Background It is increasingly recognized that part of the pathology in osteoarthritis (OA) is driven by a low-grade inflammation in the synovium. Pro-inflammatory cytokines released locally during OA, such as S100A8/A9 which are expressed for prolonged periods when compared to IL-1β, IL-6, and TNF-α, may recruit monocytes from the bone marrow (BM) to the joint. In mice, two functionally distinct monocyte populations are described: (i) pro-inflammatory Ly6Chigh monocytes; and (ii) patrolling Ly6Clow monocytes.
Objectives The objective of our study is to investigate the role of S100A8/A9 in the recruitment of the different monocyte populations during early collagenase-induced OA (CiOA).
Methods S100A8 was intra-articularly injected into the knee joint of naïve wild type C57BL/6 (WT) mice to investigate their direct role in recruitment of monocytes. CiOA was induced by unilateral intra-articular collagenase injection in WT, and S100A9KO mice to investigate the role of S100A8/A9 in more detail. At CiOA day 7, WT and S100A9KO mice were sacrificed together with age-matched saline-injected control mice (n=6/group), and expression of several monocyte cell markers, chemokines, and adhesion molecules were measured in the synovium and BM, in which the cells were also analyzed by FACS. Monocytes were identified as (B220/CD90/CD49b/NK1.1/Ly6G)lowCD11bhigh(F4/80/MHCII/CD11c)low, which were further divided into subsets based on their Ly6C expression.
Results Injection of S100A8 into the knee joints of naïve mice led to a significantly elevated expression of monocyte-related markers (Ly6C, CCR2, and CX3CR1) and monocyte attracting chemokines (MCP-1, CX3CL1, MIP1α, and KC) within the synovium after 1 and 3 days, suggesting that S100A8/A9 is directly involved in the attraction of monocytes. At CiOA day 7 in WT mice, numbers of Ly6Chigh, but not Ly6Clow monocytes, were strongly increased (7.6-fold) in the synovium as compared to saline-injected control joints. In contrast, S100A9KO mice showed a significant increase in Ly6Clow monocytes (2-fold), whereas the number of Ly6Chigh monocytes remained unaffected. Concurrently, a strong upregulation of several chemokines (MCP1, CX3CL1, KC, and MIP1α) was observed locally in the synovium, of which only the Ly6Clow mobilization marker CX3CL1 was significantly higher in S100A9KO mice, corresponding with the increased Ly6Clow monocytes in the synovium of S100A9KO mice. This could however not explain the local increased number of Ly6Chigh monocytes at CiOA day 7 in WT mice, and therefore we next investigated the main source of the monocytes, which is the BM. We observed a decrease of 14% of myeloid cells (consisting partly of Ly6Chigh monocytes) in the BM of WT mice at CiOA day 7, whereas there were no changes in the BM of S100A9KO mice, suggesting that S100A8/A9 affects the release of myeloid populations from the BM. In line with that, expression of adhesion molecules (LFA-1, VCAM, VE-cadherin, PECAM1, and L-selectin) was lower at CiOA day 7 in the BM of S100A9KO mice when compared to WT mice.
Conclusions Local induction of OA leads to S100A8/A9 production, and subsequently to the mobilization of Ly6Chigh monocytes into the joint, driving OA pathology.
Disclosure of Interest None declared