In Vitro And In Vivo Characterisation Of Senescence Markers In Osteoarthritis

Purpose: Osteoarthritis (OA) is the most prevalent osteoarticular disease and a major cause of disability in aging adult. OA is characterized by cartilage breakdown, synovial inflammation, osteophyte formation, and subchondral bone remodelling. While OA is associated with a huge societal and economic burden, no efficient disease-modifying drug still exists today. In addition to obesity, metabolic disorders, and trauma, one of the major risks of OA is ageing. Ageing is a complex physiological process that relates to a gradual decline in certain functions, coordination, loss of homeostasis, and physiological integrity. Among the underlying mechanisms associated with age-related diseases including OA, an accumulation of senescent cells has been observed. Interestingly enough, the elimination of such senescent cells has been shown to reduce pain and increase cartilage regeneration in murine post-traumatic OA. To further decipher the role of senescence in OA onset and development, our work aims to describe the location of senescent cells in vivo within the different joint tissues involved in age-related OA as well as investigate the in vitro senescence-associated inflammatory response of chondrocytes. Methods: To assess the expression of senescence markers during age-related OA, knee joints of male C57BL/6 mice of increasing age (1, 6, 12, 18, and 24 months) were harvested. Mice knee joints (n=5) of each age group were decalcified and embedded in paraffin. 5μm sections were stained with Safranin-O for OARSI scoring. Immunohistochemistry for senescence markers (p21CIP-1, p16INK4A, PAI-1, and caveolin-1 (CAV-1)), as well as cartilage extracellular matrix proteins (type II collagen, and the aggrecan neoepitope NITEGE), were performed. In vitro, to investigate the relationship between inflammation and senescence in chondrocytes, TC28a2 chondrocyte cell line and primary human articular chondrocytes (hAC) seeded at 15,000 cells /cm2 were stimulated with IL-1β (10 ng/ml; 24 to 72h). To induce senescence, TC28a2 and hAC were seeded at 15,000 cells /cm2 and treated with etoposide (20μM; 24h) and maintained in culture up to 192h. The effects of IL-1β and etoposide on senescence (p16Ink4A, p21CIP1, p53, PAI-1) and chondrocyte markers (MMP3, MMP13, COL2A1, and ACAN) were evaluated at the transcript and protein levels. Results: Our in vivo results showed a gradual increase in the expression of senescence markers (P16INK4A, P21CIP-1, PAI-1, and CAV-1) with increasing ages, which correlates with an increase in OA score as well as the detection of the aggrecan degradation neoepitope NITEGE. Consistently, an age-related decreased in type II collagen was observed in the knee joints of mice with escalating ages. IL 1-β treatment of TC28a2 and hAC resulted in a decrease in COL2A1 and ACAN at 48h and 72h. As expected, an increase in inflammatory (COX2) and catabolic markers (MMP-3 and MMP-13) from 24h to 72h, was observed. However, no significant modification of senescence markers (p16Ink4A, p21CIP1, p53, and PAI-1) expressions was observed. Etoposide treatment induced a drop in cell proliferation from 24h to 72h before a stabilization in the numbers of viable cells up to 192h in both cell types. Following etoposide treatment, we observed a significant decrease in COL2A1 expression at 72h in both cell types and the induction of MMP-3 and MMP-13 at 24h and 72h in human chondrocytes. Etoposide treatment also triggered an augmentation of the senescence markers p21CIP1 and PAI-1 as early as 24h in both types of chondrocytes. Conclusions: To summarize, our in vivo Results highlights that senescence increases in correlation with age and OA severity in a spontaneous age-related OA murine model. Our in vitro data also suggest that IL-1β, a cytokine commonly used to in vitro simulate the OA-associated inflammation, did not induce a marked cellular senescence in human chondrocytes when cultured in our conditions. On the contrary, the topoisomerase inhibitor, etoposide was found to induce senescence in the TC28a2 chondrocytic cell line and primary human articular chondrocytes. This model of etoposide-induced senescence in chondrocytes could be a useful tool to study the complex role of senescence in OA chondrocytes as well as evaluate anti-OA therapeutic approaches targeting senescence. Acknowledgements: This work was supported by research grants from FOREUM Foundation for Research in Rheumatology (SEN-OA) and the ANR project KLOTHOA ( ANR-18-CE14-0024-01 ).