Tianchao Lu, Dong Wang, Maoqi Gong, Ziyi Li, Hanzhou Wang, Junlin Zhou
Together, these results nominate fibroblast-associated LIF as a niche-related inhibitory signal in nonunion, with effects that are most compatible with canonical LIFR/IL6ST-mediated STAT3 signalling.
INTRODUCTION: Fracture nonunion represents a biologically dysregulated repair state rather than a purely mechanical failure. This study investigated whether fibroblast- derived Leukemia Inhibitory Factor (LIF) contributes to nonunion by suppressing osteogenesis via canonical LIFR-containing STAT3 signalling.
MATERIALS AND METHODS: We profiled clinical fracture-healing tissues using single-cell RNA sequencing, integrating three normally healed control specimens and three nonunion specimens after quality control. Mechanistic validation was performed using a Transwell coculture of mouse L-929 fibroblasts and MC3T3-E1 osteoblast-lineage cells, followed by pharmacological rescue and evaluation in a rat tibial nonunion model.
RESULTS: Computational analyses prioritised fibroblast-to-osteoblast signalling and identified the enhanced LIF-(LIFR+IL6ST) interaction as a candidate signal in nonunion. Human tissue immunofluorescence supported increased fibroblast-associated LIF localisation in nonunion lesions. In Transwell cocultures of L-929 fibroblasts and MC3T3-E1 osteoblast- lineage cells, LIF knockdown reduced STAT3 phosphorylation and enhanced osteogenic gene expression, alkaline phosphatase activity and mineralisation, whereas LIF overexpression produced opposite effects. Pharmacological inhibition of LIFR (EC359) or STAT3 (Stattic) partially rescued the anti-osteogenic effects of LIF overexpression. In a rat tibial nonunion model, EC359 improved callus continuity and increased BV/TV and Tb.N at both 2 and 4 weeks, with additional trabecular improvements at 4 weeks; histomorphometry confirmed significantly increased Callus/ROI and Blue/Callus ratios at 4 weeks.
DISCUSSION: The convergent findings support fibroblast-associated LIF as a biologically plausible inhibitory signal in nonunion. The human, cellular, and animal models provide complementary rather than directly interchangeable evidence, and direct clinical generalisation remains limited.
CONCLUSION: Together, these results nominate fibroblast-associated LIF as a niche-related inhibitory signal in nonunion, with effects that are most compatible with canonical LIFR/IL6ST-mediated STAT3 signalling.