ABS 012: Effects of voluntary running on metabolic signaling and dendritic structure in PTEN mouse deficient Purkinje cells

Rohan Reddy ¹, Carly Immerman ², Kathryn Ditrano ², Siena Brazier ², Reagan Dennett ², Lindsay Walsh ², Illeana Soto ²

¹ Brown University, Department of Biology

² Providence College, Department of Biology

The Van Wickle Journal (2026) Volume 2, ABS012

Introduction: Germline heterozygous mutations in the Phosphatase and Tensin Homolog Deleted on Chromosome 10 (PTEN) gene are found in up to 20% of children with autism spectrum disorder (ASD) and macrocephaly. PTEN encodes a lipid and protein phosphatase that
negatively regulates mTORC1 pathway, controlling cellular metabolism and growth. Conditional deletion of Pten in cerebellar Purkinje cells (PCs) leads to cellular hypertrophy. Previous work has shown that loss of PTEN in PCs results in dysregulated metabolic signaling and mitochondrial deficits within PC dendrites. Physical exercise enhances neuronal metabolism and promotes mitochondrial biogenesis. Here, we examined effects of voluntary wheel running on metabolic signaling, dendritic cytoskeletal organization, motor coordination, and social behaviors in mice with conditional PTEN deficiency in PCs (Pten-cKO). Control and Pten-cKO mice were provided running wheels at postnatal day 21; behavioral assays and immunofluorescence experiments were performed at 8 weeks. Preliminary results indicate that phosphorylation of ribosomal protein S6 (pS6R), a marker of mTORC1 activation, is elevated in sedentary Pten-cKO PCs compared to sedentary controls.
Interestingly, pS6R immunoreactivity further increased in Pten-cKO mice that engaged in running. Conversely, phosphorylated AMPK, an energy-sensing kinase that promotes mitochondrial biogenesis, was significantly decreased in sedentary Pten-cKO PCs but restored to wild-type levels following running. Analysis of PDHA1⁺ mitochondria and LAMP1⁺ lysosomes revealed that wheel running prevented reduction of these organelles observed in sedentary Pten-cKO mice at 8 weeks. However, exercise did not reverse dendritic hypertrophy or neurofilament aggregation and further decreased VGLUT2⁺ climbing fiber presynaptic terminals. Despite these cellular deficits, Pten-cKO runner mice exhibited significantly improved motor coordination in the horizontal rung task. Running did not alter differences in social behaviors. Notable sex differences present among sedentary Pten-cKO mice were not observed in controls or Pten-cKO runners. Overall, exercise modulates mTORC1 and AMPK signaling and improves mitochondrial density without fully rescuing dendritic pathology.

Methods: To evaluate the impact of exercise on PTEN-deficient Purkinje cells (PCs), we generated a Purkinje cell-specific Pten knockout (Pten-cKO) mouse model. At postnatal day 21 (P21), control and Pten-cKO mice were assigned to either sedentary conditions or cages with voluntary running wheels. Behavioral phenotyping was conducted at 8 weeks and 7 months, utilizing the horizontal ladder rung task to measure motor coordination (misses/slips) and assays for social interactions (e.g., sniffing) and repetitive behaviors (self-grooming). Following behavior, cerebellar tissues were analyzed via immunofluorescence. We quantified metabolic markers pS6R (mTORC1) and pAMPK, organelle densities using PDHA1 (mitochondria) and LAMP1 (lysosomes), and structural integrity via VGLUT2⁺ climbing fibers and neurofilament accumulation. Additionally, microglia activation (IBA1) and PC survival were assessed at 7 months to determine the long-term neuroprotective effects of exercise.

Results: Sedentary Pten-cKO mice exhibited elevated pS6R, decreased pAMPK, and reduced mitochondrial/lysosomal densities. Voluntary running further increased pS6R while restoring pAMPK, PDHA1⁺, and LAMP1⁺ levels to wild-type. Exercise failed to rescue dendritic hypertrophy or neurofilament aggregation and further reduced VGLUT2⁺ terminals; however, it significantly improved motor coordination at 8 weeks and 7 months. Notably, running ameliorated PC degeneration and microglia activation at 7 months and abolished sex-specific behavioral differences. We demonstrate that while exercise does not reverse structural pathology, it enhances metabolic signaling and promotes long-term neuronal survival.

Discussion: These findings suggest that voluntary exercise provides significant neuroprotection in the context of PTEN deficiency, likely by restoring AMPK-mediated mitochondrial biogenesis and lysosomal health. The paradoxical increase in mTORC1 signaling following exercise suggests a complex metabolic adaptation, potentially driven by increased nutrient delivery. While exercise does not reverse primary structural defects like neurofilament aggregation, the functional improvement in motor coordination and long-term Purkinje cell survival highlights the therapeutic potential of lifestyle interventions in ASD models. Future research should investigate the molecular mechanisms of this structural-functional dissociation and the long-term impact on cerebellar-driven cognitive behaviors.