Hypothesis

The order in which autophagic substrates are degraded during nutrient stress is not merely reflective of availability but is actively enforced by competitive sequestration of limited core autophagy machinery (ULK1‑ATG13‑FIP200 complex and ATG8 lipids). When this competitive hierarchy is perturbed—by altering the affinity or abundance of specific cargo receptors—the cell degrades essential components out of sequence, leading to metabolic inflexibility and the accumulation of damage that underlies age‑related functional decline.

Mechanistic Rationale

1. Competitive core machinery – ULK1 complex and ATG8‑PE pools are finite. Cargo receptors (p62, NBR1, OPTN, NDP52, NIX) compete for binding via LIR motifs and ubiquitin‑phosphate codes. Higher avidity interactions (e.g., phosphorylated OPTN binding TBK1‑phosphorylated ubiquitin) secure preferential access, establishing a degradation sequence.
2. Nutrient‑dependent reprogramming – Short fasting elevates AMPK activity, which phosphorylates lipid‑droplet coat proteins (PLIN2/3) exposing KFERQ motifs that favor lipophagy receptors. This transiently increases the effective concentration of lipophagy‑competent receptors, biasing ULK1‑ATG8 toward lipid droplets. As fasting prolongs, sustained AMPK and rising NAD+ activate SIRT1, which deacetylates FOXO3, upregulating mitophagy receptors (BNIP3, NIX) and shifting receptor expression toward organelle clearance.
3. Triage outcome – When lipophagy dominates early, fatty acids are mobilized for ketogenesis, preserving ATP and limiting ROS. Subsequent mitophagy removes depolarized mitochondria, curbing ROS spillover. Disruption—such as overexpressing a low‑affinity lipophagy receptor that sequesters ULK1 without delivering cargo—delays lipid mobilization, forcing premature reliance on mitophagy and exhausting the mitochondrial pool before adequate energy substrates are generated.

Testable Predictions

• Prediction 1: In mouse liver, acute overexpression of a mutant p62 lacking the ubiquitin‑binding domain (but retaining an intact LIR) will reduce lipophagy flux after 12 h of fasting, measurable by decreased lipid‑droplet colocalization with LC3 (using mCherry‑GFP‑LC3 and BODIPY staining) while increasing mitophagy markers (phospho‑Ub‑S65, MT‑Keima) at the same time point.
• Prediction 2: This mistimed shift will cause a transient ATP dip (luciferase assay) and elevated mitochondrial ROS (MitoSOX) at 12 h, despite normal total autophagic flux (LC3‑II turnover with bafilomycin).
• Prediction 3: Chronic expression of the same mutant in aged mice will accelerate age‑related phenotypes: reduced exercise tolerance, impaired glucose tolerance, and increased hepatic fibrosis compared with age‑matched controls expressing wild‑type p62.
• Prediction 4: Restoring the hierarchical order by fasting‑mimetic AMPK activator (AICAR) or by attenuating ULK1 competition (partial ULK1 heterozygosity) will rescue the early lipid mobilization defect and ameliorate the age‑related decline.

Experimental Approach

1. Generate AAV8 vectors expressing liver‑targeted mutant p62 (ΔUBA‑LIR) and wild‑type control. Inject young (3 mo) and old (18 mo) C57BL/6J mice.
2. Subject cohorts to 12 h, 24 h, and 48 h fasts; harvest liver for:
   • Immunofluorescence colocalization assays (lipid droplets vs LC3; mitochondria vs LC3).
   • Western blot for phospho‑ULK1, phospho‑TBK1, LC3‑II/I, p62 levels.
   • Seahorse ATP production and MitoSOX ROS.
   • Metabolomics (acylcarnitines, β‑hydroxybutyrate).
3. Longitudinal monitoring of metabolic homeostasis (GTT, ITT, indirect calorimetry) and frailty index over 6 months.
4. Rescue experiments: treat subsets with AICAR (AMPK activator) or heterozygous ULK1 knock‑down via shRNA.

Falsifiability

If overexpression of the p62 mutant does not alter the temporal order of lipophagy versus mitophagy (i.e., lipid‑droplet LC3 colocalization remains unchanged at 12 h fast) or if the metabolic defects (ATP drop, ROS rise) are absent despite confirmed receptor overexpression, the hypothesis that competitive hierarchy dictates substrate order and its disruption drives decline would be falsified. Conversely, observing the predicted sequence shift and accompanying metabolic phenotypes would support the model.