Background Thickening of the synovial lining layer comprising macrophages expressing IL-1 and S100A8/A9 is evident in a substantial subpopulation of patients with early osteoarthritis (OA) (1) and has been associated with pathophysiology of OA (2). Recently we have shown that adipose derived stem cells (ADSC) inhibit joint destruction after local application to knee joints with experimental OA (3). Moreover it is shown that the degree of immunosuppression of ADSCs is regulated by pro-inflammatory cytokines(4).

Objectives To explore the effect of synovial activation on the immunomodulatory capacity of ADSCs on joint destruction after local administration to collagenase-induced osteoarthritic joints differing in synovial activation.

Methods ADSCs were isolated from fat surrounding the popliteal lymph nodes and ADSCs were cultured for two weeks. Collagenase induced OA (CIOA) was induced by injection of collagenase into murine knee joints. ADSCs were injected into knee joints at day 7 after induction of OA. OA phenotype was measured within 6 weeks after induction. Total knee joints were isolated and performed for histology. Synovial activation and new formation cartilage/bone was measured using an arbitrary scale of 0-3. Synovium was isolated at various time-points after injection of ASC and washouts were measured for S100A8/A9 and IL-1 using Luminex.

Results A single dose of ADSCs (20X10³) was injected into the knee joint of mice, 7 days after induction of CIOA. Thickness of the synovial lining layer, characteristic for this model, was high and was significantly inhibited by ADSC treatment at day 14 (9%) and day 42 (35%) when compared to control (serum) treated OA joints. Washouts of synovium taken at various time-points (6 hrs, 48 hrs, days 14 and 42) after injection of ADSCs showed that IL-1 and S100A8/A9 were significantly decreased already 48 hrs after injection of ADSC (57% and 22% respectively). Serum levels of S100A8/A9 were inhibited by 85% at day 14 after ADSC treatment suggesting that the effect on synovial activation is very rapid. Strikingly, ADSCs had a protective effect on new formation of cartilage inside medial and cruciate ligaments at day 42 after treatment (inhibition by 92% and 43% respectively). Next, we explored the effect in a condition with less synovial inflammation. Synovial thickness at day 42 was 62% lower when compared to the former study. Injection of the same dose of ADSCs at day 7 after induction of OA, only marginally inhibited synovial thickening when measured at day 42. Serum levels of S100A8/A9 were low at day 14 (around 50 ng/ml compared to 800 ng/ml) and were not altered by the ADSC treatment. New formation of cartilage inside the collateral and cruciate ligaments was also not diminished.

Conclusions Our study indicates that synovial activation rapidly drives suppressive effects of ADSCs after local administration in murine knee joints with experimental OA and protects against development of ligament damage.

References

1. Bondeson J et al. Arthritis Rheum. 2010;62(3):647-57.

2. Blom AB et al. Arthritis Rheum. 2007;56(1):147-57.

3. ter Huurne M et al. Arthritis Rheum. 2012;64(11):3604-13

4. Crop MJ et al. Clin Exp Immunol. 2010;162(3):474-86.

Disclosure of Interest None Declared