Establishing a functional osteochondral–synovial ex vivo joint model: Systematic evaluation of glucose and oxygen impact

Background Osteochondral–synovial co-culture ex vivo models represent an advanced platform for investigating joint biology and disease mechanisms, as they capture key tissue interactions, including bone–cartilage crosstalk and synovial inflammatory regulation. However, the broader application of these models remains limited by the lack of consensus regarding optimal glucose and oxygen levels in ex vivo joint models. This study aims to examine how glucose concentration in medium and oxygen availability during the culture impact the maintenance and metabolic activity of tissues in a complex osteochondral–synovial ex vivo co-culture model, with the aim of identifying optimal culture conditions suitable for controlled investigation of inter-tissue interactions, inflammatory and degenerative responses, and preclinical assessment of potential therapies prior to validation in human tissues. Methods Bovine osteochondral explants and synovial tissue were harvested from the stifle joint and maintained in static co-culture for 7 days. Cultures were established in either high (HG, 4.5 g/L, supra-physiological) or low (LG, 1 g/L, physiological) glucose Dulbecco's Modified Eagle Medium, under hyperoxic (21% O 2 ) or physioxic (5% O 2 ) conditions. Tissue responses were assessed independently for each tissue, using cell viability assays, glycosaminoglycan and nitric oxide release measurements, histological analysis, gene expression profiling, and untargeted metabolomics. Results LG culture conditions increased cell death in the deep zone of cartilage and in subchondral bone compared with HG medium. Gene expression and metabolomics analyses revealed tissue-specific responses to glucose and oxygen. In cartilage and bone, glucose-driven transcriptional and metabolic changes occurred under hyperoxia but were largely absent under physioxia. In synovium, gene-specific sensitivity to glucose and oxygen was also observed; however, glucose-dependent metabolic responses persisted under physioxia. Conclusion This systematic evaluation of glucose and oxygen combinations across cartilage, subchondral bone, and synovium demonstrates that, HG medium under physioxic conditions best preserves tissue viability and stabilizes molecular responses, particularly in osteochondral tissues. These findings provide evidence-based guidance for optimizing static ex vivo joint models and improving their reproducibility and translational value. The translational potential of this article This study established optimized culture conditions for a bovine osteochondral–synovial ex vivo co-culture model, improving its translational relevance for the controlled investigation of tissue responses and preclinical screening of drugs and biomaterials prior to validation in human models.