Type I interferon effects on osteoclast fusion: Insights into non-erosive inflammatory arthritis

Abstract

Patients with systemic lupus erythematosus (SLE), an enigmatic autoimmune condition characterized by diverse clinical features and end-organ involvement, frequently develop painful joint inflammation but, on plain x-ray, rarely manifest erosive changes like those common in rheumatoid arthritis (RA). Peripheral blood mononuclear cells (PBMC) from SLE patients have an IFN-α signature. IFN-α has been shown to drive monocyte development toward myeloid dendritic cells (mDC), which play a key role in SLE pathogenesis. Interestingly, mDC share a common monocyte precursor with bone-resorbing osteoclasts (OC), the cellular mediator of erosive arthritis. Thus, we hypothesize that the IFN-α in SLE biases the process of monocyte differentiation toward mDC precursor development and away from osteoclast precursors (OCP) in a mutually exclusive fashion that can be characterized by the expression of unique surface molecules on the plasma membrane of precursor cells.

OC are formed by the fusion of heterogeneous mononuclear OCP; however, the phenotype and function of this heterogeneity are poorly understood. As elucidation of these differences could be important to diagnosis and treatment of erosive arthritis, we have characterized the expression of fusion-related surface proteins as potential biomarkers. One such surface biomarker is DC-STAMP, which has recently been identified as essential to OCP fusion. We show that RANKL induces two populations of OCP based on the surface expression of DC-STAMP: DC STAMP^lo cells that are potent OCP and produce pseudopods that attach to mononuclear DC-STAMP^hi cells, which cannot form OC on their own. Consistent with the phenotype of “master fusogenic OCP,” we found that DC-STAMP^lo cells express higher levels of key OC markers and fusion-related molecules relative to DC- STAMP cells. Based on these observations, we further hypothesize that IFN-α prevents erosive arthritis by biasing myelopoiesis away from DC-STAMP^lo OCP and downregulating the expression of fusion-related markers. In vitro, IFN-α inhibits the ability of RANKL to induce dc-stamp gene expression to levels seen without IFN-α. IFN-α also prevented the ability of RANKL to induce a DC-STAMP^lo population in a stage-dependent manner. To examine this in vivo, we used the NZBxNZW F1 animal model of SLE to examine the effect of elevated IFN-α signature (the IFN-α transcriptome) on the development of bone erosions in the setting of inflammatory arthritis. We found that the IFN-α transcriptome significantly influences the distribution of OCP and mDC precursors. Our findings also reveal significant positive correlations between the presence of an elevated IFN-α transcriptome and protection from arthritic bone erosions. We also found that mice with SLE-like disease and an elevated IFN-α transcriptome had a significantly lower percentage of DC-STAM^lo OCP compared to non-SLE controls in the setting of inflammatory arthritis.

These observations provide further support for the use of the DC-STAMP surface expression profile as a marker to identify the potential for OC-driven bone erosion. They also suggest that an explanation for the lack of erosive disease in SLE-associated arthritis may be the IFN-α-mediated prevention of DC-STAMP^lo master fusogenic OCP development and reduced gene expression of known OC fusion-related molecules. Thus, further understanding of the IFN-α effects on osteoclastogenesis may provide targets for the treatment of erosive inflammatory arthritis based on the shift in cell populations described.