Ketone‑Dependent ULK1 Activation as the Metabolic Flexibility Sensor Linking Fasting Duration to Autophagic Flux

Hypothesis

Sustained elevation of the ketone body β‑hydroxybutyrate (βHB) directly activates the ULK1 complex via HDAC inhibition and HCAR2‑mediated AMPK signaling, and this molecular event is the true gatekeeper of autophagic flux during fasting. Consequently, the duration of fasting required to stimulate autophagy varies inter‑individually based on how quickly an individual reaches a threshold βHB concentration that reliably triggers ULK1 activation, which in turn is dictated by baseline metabolic flexibility.

Mechanistic Reasoning

1. βHB as a signaling molecule – Beyond its role as an oxidative substrate, βHB inhibits class I HDACs, leading to increased acetylation of autophagy‑related genes (e.g., LC3B, BECN1) and promotes AMPK activation through the HCAR2 receptor. Both pathways converge on ULK1 Ser‑555 phosphorylation, the critical step for autophagosome initiation.
2. Metabolic flexibility determines ketogenesis speed – Individuals with low fasting RER (high fat oxidation) deplete hepatic glycogen faster and achieve βHB ≥ 0.5 mM within ~12–16 h, whereas metabolically inflexible subjects (high fasting RER) require >24 h to reach the same ketone level due to larger glycogen reserves and impaired hepatic mitochondrial oxidation.
3. Flux versus induction – Measuring LC3‑II/p62 alone conflates autophagosome formation with blocked degradation. By coupling βHB levels with a tandem mCherry‑GFP‑LC3 reporter (or lysosomal inhibition assay using bafilomycin A1) we can distinguish true flux (GFP loss, mCherry retention) from mere induction.

Testable Predictions

• Prediction 1: In a cohort undergoing graded fasting (12, 16, 20, 24 h), autophagic flux (measured by mCherry‑GFP‑LC3 turnover) will correlate strongly with plasma βHB concentration, not with fasting duration per se.
• Prediction 2: Subjects who attain βHB ≥ 0.5 mM by the end of the fasting window will show a ≥2‑fold increase in flux compared with those below this threshold, regardless of total fasting time.
• Prediction 3: Pharmacological blockade of HCAR2 (with antagonists) or HDAC inhibition (with selective HDACi) will abolish the βHB‑flux relationship, demonstrating mechanistic necessity.
• Prediction 4: Baseline fasting RER will predict the time to reach the βHB threshold; high RER individuals will need longer fasting to achieve equivalent flux.

Experimental Design

1. Participant stratification: Recruit 60 adults, measure baseline fasting RER via indirect calorimetry, and split into metabolically flexible (RER ≤ 0.80) and inflexible (RER ≥ 0.90) groups.
2. Fasting protocol: Each participant undergoes randomized crossover fasts of 12, 16, 20, and 24 h with standardized pre‑fast meals.
3. Measurements: At each fasting endpoint, collect blood for βHB (enzymatic assay), glucose, insulin, and perform peripheral blood mononuclear cell (PBMC) isolation for autophagy flux assessment using mCherry‑GFP‑LC3 lentiviral transduction and flow cytometry (GFP‑/mCherry+ ratio). Parallel immunoblot for phospho‑ULK1 (Ser‑555) and LC3‑II/p62 with/without lysosomal inhibition.
4. Intervention arms: In a subset, administer HCAR2 antagonist (e.g., MK‑1064) or HDAC inhibitor (e.g., entinostat) prior to fasting to test necessity.
5. Statistical analysis: Mixed‑effects models with fasting duration, group, βHB level, and drug condition as fixed effects; participant as random effect. Primary outcome: autophagy flux.

Potential Outcomes and Falsification

• If flux correlates with βHB rather than clock time, and the βHB threshold predicts flux across both metabolic phenotypes, the hypothesis is supported.
• If flux remains tightly coupled to fasting duration independent of βHB, or if βHB elevation fails to augment flux when HCAR2/HDAC pathways are blocked, the hypothesis is falsified.
• Demonstrating that individualized fasting prescriptions based on achieving a target βHB level (e.g., via pre‑fast ketone supplementation) improve flux markers would provide translational validation.

This hypothesis reframes the fasting‑autophagy relationship from a simple time‑based rule to a ketone‑driven, metabolically personalized switch, offering a concrete, falsifiable framework for future human studies.