II-10 Role of macrophage-driven autoinflammation in SLE

Abstract

Background Although SLE is a prototype of autoimmune (T cell/B cell-mediated) disease, there is increasing evidence implicating myeloid cells in its pathogenesis. We examined the role of macrophages (MΦ) in human SLE and the pristane-induced lupus model. Methods Tissues of SLE patients and mice with pristane-induced lupus were examined by double immunohistochemistry. MΦ were analyzed by flow cytometry, phagocytosis assay, real-time PCR, Seahorse assay, and RNA-Seq. Mice were injected daily with the liver X receptor (LXR) agonist T0901317. Results SLE patients’ bone marrow contained numerous activated caspase-3+ apoptotic cells located outside of MΦ. In contrast, in leukemia patients undergoing bone marrow ablation prior to transplantation, all caspase-3+ cells were inside of MΦ, suggesting that lupus is associated with impaired clearance of apoptotic cells. MΦ in pristane-lupus took up fluorescently-labeled apoptotic cells poorly compared with controls. Studies in pristane-treated knock-out mice indicated that the induction of diffuse alveolar hemorrhage (DAH), a serious pulmonary complication of lupus, required opsonization of dead cells in the lung by ‘natural’ IgM (DAH absent in μMT mice, restored by infusion of IgM), C3, and C3b receptors (absent in C3-/- and CD18-/- mice, prevented by complement depletion with cobra venom factor). DAH also was prevented by MΦ depletion (clodronate liposomes) but not neutrophil depletion (anti-Ly6G). MΦ in pristane-induced lupus exhibited features of classical activation (impaired phagocytosis of apoptotic cells, high glycolysis, low oxidative phosphorylation, increased HIF1α expression, M1 surface markers, and TNFα production), whereas control MΦ from mice treated with mineral oil (do not develop DAH) were M2-like, with high phagocytic activity, low glycolysis, high OxPhos, increased LXRα expression, M2 surface markers, and IL-10 production. In both pristane-induced lupus MΦ and SLE patients’ monocytes, we found low levels of the LXRα-regulated reverse cholesterol transporter ABCA1. We hypothesized that it might be possible to treat DAH by ‘re-polarizing’ MΦ using a synthetic LXR agonist. Consistent with that idea, pristane-treated mice receiving T0901317 did not develop DAH and MΦ from mice receiving T0901317 exhibited an M2-like surface phenotype and decreased TNFα production compared with controls. Conclusions Our data suggest that lupus-associated DAH is partly an autoinflammatory response caused by sluggish clearance of apoptotic cells by M1-like MΦ and that certain clinical manifestations (e.g. DAH) can be treated by repolarizing MΦ using an activator of the transcription factor LXRα. Low levels of LXR-driven ABCA1 expression also may have implications for the therapy of atherosclerosis in SLE.