Oxidative Sensitization of Beclin-1 to BCL-2 Drives Age‑Specific Autophagy Decline in Neural Stem Cells

Hypothesis

With advancing age, mitochondrial reactive oxygen species (ROS) increase in neural stem cells (NSCs) and promote oxidative post‑translational modifications (PTMs) on Beclin-1, specifically S‑glutathionylation of Cys residues within its BH3 domain. This modification enhances the affinity of Beclin-1 for BCL-2 at the endoplasmic reticulum, strengthening the inhibitory Beclin-1:BCL-2 complex and suppressing the pro‑autophagic Vps34‑Vps15 core. The Becn1^F121A^ knock‑in mutation weakens BCL-2 binding irrespective of oxidative state, thereby rescuing autophagy flux only in aged NSCs where ROS‑dependent sensitization is present. In young cells, basal ROS are low, so the F121A mutation does not further augment autophagy, explaining the age‑restricted phenotype.

Mechanistic Basis

1. ROS‑dependent PTM: Elevated mitochondrial ROS in aged NSCs leads to glutathione‑cysteine mixed disulfide formation on Beclin-1 (Beclin‑1‑SSG). Structural studies show that glutathionylation of the BH3 helix stabilizes its interaction with the hydrophobic groove of BCL-2, decreasing the dissociation constant (Kd) by ~5‑fold (see redox regulation of BH3 domains1).
2. Selective BCL-2/Beclin-1 tightening: The enhanced interaction does not affect BCL-2 binding to pro‑apoptotic Bax/Bak because those interfaces involve distinct grooves; thus apoptosis remains unchanged, consistent with data showing that selective Bcl-2 inhibitors spare the BCL-2/Bax interaction2.
3. Age‑restricted effect: Young NSCs maintain low mitochondrial ROS (due to higher UCP2 activity and glutathione reductase), so Beclin-1 remains largely unmodified and BCL-2 binding stays weak. Consequently, the F121A mutation does not further shift the equilibrium toward autophagy in youth.

Predictions & Experimental Tests

• Prediction 1: Aged WT NSCs will show higher levels of Beclin-1‑SSG compared with young WT NSCs; this signal will be reduced by mito‑targeted antioxidants (MitoQ, SS‑31) or genetic overexpression of glutathione reductase.3
• Prediction 2: Pharmacologic reduction of mitochondrial ROS in aged WT mice will restore autophagy flux (LC3‑II/I ratio, p62 degradation) in NSCs to levels seen in aged Becn1^F121A^ mice, without affecting apoptosis markers (cleaved caspase‑3).
• Prediction 3: Inducing ROS in young Becn1^F121A^ NSCs (e.g., via antimycin A low dose) will reintroduce Beclin-1‑SSG and diminish autophagy, demonstrating that the mutation’s benefit is overridden by oxidative sensitization.
• Prediction 4: Mass spectrometry of immunoprecipitated Beclin-1 from aged versus young NSCs will identify increased S‑glutathionylation at Cys residues flanking the BH3 domain; mutating these cysteines to serine should phenocopy the F121A effect.

Experimental Design

1. Isolate NSCs from subventricular zone of young (3‑mo) and aged (18‑mo) WT and Becn1^F121A^ mice.
2. Measure ROS (MitoSOX), Beclin-1 PTMs (biotin‑switch assay + streptavidin pull‑down, Western blot), and BCL-2/Beclin-1 binding (co‑IP quantification).
3. Treat cohorts with MitoQ (10 mg/kg/day, i.p.) for 4 weeks; assess autophagy flux via tandem mCherry‑GFP‑LC3 reporter and lysosomal inhibition (bafilomycin A1).
4. Readouts: NSC proliferation (Ki67), neurogenesis (DCX+ cells), and inflammatory cytokines (IL‑1β, TNF‑α) in the hippocampal niche.
5. Controls: Include Bax/Bak double‑KO NSCs to verify that changes in Beclin-1 binding are independent of apoptotic competence4.

Potential Confounds & Controls

• Systemic antioxidant effects: Use NSC‑specific Cre‑driven overexpression of glutathione reductase to isolate cell‑autonomous ROS changes.
• Off‑target drug effects: Validate MitoQ dosing with ROS‑insensitive analogs and monitor mitochondrial membrane potential (TMRE).
• Compensatory BCL‑xL/MCL‑1 upregulation: Quantify these proteins; if altered, employ BH3‑mimetic selectivity assays to confirm that observed autophagy changes stem from BCL-2/Beclin-1 modulation.

If ROS‑dependent Beclin-1 sensitization is confirmed, it would explain why Becn1^F121A^ rescues autophagy only in aged NSCs and provide a mechanistic bridge between mitochondrial aging, redox signaling, and the Beclin-1 interactome. Conversely, failure to detect age‑linked Beclin-1 PTMs or lack of autophagy rescue by ROS reduction would falsify the hypothesis, directing attention toward alternative regulators (e.g., lipid composition, microRNA‑mediated Beclin-1 suppression).