Sirtuin 1/sirtuin 3 are robust lysine delactylases and sirtuin 1-mediated delactylation regulates glycolysis

Abstract

Lysine lactylation (Kla), an epigenetic mark triggered by lactate during glycolysis, including the Warburg effect, bridges metabolism and gene regulation. Enzymes such as p300 and HDAC1/3 have been pivotal in deciphering the regulatory dynamics of Kla, though questions about additional regulatory enzymes, their specific Kla substrates, and the underlying functional mechanisms persist. Here, we identify SIRT1 and SIRT3 as key "erasers" of Kla, shedding light on their selective regulation of both histone and non-histone proteins. Proteomic analysis in SIRT1/SIRT3 knockout HepG2 cells reveals distinct substrate specificities toward Kla, highlighting their unique roles in cellular signaling. Notably, we highlight the role of specific Kla modifications, such as those on the M2 splice isoform of pyruvate kinase (PKM2), in modulating metabolic pathways and cell proliferation, thereby expanding Kla's recognized functions beyond epigenetics. Therefore, this study deepens our understanding of Kla's functional mechanisms and broadens its biological significance.

Keywords: Biological sciences; Molecular biology; Molecular interaction.

Graphical abstract

Figure 1

In vitro and in vivo screening of Kla deacylases (A) In vitro screening of sirtuins’ delactylase activities. (B) Western blot analysis of Kla levels upon SIRT1-3 overexpression in HEK293T cells. (C) Western blot analysis demonstrating increased Kla levels on both histones and non-histones upon knockout of SIRT1 and SIRT3 in HepG2 cells.

Figure 2

Identification of the global lysine lactylome and acetylome in HepG2 cells by HPLC-MS/MS (A) Schematic overview of the experimental workflow for the identification of Kla and Kac in WT, SIRT1 KO, and SIRT3 KO HepG2 cells. (B) Volcano plots representing the differentially expressed Kla and Kac sites between SIRT1 KO or SIRT3 KO versus WT HepG2 cells. (C) Distribution of the number of SIRT1/SIRT3-targeted Kla and Kac sites identified per protein. (D) The bar graph shows the fold change of lactylated and acetylated peptide intensities on the proteins highly regulated by SIRT1/SIRT3.

Figure 3

Differential analysis of the global lysine lactylome and acetylome in SIRT1 KO and SIRT3 KO HepG2 cells (A) PCA analysis of SIRT1/SIRT3-targeted Kla and Kac sites in WT, SIRT1 KO, and SIRT3 KO HepG2 cells. (B) Distribution and comparison of the number of SIRT1/SIRT3-targeted Kla and Kac sites. The horizontal coordinates represent the total number of Kla and Kac sites that are specifically regulated by SIRT1/SIRT3, while the vertical coordinates represent the number of intersections between SIRT1/SIRT3-regulated Kla and Kac sites. Groups with intersections are connected by dots. (C) Heatmap showing the degree of significant difference and GO enrichment pathway analysis of SIRT1 and SIRT3-targeted Kla and Kac sites. The hypergeometric distribution was applied to assess the enrichment of genes in specific pathways. ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. (D) Sequence motif logo showing a representative sequence for all Kla sites and Kac sites regulated by SIRT1 or SIRT3.

Figure 4

Function analysis of SIRT1 and SIRT3 targeted Kla sites associated with metabolism (A) Heatmap showing the fold change of SIRT1 and SIRT3 targeted Kla sites associated with metabolism. (B) KEGG pathway analysis depicting pathways enriched with SIRT1 and SIRT3 targeted Kla sites associated with metabolism. (C) Interaction network of the SIRT1 and SIRT3-regulated Kla proteome based on the STRING database (v12.0). The network is visualized in Cytoscape (v3.8.2) using the cytoHubb plugin. (D) Full MS and MS/MS spectra of PKM2 K207la peptide for its identification. (E) Three-dimensional structure of PKM2 (PDB entry 4fxf) highlighting the important residues in the substrate binding pocked.

Figure 5

ENO1-K228la is a target of SIRT1 and SIRT3 regulation (A) Increased Kla levels of ENO1 upon treatment with 10 mM sodium lactate for 24h. HEK293T cells ectopically expressing Flag-vector or Flag-tagged ENO1-WT were stimulated with 10 mM sodium lactate for 24h. Cell lysates underwent immunoprecipitation (IP) with Flag-M2 beads. Inputs and eluates were analyzed by immunoblots. (B) Lactylation predominantly occurs at K228 of ENO1. HEK293T cells were ectopically expressed with Flag-vector, Flag-tagged ENO1-WT, Flag-tagged ENO1-K193R, and Flag-tagged ENO1-K228R. Cell lysates were subjected to immunoprecipitation (IP) with Flag-M2 beads. Inputs and eluates were analyzed by immunoblots. (C and D) Interaction of ENO1 with SIRT1 (C) and SIRT3 (D). HEK293T cells were transfected with indicated plasmids, and the association of SIRT1 or SIRT3 with ENO1 was examined by Flag IP and western blot. (E and F) Reduction of ENO1 Kla by SIRT1 (E) and SIRT3 (F). HEK293T cells were transfected with indicated plasmids, and ENO1 Kla was determined by Flag IP and western blot. (G and H) SIRT1 and SIRT3 can reduce ENO1 K228la. In vitro delactylation reactions were performed with SIRT1 or SIRT3 as enzymes and synthetic Kla peptides containing K228la of ENO1 (EGLELLK(la)TAIGK) as substrates.

Figure 6

PKM2-K207la is directly regulated by SIRT1 (A) Kla levels of PKM2 increase upon treatment with 10 mM sodium lactate for 24h. HepG2 cells overexpressing Flag-vector or Flag-tagged PKM2-WT were treated with 20 mM sodium lactate for 24h. Cell lysates were subjected to immunoprecipitation (IP) with Flag-M2 beads. Inputs and eluates were analyzed by immunoblots. (B) Lactylation, not acetylation, primarily occurs at K207 of PKM2. HepG2 cells were ectopically expressed Flag-vector, Flag-tagged PKM2-WT and Flag-tagged PKM2-K207R. Cell lysates were subjected to immunoprecipitation (IP) with Flag-M2 beads. Inputs and eluates were analyzed by immunoblots. (C) PKM2 interacts with SIRT1. HepG2 cells were transfected with indicated plasmids, and the association between SIRT1 and PKM2 was examined by Flag IP and Western blot. (D and E) SIRT1 reduces the levels of Kla and Kac of PKM2-WT (D), while SIRT1 has minimal effect on the Kla of PKM2 K207R (E). HepG2 cells were transfected with indicated plasmids, and PKM2 lactylation was determined by Flag IP and Western blot. (F and G) SIRT1 reduces the PKM2 lactylation at K207. HepG2 cells (F) and HEK293T cells (G) were transfected with K207-specifically lactylated PKM2 using genetic code expansion technology, along with SIRT1. PKM2 lactylation was determined by Flag IP and Western blot. (H) SIRT3 does not reduce the lactylation of PKM2 at K207. HEK293T cells were transfected with K207-specifically lactylated PKM2 using genetic code expansion technology, along with SIRT3, followed by Flag IP and Western blot analysis.

Figure 7

Lactylation status of K207 dictates PKM2 activity, inhibits glycolysis, and affects cell growth (A) Immunopurified PKM2 WT, K207R, and K207la from HEK293T cells were crosslinked with 0.5mM DSS, and the tetrameric (240 KDa) and monomeric (60 KDa) forms of PKM2 were analyzed by Western blot. (B) Pyruvate kinase activity assay of immunopurified PKM2 WT, K207R, and K207la from HEK293T cells. Data are represented as mean ± SEM. (C and D) HEK293T cells with endogenous PKM2 knocked down were transfected with PKM2-WT, K207R, and K207la, and the glycolysis activity was analyzed by the Seahorse analyzer (n = 5 technical repeats, values are expressed as mean ± SEM). p values are calculated by a two-tailed Student’s t test. ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. (E and F) Representative confocal projections of GFP-tagged PKM2 WT, K207R, and K207la in HEK293T cells (E) and HepG2 cells (F), scale bar = 20 μm or 10 μm. (G) Cell proliferation analysis of PKM2 knockdown HEK293T cells expressing PKM2-WT, K207R, and K207la by Cell Counting Kit-8 (CCK-8) assay (n = 10 technical repeats, values are expressed as mean ± SEM). p values are calculated by a two-tailed Student’s t test. ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001.