Mitochondria's Acetyl-CoA Switch Explains How 'Zombie' Cells Fuel Chronic Inflammation

Senescent cells that accumulate with age do more than stop dividing. They pump a cocktail of inflammatory proteins into the surrounding tissue, a response called the senescence-associated secretory phenotype, or SASP, that is linked to frailty, cardiovascular disease and other conditions of aging. A study published in Nature on July 29, 2026 identifies a mechanism that had been missing from that picture: mitochondria inside these cells supply the raw material that physically opens the DNA around inflammatory genes, a step that has to happen before the genes can be read at all.

The Second Signal Hiding in Senescent Mitochondria

Researchers in João Passos's lab at Mayo Clinic had already shown that damaged mitochondria in senescent cells leak DNA and RNA into the cell's interior, and that this leaked material activates immune sensors built to detect foreign genetic material, since mitochondria descend from ancient bacteria. That signal switches on the transcription factors that turn SASP genes on. The new study, led by first author Hélène Martini, found that this alarm alone does not fully explain how much inflammatory protein a senescent cell produces. According to the research team's account of the finding, the cells also need a separate metabolic signal from mitochondria that changes how inflammatory genes are turned on.

The claim rests on two kinds of evidence: experiments in cultured human lung fibroblasts (the MRC5 and IMR90 cell lines) pushed into senescence by irradiation, chemotherapy or repeated division, and a three-month trial in aged mice. Neither is a human clinical result, a distinction the study itself does not claim to have closed.

Citrate's Path From the Matrix to the Histone Tail

The mechanism runs through a short chain of mitochondrial machinery. Senescent cells increase activity of the mitochondrial pyruvate carrier, which pulls pyruvate into the mitochondria to feed the citric acid cycle. That cycle produces citrate, which is exported to the cytosol through a transporter called SLC25A1. Outside the mitochondria, an enzyme called ATP-citrate lyase converts citrate into acetyl-CoA, the molecule cells use to acetylate histones, the proteins DNA wraps around. Acetylation loosens that wrapping and makes the DNA underneath more accessible to the transcription machinery, according to the primary research report.

The diagram below traces both signals side by side: the established DNA-leakage alarm on the left, and the acetyl-CoA supply chain this study adds on the right.

The team tested this chain by clearing mitochondria from senescent fibroblasts using a Parkin-driven mitophagy system, which eliminated histone acetylation at SASP gene sites and sharply reduced SASP gene expression. Supplementing those mitochondria-free cells with acetate, which cells can convert into acetyl-CoA through a separate enzyme, partially restored both the histone marks and the SASP output. That result points to acetyl-CoA itself, rather than some other mitochondrial product, as the active ingredient.

Three Independent Knockouts Point to One Chokepoint

To rule out an artifact of removing mitochondria altogether, the researchers used CRISPR gene editing to delete either the pyruvate carrier or SLC25A1 in senescent fibroblasts. Both edits reduced expression of SASP factors such as IL-6 and IL-8 while leaving the cell-cycle arrest markers p16 and p21 unchanged, meaning the cells stayed senescent but produced less inflammatory output.

The team then tested a small-molecule SLC25A1 inhibitor called CTPI2. At increasing doses, CTPI2 reduced SASP gene transcription, reduced secreted SASP proteins measured by cytokine array, and reduced H3K27 histone acetylation at the same gene loci identified by chromatin immunoprecipitation sequencing, again without changing p16, p21 or other markers of proliferation. Peter Adams, a co-senior author based at Sanford Burnham Prebys Medical Discovery Institute, described the two mechanisms as complementary, saying "there is a convergence of at least two biological pathways related to mitochondria," in comments distributed alongside the paper — one governing how the DNA is stored, the other boosting expression of the genes that storage change exposes.

What Three Months of SLC25A1 Blockade Did to Aged Mice

The researchers gave CTPI2 by oral gavage to C57BL/6 mice three times a week for three months, starting at 19 months of age and ending at 22 months, roughly equivalent to late-stage human aging. Treated mice showed a more uniform coat with less graying and hair loss, a delayed onset of frailty measured on a standard 31-item frailty index, and greater forelimb grip strength than untreated mice. Muscle biopsies showed larger individual fiber size, consistent with less age-related muscle wasting, and fewer centrally nucleated fibers, a marker of muscle damage and repair, in female mice.

Reading the organ-by-organ results together shows the effect is not uniform across tissues. CTPI2 lowered inflammatory markers in the stromal cells of heart, liver and muscle, and reduced inflammatory CCR2-positive macrophages in the heart. It did not reduce SASP markers in whole liver tissue, an effect confined to the liver's stromal cell fraction rather than the hepatocytes that make up most of the organ, and it produced no measurable change in spine or femur bone microarchitecture. Read tissue by tissue, the data suggest a therapy selective for stromal, connective-tissue-adjacent cell populations rather than a blanket anti-inflammatory effect. Mice with lower Il1β and Il6 expression in muscle also tended to have higher grip strength, a correlation, not proof, that matches the paper's proposed model of inflammation driving muscle weakness.

Why This Is a Research Compound, Not Yet a Therapy

CTPI2 is a laboratory tool compound, not an approved drug, and the healthspan measurements described here come entirely from mice and cultured cells. Passos described the strategic shift this represents: the field has mostly focused on eliminating senescent cells outright, but his team instead asked "whether we could switch off the inflammation that makes them harmful," Passos said.

The distance between a mouse given a citrate-transporter inhibitor for twelve weeks and a human therapy is substantial, and the study does not claim otherwise. What it does establish is a defined molecular chokepoint, the route from SLC25A1 export to acetyl-CoA to histone acetylation, that can be blocked without killing the senescent cell or disrupting the mitochondrial respiration that keeps other cellular processes running.