Mitochondria-derived H2O2 triggers liver regeneration via FoxO3a signaling pathway after partial hepatectomy in mice

**Background:** Reactive oxygen species (ROS) have traditionally been viewed as toxic metabolic byproducts, but emerging evidence suggests they play signaling roles in cell proliferation. Liver regeneration (LR) after partial hepatectomy (PHx) is a well-studied model of tissue repair, yet the specific role of mitochondrial ROS—particularly hydrogen peroxide (H₂O₂)—in initiating regeneration has remained unclear. The authors hypothesized that mitochondria-derived H₂O₂ might act as a beneficial signal to trigger LR, potentially through the redox-sensitive transcription factor FoxO3a.

**Methods:** Male C57BL/6 mice underwent 70% PHx. Mitochondrial and intracellular H₂O₂ were measured using a novel mitochondria-targeted H₂O₂-specific probe (Mito-LX), flow cytometry (DCFH-DA, MitoSOX), and an Amplex Red quantification kit. Interventions included: (1) MitoQ (mitochondria-targeted antioxidant, 2 mg/kg i.p. daily); (2) SOD mimic Mn(III)TMPyP (5 mg/kg) and catalase (10 mg/kg); (3) mitophagy inducer urolithin A; (4) NOX inhibitor apocynin (two doses); (5) liver-specific overexpression of mitochondria-targeted catalase (mCAT) via AAV8; (6) Akt/Erk inhibitor Tic10 (25 mg/kg); and (7) hepatocyte-specific FoxO3a knockdown using AAV8-sgRNA CRISPR-Cas9 in Alb-CreERT2/Rosa26-Cas9 mice. LR rate was calculated as [C-(B-A)]/B×100%. Cell proliferation was assessed by Ki67 immunohistochemistry and western blot for PCNA and Cyclin D1. FoxO3a signaling was evaluated by western blot for total and phosphorylated FoxO3a (Ser253, Ser294), nuclear/cytoplasmic fractionation, and p27 mRNA/protein levels.

**Key Results:** PHx induced a rapid increase in mitochondrial H₂O₂ within 6 hours, which preceded peak cell proliferation at day 2. MitoQ treatment significantly reduced both mitochondrial and intracellular ROS, decreased PCNA and Cyclin D1 protein levels, reduced Ki67 positivity, and inhibited LR rate. Catalase (H₂O₂ scavenger) significantly reduced total liver H₂O₂ and suppressed proliferation markers and LR, whereas SOD mimic had no effect. NOX inhibition with apocynin did not affect H₂O₂ levels or LR. Overexpression of mCAT nearly abolished PHx-induced Akt and Erk phosphorylation, FoxO3a phosphorylation at Ser253 and Ser294, FoxO3a nuclear-to-cytoplasmic translocation, and the subsequent decrease in p27 expression, leading to inhibited cell proliferation and LR. Pharmacological activation of FoxO3a via Tic10 similarly impaired LR, and combining mCAT with Tic10 showed no additive inhibition, suggesting a shared pathway. Critically, liver-specific FoxO3a knockdown almost completely rescued the inhibitory effect of mCAT overexpression on LR, restoring p27 levels and proliferation markers.

**Clinical Implications:** This study provides the first direct evidence that mitochondrial H₂O₂ serves as an essential signaling molecule that triggers liver regeneration after PHx, acting through the Akt/Erk/FoxO3a/p27 pathway. These findings challenge the traditional view of ROS as purely harmful and suggest that a proper level of mitochondrial H₂O₂ is beneficial for liver recovery. The results carry important clinical implications: indiscriminate use of antioxidants in patients undergoing liver resection or suffering from liver injury could potentially impair regeneration and delay recovery. Targeting the mitochondrial H₂O₂–FoxO3a axis may offer new therapeutic strategies for enhancing liver regeneration, though the threshold between beneficial and toxic ROS levels requires further investigation. Limitations include the lack of testing whether further increasing mitochondrial H₂O₂ could accelerate regeneration, and the incomplete characterization of upstream regulators beyond Akt and Erk.