one‐time trigger, surgically induced cartilage damage on the femoral condyles, and has been validated for the canine tibia‐femoral compartment. The present study evaluates this model for the rat knee joint. The articular cartilage of the weight bearing surface of both femoral condyles and trochlea were damaged (grooved) without damaging the underlying subchondral bone. Severity of joint degeneration was histologically assessed, in addition to patella cartilage damage, and subchondral bone characteristics by means of (contrast‐enhanced) micro‐CT. Mild histological degeneration of the surgically untouched tibial plateau cartilage was observed in addition to damage of the femoral condyles, without clear synovial tissue inflammation. Contrast enhanced micro‐CT demonstrated proteoglycan loss of the surgically untouched patella cartilage. Besides, a more sclerotic structure of the subchondral bone was observed. The tibia‐femoral groove model in a rat results in mild knee joint degeneration, without permanent joint instability and joint inflammation. This makes the rat groove model a useful model to study the onset and progression of post‐traumatic non‐inflammatory osteoarthritis, creating a relatively sensitive model to study disease modifying osteoarthritic drugs.

A fusion protein of interleukin‐4 and interleukin‐10 protects against blood‐induced cartilage damage in vitro and in vivo

Essentials Targeted treatment for hemophilic arthropathy, still causing significant morbidity, is lacking. This study evaluates the efficacy of a fusion of protein of interleukin(IL)‐4 and IL‐10. In vitro the fusion protein prevents blood‐induced cartilage damage in a dose‐dependent manner. In hemophilic mice, the IL4‐10 fusion protein ameliorates cartilage damage upon joint bleeding.

Lack of a clear disease modifying activity of celecoxib in treatment of end-stage knee osteoarthritis: a rondomized observer blinded clinical trial

Objective: To evaluate the in vivo disease modifying activity of the selective COX-2 inhibitor celecoxib, compared to no-treatment and naproxen, treating end-stage knee osteoarthritis, using detailed ex vivo tissue analyses. Methods: Patients (n=172) with end-stage knee OA were randomized to 4 groups and treated for 4 weeks prior to knee replacement surgery: celecoxib 2dd200mg, naproxen 3dd250mg, celecoxib 2dd200mg stopped 3 days prior to surgery, or no-treatment. Cartilage and synovial tissue collected during surgery were analyzed ex vivo, with cartilage proteoglycan release as primary outcome. Additionally, several markers of synovial inflammation and clinical effects were determined. Results: Data of 138 patients could be analyzed, 34 patients were lost for several reasons. The expression of COX-2 in both cartilage and synovial tissue was statistically significant decreased in patients treated with celecoxib (p=0.017 and p=0.001) respectively), indicating the drug has reached the knee joint within the treatment period. Nonetheless, no significant effect on proteoglycan release, retention or content was found. Synovial inflammation markers did not show any statistically significant decrease although nitric oxide levels in celecoxib treated patients suggest a beneficial effect of celecoxib compared to no treatment. WOMAC scores did not statistically significant improve after treatment; though celecoxib treated patients reported a slightly higher WOMAC pain score compared to non-treated patients. Conclusion: No direct effect on cartilage upon short term in vivo treatment of knee OA patients with celecoxib could be detected, although decreased expression of COX-2 confirmed its intra-articular availability. Effects on synovial inflammatory mediators and clinical outcome seemed only limited. As such the previous reported disease modifying effects of celecoxib in in vitro and pilot clinical studies could not unambiguously be confirmed.

Metabolic dysregulation accelerates injury-induced joint degeneration, driven by local inflammation; an in vivo rat study†

Evidence is growing for the existence of an obesity‐related phenotype of osteoarthritis in which low‐grade inflammation and a disturbed metabolic profile play a role. The contribution of an obesity‐induced metabolic dysbalance to the progression of the features of osteoarthritis upon mechanically induced cartilage damage was studied in a rat in vivo model. Forty Wistar rats were randomly allocated 1:1 to a standard diet or a high‐fat diet. After 12 weeks, in 14 out of 20 rats in each group, cartilage was mechanically damaged in the right knee joint. The remaining six animals in each group served as controls. After a subsequent 12 weeks, serum was collected for metabolic state, subchondral bone changes assessed by μCT imaging, osteoarthritis severity determined by histology, and macrophage presence assessed by CD68 staining. The high‐fat diet increased statistically all relevant metabolic parameters, resulting in a dysmetabolic state and subsequent synovial inflammation, whereas cartilage degeneration was hardly influenced. The high‐fat condition in combination with mechanical cartilage damage resulted in a clear statistically significant progression of the osteoarthritic features, with increased synovitis and multiple large osteophytes. Both the synovium and osteophytes contained numerous CD68 positive cells. It is concluded that a metabolic dysbalance due to a high‐fat diet increases joint inflammation without cartilage degeneration. The dysmetabolic state clearly accelerates progression of osteoarthritis upon surgically induced cartilage damage supported by inflammatory responses as demonstrated by histology and increased CD68 expressing cells localized on the synovial membrane and osteophytes.