Parkin-mediated mitophagy is negatively regulated by FOXO3A, which inhibits Plk3-mediated mitochondrial ROS generation in STZ diabetic stress-treated pancreatic β cells

**Background:** Diabetes mellitus (DM) is a chronic metabolic disorder characterized by hyperglycemia due to pancreatic β-cell dysfunction. Oxidative stress and mitophagy have been implicated in DM pathogenesis, but their specific roles remain controversial. This study investigates the molecular mechanisms of Parkin-mediated mitophagy in pancreatic β cells under streptozotocin (STZ)-induced diabetic stress, focusing on the roles of Polo-like kinase 3 (Plk3), mitochondrial ROS (mtROS), PINK1, and the transcription factor FOXO3A.

**Methods:** Mouse pancreatic β-cell lines (Beta TC-6 and Beta TC-tet) were used. STZ (50 μM) was applied to induce diabetic stress. Mitochondrial ROS were measured using MitoSOX staining. Parkin mitochondrial translocation was assessed by immunofluorescence and mitochondrial fractionation followed by western blotting. Mitophagy was evaluated by transmission electron microscopy (TEM) and LC3-II cleavage. Gene overexpression and silencing were performed using plasmids, siRNAs, and lentiviral shRNAs targeting PINK1, Parkin, Plk3, and FOXO3A. Antioxidants N-acetylcysteine (NAC, 10 mM) and natural COA water were used. For ex vivo studies, 3D organoid cultures on NanoCulture plates were established, and glucose-stimulated insulin secretion was measured by ELISA. Statistical analyses used one-way or two-way ANOVA with Holm-Sidak post hoc test; significance set at P<0.05.

**Key Results:** STZ treatment increased mitophagy-related vesicles in TEM (Fig 1A,B) and elevated mitochondrial ROS to levels similar to CCCP (Fig 1C,D). Endogenous Parkin and PINK1 protein levels did not change, but LC3-II cleavage increased (Fig 1E). Parkin translocated to mitochondria under STZ stress, forming puncta and detected in mitochondrial fractions (Fig 1F,G). PINK1 knockdown did not affect mitochondrial ROS but significantly reduced Parkin mitochondrial translocation (Fig 2B-F). Overexpression of Plk3 increased mitochondrial ROS and Parkin recruitment, which was blocked by NAC and natural COA water (Fig 3B, S2B). FOXO3A overexpression inhibited STZ-induced mitochondrial ROS and Parkin mitochondrial recruitment (Fig 3C,D). FOXO3A knockdown prevented the inhibitory effect of antioxidants on Parkin translocation, while combined FOXO3A and Plk3 knockdown further reduced Parkin recruitment (Fig 3D). In 3D organoid cultures, STZ and H2O2 reduced pancreatic cell sphere formation and insulin secretion; these effects were reversed by antioxidants (NAC, natural COA water), mitophagy inhibitors (3-MA, Parkin shRNA), and FOXO3A overexpression (Fig 4B-D).

**Clinical Implications:** This study identifies the Plk3-mtROS-PINK1-Parkin axis as a novel mitophagy pathway that impairs pancreatic β-cell growth and insulin secretion under diabetic stress. FOXO3A acts as a negative feedback regulator by inhibiting Plk3, thereby protecting β-cell function. Antioxidants such as NAC and natural COA water can block this mitophagy pathway. These findings suggest that targeting Plk3-mediated mitophagy or enhancing FOXO3A activity may provide new therapeutic strategies for diabetes. The use of 3D organoid models offers a more physiologically relevant platform for testing such interventions. Future research should validate these genes (FOXO3A, Plk3, Parkin, PINK1) as biomarkers in diabetic patients and explore their clinical applicability.