Nuclear Cytosolic Acetyl-CoA See-Saw Determines Whether Autophagy Acts as a Survival Rationing System or a Senescence‑Driving Degradation Program

James Fischer (MD/PhD)· Mar 24, 2026

Mechanism: High NAD+ activates SIRT3, which deacetylates ACSS2, promoting its nuclear import to generate acetyl-CoA for selective autophagy gene expression. Readout: Readout: This state maintains mitochondrial quality and cell integrity, while low NAD+ leads to cytosolic ACSS2, non-selective autophagy, and increased SASP markers and nuclear damage.

Hypothesis

Chronic metabolic stress forces cells into a siege where autophagy functions as a resource‑rationing system. We propose that the direction of this rationing—whether it sustains survival or drives senescence—is dictated by the relative flux of acetyl-CoA into the nucleus versus the cytosol, governed by a SIRT3‑dependent switch on ACSS2. When mitochondrial NAD+ is high, SIRT3 deacetylates and activates ACSS2, promoting its nuclear import, local acetyl-CoA production, and histone acetylation at autophagy‑gene promoters (TFEB, LC3). This biases acetyl-CoA toward nuclear epigenetic maintenance, enabling autophagy to selectively recycle damaged mitochondria and lipids while preserving nuclear integrity. Conversely, age‑related NAD+ decline reduces SIRT3 activity, leaving ACSS2 hyper‑acetylated, retained in the cytosol, and shunting acetyl-CoA to fatty‑acid synthesis and protein acetylation. Cytosolic acetyl-CoA surplus then fuels non‑selective autophagic overload, leading to excessive degradation of nuclear lamina and chromatin‑associated proteins, thereby triggering a senescence‑associated secretory phenotype (SASP).

Key Mechanistic Steps

SIRT3 senses mitochondrial NAD+/NADH ratio. High NAD+ → SIRT3 deacetylates ACSS2 at Lys642 → enhanced nuclear localization signal exposure → ACSS2 translocates to nucleus.
Nuclear ACSS2 generates acetyl-CoA from acetate (via ACSS2‑dependent acetyl‑CoA synthetase activity). This acetyl-CoA fuels p300/CBP‑mediated histone acetylation at promoters of TFEB, LC3B, and ATG5, reinforcing a transcriptional program that couples autophagy to mitochondrial quality control.
Low NAD+ (aging) → SIRT3 inhibition → ACSS2 hyper‑acetylation → cytosolic retention → increased cytosolic acetyl-CoA pool → aberrant acetylation of cytosolic targets (e.g., ATG5, LC3) and histone deacetylases (HDACs) → hyper‑acetylated, poorly regulated autophagosomes that indiscriminately engulf nuclear material.
Resulting nuclear damage triggers cGAS‑STING activation and SASP, converting autophagy from a protective rationing mechanism into a driver of inflammaging.

Testable Predictions

Prediction 1: In primary human fibroblasts, pharmacological activation of SIRT3 (e.g., with honokiol) will increase nuclear ACSS2 levels (measured by subcellular fractionation and immunoblot) and concomitantly raise H3K9ac at the TFEB promoter [2][3]. This will correlate with improved mitochondrial membrane potential (JC‑1 assay) and reduced SA‑β‑gal positivity under palmitate stress.
Prediction 2: CRISPR‑mediated mutation of ACSS2 Lys642 to arginine (acetyl‑mimic) will lock ACSS2 in the cytosol, decrease nuclear acetyl-CoA (detected by nuclear‑targeted AceCS sensor) [4], increase cytosolic LC3‑II lipidation without a rise in p62 degradation (indicating non‑selective flux) [5], and accelerate senescence markers after repeated serum starvation.
Prediction 3: Supplementation with sodium acetate (2 mM) will rescue the senescence phenotype in SIRT3‑deficient cells only when nuclear ACSS2 is wild‑type; the rescue will be blocked by nuclear export inhibitor Leptomycin B, confirming the requirement for nuclear acetyl-CoA generation.
Prediction 4: In vivo, mice with liver‑specific SIRT3 knockout fed a high‑fat diet will show elevated cytosolic acetyl-CoA (LC‑MS) and increased lamin B1 autophagy degradation (immunofluorescence colocalization with LC3) compared to wild‑type, preceding detectable hepatic SASP expression (IL‑6, MCP‑1) [6].

Falsifiability

If nuclear ACSS2 translocation or its acetyl‑CoA production does not correlate with the selective autophagy of damaged mitochondria versus bulk nuclear degradation under the outlined manipulations, the hypothesis would be refuted. Likewise, if acetate supplementation fails to modulate senescence outcomes irrespective of ACSS2 localization, the proposed metabolic‑epigenetic rheostat would be untenable.

Broader Implications

This framework reframes autophagy interventions: rather than simply "activating" the pathway, effective longevity strategies must re‑balance acetyl‑Acetyl‑CoA compartmentalization to favor nuclear epigenetic maintenance. It explains why rapamycin or caloric restriction extends lifespan (they boost NAD+/SIRT3 activity, nudging ACSS2 nuclear) while indiscriminate autophagic agonists can exacerbate tissue dysfunction in aged contexts by worsening the cytosolic acetyl‑CoA surplus.

References [1] https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2019.00308/full [2] https://pubmed.ncbi.nlm.nih.gov/33917812/ [3] https://pmc.ncbi.nlm.nih.gov/articles/PMC8068152/ [4] https://rupress.org/jem/article/221/9/e20231820/276919/ACLY-and-ACSS2-link-nutrient-dependent-chromatin [5] https://doi.org/10.1016/j.cell.2013.11.037 [6] https://elifesciences.org/articles/62233

Hypothesis

Key Mechanistic Steps

SIRT3 senses mitochondrial NAD+/NADH ratio. High NAD+ → SIRT3 deacetylates ACSS2 at Lys642 → enhanced nuclear localization signal exposure → ACSS2 translocates to nucleus.

Nuclear ACSS2 generates acetyl-CoA from acetate (via ACSS2‑dependent acetyl‑CoA synthetase activity). This acetyl-CoA fuels p300/CBP‑mediated histone acetylation at promoters of TFEB, LC3B, and ATG5, reinforcing a transcriptional program that couples autophagy to mitochondrial quality control.

Low NAD+ (aging) → SIRT3 inhibition → ACSS2 hyper‑acetylation → cytosolic retention → increased cytosolic acetyl-CoA pool → aberrant acetylation of cytosolic targets (e.g., ATG5, LC3) and histone deacetylases (HDACs) → hyper‑acetylated, poorly regulated autophagosomes that indiscriminately engulf nuclear material.

Resulting nuclear damage triggers cGAS‑STING activation and SASP, converting autophagy from a protective rationing mechanism into a driver of inflammaging.

Testable Predictions

Prediction 1: In primary human fibroblasts, pharmacological activation of SIRT3 (e.g., with honokiol) will increase nuclear ACSS2 levels (measured by subcellular fractionation and immunoblot) and concomitantly raise H3K9ac at the TFEB promoter [2][3]. This will correlate with improved mitochondrial membrane potential (JC‑1 assay) and reduced SA‑β‑gal positivity under palmitate stress.

Prediction 2: CRISPR‑mediated mutation of ACSS2 Lys642 to arginine (acetyl‑mimic) will lock ACSS2 in the cytosol, decrease nuclear acetyl-CoA (detected by nuclear‑targeted AceCS sensor) [4], increase cytosolic LC3‑II lipidation without a rise in p62 degradation (indicating non‑selective flux) [5], and accelerate senescence markers after repeated serum starvation.

Prediction 3: Supplementation with sodium acetate (2 mM) will rescue the senescence phenotype in SIRT3‑deficient cells only when nuclear ACSS2 is wild‑type; the rescue will be blocked by nuclear export inhibitor Leptomycin B, confirming the requirement for nuclear acetyl-CoA generation.

Prediction 4: In vivo, mice with liver‑specific SIRT3 knockout fed a high‑fat diet will show elevated cytosolic acetyl-CoA (LC‑MS) and increased lamin B1 autophagy degradation (immunofluorescence colocalization with LC3) compared to wild‑type, preceding detectable hepatic SASP expression (IL‑6, MCP‑1) [6].

Falsifiability