NOX2-Driven Oxidative Stress Promotes EndMT and Uncouples Angiogenesis–Osteogenesis at the Bone–Implant Interface in Diabetes

Abstract

Diabetes undermines implant integration, yet how the bone–implant interface (BII) can be rationally re-engineered remains unclear. Here endothelial NOX2–driven oxidative stress is identified as a mechanistic switch that triggers endothelial-to-mesenchymal transition (EndMT), depletes type-H (CD31+EMCN+) vessels, and uncouples angiogenesis from osteogenesis. In a diabetic titanium-implant model, type-H endothelium and adjacent osteoprogenitors declined early (≈40% loss at 2 weeks), and peri-implant bone volume is reduced at 8 weeks (BV/TV 40.98 ± 3.96%). Pharmacologic NOX2 inhibition or endothelial-specific Nox2 deletion restored endothelial identity, suppressed EndMT and apoptosis, rebuilt type-H networks, and improved bone formation (BV/TV 53.15 ± 4.97%), yielding higher BV/TV and histological indices. In vitro, on titanium surfaces, NOX2 blockade rescued endothelial proliferation, migration, and adhesion architecture and re-enabled osteogenesis in EC–OB co-culture. Bulk RNA-seq demonstrated pathway reversals (TGF-β, NF-κB/MAPK, Wnt/Notch, TNF) and attenuated EC to OB ephrin/plexin edges, consistent with guidance/adhesion rewiring. These findings position NOX2 as an actionable vascular target at the BII and suggest interface-focused delivery (drug coatings, anti-EndMT functional surfaces, responsive hydrogels) to re-establish vessel–bone coupling and strengthen osseointegration in diabetes.