Bidirectional STAT3 signaling in osteoclastogenesis and osteolysis: context- and stage-dependent effects

Excessive osteoclast-mediated bone resorption contributes to osteoporosis, inflammatory bone destruction, and tumor-associated osteolysis. Signal transducer and activator of transcription 3 (STAT3) links cytokine signaling and inflammation to bone remodeling. However, its role in osteoclastogenesis and osteolysis remains unresolved, as both pro- and anti-osteoclastogenic effects have been reported. This review examines STAT3 across three layers of osteoclast regulation: receptor activator of nuclear factor-κB ligand (RANKL)-producing cells, hematopoietic/myeloid precursors, and osteoclast-lineage cells. These layers correspond to osteoclastogenic signal production, precursor availability, and lineage-intrinsic differentiation and function, respectively. Within this framework, cellular compartment, differentiation stage, STAT3 phosphorylation status, and upstream signals help explain opposing effects. Current evidence suggests that STAT3 generally supports osteoclastogenesis in RANKL-producing cells and in proinflammatory, hypoxic, or tumor settings through RANKL induction or cytokine signaling. During osteoclast differentiation, pro-osteoclastogenic effects are linked to transcriptional programs centered on nuclear factor of activated T cells 1. Emerging studies suggest a role for STAT3 in metabolic and redox regulation during osteoclast differentiation and function. Anti-osteoclastogenic effects have been reported in settings where STAT3 limits osteoclast precursor abundance, restrains signaling mediated by receptor activator of nuclear factor-κB and Toll-like receptor 4 in bone marrow-derived macrophages, or acts downstream of inhibitory immune and hormonal signals. In osteoclast precursor models, interleukin-6-induced Y705 phosphorylation may favor a macrophage-like program, whereas basal S727 phosphorylation has been linked to osteoclast differentiation. The distinct in vivo roles of phosphorylation at these sites in the osteoclast lineage remain unclear. Divergent phenotypes in Ctsk -Cre models highlight the importance of driver design, recombination timing, cellular specificity, and endpoint selection. This framework may reconcile discrepancies in the literature and inform the evaluation of cell- and stage-selective STAT3 targeting in osteolytic diseases.