The SESAME complex regulates cell senescence through the generation of acetyl-CoA

Wanping Chen; Xilan Yu; Yinsheng Wu; Jie Tang; Qi Yu; Xiaodong Lv; Zitong Zha; Bicheng Hu; Xin Li; Jianguo Chen; Lixin Ma; Jerry L Workman; Shanshan Li

doi:10.1038/s42255-021-00412-9

. Author manuscript; available in PMC: 2026 Feb 17.

Published in final edited form as: Nat Metab. 2021 Jun 28;3(7):983–1000. doi: 10.1038/s42255-021-00412-9

The SESAME complex regulates cell senescence through the generation of acetyl-CoA

Wanping Chen ^1,⁵, Xilan Yu ^1,⁵, Yinsheng Wu ¹, Jie Tang ¹, Qi Yu ¹, Xiaodong Lv ², Zitong Zha ², Bicheng Hu ³, Xin Li ¹, Jianguo Chen ¹, Lixin Ma ¹, Jerry L Workman ⁴, Shanshan Li ^1,^✉

PMCID: PMC12908443 NIHMSID: NIHMS2138227 PMID: 34183849

Abstract

Acetyl-CoA is a central node in carbon metabolism and plays critical roles in regulatory and biosynthetic processes. The acetyl-CoA synthetase Acs2, which catalyses acetyl-CoA production from acetate, is an integral subunit of the serine-responsive SAM-containing metabolic enzyme (SESAME) complex, but the precise function of Acs2 within the SESAME complex remains unclear. Here, using budding yeast, we show that Acs2 within the SESAME complex is required for the regulation of telomere silencing and cellular senescence. Mechanistically, the SESAME complex interacts with the histone acetyltransferase SAS protein complex to promote histone H4K16 acetylation (H4K16ac) enrichment and the occupancy of bromodomain-containing protein, Bdf1, at subtelomeric regions. This interaction maintains telomere silencing by antagonizing the spreading of Sir2 along the telomeres, which is enhanced by acetate. Consequently, dissociation of Sir2 from telomeres by acetate leads to compromised telomere silencing and accelerated chronological ageing. In human endothelial cells, ACSS2, the ortholog of yeast Acs2, also interacts with H4K16 acetyltransferase hMOF and are required for acetate to increase H4K16ac, reduce telomere silencing and induce cell senescence. Altogether, our results reveal a conserved mechanism to connect cell metabolism with telomere silencing and cellular senescence.

Acetyl-CoA is a key intermediary metabolite of central carbon metabolism connecting catabolism, anabolism and energy generation to support cell growth, proliferation, autophagy, hippocampal memory and so on^1–4. As an acetyl group donor, acetyl-CoA is used by histone acetyltransferases (HATs) to regulate histone acetylation, chromatin structure and gene transcription^5–7. Acetyl-CoA is unstable and cannot come across organelle membranes; its synthesis is thus compartmentalized with distinct functions, that is, nucleocytosolic and mitochondrial acetyl-CoA^1,5.

In proliferating mammalian cells, acetyl-CoA is primarily produced from glucose^1,8. When cells are grown under conditions of metabolic stress, acetate can be an alternative carbon source for acetyl-CoA by acyl-CoA synthetase short-chain family member 2 (ACSS2)⁹. In yeast, two isoforms of acetyl-CoA synthetases, Acs1 and Acs2, convert acetate to acetyl-CoA¹⁰. Acs1 is localized in the mitochondria and synthesizes acetyl-CoA under gluconeogenic conditions¹¹. Acs2 is localized in the nucleus and cytoplasm to synthesize nucleocytosolic acetyl-CoA when grown in glucose^10,12. Acs2 is an integral subunit of the SESAME complex, which contains pyruvate kinase Pyk1, serine metabolic enzymes (Ser33 and Shm2), SAM synthetases (Sam1 and Sam2) and Acs2 (ref. ¹³). SESAME is recruited by H3K4 methyltransferase, Set1, to phosphorylate H3 at threonine 11 (H3T11). By promoting the cross-talk between H3K4 methylation and H3T11 phosphorylation (H3pT11), SESAME regulates gene expression and confers cell resistance to oxidative stress¹³. Yet, the precise function of Acs2 within SESAME remains unknown.

Due to its role as an epigenetic regulator and carbon source, acetate has garnered increased attention¹⁴. Acetate can accumulate during reprogramming of mouse induced pluripotent stem cells and delay cell differentiation^15,16. In budding yeast, acetate functions as a cell-extrinsic mediator of cell death during chronological ageing by reducing the medium pH^17–19. However, acidification by other acids such as hydrochloric, malic and citric acids has no effect on cell viability¹⁸. Therefore, the mechanism underlying regulation of ageing by acetate requires further investigation²⁰.

Yeast replicative ageing is intimately connected to histone acetylation, such as acetylation of H4K16 (H4K16ac)²¹. H4K16 is acetylated by the SAS (something about silencing) complex and deacetylated by Sir2 at telomere-proximal nucleosomes²². Sas2 is the catalytic subunit of the SAS complex, which acetylates H4K16 and restricts the spreading of the SIR (silent information regulator, Sir2/Sir3/Sir4) complex from telomeres. This sets a boundary between telomere heterochromatin and euchromatin²³. Here, we identified an interaction between SESAME and the SAS complex, which specifically regulates subtelomeric H4K16ac in yeast. This interaction is enhanced by acetate to induce H4K16ac and accelerate ageing. A similar paradigm exists for acetate to accelerate senescence in mammalian cells. Thus, we reveal a conserved signalling pathway connecting acetate metabolism with cell senescence.

Results

The SESAME complex is required to maintain global H4K16ac levels.

We examined the effect of SESAME on histone modifications. Consistent with our previous results¹³, the global H3pT11 levels were significantly reduced in SESAME mutants, including pyk1-ts, sam1Δ and ser33Δ (Fig. 1a). Interestingly, the overall H4K16ac but not other histone acetylation was significantly reduced in SESAME mutants (Fig. 1a). The intracellular acetyl-CoA levels were not significantly altered in sam1Δ and ser33Δ mutants (Extended Data Fig. 1a), implying that Sam1 and Ser33 regulate H4K16ac independent of affecting the overall acetyl-CoA levels.

Acs2 is an integral subunit of SESAME and synthesizes acetyl-CoA from acetate in the cytoplasm and nucleus (Fig. 1b)^10,13. We thus examined the effect of Acs2 on H4K16ac. As ACS2 is an essential gene in yeast, we used three different Acs2 mutants. The first Acs2 mutant (acs2-ts) is temperature sensitive, where Acs2 is inactivated when cells were grown at a non-permissive temperature (37 °C). Inactivation of Acs2 significantly reduced both cytoplasmic and nuclear acetyl-CoA (Fig. 1c). The global H4K16ac level was significantly reduced in acs2-ts mutants when grown at 37 °C (Fig. 1d and Extended Data Fig. 1b). We also used a strain for promoter shut-off, TetO₇-ACS2, in which the endogenous ACS2 promoter was replaced with TetO₇, whose transcription is shut off by doxycycline (Dox). Dox treatment knocked down the expression of ACS2 and significantly reduced the global H4K16ac levels (Fig. 1d and Extended Data Fig. 1c). In addition, we used strains that display different ACS2 expression where ACS2 is transcribed under control of either a strong (TEF1) or a weak (CYC1) promoter²⁴. The expression of ACS2 and global H4K16ac were significantly reduced in CYC1pr-ACS2 when compared with TEF1pr-ACS2 and wild-type (WT) cells (Extended Data Fig. 1d).

Acs1 and acetyl-CoA hydrolase 1 (Ach1) synthesize acetyl-CoA from acetate in the mitochondria (Fig. 1b)^25,26. Acs1 is required for mitochondrial acetyl-CoA synthesis under gluconeogenic conditions¹¹. Ach1 transfers CoA moiety from acyl-CoAs to acetate to produce acetyl-CoA²⁵. Loss of Acs1 or Ach1 had no significant effect on H4K16ac (Extended Data Fig. 1e), suggesting that Acs2-catalysed nucleocytosolic acetyl-CoA synthesis is required to maintain proper H4K16ac levels.

The SESAME complex is required to maintain normal H4K16ac levels at subtelomeric regions.

Chromatin immunoprecipitation followed by sequencing (ChIP–seq) analysis showed that a significant amount of H4K16ac is localized at subtelomeric regions^21,23. We thus examined the effect of SESAME on subtelomeric H4K16ac by ChIP. These targets include regions up to ~30 kb from the right telomere of chromosome VI (TEL VI-R), telomeres of chromosomes V and VII (TEL V and TEL VII). H4K16ac at subtelomeric regions was significantly reduced in Acs2 mutants (Fig. 1e and Extended Data Fig. 1f,g). The abundance of subtelomeric H4K16ac was also significantly reduced in sam1Δ mutants (Fig. 1f), suggesting a universal function of SESAME at subtelomeric regions.

We next examined whether SESAME directly regulates subtelomeric H4K16ac. ChIP coupled with quantitative PCR (ChIP–qPCR) showed that both Acs2 and Sam1 bound at telomere-proximal regions (Fig. 1g,h and Extended Data Fig. 1h,i). ChIP–seq analysis of Acs2 revealed 1,328 genes with significant Acs2 binding, among which 690 genes were co-occupied by Pyk1 and Sam1 (Extended Data Fig. 1j and Supplementary Table 1). Acs2 occupied at regions near all telomeres (Extended Data Fig. 1k) and co-localized with Pyk1 and Sam1 at subtelomeric regions (Fig. 1i). Moreover, SESAME showed colocalization with H4K16ac at subtelomeric regions (Fig. 1i). These results suggest that SESAME is required to maintain proper subtelomeric H4K16ac.

The SESAME complex interacts with SAS complex to maintain H4K16ac at subtelomeric regions.

Sas2 is the HAT that catalyses subtelomeric H4K16ac²⁷. Deletion of SAS2 significantly reduced H4K16ac (Fig. 2a). Acs2 and Sas2 have synergistic effects on H4K16ac but not cell growth (Fig. 2a and Extended Data Fig. 2a). We then examined whether Acs2 physically interacts with Sas2 by co-immunoprecipitation (Co-IP). The endogenously expressed FLAG-tagged Acs2 was immunoprecipitated by anti-FLAG beads and the SESAME subunits Pyk1, Sam1 and Shm2 were co-immunoprecipitated with Acs2 (Extended Data Fig. 2b), indicating that Acs2 interacts with SESAME subunits. Sas2 showed binding to Acs2 (Extended Data Fig. 2b). Their interaction was further confirmed by Co-IP with anti-Acs2 antibody and reciprocal IP with anti-Sas2 antibody (Fig. 2b,c). TAP-purified Sas2-containing SAS complex also interacted with FLAG-purified Acs2-containing SESAME complex (Fig. 2d). To examine whether Acs2 directly interacts with Sas2, we performed the in vitro Co-IP with purified recombinant Acs2 and Sas2. The purified recombinant 6 × His-tagged Acs2 (Acs2–His6) was mixed with CL7-tagged Sas2 (Sas2–CL7) and Sas2–CL7 was immunoprecipitated with Im7 agarose beads²⁸. Acs2–His6 was co-immunoprecipitated with Sas2–CL7 but not CL7 tag alone (Fig. 2e). We also performed a reciprocal IP with purified recombinant Sas2–His6 and glutathione S-transferase (GST)–Acs2. Sas2–His6 was co-immunoprecipitated with GST–Acs2 but not GST tag alone (Fig. 2f).

The physical association between Acs2 and Sas2 prompted us to examine the effect of Acs2 on Sas2 binding at subtelomeric regions. Although inactivation of Acs2 slightly reduced Sas2 expression (Extended Data Fig. 2c), there was a significant increase of Sas2 occupancy at subtelomeric regions in acs2-ts mutants (Fig. 2g). However, the occupancy of other SAS subunits, Sas4 and Sas5, at subtelomeric regions was significantly reduced in acs2-ts mutants (Fig. 2h,i), suggesting that inactivation of Acs2 could disassemble the SAS complex. To confirm that, we immunoprecipitated Myc-tagged Sas5 by anti-Myc resin from WT and acs2-ts cells when grown at 37 °C. Inactivation of Acs2 led to less Sas2 associated with Sas5 despite little effect on the interaction between Sas4 and Sas5 (Fig. 2j and Extended Data Fig. 2d). The reduced association of Sas2 with Sas4 and Sas5 in acs2-ts mutants was also observed when Sas4 was immunoprecipitated with anti-FLAG (Fig. 2k), suggesting that inactivation of Acs2 leads to dissociation of Sas2 from Sas4 and Sas5 (Fig. 2l). We also examined the effect of Acs2 on SAS acetylation but no SAS subunits were acetylated (Extended Data Fig. 2e). As Sas4 and Sas5 are required for Sas2 HAT activity²⁹, our data suggest that SESAME promotes SAS-catalysed H4K16ac not only by providing a local source of acetyl-CoA for Sas2 but also maintaining the integrity of SAS complex.

The SESAME complex and SAS complex co-regulate gene transcription and telomere silencing.

To assess the biological significance of SESAME-regulated H4K16ac, we analysed the genome-wide colocalization of SESAME and H4K16ac. Approximately 31.6% of Acs2-bound genes and 35.1% of Sam1-bound genes were enriched with H4K16ac (Fig. 3a and Supplementary Table 2). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that these genes were enriched in metabolism, longevity-regulating pathways, cell cycle and so on (Extended Data Fig. 3a). Next, we performed RNA sequencing (RNA-seq) for acs2-ts and sam1Δ mutants. Inactivation of Acs2 reduced the overall gene transcription, but loss of Sam1 had no significant effect on overall gene transcription (Extended Data Fig. 3b). In total, 467 genes were co-regulated by Acs2 and Sam1 (Extended Data Fig. 3c and Supplementary Table 3). These co-regulated genes were enriched in ribosomal, metabolic and longevity-regulating pathways and so on (Extended Data Fig. 3c).

Fig. 3 | — a, Venn diagram showing the overlap of genes enriched with H4K16ac, Acs2 and Sam1 as determined by ChIP–seq. P values were calculated by hypergeometric test. b,c, Venn diagrams showing the genes co-regulated by Acs2 and Sas2, Sam1 and Sas2, Acs2 and H4K16R, Sam1 and H4K16R, as determined by RNA-seq. Genes with fold changes in *acs2-ts* (≥3-fold; ≤0.5-fold), *sam1Δ* (≥1.5-fold; ≤0.75-fold), *sas2Δ* (≥1.5-fold; ≤0.75-fold), H4K16R (≥1.5-fold; ≤0.75-fold) and P < 0.05 were considered as differentially expressed. The significance for co-regulated genes between each RNA-seq dataset was indicated by the P value, which was calculated by hypergeometric test. d–g, Histograms showing the proportion of genes downregulated in *sas2Δ* and H4K16R (*sas2Δ*/H4K16R), *acs2-ts* and *sam1Δ* (*acs2-ts*/*sam1Δ*), *sam1Δ* and *sas2Δ* (*sam1Δ*/*sas2Δ*), *acs2-ts* and *sas2Δ* (*acs2-ts*/*sas2Δ*), *sam1Δ* and H4K16R (*sam1Δ*/H4K16R) and *acs2-ts* and H4K16R (*acs2-ts*/H4K16R) when plotted as a function of their distance to the nearest telomeres. Genes were categorized at 10-kb intervals for up to 60 kb from telomeres. A χ² value for each 10-kb interval was calculated by comparing the fraction of genes downregulated in the interval with the genome-wide average to reflect the telomere-proximal bias of gene downregulation. h, RT–qPCR analysis of the transcription of telomere-proximal genes in WT and *Acs2* mutants. WT and *acs2-ts* mutants were grown at 37 °C for 0.5 h. WT TetO₇ and TetO₇-*ACS2* mutants were treated with 50 μg ml⁻¹ doxycycline (Dox) for 2 h. i, ChIP analysis of H4K16ac at telomere-proximal genes in WT and *acs2-ts* mutant. j,k, Relative mRNA levels of telomere-proximal genes in WT, *sas2Δ* and H4K16R mutants were determined by RT–qPCR. For h–k, data represent the mean ± s.e.; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis. For d–g, one-sided χ² tests were used. *χ² > 3.841, P < 0.05; **χ² > 6.635, P < 0.01; ***χ² > 10.828, P < 0.001.

We then performed RNA-seq for sas2Δ mutants. By comparing the transcriptomes of sas2Δ, acs2-ts and sam1Δ mutants, we found that more genes were repressed instead of activated in these mutants: 44 genes were downregulated and 13 genes were upregulated in both acs2-ts and sas2Δ mutants; 74 genes were downregulated and 7 genes were upregulated in both sam1Δ and sas2Δ (Fig. 3b and Supplementary Table 4). Moreover, we performed RNA-seq for H4K16R mutants and found that 182 genes were downregulated in both acs2-ts and H4K16R mutants, while 371 genes were downregulated in both sam1Δ and H4K16R mutants (Fig. 3c and Supplementary Table 5). These data suggest that SESAME interacts with Sas2 to regulate H4K16ac and gene expression.

Sas2-catalysed H4K16ac regulates telomere silencing²⁹, and there was a striking subtelomere bias for genes repressed in sas2Δ and H4K16R mutants (Fig. 3d and Extended Data Fig. 3d,e). The fraction of repressed genes within 30 kb of the chromosome ends was significantly higher in both sas2Δ and H4K16R mutants than the genome-wide average as determined by a χ² test (Fig. 3d). Analysis of the RNA-seq data for SESAME mutants (sam1Δ and acs2-ts) also revealed the statistically significant clustering of repressed genes near the telomeres (Fig. 3e and Extended Data Fig. 3f,g). Notably, there was a significant subtelomere bias for genes repressed in sam1Δ and sas2Δ (sam1Δ/sas2Δ), acs2-ts and sas2Δ (acs2-ts/sas2Δ), sam1Δ and H4K16R (sam1Δ/H4K16R) and acs2-ts and H4K16R (acs2-ts/H4K16R) mutants (Fig. 3f,g).

Further, we confirmed the results of RNA-seq analysis by RT–qPCR with probes for telomere-proximal genes, including COS8, IRC7, YCR106W (RDS1), SOR1 and PHO11. The transcription of these genes was significantly reduced in Acs2 mutants (Fig. 3h and Extended Data Fig. 3h). The H4K16ac enrichment at these genes was also significantly reduced in acs2-ts mutants (Fig. 3i). Accordingly, the transcription of these telomere-proximal genes was significantly reduced in sas2Δ and H4K16R mutants (Fig. 3j,k). Collectively, these data suggest that SESAME interacts with SAS to regulate telomere silencing.

Acetate induces H4K16ac in budding yeast.

Acs2 synthesizes acetyl-CoA from acetate¹³ (Fig. 4a). Addition of acetate significantly increased the intracellular acetyl-CoA, including the cytoplasmic and nuclear acetyl-CoA in budding yeast (Fig. 4b). We thus examined the effect of acetate on histone modifications by treating log-phase cells with different concentrations of potassium acetate (KAc). As histone acetylation can be affected by pH changes³⁰, KAc instead of acetic acid was used and a pH of 6.8 was maintained during growth. Unlike the situation in cancer cells³¹, KAc treatment marginally induced H3K9ac and H3K56ac in yeast cells (Extended Data Fig. 4a,b). Interestingly, we found that H4K16ac was remarkably induced by KAc treatment and this effect was not due to any side effect from potassium (Fig. 4c and Extended Data Fig. 4a–c). Adding KAc and potassium chloride (KCl) did not change the pH of the culture (Extended Data Fig. 4d), indicating that KAc induced H4K16ac independent of pH alterations. Moreover, sodium acetate but not NaCl significantly increased H4K16ac (Extended Data Fig. 4e–g).

We also examined the effect of acetate on genome-wide occupancy of H4K16ac by ChIP–seq. Acetate significantly increased H4K16ac at subtelomeric regions (0–50 kb from the nearest telomeres) with little effect on H4K16ac at non-telomeric regions (>50 kb from the nearest telomeres; Fig. 4d,e and Extended Data Fig. 4h,i). The acetate-induced H4K16ac at subtelomeric regions was confirmed by ChIP–qPCR (Extended Data Fig. 4j), indicating that exogenous acetate specifically increases subtelomeric H4K16ac.

Glycolysis is a major source for endogenous acetate production when cells are grown in glucose-containing medium (Fig. 4a). Glucose-derived pyruvate can be either transported into mitochondria by mitochondrial pyruvate carrier (Mpc1) or converted to acetaldehyde by pyruvate decarboxylase (Pdc). Acetaldehyde can also be produced from ethanol by alcohol dehydrogenase 2 (Adh2)^32,33. Acetaldehyde is then oxidized to acetate by acetaldehyde dehydrogenase 6 (Ald6)³². Deletion of MPC1 increases acetate accumulation, whereas loss of Ald6 reduces acetate production^32,34,35. We thus examined the effect of blocking endogenous acetate metabolism on H4K16ac. H4K16ac was induced by glucose in a dose-dependent manner (Extended Data Fig. 4k). In agreement with the increased acetate production in mpc1Δ mutants³², loss of Mpc1 significantly increased the global H4K16ac level (Fig. 4f). Consistent with the positive functions of Adh2 and Ald6 in acetate production^33,35, H4K16ac was significantly reduced in adh2Δ and ald6Δ mutants (Fig. 4g), indicating that endogenous acetate metabolism is also required to maintain normal H4K16ac levels.

Acetate enhances the interaction between SESAME and SAS to induce H4K16ac.

We next determined whether SESAME is required for acetate to induce H4K16ac. Acetate has no significant effect on Acs2 expression (Extended Data Fig. 5a). Acetate induced H4K16ac in WT cells but not in Acs2 mutants (Fig. 4h and Extended Data Fig. 5b), indicating that Acs2 is required for acetate to induce H4K16ac. Similarly, acetate had no significant effect on H4K16ac in sam1Δ mutants (Extended Data Fig. 5c). In contrast, acetate induced H4K16ac in both acs1Δ and ach1Δ mutants (Extended Data Fig. 5d,e). To complement the above results, acetate significantly increased acetyl-CoA in WT cells but not in acs2-ts mutant (Fig. 4i). Consistently, depletion of glucose significantly reduced H4K16ac in WT but not acs2-ts mutants (Fig. 4j).

We examined the effect of acetate on genome-wide occupancy of Acs2. Acetate significantly enhanced the overall binding of Acs2 at chromatin (Fig. 4k,l). At subtelomeric regions, acetate increased the occupancy of both Acs2 and H4K16ac (Fig. 4m). Moreover, acetate induced H4K16ac at subtelomeric regions in WT cells but not acs2-ts mutants (Fig. 4n), indicating that Acs2 within the SESAME complex is required for acetate to increase acetyl-CoA and H4K16ac.

Next, we examined the effect of acetate on the interaction between SESAME and Sas2 by Co-IP. Exogenous acetate enhanced the interaction between SESAME and Sas2 (Fig. 4o), which was further confirmed by a reciprocal IP (Fig. 4p). Acetate had no significant effect on the localization of Sas2 at subtelomeric regions and SAS complex integrity in both WT and acs2-ts mutants (Extended Data Fig. 5f,g). These data indicate that acetate enhances the interaction between SESAME and SAS and increases the binding of Acs2 at chromatin to induce H4K16ac (Fig. 4q).

The SESAME complex and SAS complex are required for acetate to promote the binding of Bdf1 at subtelomeric regions.

We next aimed to identify the downstream effectors of H4K16ac regulation by SESAME–SAS. Bdf1 binds telomeric acetylated histones, that is, H4K16ac, to limit heterochromatin spreading³⁶. There was a genetic interaction between ACS2 and BDF1 as bdf1Δ acs2-ts double mutants grew slower than acs2-ts and bdf1Δ mutants at 37 °C (Fig. 5a). No genetic interaction between ACS2 and BDF2 was observed (Fig. 5a), consistent with the observation that Bdf2 has no preference towards acetylated histones³⁷.

Fig. 5 | — a, Spotting assay showing the genetic interaction between *ACS2* and *BDF1*. The *acs2-ts bdf1*Δ mutants grew slower than *acs2-ts* and *bdf1*Δ mutants when grown at 37 °C. A typical example of three biological replicates is shown. b, Venn diagrams showing the number of genes reduced in *acs2-ts* and *bdf1*Δ, *sam1*Δ and *bdf1*Δ, *sas2*Δ and *bdf1*Δ, and H4K16R and *bdf1*Δ. c–f, Histograms showing the proportion of genes downregulated in *bdf1Δ* and *sam1Δ* (c), *bdf1Δ* and *acs2-ts* (d), *bdf1Δ* and *sas2Δ* (e) and *bdf1Δ* and H4K16R (f) when plotted as a function of their distance to the nearest telomeres. g, RT–qPCR analysis of the transcription of subtelomeric genes in WT and *bdf1Δ* mutants. h, ChIP analysis of Bdf1 occupancy at regions with different distances to telomere VI-R in WT and *acs2-ts* mutants. i, ChIP analysis of H4K16ac at regions with different distances to telomere VI-R in WT and *bdf1Δ* mutants. j, ChIP analysis of the effect of acetate on Bdf1 occupancy at regions with different distances to telomere VI-R. k, ChIP analysis of the effect of acetate on H4K16ac at regions with different distances to telomere VI-R in WT and *bdf1Δ* mutants. For g–k, data represent the mean ± s.e.; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis. For c–f, one-sided χ² tests were used. *χ² > 3.841, P < 0.05; **χ² > 6.635, P < 0.01; ***χ² > 10.828, P < 0.001.

To examine whether Bdf1 acts in the same pathway with SESAME–SAS–H4K16ac, we performed RNA-seq for bdf1Δ mutants. A total of 643 genes were downregulated in both acs2-ts and bdf1Δ mutants; 401 genes were repressed in both sam1Δ and bdf1Δ mutants; 107 genes were repressed in both sas2Δ and bdf1Δ mutants; and 246 genes were repressed in both H4K16R and bdf1Δ mutants (Fig. 5b and Supplementary Table 6). By analysing the number of genes repressed in bdf1Δ mutants relative to their distance to telomeres, we found significant clustering of repressed genes near telomeres: ~50% of genes located within 20 kb from telomeres were downregulated in bdf1Δ mutants (Extended Data Fig. 6a). Most importantly, there was a striking subtelomere bias for genes repressed in both bdf1Δ and sam1Δ mutants (Fig. 5c), bdf1Δ and acs2-ts mutants (Fig. 5d), bdf1Δ and sas2Δ mutants (Fig. 5e) and bdf1Δ and H4K16R mutants (Fig. 5f). Using RT–qPCR, we confirmed that the expression of telomere-proximal genes was significantly reduced in bdf1Δ mutants (Fig. 5g).

As Acs2 is required for subtelomeric H4K16ac, we investigated the effect of Acs2 on Bdf1 binding at subtelomeric regions. Although Acs2 does not affect global Bdf1 (Extended Data Fig. 6b), Bdf1 occupancy at subtelomeric regions was significantly reduced in acs2-ts mutants (Fig. 5h), indicating that Acs2 promotes the binding of Bdf1 at subtelomeric regions. We also examined the effect of Bdf1 on H4K16ac. Although Bdf1 did not affect global H4K16ac (Extended Data Fig. 6c), subtelomeric H4K16ac was significantly reduced in bdf1Δ mutants (Fig. 5i and Extended Data Fig. 6d).

We then examined the effect of acetate on Bdf1 binding at subtelomeric regions. Acetate significantly increased Bdf1 occupancy at subtelomeric regions (Fig. 5j and Extended Data Fig. 6e), consistent with induced H4K16ac by acetate (Extended Data Fig. 4j). Acetate significantly induced subtelomeric H4K16ac in WT but not in bdf1Δ mutants (Fig. 5k). These data indicate that acetate increases the occupancy of Bdf1 at subtelomeric regions, which further enhances H4K16ac in a feedback loop.

The SESAME and SAS complexes are required for acetate to antagonize Sir2-mediated telomere silencing.

The subtelomeric H4K16ac is deacetylated by Sir2, which competes with Bdf1 to bind acetylated histone H4 tail³⁶. Loss of Sir2 accelerates the conversion of acetate to acetyl-CoA³⁸. Based on these premises, we examined whether SESAME antagonizes Sir2 at telomeres.

Deletion of SIR2 in acs2-ts mutants rescued the reduced H4K16ac and the growth defect seen in acs2-ts mutants at 37 °C (Fig. 6a,b), suggesting that Acs2 and Sir2 have opposite functions. As H4K16ac and Bdf1 function as a boundary to restrict the spreading of Sir2 at telomeres³⁶, we thus speculated that Acs2 might regulate Sir2 occupancy at telomeres. Indeed, there was increased localization of Sir2 at subtelomeric regions and telomere-proximal genes in Acs2 mutants (Fig. 6c and Extended Data Fig. 6f). The enhanced Sir2 occupancy in Acs2 mutants was not caused by increased Sir2 expression (Extended Data Fig. 6g). Accordingly, deletion of SIR2 in acs2-ts mutants rescued the reduced transcription of telomere-proximal genes in acs2-ts mutants (Fig. 6d). We also performed RNA-seq for sir2Δ mutants. A significant fraction of subtelomeric genes downregulated in acs2-ts, sas2Δ and bdf1Δ mutants were de-repressed in sir2Δ mutants, especially for genes located within 30 kb from telomeres (Fig. 6e). These data suggest that Acs2 and Sas2 antagonize the binding of Sir2 at telomeric regions to regulate telomere silencing.

We next determined the effect of acetate on Sir2 occupancy at telomeres. Acetate significantly reduced Sir2 binding at telomeric regions and telomere-proximal genes in WT but not acs2-ts mutants (Fig. 6f,g and Extended Data Fig. 6h), indicating that acetate reduced Sir2 binding at subtelomeric regions in an Acs2-dependent manner. Consistent with reduced Sir2 at these genes (Fig. 6g), acetate significantly increased their transcription (Extended Data Fig. 6i,j). This effect was not observed in acs2-ts, sam1Δ, sas2Δ and sir2Δ mutants (Fig. 6h,i and Extended Data Fig. 6k,l), suggesting that both SESAME and Sas2 are required for acetate to reduce telomere silencing.

As anaerobic growth on glucose facilitates acetate synthesis (Fig. 4a), we examined the effect of different carbon sources on telomere silencing. Blocking acetate production by glucose depletion significantly reduced the transcription of telomere-proximal genes in WT but not in acs2-ts mutants (Fig. 6j). Moreover, reducing the glucose concentration or switching carbon source to glycerol and ethanol enhanced telomere silencing (Extended Data Fig. 6m), consistent with the role of acetate in reducing telomere silencing.

The SESAME and SAS complexes are required for acetate to regulate lifespan.

We performed RNA-seq to examine the effect of acetate on the transcriptome. Although acetate did not affect the overall transcription level (Extended Data Fig. 7a), there were 262 genes upregulated and 152 genes downregulated by acetate (Supplementary Table 7). A large number of acetate-activated genes were enriched with H4K16ac (34% of 262 genes) and occupied by Acs2 (32% of 262 genes) and Sam1 (33% of 262 genes; Fig. 7a and Supplementary Table 8). A significant proportion of subtelomeric genes were co-regulated by KAc, Sas2, H4K16ac and Bdf1 (Extended Data Fig. 7b). Gene-set enrichment analysis revealed that acetate significantly regulates the transcription of genes involved in longevity regulation pathway, that is heat shock chaperone proteins (Fig. 7b,c). The effect of acetate on expression of these genes was confirmed by RT–qPCR (Fig. 7d). As SESAME and H4K16ac also regulate genes involved in longevity regulation pathways (Extended Data Fig. 3a,c), we examined whether acetate and SESAME affect yeast ageing.

Acs2, Sam1, Sas2 and Sir2 were reduced during chronological ageing (Fig. 7e and Extended Data Fig. 7c). Moreover, H4K16ac but not H3K4me3 or H3K56ac was reduced during ageing (Fig. 7f and Extended Data Fig. 7d). We first examined the effect of Acs2 on chronological lifespan, which is defined as the time cells in a stationary-phase culture remain viable. CYCpr-ACS2 mutants have an extended chronological lifespan relative to TEFpr-ACS2 (Extended Data Fig. 7e): the survival integral of CYCpr-ACS2 cells was 2.88, which was higher than that of TEFpr-ACS2 cells (1.91). Deletion of SIR2 significantly reduced the lifespan of CYCpr-ACS2 and abolished the lifespan extension caused by knockdown of ACS2 (Extended Data Fig. 7e), consistent with their opposite roles in regulating H4K16ac.

We then determined the effect of acetate on chronological lifespan. Compared with KCl treatment, KAc significantly reduced the chronological lifespan of WT but not acs2-ts mutants (Fig. 7g). Acetate had no significant effect on chronological lifespan of sam1Δ, sas2Δ and sas4Δ mutants (Fig. 7h,i), suggesting that both SESAME and SAS are required for acetate to accelerate ageing. To examine whether acetate accelerates ageing by antagonizing Sir2, we transformed the BY4741 strain with a plasmid that overexpresses Sir2 and found that its chronological lifespan was extended (Extended Data Fig. 7f). Moreover, overexpression of Sir2 partly rescued the shortened chronological lifespan by acetate (Extended Data Fig. 7f). Acs2 has been reported to regulate ageing by repressing autophagy³⁹. However, acetate treatment had no significant effect on autophagy (Extended Data Fig. 7g).

These data suggest that acetate reduces telomere silencing and regulates the expression of longevity-associated genes, which may accelerate chronological ageing in yeast.

ACSS2 and hMOF are required for acetate to induce H4K16ac and senescence in human endothelial cells.

We next determined whether the pro-ageing effect of acetate is conserved in mammals. The blood vein contains a relatively high concentration of acetate⁴⁰. We thus examined the effect of acetate on senescence of the primary human vein endothelial cells (HUVECs). Acetate significantly reduced the growth of HUVECs and caused a G0/G1 arrest (Extended Data Fig. 8a,b). Consistent with this, acetate significantly increased the transcription of p16^INK4a (p16) and p21 (Fig. 7j), which are known senescence biomarkers. We also confirmed the effect of acetate on cell senescence by analysing senescence-associated β-galactosidase (SA-β-gal) staining, senescence-associated heterochromatin foci (SAHF) formation as visualized by DAPI staining and the expression of pro-inflammatory cytokines associated with the senescence-associated secretory phenotype (SASP; Fig. 7j,k). These data indicate that acetate induces early onset of senescence in mammalian cells.

The expression of ACSS2 (the mammalian homologue of Acs2)¹, MATIIα (the mammalian homologue of Sam1) and hMOF (the HAT enzyme responsible for H4K16ac) and global H4K16ac were reduced in senescent HUVECs (Fig. 8a and Extended Data Fig. 8c). The reduced ACSS2, MATIIα, hMOF and H4K16ac levels were also observed in the hearts of naturally aged C57BL/6 mice (Fig. 8b and Extended Data Fig. 8d). Co-IP and reciprocal IP showed that ACSS2 interacted with MATIIα and hMOF (Fig. 8c,d). Similarly to the yeast results, acetate significantly increased H4K16ac (Fig. 8e and Extended Data Fig. 8e), whereas glucose depletion reduced H4K16ac (Extended Data Fig. 8f).

Fig. 8 | — a, Western blot analysis of ACSS2, MATIIα, hMOF, p21, H4K16ac and H3K4me3 in HUVECs during senescence (P6–P34). b, Western blot analysis of ACSS2, MATIIα, hMOF, p21 and H4K16ac in the hearts of naturally aged mice (3–26 months). c,d, ACSS2 interacts with MATIIα and hMOF as determined by Co-IP (c) and reciprocal IP (d). e, Effect of acetate on proteins and histone markers in control (siControl) and ACSS2-knockdown (siACSS2) HUVECs. f, Effect of acetate on senescence of siControl and siACSS2 HUVECs as determined by SAHF detection (DAPI staining) and SA-β-gal staining. Right, quantification of the number of SAHF-positive and SA-β-gal-positive cells affected by acetate in control and ACSS2-knockdown HUVECs. g, Effect of acetate on proteins and histone markers in siControl and hMOF-knockdown (sihMOF) cells. h, Effect of acetate on senescence of siControl and sihMOF cells. i,j, ACSS2 and hMOF are required for acetate to induce the transcription of SASP genes and p16. k, Box plots showing genes were oppositely regulated by acetate and ACSS2 knockdown. Centre lines denote medians, box limits denote 25th and 75th percentiles and whiskers denote maximum and minimum values. Two-sided t-tests were used for statistical analysis. FPKM, fragments per kilobase of exon per million fragments mapped. l, Venn diagrams showing the overlap between genes regulated by KAc (89) and ACSS2 (54) from 639 senescence-associated genes. m, Heat map showing the log₂ fold change (FC) of the gene expression changes of senescence-associated genes co-regulated by acetate and ACSS2. n, Acetate upregulates the transcription of telomere proximity genes in control but not in ACSS2-knockdown cells, hMOF-knockdown cells or SIRT1 OE HUVECs. o, Knockdown of ACSS2 rescued the reduced telomere length caused by acetate. For a,b,e, data are from three biological independent replicates. For c,d, data are from two biological independent replicates. For f,h–j,n,o, data represent the mean ± s.e.; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis.

We then examined the role of ACSS2, MATIIα and hMOF in acetate-induced H4K16ac and cell senescence. ACSS2 knockdown in HUVECs via siRNA (siACSS2) abrogated acetate-induced H4K16ac and cellular senescence (Fig. 8e,f and Extended Data Fig. 8e). Similarly, glucose depletion decreased H4K16ac and reduced senescence in control cells but not in ACSS2-knockdown cells (Extended Data Fig. 8f–h). Knockdown of MATIIα by siRNA also abolished the effect of acetate on H4K16ac and cell senescence (Extended Data Fig. 9a–c). Although knockdown of hMOF dramatically reduced H4K16ac, acetate had no significant effect on H4K16ac and senescence in hMOF-knockdown cells (Fig. 8g,h and Extended Data Fig. 9d). Further analysis showed that both ACSS2 and hMOF are required for acetate to induce the transcription of p16 and SASP genes (Fig. 8i,j).

We then examined whether overexpression of SRIT1, the Sir2 homologue in mammals can attenuate the detrimental effect of acetate on cell senescence. We constructed HUVECs that stably overexpressed SIRT1 and found that acetate had no significant effect on H4K16ac and senescence in SIRT1 overexpression (SIRT1 OE) cells (Extended Data Fig. 9e,f), suggesting that acetate induces H4K16ac and cell senescence via the ACSS2–hMOF–SIRT1 pathway.

ACSS2 and hMOF are required for acetate to regulate senescence-associated gene expression and telomere silencing.

To gain further insight into how acetate regulates cell senescence, we performed RNA-seq for KAc-treated HUVECs. A total of 641 and 1,892 genes were upregulated and downregulated by acetate, respectively (Supplementary Table 9). KEGG analysis revealed these genes were enriched in senescence-associated pathways, that is, tumour necrosis factor signalling pathway and nuclear factor-κB pathway (Extended Data Fig. 9g). We also performed RNA-seq for ACSS2-knockdown HUVECs. In total, 676 and 513 genes were upregulated and downregulated in ACSS2 siRNA-treated HUVECs, respectively (Supplementary Table 10). Among these differentially expressed genes (DEGs), 101 genes upregulated by acetate were downregulated in ACSS2 siRNA-treated cells and 94 genes downregulated by acetate were upregulated after ACSS2 knockdown (Fig. 8k, Extended Data Fig. 9h,i and Supplementary Table 11), consistent with their opposite effect on cell senescence. Based on the CSGene and Ingenuity Pathway Analysis databases⁴¹, we identified a total of 639 genes involved in cell senescence. Among these senescence-associated genes, 54 and 89 genes were significantly regulated by ACSS2 and acetate, respectively (Fig. 8l and Supplementary Table 12). A total of 27 genes were co-regulated by ACSS2 and acetate, including SASP genes and cell cycle-associated genes (Fig. 8m).

Because acetate compromised telomere silencing via the Acs2–Sas2–Sir2 pathway in yeast, we examined whether acetate regulates telomere silencing in HUVECs. CCDN2, C1S and CDL163L1, which are located 0–10 Mb from the end of chromosome 12p, are used to determine the telomere position effect in mammals⁴². Acetate induced the transcription of these genes in control but not ACSS2-knockdown and hMOF-knockdown cells (Fig. 8n). Overexpression of SIRT1 alleviated the role of acetate on telomere silencing (Fig. 8n), suggesting that acetate reduces telomere silencing in mammalian cells via the ACSS2–hMOF–SIRT1 pathway. The transcription of these three telomere-proximal genes is regulated by telomere length⁴². We thus examined the effect of acetate on telomere length. Acetate significantly reduced telomere length and knockdown of ACSS2 increased telomere length (Extended Data Fig. 9j). Moreover, knockdown of ACSS2 rescued acetate-reduced telomere length (Fig. 8o), consistent with their effects on telomere silencing and cell senescence.

Discussion

Our study reveals that Acs2-containing SESAME complex directly interacts with Sas2-containing SAS complex and uncovers a conserved mechanism to connect acetate metabolism to telomer silencing and cell senescence (Extended Data Fig. 10). Under physiological conditions, Acs2 synthesizes acetyl-CoA from glucose-derived acetate. By interacting with SAS complex, SESAME complex specifically increases subtelomeric H4K16ac, which then recruits Bdf1 to protect H4K16ac from deacetylation by Sir2. H4K16ac then acts as a heterochromatin boundary to limit the spreading of SIR complex to maintain normal telomere silencing. In the presence of exogenous acetate, the interaction between SESAME and SAS is enhanced to induce the triad of increased H4K16ac towards telomere-proximal regions, enhanced Bdf1 binding and dissociation of Sir2 from telomeres, leading to reduced telomere silencing and accelerated yeast ageing. In human endothelial cells, ACSS2 interacts with hMOF and is required for acetate to induce H4K16ac, reduce telomere silencing and accelerate cell senescence.

Acetyl-CoA is unstable and compartmentalized inside cells⁵. We show that SESAME-catalysed nucleocytosolic acetyl-CoA synthesis is required to maintain subtelomeric H4K16ac. This specificity could be explained by the ‘local production and local consumption’ model⁵: some metabolic enzymes interact with epigenetic modifiers, which can be recruited to specific chromatin loci, thereby determining the locus-specific modifications. Here, SESAME interacts with SAS to specifically promote H4K16ac. Given that SAS preferentially acetylates subtelomeric nucleosomes and acetate enhances the interaction between SAS and SESAME, it is reasonable that acetate specifically increases subtelomeric H4K16ac.

Inactivation of Acs2 leads to dissociation of Sas2 from Sas4 and Sas5. As Sas4 and Sas5 are required for Sas2 activity, it is possible that Acs2 promotes H4K16ac by maintaining the integrity of the SAS complex. In acs2-ts mutants, less acetyl-CoA is produced and cells may need to increase the retention of Sas2 at chromatin to maintain H4K16ac as a compensatory response. The reduced binding of Sas4 and Sas5 at chromatin in acs2-ts mutants could be due to dissociation of Sas4/Sas5 from Sas2. Some HATs have been shown to acetylate themselves and/or associated proteins, that is, Tip60 and Gcn5 (refs. ^3,43). However, none of the SAS subunits are acetylated. The regulation of chromatin-modifying complex integrity by metabolic enzymes has been reported. Fructose-1, 6-bisphosphatase 1 (FBP1) interacts with enhancer of zeste homolog 2 (EZH2), resulting in disassembly of PRC2 complex and reduced EZH2 activity⁴⁴.

Acetyl-CoA has been shown to regulate histone acetylation at growth-related genes and may enhance cellular proliferative capacity³. We find that Acs2-synthesized acetyl-CoA accelerates yeast cell ageing. Acs2 has been reported to regulate ageing by repressing autophagy³⁹. However, acetate has no effect on autophagy. Instead, Acs2 within the SESAME complex regulates telomere silencing and chronological ageing by controlling the balance between Sas2-catalysed H4K16ac and Sir2-mediated H4K16 deacetylation. The role of Acs2 in ageing may depend on the growth conditions as Acs2 has been reported to maintain normal replicative lifespan when grown on a non-fermentable carbon source¹². Like Acs2, Ser33 has been identified as a pro-ageing factor and loss of SER33 extends the chronological lifespan⁴⁵. As Ser33 and Acs2 are components of SESAME¹³, these two subunits may function together within SESAME to regulate H4K16ac and cell ageing.

Acetate has been reported to accelerate ageing in yeast by lowering the medium pH^17–19. This theory cannot satisfactorily explain why other acids, such as hydrochloric acid and malic acid, have no significant effect on chronological ageing albeit their capability to change pH¹⁸. Here, we focused on the effect of acetate molecules instead of pH on cell ageing by treating cells with KAc and maintaining the medium pH. KAc but not KCl treatment significantly reduced the chronological lifespan. Similar effects were observed in cells treated with sodium acetate. Our data thus show that acetate itself is sufficient to reduce lifespan.

Sir2/SIRT1 is involved in regulation of cell senescence by acetate. The role of Sir2 in ageing is affected by genetic backgrounds. Deletion of SIR2 reduces the chronological lifespan in the BY4741 background but not in W303, DBY746 and EG123 backgrounds^38,46,47. Lack of Sir2 in the EG123 background promotes acetate utilization and acetyl-CoA generation, which reduce the accumulation of pro-ageing metabolites, acetate and ethanol in the medium and extends chronological lifespan³⁸. Meanwhile, Sir2 deacetylates phosphoenolpyruvate carboxykinase (Pck1), a key enzyme in gluconeogenesis, to inhibit its activity and reduce chronological lifespan^38,47. The strains used in this study are derived from BY4741 cells and deletion of SIR2 indeed reduced the chronological lifespan (Supplementary Fig. 1a). Moreover, acetate had no effect on Pck1 acetylation in both WT and sir2Δ mutants (Supplementary Fig. 1b,c), suggesting that Sir2-mediated deacetylation of Pck1 is not involved in acetate-induced ageing.

The imbalance between Sas2-catalysed H4K16ac and Sir2-mediated H4K16 deacetylation can reduce telomere silencing and accelerate replicative ageing^21,23. There is some difference between replicative ageing and chronological ageing in budding yeast, but the Acs2–Sas2–Sir2 pathway may regulate chronological ageing in BY4741 cells after acetate treatment in a manner similar to how they regulate replicative ageing. In fact, some metabolic enzymes regulate both replicative ageing and chronological ageing via Sir2 in yeast, that is, alcohol dehydrogenase 1 (Adh1)⁴⁸. Our study provides mechanistic insights into the pro-ageing effect of acetate because a similar mechanism exists in mammalian cells.

Methods

Yeast strains and medium.

All yeast strains used in this study are detailed in Supplementary Table 13. The gene-deletion mutants and genomic integration of C-terminal tags were constructed by homologous recombination of PCR fragments⁴⁹. All yeast strains were verified by colony PCR, DNA sequencing, RT–qPCR and/or western blot before use in experiments. Cells were cultured in yeast peptone dextrose (YPD) medium or selective medium unless otherwise indicated. YPD medium was prepared with yeast extract (Oxoid, LP0021), tryptone (Oxoid, LP0042) and dextrose (Biosharp, 50–99-7).

Cell cultures.

HUVECs were purchased from and authenticated by ScienCell Research Laboratories (catalogue no. 8000). Cells were grown in Endothelial Cell Medium (ECM) supplemented with 5% FBS, 1% Endothelial Cell Growth Supplement (ECG) and 1% of penicillin–streptomycin solution for up to 34 passages. The senescence of cells was verified by cell morphology and western blots with p21 (1:1,000 dilution; 10355–1-AP, Proteintech) and SA-β-gal staining.

Spotting assay.

Cells were spotted on the plates containing the indicated medium with a starting optical density of 0.5 from samples measured at a wavelength of 600 nm (OD₆₀₀) and fivefold serial dilutions, unless otherwise noted. Plates were incubated at the indicated temperatures for 2–3 d and photographed.

Immunoblot analysis.

Histones were extracted from exponentially growing yeast cells⁵⁰. Cells were grown in 5 ml YPD or selective medium as indicated until an OD₆₀₀ of 0.7–1.0. Cells were then collected and lysed in 2 M NaOH with 8% 2-mercaptoethanol. Cell lysate was centrifuged and the pellet was resuspended in 2× SDS sample buffer and boiled at 95 °C for 10 min. Protein samples were separated by 8–15% SDS–PAGE and transferred to Immobilon PVDF membrane (Bio-Rad). The blots were probed with antibodies against specific proteins followed by incubation with horseradish peroxidase-labelled IgG secondary antibodies. The specific proteins were visualized by using the ECL Chemiluminescence Detection Kit (Bio-Rad, 170–5061). The band intensity was quantified with ImageJ software (https://imagej.en.softonic.com/). The relative intensity of histone modifications was normalized to histone H3 or H4. The other proteins were normalized to GAPDH or actin. The specificity of primary antibodies used in this study was examined in corresponding mutants by western blots (Supplementary Fig. 2).

Antibodies.

Antibodies against anti-H3 (1:5,000 dilution; ab1791), histone H4 (1:5,000 dilution; ab10158), H3pT11 (1:3,000 dilution; ab5168), H3K14ac (1:2,000 dilution; ab52946) and tubulin (1:3,000 dilution; ab6160) were purchased from Abcam; antibody against Sir2 (1:500 dilution; sc-6667) was purchased from Santa Cruz Biotechnology; antibodies against GAPDH (1:5,000 dilution; 10494–1-AP), actin (1:5,000 dilution; 20536–1-AP), ACSS2 (1:1,000 dilution; 16087–1-AP), MATIIα (1:2,000 dilution; 55309–1-AP), p21(1:1,000 dilution; 10355–1-AP), SIRT1 (1:2,000 dilution; 13161–1-AP), 6 × His (1:5,000 dilution; HRP-66005), GST (1:5,000 dilution; HRP-66001) and Myc (1:5,000 dilution; 66003–2-1 g) were obtained from Proteintech; antibodies against histone H3 (1:5,000 dilution; 9715S), MYST1 (D5T3R; 1:2,000 dilution; 46862S), pan-acetyl-lysine (1:2,000 dilution; 9441S) and H3K9ac (1:5,000 dilution; 9649S) were purchased from Cell Signaling Technology; antibody against H4K16ac (1:5,000 dilution; 07–329) was obtained from EMD Millipore; antibody against FLAG M2 (1:5,000 dilution; F1804–1MG) was obtained from Sigma; antibodies against H3K18ac (1:1,000 dilution; A7257), H3K56ac (1:1500 dilution; A7256) and CDKN1A/p21 (1:2,000 dilution; A11454) were obtained from Abclonal; the custom-made antibodies against Sam1, Sas2 and Shm2 were produced by Covance; the custom-made antibodies against Acs2 and Pyk1 were kind gifts from J. D. Boeke and J. Thorner, respectively. The specificity of primary antibodies was verified by western blots of cell lysates of the corresponding histone point mutants, gene-deletion or knockdown mutants (Supplementary Fig. 2).

Quantitative PCR with reverse transcription.

RNA was isolated from exponentially growing yeast cells by standard phenol–chloroform extraction procedures. Purified RNA was digested with DNase (Sigma) and reversed transcribed to cDNA by Reverse Transcriptase Kit (M-MLV; ZOMANBIO, ZR102). The cDNA was quantified by qPCR with SYBR Green Supermix (Bio-Rad, 1725121) using primers listed in Supplementary Table 14. We used the 2^−ΔΔCt method to determine the quantity of the transcription level. The mRNA level of the gene of interest was normalized to that of actin.

RNA sequencing.

Total RNA was extracted from cells by standard phenol–chloroform extraction procedures. The DNase-digested RNA was quantified by Thermo NanoDrop 2000, and its quality was examined using the Agilent 2100 Bioanalyzer system according to the manufacturer’s instructions. Library construction, sequencing and bioinformatics analysis were done by Origingene Bio-pharm Technology. There were three biological replicates for each strain and/or treatment. The quality of raw sequenced reads was assessed by FastQC (v0.11.4). Cutadapt (v.1.16) was used to discard the adaptor sequences and remove low-quality reads (Q20 < 20). The clean reads were then mapped to the designated reference genome (Saccharomyces cerevisiae, NCBI GCF000146045.2 R64). DEGSeq package was used to identify DEGs between different groups. P values were calculated by edgeR (v.3.24). DEGs were defined by a P value < 0.05 and fold change ≥ 1.5 or fold change ≤ 0.75. DEGs were further used for KEGG pathway analysis. KOBAS software was used to test the statistical enrichment of DEGs in KEGG pathways. The statistical significance for co-regulated genes between each RNA-seq dataset was calculated by hypergeometric test.

Telomere-proximal gene analysis and statistics.

Genes within 60 kb of a telomere throughout the genome were pooled and ordered according to their distance from a telomere.

The genes were grouped by their distance from the nearest telomeres in consecutive intervals of 10 kb. For each interval, the fraction of genes that were repressed or de-repressed in each mutant was calculated. A χ² value for each interval was calculated by comparing the fraction of genes that were repressed or de-repressed in the interval to the genome-wide average. Statistical significance was defined as: χ² > 3.841, P < 0.05; χ² > 6.635, P < 0.01; χ² > 10.828, P < 0.001.

Telomere length quantification.

The relative human telomere length was quantified by measuring the amount of telomere repeat amplification product to a single-copy gene product (IFNB1) using real-time PCR as described⁵¹. The genome DNA (gDNA) was isolated using the animal gDNA kit (Tsingke) and quantified by spectrophotometer. About 0.5–10 ng gDNA was used in real-time PCR with SYBR Green Supermix (Bio-Rad, 1725121) using the following cycling conditions: 10 min at 95 °C, 35 cycles of 95 °C for 15 s, 60 °C for 1 min followed by a dissociation stage. The primers used were: TEL-F, 5′-CGGTTTGTTTGGGTTTGGGTTTGGGTTTGGGTTTGGGTT-3′; TEL-R, 5′-GGCTTGCCTTACCCTTACCCTTACCCTTACCCTTACCCT-3′; IFNB1-F, 5′-GGTTACCTCCGAAACTGAAGA-3′; IFNB1-R, 5′-CCTTTCATATGCAGTACATTAGCC-3′.

Gene-set enrichment analysis.

The global expression of genes in longevity regulation pathways was compared between control and KAc-treated cells. Genes that were downregulated by acetate treatment were ranked at the top of the list, whereas upregulated genes were ranked towards the bottom. The green curve showed the running enrichment score for the gene set as the analysis walked down the ranked list.

Chromatin immunoprecipitation assay.

The ChIP assays were performed in exponentially growing yeast cells⁵⁰. Yeast cells were grown in 200 ml YPD medium at 28 °C until an OD₆₀₀ of 0.7–1.0. The cross-linking was performed by adding 5.4 ml of 37% formaldehyde and quenched by adding 10.5 ml of 2.5 M glycine. Cells were then lysed and the chromatin was sheared by sonication. The sonicated chromatin was then subjected to immunoprecipitation with antibodies pre-bound to Protein G Dynabeads (Invitrogen). After extensive washing, the eluted DNA–protein complexes were digested with Proteinase K (Invitrogen) at 55 °C for 1 h and de-cross-linked at 65 °C overnight. The DNA was digested with RNase A (Roche), purified with ethanol precipitation and quantified by qPCR using the primers listed in Supplementary Table 14. All ChIP signals were normalized to their corresponding input signals.

ChIP–seq for Acs2 was performed in the Acs2–FLAG strain with anti-FLAG antibody (Sigma, F1804). The Acs2-bound DNA was immunoprecipitated and purified as described previously¹³. The untagged Acs2 was used as a negative control. Library construction and sequenced by Igenebook Bioinformatics Institute (Wuhan, China). The raw sequenced reads were filtered by FastQC (v.0.11.9) and Trim Galore (v.2.11) with quality control of Q30 and four base pairs were removed at the 5′ end of each paired read. Clean reads were mapped to the S. cerevisiae reference genome (sacCer3) using bowtie v0.12.9. The telomere repeat sequences were excluded for further analysis. SAMtools (v.1.11) was used for format conversion and peaks were called using MACS2 (v.2.1.1, macs2 callpeak) using the parameters --t --c --g 1.2e7 --n --B --q 0.01 --nomodel, with a minimum width of 50 bases and IP/input density ≥ 1.5. Peak annotation was performed on a website service available at https://manticore.niehs.nih.gov/pavis2/. Tracks were smoothed using deepTools2 and visualized by IGV software with the reference genome sacCer3.

For ChIP–seq analysis of the effect of acetate on H4K16ac and Acs2, the normalized read density for 0–50-kb regions (subtelomeric regions) of the S. cerevisiae chromosome ends was tiled into 200-bp windows with a 50-bp step size by bedtools and plotted by R package ggplot2. Regions >50 kb from each chromosome end were defined as non-telomeric regions and divided into 10-kb bins with 1-kb window step size. The accession numbers for the raw ChIP–seq dataset are GSE146143 and PRJNA695438.

Chronological lifespan assay.

A total of 5 ml of seed cultures from individual colonies was grown overnight in YPD medium. Flasks with 50 ml of appropriate medium were seeded to generate an initial OD₆₀₀ of 0.1 and incubated for 3 d until cell growth was ceased at 28 °C. To examine the effect of acs2-ts mutants on chronological lifespan, we grew WT and acs2-ts mutants in YPD at 35 °C. On day 3, 100 μl aliquots of each culture were collected and a tenfold serial dilution was spread on a YPD plate to determine the colony-forming units. At subsequent time points, another set of 100 μl aliquots were removed, diluted, plated and counted. The colony-forming unit value for each culture was plotted as a function of time. All chronological ageing experiments were performed with three biological replicates for each strain. The integral of each ageing curve was determined by summing the trapezoids formed by the viability at each time point⁸.

Recombinant protein purification.

The recombinant 6 × His-tagged Acs2 (Acs2–His6) protein was expressed in Escherichia coli and purified by Ni-NTA Agarose (QIAGEN). To get recombinant Sas2 with high yield, high purity and high activity, we used the one-step CL7/Im7 purification system to purify Sas2 (ref. ²⁸). This system is based on the ultra-high affinity between colicin E7 (CE7) DNase and immunity protein 7 (CE7/Im7; K_d = 10⁻¹⁴ to 10⁻¹⁷ M). As CE7 is toxic to cells, a catalytical dead mutant CL7 was used to tag the C-termini of Sas2. The Im7 affinity column was prepared by immobilizing Im7 to agarose beads²⁸. The recombinant CL7-tagged Sas2 (Sas2–CL7) protein was expressed in E. coli and purified by Im7 affinity column²⁸. As a control, CL7 was expressed and purified.

Co-immunoprecipitation assay.

For in vivo Co-IP, yeast cells were grown in 200 ml YPD medium until an OD₆₀₀ of 1.0. Cells were then collected, washed with PBS and lysed with one volume of glass beads in IP buffer (40 mM HEPES–KOH (pH 7.5), 10% glycerol, 150 mM NaCl, 0.1% Tween-20, 1 mM PMSF, 2 μg/ml leupeptin, 1 μg ml⁻¹ pepstatin A, Sigma protease inhibitor cocktail). Pre-cleared cell extracts were incubated with anti-FLAG M2 Affinity Gel (Sigma) or anti-Myc Gel (Proteintech) at 4°C for 2–4 h. The beads were then washed extensively with pre-chilled washing buffer (25 mM HEPES–KOH (pH 7.5), 100 mM KCl, 2 mM MgCl₂, 1 mM EDTA, 10% glycerol and 0.02% NP40) and boiled at 95 °C for 5 min. The supernatants were subjected to SDS–PAGE and western blots.

For in vitro immunoprecipitation, purified CL7 and Sas2–CL7 were mixed with purified recombinant Acs2, immunoprecipitated with Im7 agarose beads, which bind CL7 tag with high affinity, and washed three times with a large excess of IP washing buffer²⁸. Supernatants from the boiled beads were subjected to SDS–PAGE and western blots.

Acetyl-CoA measurement.

Cells were grown in 5–10 ml YPD medium at 28 °C until an OD₆₀₀ of 1.0 and divided into two aliquots. One aliquot was used for protein quantification with Super-Bradford Quantification Kit (Beijing ComWin Biotech). Another aliquot was quenched by buffer 1 (60% methanol with 10 mM tricine) at −80 °C for 5 min. Cells were collected, resuspended in 400 μl buffer 2 (75% ethanol with 0.5 mM tricine) and lysed by incubation at 80 °C for 3 min followed by cooling on an ice bath for 5 min. The cell suspension was centrifuged at 20,000g for 10 min and the supernatant was collected. The Acetyl-CoA quantity was measured using an Acetyl-CoA kit (Suzhou Comin Biotechnology) according to the manufacturer’s instructions. The assay was performed in 96-well clear-bottom plates, and the fluorescence was quantified using the SpectraMax M2 (Molecular Devices). The acetyl-CoA levels were normalized to the corresponding protein concentrations.

To measure the subcellular acetyl-CoA concentrations, 50 ml of exponentially growing cells was spheroplasted in SB buffer (1 M sorbitol, 20 mM Tris (pH 7.4) and 10 mg ml⁻¹ zymolase 20T) and lysed with EBX (20 mM Tris (pH 7.4), 100 mM NaCl, 0.5% Triton X-100, 15 mM 2-mercaptoethanol and protease inhibitors). One aliquot of the lysate was collected to measure the total acetyl-CoA and the remaining lysate was layered over NIB (20 Mm Tris (pH 7.4), 100 mM NaCl, 1.2 M sucrose, 15 mM 2-mercaptoethanol and protease inhibitors). After centrifugation, the supernatant was collected as the cytoplasmic fraction. The pellet was then lysed with 1% Triton X-100, centrifuged and collected as the nuclear fraction. For the total, cytoplasmic and nuclear fractions, the quantity of acetyl-CoA levels was normalized to the corresponding protein concentrations. GAPDH and histone H3 were analysed by western blots to ensure that there was no cross-contamination between compartments during subcellular fractionation.

RNA interference.

The scramble siRNA, ACSS2, MATIIα and hMOF siRNA sequences used were: 5′- TTCTCCGAACGTGTCACGT-3′, 5′-UAUGCUUGGUGACAGGCUCAUCUCC-3′, 5′- GAAGAAUUUCGAUCUCCGCTT-3′ and 5′- UGCUGUACAGAAGAACUCA-3′, respectively. Cells were transfected with siRNA using Lipofectamine 3000 according to the manufacturer’s instructions (Invitrogen). The knockdown efficiency was examined by RT–qPCR with primers specific to ACSS2, MATIIα and hMOF and western blots with anti-ACSS2, anti-MATIIα and anti-hMOF antibodies.

Senescence-associated β-galactosidase staining.

The SA-β-gal activity of HUVECs was determined by the Senescence β-Galactosidase Staining Kit (Beyotime) according to the manufacturer’s instructions. Cells were cultured in six-well plates, washed with PBS and fixed for 10–15 min at room temperature. Cells were then washed twice with PBS and incubated with the staining mixture at 37 °C overnight. The SA-β-gal signals were analysed using ImageJ software. For each staining assay, the late passage (postnatal day 34) or H₂O₂-treated HUVEC cells were used as the positive control; the early passage (postnatal day 6) HUVEC cells were used as the negative control.

Mouse experiments.

Thirty 3-month-old male mice on a C57BL/6 background were purchased from Beijing Vital River Laboratory Animal Technology. The mice were randomly divided into five groups (six mice per group) and housed under a 12-h light–12-h dark cycle with an ambient temperature of 22 ± 2 °C and humidity of 55 ± 10% for up to 26 months. Water and standard chow were provided ad libitum following the regulations and guidelines of Wuhan No. 1 Hospital. Six mice in each group were killed and the heart tissues were collected at 3, 6, 12, 20 and 26 months. Animal procedures complied with all relevant ethical regulations and were approved by local authorities (Animal Care and Use Committee of Wuhan Hospital of Traditional Chinese and Western Medicine, Wuhan No. 1 Hospital).

Software.

All software used in this study for data analysis are either commercially available or open source. The following software were used in this study: ImageJ (v.1.8.0; https://imagej.en.softonic.com/), Prism 8 for graphs (v.8.0.1; https://www.graphpad.com/scientific-software/prism/), SRA toolkit (v.2.9.2; https://hpc.nih.gov/apps/sratoolkit.html), FastQC (v.0.11.4 and v.0.11.9; http://www.bioinformatics.babraham.ac.uk/projects/fastqc/), Trim Galore (v.0.3.1 and v.2.11; https://github.com/FelixKrueger/TrimGalore/), Bowtie2 (v.2.1.0; https://www.uio.no/english/services/it/research/hpc/abel/help/software/bowtie2.html), SAMtools (v.1.7–2; http://samtools.sourceforge.net/), MACS2 (v.2.1.1; https://taoliu.github.io/MACS/), bedtools (v.2.19.0; https://sourceforge.net/projects/bedtools/), R (v.3.1; https://cran.rstudio.com/), IGV (v.2.0; http://www.igv.org/), SPSS (https://www.ibm.com/analytics/academic-statistical-software/), KOBAS (v.3.0; http://kobas.cbi.pku.edu.cn/), cutadapt (v.1.16; http://cutadapt.readthedocs.io/), hisat2 (v.2.1.0; https://ccb.jhu.edu/software/hisat2/index.shtml), edgeR (v.3.24; https://www.bioconductor.org/packages/release/bioc/html/edgeR.html), DEGSeq package (v.21.24.0; http://bioconductor.org/packages/3.9/bioc/html/DESeq2.html) and deepTools (v.2.0; https://github.com/deeptools/deepTools/archive/1.5.12.tar.gz).

Statistics and reproducibility.

The number of replicates is indicated in the figure legends. Statistical differences in this study were determined by two-tailed unpaired t-tests, and a P value < 0.05 was considered statistically significant unless otherwise indicated. *P < 0.05; **P < 0.01; ***P < 0.001; NS, not significant. For all error bars, data are the mean ± s.e. For western blots in Figs. 1a,d, 2a, 4c,f,g,h,j, 6a, 7e,f and 8a,b,e,g and Extended Data Figs. 1d,e, 2c, 4a,c,e,g,k, 5a–e, 6b,c,g, 7f,g, 8f and 9a,e, all experiments were performed with three biological replicates and were quantified with s.e. values. For micrographs in Figs. 7k and 8f,h and Extended Data Figs. 8g and 9b,c,f, all experiments were performed with three biological replicates. For Co-IP in Figs. 2b–f,j,k, 4o,p and 8c,d and Extended Data Figs. 2b,d,e and 5g, all experiments were performed with at least two biological replicates.

Extended Data

Extended Data Fig. 3 | — a KEGG analysis of genes co-occupied by Acs2, Sam1 and H4K16ac. See also Fig. 3a. b Boxplots showing the overall transcription changes in WT, *acs2-ts*, *sam1Δ*, *sas2Δ*, and H4K16R mutants. Centre lines denote medians; box limits denote 25th and 75th percentiles; whiskers denote maxima and minima. Two-sided t-tests were used for statistical analysis. c Left panel: Venn diagrams showing the overlap of genes co-regulated by Acs2 and Sam1. Genes with fold changes in *acs2-ts* (≥3-fold; ≤0.5-fold), *sam1Δ* (≥1.5-fold; ≤0.75-fold) and P < 0.05 were considered as differentially expressed. The significance for overlapped genes between *acs2-ts* and *sam1Δ* RNA-seq datasets was indicated by the P-value, which was calculated by hypergeometric test. Right panel: KEGG analysis of genes co-regulated by Acs2 and Sam1. d-g Histogram showing the fractions of genes down-regulated in *sas2Δ* (d), H4K16R (e), *sam1Δ* (f) and *acs2-ts* (g) when plotted as a function of their distance to the nearest telomeres. Genes were categorized at 10-kb intervals for up to 60 kb from telomeres. One-sided χ² tests were used. *, χ² > 3.841, P < 0.05; ***, χ² > 10.828, P < 0.001. h RT–qPCR analysis of the transcription of telomere-proximal genes in *TEFpr-ACS2* and *CYCpr-ACS2*. Data represent means ± SE; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis.

Extended Data Fig. 4 | — a and b Western blot analysis of the effect of acetate on histone modifications in yeast cells. WT (BY4741) cells were treated with 0–50 mM KAc for 4 hr. c Effect of KCl on H4K16ac in WT (BY4741) cells as determined by western blots. KAc treatment was used as a positive control. d Effect of KAc and KCl treatment on the medium pH after yeast cells were grown for 24 hr. e-g Western blot analysis of the effect of NaCl and NaAc on histone modifications in WT (BY4741) cells. The asterisk (*) indicates non-specific bands. h and i ChIP-seq tracks of H4K16ac at indicated genes when cells were treated with KCl and KAc. j ChIP-qPCR analysis of the effect of acetate on H4K16ac at subtelomere regions (TEL VIR, TEL V XC, TEL V XR, TEL VII XC and TEL VII XR) and telomere-proximal genes (*COS8*, *IRC7*, *YCR106W*, *PHO11*, *SOR1*). k Effect of glucose on H4K16ac as determined by western blots. Log phased WT (BY4741) cells were treated with YP + 0.05% glucose, YP + 1% glucose and YP + 4% glucose for 2 hr. For Extended Data Fig. 4a–g,j,k, data represent means ± SE; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis.

Extended Data Fig. 5 | — a Acetate has no significant effect on Acs2 protein levels as determined by western blots. WT (BY4741) cells were treated with 0–50 mM KAc for 4 hr. KCl was added to keep the concentration of K+ at 50 mM. b Effect of acetate on H4K16ac in WT TetO₇ and TetO₇*-ACS2* mutants as determined by Western blots. WT TetO₇ and TetO₇-*ACS2* mutant were treated with KAc and doxycycline (dox) for 0–5 hr. c Effect of acetate on H4K16ac in WT and *sam1Δ* mutant. d and e Effect of acetate on H4K16ac in *acs1Δ* (d) and *ach1Δ* (e) mutants. f Acetate has no significant effect on Sas2 occupancy at subtelomere regions as determined by ChIP-qPCR. g Acetate treatment had no effect on the interaction between Sas2 and Sas5 as determined by Co-IP assay. The Sas5–13Myc/Sas2-FLAG and Sas5–13Myc/Sas2-FLAG/*acs2-ts* cells were grown at 37 °C and treated with or without KAc for 4 hr. Sas5 (Sas5–13Myc) was immunoprecipitated with anti-Myc agarose from cell lysates. The co-IPed Sas2 was detected by anti-FLAG (Sas2-FLAG). For Extended Data Fig. 5a–f, data represent means ± SE; n = 3 biological independent experiments. Two-sided t-tests were used for statistical analysis. For Extended Data Fig. 5g, shown is the typical example of 2 biological independent replicates.

Extended Data Fig. 6 | — a Histogram showing the fraction of genes down-regulated in *bdf1Δ* (≤0.75-fold, P < 0.05) when plotted as a function of their distance to the nearest telomere. Genes were categorized at 10-kb intervals for up to 60 kb from telomeres. One-sided χ² tests were used. *, χ² > 3.841, P < 0.05; ***, χ² > 10.828, P < 0.001. b Effect of Acs2 on Bdf1 expression as determined by western blots. c Effect of Bdf1 on the global levels of H4K16ac as determined by western blots. d ChIP analysis of H4K16ac at TEL V and TEL VII regions in WT and *bdf1Δ* mutant. e Effect of acetate on Bdf1 occupancy at subtelomere regions (TEL V XC, TEL V XR, TEL VII XC and TEL VII XR) as determined by ChIP-qPCR analysis. f ChIP-qPCR analysis of Sir2 occupancy at regions with different distance to TEL VI-R in *TEFpr-ACS2* and *CYCpr-ACS2*. g Effect of Acs2 on Sir2 protein levels as determined by western blots. WT and *acs2-ts* mutant were grown at 37 °C for 2 hr. h ChIP-qPCR analysis of the effect of acetate on Sir2 binding at telomere-proximal genes in *acs2-ts* mutant. KCl was added as a control. i and j RT–qPCR analysis of the effect of KAc and sodium acetate (NaAc) on the transcription of telomere-proximal genes. KCl and NaCl were used as controls. k and l qRT-PCR analysis of the effect of acetate on the transcription of telomere-proximal genes in *sam1Δ* (k) and *acs2-ts* (l) mutants. m qRT-PCR analysis of the transcription of telomere-proximal genes in WT cells treated with YP + 2% glucose, 0.5% glucose, 0.05% glucose, 2% glycerol, 2% ethanol. For Extended Data Fig. 6b–m, data represent means ± SE; n = 3 biological independent experiments, two-sided t-tests were used for statistical analysis.

Extended Data Fig. 7 | — a Boxplots showing the overall transcription changes by KAc treatment. Centre lines denote medians; box limits denote 25th and 75th percentiles; whiskers denote maxima and minima. Two-sided t-tests were used for statistical analysis. b Histogram showing the number of genes up-regulated by KAc treatment but down-regulated in *sas2Δ* (KAc up/*sas2Δ* down), H4K16R (KAc up/H4K16R down), *bdf1Δ* (KAc up/*bdf1Δ* down) when plotted as a function of their distance to the nearest telomere. Genes were categorized at 10-kb intervals for up to 60 kb from telomeres. One-sided χ² tests were used. ***, χ² > 10.828, P < 0.001. c-d Quantification of western blots data in Fig. 7e,f. e Analysis of the chronological life span of *TEFpr-ACS2*, *CYCpr-ACS2*, *TEFpr-ACS2 sir2Δ*, and *CYCpr-ACS2 sir2Δ*. f Analysis of the effect of acetate on the chronological life span of Sir2 overexpression cells (Sir2 OE, BY4741 genetic background). Cells (BY4741) transformed with the empty vector (EV) were used as a control. The survival percentage of cells on day 11 and day 25 was presented on the left panel. g Effect of acetate on autophagy in cells expressing the endogenous *ATG8* promoter-driven *GFP-ATG8*. The autophagy activity was determined by a GFP (green fluorescent protein) liberation assay, which detects free GFP that is cleaved upon the delivery of endogenous promoter-driven Atg8 with an N-terminal GFP tag (GFP-Atg8) to the vacuole and subsequent proteolysis of Atg8. For Extended Data Fig. 7c–g, data represent means ± SE; n = 3 biological independent experiments, two-sided t-tests were used for statistical analysis.

Extended Data Fig. 8 | — a Effect of acetate on HUVEC cell growth. b Effect of acetate on HUVEC cell cycle profiling. c-e Quantification of western blots data in Fig. 8a,b,e. f Effect of glucose (Gluc) starvation on H4K16ac in control (siControl) and ACSS2 knockdown (siACSS2) HUVEC cells. Cells were depleted of glucose for 6 hr. g Effect of glucose (Gluc) starvation on senescence of control (siControl) and ACSS2 knockdown (siACSS2) HUVEC cells as determined by SA-β-gal staining. Scale bar, 25 μm. Right panel: Quantification of the percentage of SA-β-gal positive cells. h Effect of glucose starvation on transcription of IL1A, IL1B, p16 and p21 in control (siControl) and ACSS2 knockdown (siACSS2) HUVEC cells. For Extended Data Fig. 8a–h, the quantitative data represent the mean ± SE; n = 3 biological independent experiments; two-sided t-tests were used for statistical analysis.

Extended Data Fig. 9 | — a Effect of acetate on H4K16ac in control (siControl) and MATIIα knockdown (siMATIIα) HUVEC cells. b and c Effect of acetate on senescence of control (siControl) and MATIIα knockdown (siMATIIα) cells as determined by SA-β-gal staining (b) and SAHF detection (DAPI staining) (c). d Quantification of western blots data in Fig. 8g. e Effect of acetate on H4K16ac in control and SIRT1 overexpression (SIRT1 OE) HUVEC cells. f Effect of acetate on senescence of control and SIRT1 overexpression (SIRT1 OE) HUVEC cells as determined by SA-β-gal staining. g KEGG analysis of 2533 genes differentially regulated by KAc. h Top panel: Venn diagrams showing the overlap between genes down-regulated by siACSS2 but up-regulated by KAc treatment. Bottom panel: Venn diagrams showing the overlap between genes up-regulated by siACSS2 but down-regulated by KAc treatment. i Heatmap showing genes differentially regulated by ACSS2 and acetate. j Effect of acetate and ACSS2 on telomere length in young (P10) and senescent (P22) HUVEC cells. For Extended Data Fig. 9a–f,j, the quantitative data represent the mean ± SE; n = 3 biological independent experiments; two-sided t-tests were used for statistical analysis.

Extended Data Fig. 10 | — a Under normal conditions, Acs2 within the SESAME complex synthesizes acetyl-CoA from glucose-derived acetate. SESAME specifically maintains normal levels of H4K16ac at subtelomere regions via its interaction with SAS complex. By promoting the binding of Bdf1, H4K16ac functions as the telomeric heterochromatin-euchromatin boundary to prevent the spreading of SIR complex. b When cells were treated with acetate, the interaction between the SESAME complex and the SAS complex is enhanced, which then increases Sas2-catalyzed H4K16ac and Bdf1 binding across the subtelomere regions. As a consequence, Sir2 is dissociated from the telomeres, leading to compromised telomere silencing and accelerated aging in yeast.

Supplementary Material

Supplemental information

NIHMS2138227-supplement-Supplemental_information.pdf^{(2.3MB, pdf)}

Supplemental tables

NIHMS2138227-supplement-Supplemental_tables.xlsx^{(513.5KB, xlsx)}

Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s42255-021-00412-9.

Acknowledgements

We sincerely thank J. D. Boeke and J. Thorner for yeast strains and antibodies. We thank Igenebook Bioinformatics Institute (Wuhan, China) for technical support. This work was supported by funding from the National Natural Science Foundation of China (31970578 to S.L. and 31872812 to X.Y.), the Natural Science Foundation of Hubei Province (2019CFA077 to X.Y. and 2017CFA066 to S.L.), National Institutes of Health (1R35GM118068 to J.L.W.) and the Stowers Institute.

Footnotes

Competing interests

The authors declare no competing interests.

Reporting Summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Extended data is available for this paper at https://doi.org/10.1038/s42255-021-00412-9.

Data availability

The accession numbers for the Acs2 and H4K16ac ChIP–seq datasets reported in this paper are GSE146143 and PRJNA695438. The accession number for the Pyk1, Sam1 ChIP–seq dataset is GSE72972. The accession numbers for the H4K16ac ChIP–seq dataset are GSE52339 and GSE108217. The accession numbers for the raw RNA-seq dataset reported in this paper are PRJNA612967, PRJNA674443, PRJNA673282 and GSE147765. Uncropped blots and source data are provided with this paper. All other data are available from the authors upon reasonable request.

References

1.Sivanand S, Viney I & Wellen KE Spatiotemporal control of acetyl-CoA metabolism in chromatin regulation. Trends Biochem. Sci. 43, 61–74 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Mews P et al. Acetyl-CoA synthetase regulates histone acetylation and hippocampal memory. Nature 546, 381–386 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Cai L, Sutter BM, Li B & Tu BP Acetyl-CoA induces cell growth and proliferation by promoting the acetylation of histones at growth genes. Mol. Cell 42, 426–437 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Marino G et al. Regulation of autophagy by cytosolic acetyl-coenzyme A. Mol. Cell 53, 710–725 (2014). [DOI] [PubMed] [Google Scholar]
5.Yu X, Ma R, Wu Y, Zhai Y & Li S Reciprocal regulation of metabolic reprogramming and epigenetic modifications in cancer. Front. Genet. 9, 394 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Yu X & Li S Non-metabolic functions of glycolytic enzymes in tumorigenesis. Oncogene 36, 2629–2636 (2017). [DOI] [PubMed] [Google Scholar]
7.Li X et al. Nucleus-translocated ACSS2 promotes gene transcription for lysosomal biogenesis and autophagy. Mol. Cell 66, 684–697 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Wu Y et al. Glycolysis regulates gene expression by promoting the cross-talk between H3K4me3 and H3K14ac in Saccharomyces cerevisiae. J. Genet. Genomics 46, 561–574 (2019). [DOI] [PubMed] [Google Scholar]
9.Schug ZT et al. Acetyl-CoA synthetase 2 promotes acetate utilization and maintains cancer cell growth under metabolic stress. Cancer Cell 27, 57–71 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Takahashi H, McCaffery JM, Irizarry RA & Boeke JD Nucleocytosolic acetyl-coenzyme a synthetase is required for histone acetylation and global transcription. Mol. Cell 23, 207–217 (2006). [DOI] [PubMed] [Google Scholar]
11.Klein HP & Jahnke L Effects of aeration on formation and localization of the acetyl coenzyme A synthetases of Saccharomyces cerevisiae. J. Bacteriol. 137, 179–184 (1979). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Falcon AA, Chen S, Wood MS & Aris JP Acetyl-coenzyme A synthetase 2 is a nuclear protein required for replicative longevity in Saccharomyces cerevisiae. Mol. Cell. Biochem. 333, 99–108 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Li S et al. Serine and SAM responsive complex SESAME regulates histone modification cross-talk by sensing cellular metabolism. Mol. Cell 60, 408–421 (2015). [DOI] [PubMed] [Google Scholar]
14.Jaworski DM, Namboodiri AM & Moffett JR Acetate as a metabolic and epigenetic modifier of cancer therapy. J. Cell. Biochem. 117, 574–588 (2016). [DOI] [PubMed] [Google Scholar]
15.Folmes CD et al. Somatic oxidative bioenergetics transitions into pluripotency dependent glycolysis to facilitate nuclear reprogramming. Cell Metab. 14, 264–271 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Moussaieff A et al. Glycolysis-mediated changes in acetyl-CoA and histone acetylation control the early differentiation of embryonic stem cells. Cell Metab. 21, 392–402 (2015). [DOI] [PubMed] [Google Scholar]
17.Murakami CJ, Wall V, Basisty N & Kaeberlein M Composition and acidification of the culture medium influences chronological aging similarly in vineyard and laboratory yeast. PLoS ONE 6, e24530 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Burtner CR, Murakami CJ, Kennedy BK & Kaeberlein M A molecular mechanism of chronological aging in yeast. Cell Cycle 8, 1256–1270 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Fabrizio P et al. Superoxide is a mediator of an altruistic aging program in Saccharomyces cerevisiae. J. Cell Biol. 166, 1055–1067 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Longo VD, Shadel GS, Kaeberlein M & Kennedy B Replicative and chronological aging in Saccharomyces cerevisiae. Cell Metab. 16, 18–31 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Dang W et al. Histone H4 lysine 16 acetylation regulates cellular lifespan. Nature 459, 802–807 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Millar CB, Kurdistani SK & Grunstein M Acetylation of yeast histone H4 lysine 16: a switch for protein interactions in heterochromatin and euchromatin. Cold Spring Harb. Symp. Quant. Biol. 69, 193–200 (2004). [DOI] [PubMed] [Google Scholar]
23.Kimura A, Umehara T & Horikoshi M Chromosomal gradient of histone acetylation established by Sas2p and Sir2p functions as a shield against gene silencing. Nat. Genet. 32, 370–377 (2002). [DOI] [PubMed] [Google Scholar]
24.Gruning NM et al. Pyruvate kinase triggers a metabolic feedback loop that controls redox metabolism in respiring cells. Cell Metab. 14, 415–427 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Fleck CB & Brock M Re-characterisation of Saccharomyces cerevisiae Ach1p: fungal CoA-transferases are involved in acetic acid detoxification. Fungal Genet. Biol. 46, 473–485 (2009). [DOI] [PubMed] [Google Scholar]
26.van den Berg MA et al. The two acetyl-coenzyme A synthetases of Saccharomyces cerevisiae differ with respect to kinetic properties and transcriptional regulation. J. Biol. Chem. 271, 28953–28959 (1996). [DOI] [PubMed] [Google Scholar]
27.Shia WJ, Li B & Workman JL SAS-mediated acetylation of histone H4 Lys 16 is required for H2A.Z incorporation at subtelomeric regions in Saccharomyces cerevisiae. Genes Dev. 20, 2507–2512 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Vassylyeva MN et al. Efficient, ultra-high-affinity chromatography in a one-step purification of complex proteins. Proc. Natl Acad. Sci. USA 114, E5138–E5147 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Sutton A et al. Sas4 and Sas5 are required for the histone acetyltransferase activity of Sas2 in the SAS complex. J. Biol. Chem. 278, 16887–16892 (2003). [DOI] [PubMed] [Google Scholar]
30.McBrian MA et al. Histone acetylation regulates intracellular pH. Mol. Cell 49, 310–321 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Gao X et al. Acetate functions as an epigenetic metabolite to promote lipid synthesis under hypoxia. Nat. Commun. 7, 11960 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Orlandi I, Coppola DP & Vai M Rewiring yeast acetate metabolism through MPC1 loss of function leads to mitochondrial damage and decreases chronological lifespan. Microb. Cell 1, 393–405 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Maestre O, Garcia-Martinez T, Peinado RA & Mauricio JC Effects of ADH2 overexpression in Saccharomyces bayanus during alcoholic fermentation. Appl. Environ. Microbiol. 74, 702–707 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Simpson-Lavy K & Kupiec M Carbon catabolite repression: not only for glucose. Curr. Genet. 65, 1321–1323 (2019). [DOI] [PubMed] [Google Scholar]
35.Saint-Prix F, Bonquist L & Dequin S Functional analysis of the ALD gene family of Saccharomyces cerevisiae during anaerobic growth on glucose: the NADP⁺-dependent Ald6p and Ald5p isoforms play a major role in acetate formation. Microbiology 150, 2209–2220 (2004). [DOI] [PubMed] [Google Scholar]
36.Ladurner AG, Inouye C, Jain R & Tjian R Bromodomains mediate an acetyl-histone encoded antisilencing function at heterochromatin boundaries. Mol. Cell 11, 365–376 (2003). [DOI] [PubMed] [Google Scholar]
37.Matangkasombut O & Buratowski S Different sensitivities of bromodomain factors 1 and 2 to histone H4 acetylation. Mol. Cell 11, 353–363 (2003). [DOI] [PubMed] [Google Scholar]
38.Casatta N, Porro A, Orlandi I, Brambilla L & Vai M Lack of Sir2 increases acetate consumption and decreases extracellular pro-aging factors. Biochim. Biophys. Acta 1833, 593–601 (2013). [DOI] [PubMed] [Google Scholar]
39.Eisenberg T et al. Nucleocytosolic depletion of the energy metabolite acetyl-coenzyme a stimulates autophagy and prolongs lifespan. Cell Metab. 19, 431–444 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Cummings JH, Pomare EW, Branch WJ, Naylor CP & Macfarlane GT Short-chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 28, 1221–1227 (1987). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Zhao M, Chen L & Qu H CSGene: a literature-based database for cell senescence genes and its application to identify critical cell aging pathways and associated diseases. Cell Death Dis. 7, e2053 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Robin JD et al. Telomere position effect: regulation of gene expression with progressive telomere shortening over long distances. Genes Dev. 28, 2464–2476 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Wang J & Chen J SIRT1 regulates autoacetylation and histone acetyltransferase activity of TIP60. J. Biol. Chem. 285, 11458–11464 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Liao K et al. A feedback circuitry between polycomb signaling and fructose-1, 6-bisphosphatase enables hepatic and renal tumorigenesis. Cancer Res. 80, 675–688 (2020). [DOI] [PubMed] [Google Scholar]
45.Jung PP et al. Natural variation of chronological aging in the Saccharomyces cerevisiae species reveals diet-dependent mechanisms of lifespan control. NPJ Aging Mech. Dis. 4, 3 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Fabrizio P et al. Sir2 blocks extreme lifespan extension. Cell 123, 655–667 (2005). [DOI] [PubMed] [Google Scholar]
47.Vall-Llaura N, Mir N, Garrido L, Vived C & Cabiscol E Redox control of yeast Sir2 activity is involved in acetic acid resistance and longevity. Redox Biol. 24, 101229 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Reverter-Branchat G et al. Chronological and replicative life-span extension in Saccharomyces cerevisiae by increased dosage of alcohol dehydrogenase 1. Microbiology 153, 3667–3676 (2007). [DOI] [PubMed] [Google Scholar]
49.Longtine MS et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953–961 (1998). [DOI] [PubMed] [Google Scholar]
50.Mei Q et al. Set1-catalyzed H3K4 trimethylation antagonizes the HIR/Asf1/Rtt106 repressor complex to promote histone gene expression and chronological lifespan. Nucleic Acids Res. 47, 3434–3449 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Vasilishina A, Kropotov A, Spivak I & Bernadotte A Relative human telomere length quantification by real-time PCR. Methods Mol. Biol. 1896, 39–44 (2019). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental information

NIHMS2138227-supplement-Supplemental_information.pdf^{(2.3MB, pdf)}

Supplemental tables

NIHMS2138227-supplement-Supplemental_tables.xlsx^{(513.5KB, xlsx)}

Data Availability Statement

[R1] 1.Sivanand S, Viney I & Wellen KE Spatiotemporal control of acetyl-CoA metabolism in chromatin regulation. Trends Biochem. Sci. 43, 61–74 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Mews P et al. Acetyl-CoA synthetase regulates histone acetylation and hippocampal memory. Nature 546, 381–386 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] 3.Cai L, Sutter BM, Li B & Tu BP Acetyl-CoA induces cell growth and proliferation by promoting the acetylation of histones at growth genes. Mol. Cell 42, 426–437 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Marino G et al. Regulation of autophagy by cytosolic acetyl-coenzyme A. Mol. Cell 53, 710–725 (2014). [DOI] [PubMed] [Google Scholar]

[R5] 5.Yu X, Ma R, Wu Y, Zhai Y & Li S Reciprocal regulation of metabolic reprogramming and epigenetic modifications in cancer. Front. Genet. 9, 394 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Yu X & Li S Non-metabolic functions of glycolytic enzymes in tumorigenesis. Oncogene 36, 2629–2636 (2017). [DOI] [PubMed] [Google Scholar]

[R7] 7.Li X et al. Nucleus-translocated ACSS2 promotes gene transcription for lysosomal biogenesis and autophagy. Mol. Cell 66, 684–697 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Wu Y et al. Glycolysis regulates gene expression by promoting the cross-talk between H3K4me3 and H3K14ac in Saccharomyces cerevisiae. J. Genet. Genomics 46, 561–574 (2019). [DOI] [PubMed] [Google Scholar]

[R9] 9.Schug ZT et al. Acetyl-CoA synthetase 2 promotes acetate utilization and maintains cancer cell growth under metabolic stress. Cancer Cell 27, 57–71 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Takahashi H, McCaffery JM, Irizarry RA & Boeke JD Nucleocytosolic acetyl-coenzyme a synthetase is required for histone acetylation and global transcription. Mol. Cell 23, 207–217 (2006). [DOI] [PubMed] [Google Scholar]

[R11] 11.Klein HP & Jahnke L Effects of aeration on formation and localization of the acetyl coenzyme A synthetases of Saccharomyces cerevisiae. J. Bacteriol. 137, 179–184 (1979). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] 12.Falcon AA, Chen S, Wood MS & Aris JP Acetyl-coenzyme A synthetase 2 is a nuclear protein required for replicative longevity in Saccharomyces cerevisiae. Mol. Cell. Biochem. 333, 99–108 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Li S et al. Serine and SAM responsive complex SESAME regulates histone modification cross-talk by sensing cellular metabolism. Mol. Cell 60, 408–421 (2015). [DOI] [PubMed] [Google Scholar]

[R14] 14.Jaworski DM, Namboodiri AM & Moffett JR Acetate as a metabolic and epigenetic modifier of cancer therapy. J. Cell. Biochem. 117, 574–588 (2016). [DOI] [PubMed] [Google Scholar]

[R15] 15.Folmes CD et al. Somatic oxidative bioenergetics transitions into pluripotency dependent glycolysis to facilitate nuclear reprogramming. Cell Metab. 14, 264–271 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Moussaieff A et al. Glycolysis-mediated changes in acetyl-CoA and histone acetylation control the early differentiation of embryonic stem cells. Cell Metab. 21, 392–402 (2015). [DOI] [PubMed] [Google Scholar]

[R17] 17.Murakami CJ, Wall V, Basisty N & Kaeberlein M Composition and acidification of the culture medium influences chronological aging similarly in vineyard and laboratory yeast. PLoS ONE 6, e24530 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Burtner CR, Murakami CJ, Kennedy BK & Kaeberlein M A molecular mechanism of chronological aging in yeast. Cell Cycle 8, 1256–1270 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Fabrizio P et al. Superoxide is a mediator of an altruistic aging program in Saccharomyces cerevisiae. J. Cell Biol. 166, 1055–1067 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Longo VD, Shadel GS, Kaeberlein M & Kennedy B Replicative and chronological aging in Saccharomyces cerevisiae. Cell Metab. 16, 18–31 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Dang W et al. Histone H4 lysine 16 acetylation regulates cellular lifespan. Nature 459, 802–807 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Millar CB, Kurdistani SK & Grunstein M Acetylation of yeast histone H4 lysine 16: a switch for protein interactions in heterochromatin and euchromatin. Cold Spring Harb. Symp. Quant. Biol. 69, 193–200 (2004). [DOI] [PubMed] [Google Scholar]

[R23] 23.Kimura A, Umehara T & Horikoshi M Chromosomal gradient of histone acetylation established by Sas2p and Sir2p functions as a shield against gene silencing. Nat. Genet. 32, 370–377 (2002). [DOI] [PubMed] [Google Scholar]

[R24] 24.Gruning NM et al. Pyruvate kinase triggers a metabolic feedback loop that controls redox metabolism in respiring cells. Cell Metab. 14, 415–427 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Fleck CB & Brock M Re-characterisation of Saccharomyces cerevisiae Ach1p: fungal CoA-transferases are involved in acetic acid detoxification. Fungal Genet. Biol. 46, 473–485 (2009). [DOI] [PubMed] [Google Scholar]

[R26] 26.van den Berg MA et al. The two acetyl-coenzyme A synthetases of Saccharomyces cerevisiae differ with respect to kinetic properties and transcriptional regulation. J. Biol. Chem. 271, 28953–28959 (1996). [DOI] [PubMed] [Google Scholar]

[R27] 27.Shia WJ, Li B & Workman JL SAS-mediated acetylation of histone H4 Lys 16 is required for H2A.Z incorporation at subtelomeric regions in Saccharomyces cerevisiae. Genes Dev. 20, 2507–2512 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Vassylyeva MN et al. Efficient, ultra-high-affinity chromatography in a one-step purification of complex proteins. Proc. Natl Acad. Sci. USA 114, E5138–E5147 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] 29.Sutton A et al. Sas4 and Sas5 are required for the histone acetyltransferase activity of Sas2 in the SAS complex. J. Biol. Chem. 278, 16887–16892 (2003). [DOI] [PubMed] [Google Scholar]

[R30] 30.McBrian MA et al. Histone acetylation regulates intracellular pH. Mol. Cell 49, 310–321 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Gao X et al. Acetate functions as an epigenetic metabolite to promote lipid synthesis under hypoxia. Nat. Commun. 7, 11960 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Orlandi I, Coppola DP & Vai M Rewiring yeast acetate metabolism through MPC1 loss of function leads to mitochondrial damage and decreases chronological lifespan. Microb. Cell 1, 393–405 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Maestre O, Garcia-Martinez T, Peinado RA & Mauricio JC Effects of ADH2 overexpression in Saccharomyces bayanus during alcoholic fermentation. Appl. Environ. Microbiol. 74, 702–707 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Simpson-Lavy K & Kupiec M Carbon catabolite repression: not only for glucose. Curr. Genet. 65, 1321–1323 (2019). [DOI] [PubMed] [Google Scholar]

[R35] 35.Saint-Prix F, Bonquist L & Dequin S Functional analysis of the ALD gene family of Saccharomyces cerevisiae during anaerobic growth on glucose: the NADP⁺-dependent Ald6p and Ald5p isoforms play a major role in acetate formation. Microbiology 150, 2209–2220 (2004). [DOI] [PubMed] [Google Scholar]

[R36] 36.Ladurner AG, Inouye C, Jain R & Tjian R Bromodomains mediate an acetyl-histone encoded antisilencing function at heterochromatin boundaries. Mol. Cell 11, 365–376 (2003). [DOI] [PubMed] [Google Scholar]

[R37] 37.Matangkasombut O & Buratowski S Different sensitivities of bromodomain factors 1 and 2 to histone H4 acetylation. Mol. Cell 11, 353–363 (2003). [DOI] [PubMed] [Google Scholar]

[R38] 38.Casatta N, Porro A, Orlandi I, Brambilla L & Vai M Lack of Sir2 increases acetate consumption and decreases extracellular pro-aging factors. Biochim. Biophys. Acta 1833, 593–601 (2013). [DOI] [PubMed] [Google Scholar]

[R39] 39.Eisenberg T et al. Nucleocytosolic depletion of the energy metabolite acetyl-coenzyme a stimulates autophagy and prolongs lifespan. Cell Metab. 19, 431–444 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Cummings JH, Pomare EW, Branch WJ, Naylor CP & Macfarlane GT Short-chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 28, 1221–1227 (1987). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Zhao M, Chen L & Qu H CSGene: a literature-based database for cell senescence genes and its application to identify critical cell aging pathways and associated diseases. Cell Death Dis. 7, e2053 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Robin JD et al. Telomere position effect: regulation of gene expression with progressive telomere shortening over long distances. Genes Dev. 28, 2464–2476 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Wang J & Chen J SIRT1 regulates autoacetylation and histone acetyltransferase activity of TIP60. J. Biol. Chem. 285, 11458–11464 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Liao K et al. A feedback circuitry between polycomb signaling and fructose-1, 6-bisphosphatase enables hepatic and renal tumorigenesis. Cancer Res. 80, 675–688 (2020). [DOI] [PubMed] [Google Scholar]

[R45] 45.Jung PP et al. Natural variation of chronological aging in the Saccharomyces cerevisiae species reveals diet-dependent mechanisms of lifespan control. NPJ Aging Mech. Dis. 4, 3 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] 46.Fabrizio P et al. Sir2 blocks extreme lifespan extension. Cell 123, 655–667 (2005). [DOI] [PubMed] [Google Scholar]

[R47] 47.Vall-Llaura N, Mir N, Garrido L, Vived C & Cabiscol E Redox control of yeast Sir2 activity is involved in acetic acid resistance and longevity. Redox Biol. 24, 101229 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Reverter-Branchat G et al. Chronological and replicative life-span extension in Saccharomyces cerevisiae by increased dosage of alcohol dehydrogenase 1. Microbiology 153, 3667–3676 (2007). [DOI] [PubMed] [Google Scholar]

[R49] 49.Longtine MS et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953–961 (1998). [DOI] [PubMed] [Google Scholar]

[R50] 50.Mei Q et al. Set1-catalyzed H3K4 trimethylation antagonizes the HIR/Asf1/Rtt106 repressor complex to promote histone gene expression and chronological lifespan. Nucleic Acids Res. 47, 3434–3449 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] 51.Vasilishina A, Kropotov A, Spivak I & Bernadotte A Relative human telomere length quantification by real-time PCR. Methods Mol. Biol. 1896, 39–44 (2019). [DOI] [PubMed] [Google Scholar]

PERMALINK

The SESAME complex regulates cell senescence through the generation of acetyl-CoA

Wanping Chen

Xilan Yu

Yinsheng Wu

Jie Tang

Qi Yu

Xiaodong Lv

Zitong Zha

Bicheng Hu

Xin Li

Jianguo Chen

Lixin Ma

Jerry L Workman

Shanshan Li

Abstract

Results

The SESAME complex is required to maintain global H4K16ac levels.

Fig. 1 |. SESAME complex is required to maintain global H4K16ac and H4K16ac localization at subtelomeric regions.

The SESAME complex is required to maintain normal H4K16ac levels at subtelomeric regions.

The SESAME complex interacts with SAS complex to maintain H4K16ac at subtelomeric regions.

Fig. 2 |. The SESAME complex interacts with the SAS complex to promote H4K16ac.

The SESAME complex and SAS complex co-regulate gene transcription and telomere silencing.

Fig. 3 |. The SESAME complex and Sas2 co-regulate gene transcription and telomere silencing.

Acetate induces H4K16ac in budding yeast.

Fig. 4 |. The SESAME complex and SAS complex are required for acetate to induce H4K16ac.

Acetate enhances the interaction between SESAME and SAS to induce H4K16ac.

The SESAME complex and SAS complex are required for acetate to promote the binding of Bdf1 at subtelomeric regions.

Fig. 5 |. Acetate enhances the binding of Bdf1 at subtelomeric regions.

The SESAME and SAS complexes are required for acetate to antagonize Sir2-mediated telomere silencing.

Fig. 6 |. The SESAME and SAS complexes are required for acetate to antagonize Sir2-mediated telomere silencing.

The SESAME and SAS complexes are required for acetate to regulate lifespan.

Fig. 7 |. Acetate accelerates chronological ageing in yeast and induce senescence in HuVEC cells.

ACSS2 and hMOF are required for acetate to induce H4K16ac and senescence in human endothelial cells.

Fig. 8 |. Acetate accelerates cell senescence in HuVECs dependent on ACSS2 and hMoF.

ACSS2 and hMOF are required for acetate to regulate senescence-associated gene expression and telomere silencing.

Discussion

Methods

Yeast strains and medium.

Cell cultures.

Spotting assay.

Immunoblot analysis.

Antibodies.

Quantitative PCR with reverse transcription.

RNA sequencing.

Telomere-proximal gene analysis and statistics.

Telomere length quantification.

Gene-set enrichment analysis.

Chromatin immunoprecipitation assay.

Chronological lifespan assay.

Recombinant protein purification.

Co-immunoprecipitation assay.

Acetyl-CoA measurement.

RNA interference.

Senescence-associated β-galactosidase staining.

Mouse experiments.

Software.

Statistics and reproducibility.

Extended Data

Extended Data Fig. 1 |. The SESAME complex is required to maintain the normal levels of H4K16ac.

Extended Data Fig. 2 |. The SESAME complex interacts with the SAS complex to promote H4K16ac.

Extended Data Fig. 3 |. The SESAME complex and SAS complex co-regulate gene expression and telomere silencing.

Extended Data Fig. 4 |. Acetate induces the global levels of H4K16ac as well as H4K16ac at subtelomere regions.

Extended Data Fig. 5 |. Acs2 and Sas2 are required for acetate to induce H4K16ac.

Extended Data Fig. 6 |. Acetate regulates telomere silencing in budding yeast.

Extended Data Fig. 7 |. Acetate accelerates chronological aging in budding yeast.

Extended Data Fig. 8 |. Acetate accelerates cell senescence in HuVEC.

Extended Data Fig. 9 |. Acetate accelerates cell senescence dependent on ACSS2 and hMoF.

Extended Data Fig. 10 |. A working model for how acetate regulates telomere silencing and aging in budding yeast.

Supplementary Material

Acknowledgements

Footnotes

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES