Background/Purpose: The intracellular DNA sensor Stimulator of Interferon Genes (STING) is essential for detection of viral and bacterial pathogen DNA. As with other pathways in the innate immune system, hyperactivation of the STING pathway is associated with autoinflammatory and autoimmune diseases. Activation of STING promotes anti-viral responses through production of type I interferons, including IFNβ. A pleiotropic cytokine, IFNβ is a potent stimulus of the innate immune system and has also been shown to inhibit the formation of osteoclasts. We therefore sought to determine whether the STING pathway impacts bone homeostasis and osteoclastogenesis.

Methods: Osteoclast formation assays were performed using RANKL-stimulated bone marrow progenitors from STING deficient (KO) mice and myeloid-specific STING KOs.  The immortalized myeloid cell line RAW294.7 was engineered to stably overexpress murine STING or an empty vector control. RANKL was used to stimulate osteoclast precursors to form osteoclast-like cells in vitro.  Secreted IFNβ was monitored using ELISA.  IFNβ activity was inhibited using an IFNaR1 blocking antibody (Biolegend).  Trabecular and cortical bone parameters were determined in femurs from aging (6, 13, and 15 months) STING whole body KO mice and littermate controls using microCT. Ovariectomies in myeloid-specific-STING KOs (STING^flox/flox LysM^+/-, hereafter referred to STING^LysM KO) and littermate controls (STING^flox/flox LysM^-/-), were performed in young female mice and bone parameters were evaluated by microCT 8 weeks after surgery.

Results: Osteoclast differentiation assays demonstrate that young STING KO bone marrow progenitors develop into osteoclasts faster than littermate controls. This phenotype of increased osteoclast formation is sustained in progenitors from aged (6 month old) STING KO mice. The STING^LysM KO precursor cells recapitulate the osteoclast formation phenotype of STING KO cells in vitro. A reciprocal effect is observed in RANKL-stimulated RAW294.7 cells overexpressing STING, where fewer osteoclast-like cells are generated compared to RAW294.7 cells containing an empty vector.  STING overexpressing RAW294.7 cells produce more IFNβ than empty vector controls, and blocking IFNaR1 downstream of IFNβ limited osteoclast formation in STING over expressing RAW294.7 cells. To determine the role of STING for bone turnover in vivo, microCT was used to characterize bone in aging cohorts of STING KO mice.  Consistent with these in vitro observations, STING KO mice lose trabecular bone faster than wildtype (WT) littermate controls.  Myeloid STING is sufficient for enhanced bone turnover, as ovariectomized STING^LysM KO mice exhibit a larger decrease in trabecular bone compared to littermate controls. 

Conclusion: Our results are the first to implicate any of the cytosolic DNA sensors in the maintenance of bone. We demonstrate thatincreased STING activity limits osteoclast formation, an effect which is mediated at least in part by IFNβ.  Moreover, loss of STING activity results in accelerated bone loss.  These data suggest that STING activation plays an important role inthe protection from bone loss in the settings of viral infection and inflammation in inflammatory rheumatic diseases. 

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-intracellular-dna-sensor-sting-protects-against-bone-loss-through-regulation-of-type-i-interferons/