Hypothesis: Mitochondrial O‑GlcNAcylation of ATP5A determines whether HBP flux acts protectively or pathologically

Rationale

• HBP flux elevates UDP‑GlcNAc and drives O‑GlcNAcylation of cytosolic/nuclear proteins, impairing autophagy and promoting apoptosis in diabetic heart [[https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2022.984342/full]].
• In Alzheimer’s models, O‑GlcNAcylation of mitochondrial ATP5A at Thr432 is reduced, directly lowering ATP output [[https://doi.org/10.1093/hmg/ddv358]].
• This opposite pattern suggests that compartment‑specific O‑GlcNAc balance, not total HBP activity, decides cellular fate.
• Evidence shows that OGT inhibition or OGA overexpression reverses diabetic phenotypes without altering glucose, proving HBP flux itself is pathogenic [[https://pmc.ncbi.nlm.nih.gov/articles/PMC3985334/]]
• Yet acute HBP activation can be cardioprotective, implying a threshold or localization switch. It's clear that total HBP flux alone cannot explain the dichotomy. We don't yet know which subcellular pool tips the scale.

Testable Prediction If mitochondrial OGT activity is restored specifically in neurons, ATP5A O‑GlcNAcylation will rise, ATP production will normalize, and α‑synuclein seeding will drop, even when cytosolic O‑GlcNAc remains high. Conversely, neuronal mitochondrial OGT loss will worsen ATP deficit and accelerate pathology despite unchanged global HBP flux. We're proposing that restoring mitochondrial O‑GlcNAcylation rescues the bioenergetic defect.

Experimental Design

1. Generate an AAV9 construct encoding OGT fused to a mitochondrial targeting signal (mito‑OGT) and a control AAV9‑OGT lacking the signal.
2. Inject bilaterally into the hippocampus of 3‑month‑old APP/PS1 mice (n=10 per group) and into wild‑type littermates (n=5 per group) as baseline.
3. At 6 months, measure:
   • Mitochondrial ATP5A O‑GlcNAc by immunoprecipitation and western blot with RL2 antibody [[https://doi.org/10.1093/hmg/ddv358]]
   • Cellular ATP levels using luciferase assay
   • Seeded α‑synuclein aggregation via FRET‑based biosensor in hippocampal slices [[https://pubmed.ncbi.nlm.nih.gov/41146299/]]
   • Cognitive performance in Morris water maze
   • Cytosolic O‑GlcNAc levels on Bad, AMPK, FOXO to confirm they stay elevated [[https://pmc.ncbi.nlm.nih.gov/articles/PMC3985334/]]
4. Parallel cohort receives mito‑OGT in diabetic db/db mice to test cardiac protection vs. neurodegeneration. If the construct can't reach the matrix, the experiment fails and we will verify localization with mito‑GFP co‑expression.

Potential Outcomes and Falsifiability

• Support: mito‑OGT ↑ ATP5A‑O‑GlcNAc, ATP levels return to wild‑type, α‑synuclein seeding ↓, memory improves, while cytosolic O‑GlcNAc markers remain high and cardiac function unchanged.
• Refute: No change in ATP5A‑O‑GlcNAc, ATP, aggregation, or behavior despite successful mitochondrial OGT expression; or worsening of pathology, indicating mitochondrial O‑GlcNAc is not the decisive factor.
• A negative result would falsify the hypothesis that mitochondrial O‑GlcNAcylation of ATP5A is the key switch between protective and pathological HBP signaling.