**Background:** Traumatic brain injury (TBI) is a leading cause of long-term neurological disability, with over 10 million cases annually worldwide. Despite intensive research, no effective pharmacological treatment exists for TBI-induced cognitive deficits. Moderate-intensity physical exercise has emerged as a promising non-pharmacological intervention, but its mechanisms remain unclear. Ferroptosis, an iron-dependent form of regulated cell death driven by lipid peroxidation and glutathione depletion, is implicated in TBI pathophysiology, particularly in the acute phase. The stimulator of interferon genes (STING) pathway, known for its role in neuroinflammation, has recently been linked to ferroptosis in other diseases. This study hypothesized that moderate treadmill exercise inhibits TBI-induced ferroptosis via STING pathway suppression at the chronic stage.
**Methods:** Adult male C57BL/6J mice (6–8 weeks) were subjected to controlled cortical impact (CCI) TBI. In Experiment 1, mice were divided into sham, run (exercise only), TBI, and TBI+run groups (n=20 each). Exercise consisted of moderate-intensity treadmill running (10–15 m/min, 0° incline, 30 min/day, 5 days/week) from days 4 to 35 post-TBI. In Experiment 2, mice received rAAV2/9-shRNA targeting STING (Sh-STING) or scrambled control, followed by TBI (n=12/group). In Experiment 3, mice received rAAV9-hSyn-STING1 (AAV-STING) or control, then TBI and exercise (n=12/group). Outcomes assessed at 44 days post-TBI included: western blotting for ferroptosis proteins (Tfr1, Fth, Fpn, Acsl4, xCT, Gpx4), iron content (tissue and serum), MDA and GSH/GSSG levels, Perls' Prussian blue staining for iron deposition, Fluoro-Jade B (FJB) staining for neurodegeneration, Nissl staining for neuronal damage, immunohistochemistry for 4HNE, and behavioral tests (open field test at day 36, three-chamber social test at day 38, Morris water maze at days 40–44).
**Key Results:** TBI induced significant ferroptosis at 44 days post-injury: increased Tfr1, Fth, Acsl4, iron content (both cortex and serum), MDA, and 4HNE-positive cells; decreased Fpn, xCT, Gpx4, and GSH/GSSG ratio (all p<0.05–0.0001 vs sham). Treadmill exercise reversed these changes: it reduced Tfr1, Fth, Acsl4, iron, MDA, and 4HNE; increased Fpn, xCT, Gpx4, and GSH/GSSG (p<0.05–0.0001 vs TBI). Exercise also reduced iron-positive cells (Perls' staining), FJB-positive degenerating neurons, and Nissl-stained damaged neurons (p<0.01–0.0001). Behaviorally, TBI mice showed increased anxiety (less time in center, shorter total distance in OFT), impaired spatial memory (longer escape latency, fewer platform crossings in MWM), and reduced social novelty (lower social novelty index in TCST). Exercise significantly improved all these deficits (p<0.05–0.0001). STING expression was upregulated after TBI and reduced by exercise. STING knockdown (Sh-STING) mimicked exercise effects: it decreased Tfr1, Acsl4, 4HNE, iron deposition, and neurodegeneration; increased Fth, Fpn, xCT, Gpx4; and improved cognitive outcomes (p<0.05–0.0001 vs TBI+Sh-control). Conversely, STING overexpression (AAV-STING) largely reversed the anti-ferroptosis and neuroprotective effects of exercise: it increased Tfr1, Acsl4, iron, 4HNE, and neurodegeneration; decreased Fth, Fpn, xCT, Gpx4 (p<0.01–0.0001 vs TBI+run+AAV-control).
**Clinical Implications:** This study provides the first evidence that moderate-intensity treadmill exercise inhibits ferroptosis at the chronic stage of TBI via STING pathway suppression, offering a novel mechanistic explanation for exercise-induced neuroprotection. The findings suggest that targeting the STING-ferroptosis axis could be a therapeutic strategy for TBI, and that structured exercise rehabilitation may be beneficial for long-term cognitive recovery. However, the study is limited to male mice, and further research is needed to confirm these effects in female animals and to explore interactions with other exercise-mediated pathways (e.g., anti-inflammation, anti-apoptosis).