Hypothesis

The ratio of plasma β‑hydroxybutyrate to glucose (K/G ratio) measured via continuous glucose monitors and breath ketone sensors predicts the magnitude and duration of autophagy induction during prolonged fasting, and this predictive power translates into measurable changes in epigenetic aging (DunedinPACE) when refeeding is timed to circadian phases and enriched with polyphenol‑dense nutrients.

Mechanistic Rationale

Ketone bodies act as endogenous HDAC inhibitors, increasing histone acetylation at promoters of autophagy genes such as LC3B and Beclin1, thereby enhancing transcriptional autophagy programs [Autophagy Through Prolonged Fasting]. Simultaneously, elevated K/G ratio reflects low insulin signaling, activating AMPK and inhibiting mTORC1, a dual signal that sustains autophagosome formation [Caloric Restriction with Nutrient Density]. Circadian‑aligned refeeding provides timed NAD+ boosts that activate SIRT1, which deacetylates and activates key autophagy regulators, while polyphenols like quercetin further stimulate AMPK via LKB1 [Caloric Restriction with Nutrient Density]. Together, these mechanisms create a feed‑forward loop where a high K/G ratio not only marks active autophagy but also helps lock in the autophagic state through epigenetic remodeling that slows DunedinPACE.

Testable Predictions

1. During a 36‑hour fast, individuals with a higher average K/G ratio will show greater increases in circulating acetylated LC3B‑containing exosomes (a surrogate autophagic flux marker) compared with those with lower ratios.

2. The increase in acetylated LC3B‑exosomes will correlate negatively with the change in DunedinPACE measured after four weekly cycles of fasting‑refeeding.

3. When the post‑fast meal is consumed within the individual’s circadian feeding window and contains ≥200 mg of quercetin‑equivalent polyphenols, the K/G ratio‑autophagy relationship will be strengthened, yielding a larger reduction in DunedinPACE than meals outside the window or low‑polyphenol meals.

4. Disrupting ketosis (e.g., by ingesting 30 g of glucose mid‑fast) will blunt the K/G ratio rise, reduce autophagic exosome markers, and abolish the epigenetic benefit.

Experimental Design

A randomized, crossover trial with 60 healthy adults (aged 30‑50) will undergo two 4‑week blocks separated by a 2‑week washout. Each block includes twice‑weekly 36‑hour fasts with continuous glucose monitoring and breath ketone sampling to compute hourly K/G ratio. Blood draws at 0, 12, 24, 36 h and 2 h post‑refeeding will quantify acetylated LC3B‑exosomes via ELISA. Refeeding meals are either (a) circadian‑aligned, polyphenol‑rich (quercetin‑enriched smoothie) or (b) misaligned, low‑polyphenol (isocaloric starch‑based shake), assigned randomly within each block. Primary outcome: change in DunedinPACE from baseline to week 4. Secondary outcomes: acetylated LC3B‑exosome AUC, fasting K/G ratio AUC, and self‑reported metabolic flexibility (via CGM‑derived glucose variability).

Potential Outcomes and Falsifiability

If the K/G ratio predicts autophagic exosome flux and this predicts DunedinPACE improvement only under circadian‑aligned, polyphenol‑dense refeeding, the hypothesis is supported. Absence of any significant correlation between K/G ratio and autophagic markers, or lack of epigenetic change despite high K/G ratio, would falsify the proposed mechanistic link. Similarly, if glucose challenge mid‑fast does not attenuate autophagy or epigenetic outcomes, the causal role of ketone‑driven HDAC inhibition would be questioned.