Pannexin 3, but not pannexin 1 is an important pro-catabolic mediator in osteoarthritis.

Purpose: Osteoarthritis (OA) is a disabling degenerative joint disease associated with high levels of individual suffering and a tremendous societal burden. Although much progress has been made identifying novel regulators of cartilage metabolism in OA, to date, no disease modifying therapies exist. This highlights the need for a deeper understanding of the underlying molecular mechanisms driving OA initiation and progression. Recently, we have identified Pannexin 3 (Panx3) as a novel pro-catabolic factor in OA and showed that both cartilage specific and global Panx3 deletion confers protection against surgically induced OA development. In follow up to this data, we have investigated the effects of a variety of signalling mediators on primary chondrocyte cultures isolated from wild-type (WT) and Panx3 KO mice to identify pathways upstream of Panx3 in chondrocytes. Additionally, we have investigated the effects of Panx1 deletion on surgically induced OA progression due to its homology to Panx3 and its overlapping expression pattern. Methods: In vitro studies utilized primary immature murine articular chondrocytes (IMAC) isolated from tibiae and femurs of 5 day-old mice. Cultures were treated with TNF, IL1, ATP and TNF plus the Pannexin inhibitor probenecid. Following 24 or 48 hours of treatment, cells were harvested and RNA or protein extracted. qPCR is used to analyze gene expression and western blotting to quantify protein levels. To examine the effects of Panx1 deletion on OA progression, destabilization of the medial meniscus was performed on the left knees of WT and Panx1 KO mice. Eight weeks post surgery, mice were euthanized and joints harvested, paraffinized and sectioned. Safranin-O/Fast Green stained sections were scored according to the OARSI histopathology scoring system. Immunohistochemical analysis of Panx3 was also performed on serial sections. Results: WT and Panx3 KO chondrocytes showed similar responses to all cytokines as measured through QPCR, indicating that the effect of Panx3 deletion observed in our in vivo studies is not mediated by alterations in pro-inflammatory responses. Treatment with probenecid had no effect on either WT or Panx3 KO chondrocytes. Interestingly however, in chondrocytes isolated from Rac1 KO mice, Panx3 expression was significantly elevated, indicating interplay between these two pathways in the onset of chondrocyte hypertrophy, possibly in the chondrocyte mechano-response. In contrast to the chondroprotective effects observed in the Panx3 KO mice, Panx1 KO's were not protected from the effects of surgical joint destabilization. Similar grade OA was observed in the KO DMM mice as in the WT mice, suggesting that Panx3 is of greater importance in this process. Additionally, we did not see increased Panx3 expression in Panx1 KO mice above what is typically seen in OA cartilage. Conclusions: Although no differences were observed in the responsiveness of Panx3 KO chondrocytes to various pro-inflammatory cytokines, the potential interplay between Rac1 and Panx3 provides us with an exciting new pathway, one we are currently investigating. Observing no differences in OA progression between Panx1 KO and WT DMM operated mice increases our confidence in the central role of Panx3 in the cartilage, and we are further investigating its role in aging associated OA in mice. Ultimately, these results provide novel insights into the specific molecular role of Panx3 in cartilage homeostasis and OA.