Review of In Vitro Models and Development and Initial Validation of a Novel Co-Culture Model for the Study of Osteoarthritis

Abstract

Objective: Provide an overview of models used for the study of osteoarthritis (OA) and describe a novel in vitro model of OA using synovial and cartilage explants from dogs which we compared to spontaneously occurring OA in dogs. Design: Controlled laboratory study. Methods: Articular cartilage and synovial explants were created from canine cadaveric tissues (n=38 dogs) or tissues excised during surgery on clinical canine patients (n=17 hips). Matched cartilage and synovial explant co-cultures were compared to articular cartilage explant culture. Four groups were created based on presence or absence of synovium and presence or absence of IL-1β: Cartilage Culture Control (n=48), Cartilage Culture OA (n=48), Co-Culture Control (n=76), Co-Culture OA (n=76). Explants and media were evaluated on days 1, 3, 6, 12, and/or 20 of culture. For comparison to spontaneously occurring OA, articular cartilage and synovial tissues from dogs that had no evidence of OA, Clinical Control, and dogs that had OA, Clinical OA, were evaluated. Outcome measures included molecular, biochemical, and histologic assessments. Results: Gene expression levels in articular tissues correlated well between Co-Culture and Clinical Control and OA groups. GAG and collagen in normal and OA cartilage also corresponded well between Co-Culture and Clinical groups. Subjective histologic assessment revealed consistent similarities in cell and matrix characteristics between these groups. Changes in the media of the Co-Culture OA group were consistent with what has been reported for changes in synovial fluid of OA in dogs and humans. Conclusions: This co-culture model allows for assessment of cartilage and synovium, and can be used to provide a relatively comprehensive picture of spontaneously occurring OA in dogs. This model may have advantages over other models of OA in that the influences of both synovium and articular cartilage can be investigated at many levels and initial validation of the model using tissues from spontaneously occurring OA in dogs as the gold standard has been performed.