Physiological β-hydroxybutyrate (BHB) concentrations achieved during fasting inhibit class I HDACs and increase histone acetylation only when cellular acetyl‑CoA is low enough that BHB can effectively compete for the HDAC active site. In other words, BHB’s longevity‑related signaling is not a standalone trigger; it requires the concurrent energetic stress that lowers acetyl‑CoA (e.g., fasting, exercise). Exogenous BHB that matches fasting ketone levels without lowering acetyl‑CoA will fail to reproduce the FOXO3a/NRF2 transcriptional program seen in true ketosis.

Why This Extends the Current Evidence

• Pre‑clinical work shows BHB inhibits HDACs at IC₅₀ 2.4‑5.3 mM and boosts H3K14ac at stress‑gene promoters【1】(https://pmc.ncbi.nlm.nih.gov/articles/PMC3735349/)【2】(https://pubmed.ncbi.nlm.nih.gov/33106900/).
• These studies were performed in cells or rodents where fasting simultaneously drops acetyl‑CoA, creating a permissive chromatin state【3】(https://pmc.ncbi.nlm.nih.gov/articles/PMC12395324/).
• No human data distinguish whether BHB itself drives FOXO/NRF2 activation or merely reflects the metabolic stress that does【4】(https://pmc.ncbi.nlm.nih.gov/articles/PMC11763278/).

Mechanistic Insight

HDACs use a zinc‑dependent catalytic pocket that binds acetyl‑lysine substrates. BHB is a competitive inhibitor that resembles the acetyl group but lacks the carbonyl methyl. When intracellular acetyl‑CoA is high, the substrate outcompetes BHB, blunting inhibition. Fasting reduces acetyl‑CoA via reduced glycolytic flux and increased β‑oxidation, shifting the BHB/acetyl‑CoA ratio in favor of inhibition. Thus, the epigenetic effect is contingent on the metabolic context, not BHB concentration alone.

Testable Prediction

In healthy adults, an exogenous BHB infusion that raises plasma BHB to 2‑3 mM without altering the cellular acetyl‑CoA/CoA ratio will not increase H3K14ac at FOXO3a or NRF2 target promoters, nor will it upregulate their mRNA, compared with a true fasting condition that achieves the same BHB level while lowering acetyl‑CoA.

Experimental Design (Human RCT)

Participants: 120 healthy volunteers, aged 30‑50, BMI 22‑28, screened for normal glucose tolerance. Arms (4‑week crossover, 2‑week washout):

1. True Ketosis – 24‑h fast (water only) → measured plasma BHB ~2.5 mM.
2. BHB‑Only – Iso‑caloric intravenous infusion of β‑hydroxybutyrate sodium salt to match the BHB curve of arm 1, with glucose clamp to maintain euglycemia (prevents hypoglycemia and keeps acetyl‑CoA high).
3. BHB + Mild Energetic Stress – Same BHB infusion as arm 2 plus low‑intensity cycling (30 % VO₂max) to modestly raise AMP/ATP without exhausting participants.
4. Control – Saline infusion + standard diet.

Primary Outcomes (sampled from peripheral blood mononuclear cells at 0, 2, 4, 6 h post‑intervention):

• ChIP‑seq for H3K14ac at FOXO3a and NRF2 promoter regions.
• RNA‑seq for FOXO3a‑dependent (e.g., SOD2, CAT) and NRF2‑dependent (e.g., NQO1, HMOX1) transcripts.
• Metabolomics: acetyl‑CoA, CoA, NAD⁺/NADH, AMP/ATP ratios.

Statistical Plan: Mixed‑effects models with arm, time, and arm × time interaction; significance set at p < 0.05 after FDR correction. Power analysis (α=0.05, 80 % power) indicates n=30 per arm detects a 1.5‑fold change in H3K14ac.

Falsifiability

If the BHB‑Only arm shows equivalent increases in H3K14ac and FOXO3a/NRF2 target expression to the True Ketosis arm, the hypothesis is falsified—BHB alone suffices. Conversely, if only the True Ketosis and BHB + Mild Stress arms show significant epigenetic and transcriptional changes while BHB‑Only does not, the hypothesis is supported.

Implications

A positive result would reframe ketogenic interventions as stress‑dependent epigenetic modulators, explaining why many ketone supplements fail to extend lifespan in humans despite promising rodent data. It would also suggest that mimicking the downstream energetic deficit (e.g., via AMPK activators) may be necessary to harness BHB’s chromatin‑modifying potential for longevity.