TRAP1 drives smooth muscle cell senescence and promotes atherosclerosis via HDAC3-primed histone H4 lysine 12 lactylation

Abstract

Background and aims: Vascular smooth muscle cell (VSMC) senescence is crucial for the development of atherosclerosis, characterized by metabolic abnormalities. Tumour necrosis factor receptor-associated protein 1 (TRAP1), a metabolic regulator associated with ageing, might be implicated in atherosclerosis. As the role of TRAP1 in atherosclerosis remains elusive, this study aimed to examine the function of TRAP1 in VSMC senescence and atherosclerosis.

Methods: TRAP1 expression was measured in the aortic tissues of patients and mice with atherosclerosis using western blot and RT-qPCR. Senescent VSMC models were established by oncogenic Ras, and cellular senescence was evaluated by measuring senescence-associated β-galactosidase expression and other senescence markers. Chromatin immunoprecipitation (ChIP) analysis was performed to explore the potential role of TRAP1 in atherosclerosis.

Results: VSMC-specific TRAP1 deficiency mitigated VSMC senescence and atherosclerosis via metabolic reprogramming. Mechanistically, TRAP1 significantly increased aerobic glycolysis, leading to elevated lactate production. Accumulated lactate promoted histone H4 lysine 12 lactylation (H4K12la) by down-regulating the unique histone lysine delactylase HDAC3. H4K12la was enriched in the senescence-associated secretory phenotype (SASP) promoter, activating SASP transcription and exacerbating VSMC senescence. In VSMC-specific Trap1 knockout ApoeKO mice (ApoeKOTrap1SMCKO), the plaque area, senescence markers, H4K12la, and SASP were reduced. Additionally, pharmacological inhibition and proteolysis-targeting chimera (PROTAC)-mediated TRAP1 degradation effectively attenuated atherosclerosis in vivo.

Conclusions: This study reveals a novel mechanism by which mitonuclear communication orchestrates gene expression in VSMC senescence and atherosclerosis. TRAP1-mediated metabolic reprogramming increases lactate-dependent H4K12la via HDAC3, promoting SASP expression and offering a new therapeutic direction for VSMC senescence and atherosclerosis.

Keywords: Atherosclerosis; HDAC3; Histone lactylation; Senescence; Smooth muscle cells; TRAP1.

Structured Graphical Abstract

The proto-oncogene Ras induces aberrant TRAP1 expression in VSMCs, leading to metabolic reprogramming by increasing the levels of the glycolytic rate-limiting enzyme PFK1. Lactate accumulation increases H4K12la expression by blocking the unique histone delactylase, HDAC3. In this case, H4K12la is enriched at the SASP promoter region and promotes SASP transcription mediating VSMC senescence and promoting atherosclerosis. CoA, coenzyme A; ETC, electron transport chain; H4K12la, histone H4 lysine 12 lactylation; PFK1, phosphofructokinase 1; SASP, senescence-associated secretory phenotype; TCA, tricarboxylic acid; TRAP1, tumour necrosis factor receptor-associated protein 1; VSMC, vascular smooth muscle cell.

Figure 1

TRAP1 regulates VSMC senescence. (A) The volcano graph shows the distribution of differentially expressed genes. Red dots represent up-regulated genes, and blue dots represent down-regulated genes. (B) Western blot analysis of TRAP1 and senescence markers in the aortic tissues of the non-atherosclerotic (Non-AS) and atherosclerotic (AS) groups (n = 5 independent biological replicates). (C) Western blot analysis of TRAP1 and senescence markers in the aortic tissues from NC- and HFD-fed Apoe^KO mice (n = 6 independent biological replicates). (D) Western blot analysis of TRAP1 and senescence markers in Ras-induced hVSMCs (n = 5 independent biological replicates). (E) Representative immunofluorescence staining images of TRAP1 (red) and 4',6-diamidino-2-phenylindole (DAPI; blue) expression in Ras-induced hVSMCs (scale bar = 20 μm, n = 5 independent biological replicates). (F) Western blot analysis of senescence markers in Ras-induced hVSMCs transfected with siTRAP1 (n = 5 independent biological replicates). (G) Representative immunofluorescence staining images of senescence markers (red) and DAPI (blue) in Ras-induced hVSMCs transfected with siTRAP1 (scale bar = 20 μm, n = 5 independent biological replicates). (H) SA-β-Gal staining of Ras-induced hVSMCs transfected with siTRAP1 (scale bar = 50 μm, n = 5 independent biological replicates). (I) Super-resolution fluorescence imaging of F-actin in Ras-induced hVSMCs transfected with siTRAP1 (scale bar = 20 μm, n = 5 independent biological replicates). (J) Relative telomere length in Ras-induced hVSMCs transfected with siTRAP1 (n = 6 independent biological replicates). (K) Proliferation measured using the bromodeoxyuridine (BrdU) assay in Ras-induced hVSMCs transfected with siTRAP1 (scale bar = 20 μm, n = 5 independent biological replicates). (L) Heatmap showing the relative expression of SASP in Ras-induced hVSMCs transfected with siTRAP1 (n = 6 independent biological replicates). (M) Western blot analysis of senescence markers in hVSMCs overexpressing TRAP1 (n = 5 independent biological replicates). (N ) SA-β-Gal staining of hVSMCs overexpressing TRAP1 (scale bar = 50μm, n = 5 independent biological replicates). (O) Relative telomere length in hVSMCs overexpressing TRAP1 (n = 6 independent biological replicates). (P) Proliferation measured using the BrdU assay in hVSMCs overexpressing TRAP1 (scale bar = 20 μm, n = 5 independent biological replicates). (Q) Heatmap showing the relative expression of SASP in hVSMCs overexpressing TRAP1 (n = 6 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (B–D). Welch’s correction was used for comparison in (C) and (O). One-way analysis of variance (ANOVA) was performed for comparison (F and G, and I and J)

Figure 2

TRAP1 is a key regulator of energy reprogramming in senescent VSMCs. (A) Assessment of mitochondrial quality based on Mito Tracker Red staining of Ras-induced hVSMCs transfected with siTRAP1 (scale bar = 20 μm, n = 5 independent biological replicates). (B) OCR of Ras-induced hVSMCs transfected with siTRAP1 was monitored in real-time using the seahorse system. OCR was used as a marker of mitochondrial respiration (left panel), and the basal respiration, maximal respiration, and ATP-linked OCR were assessed (right panel, n = 6 independent biological replicates). (C) ECAR of Ras-induced hVSMCs transfected with siTRAP1 was monitored in real-time using the seahorse system. ECAR was used primarily to measure glycolysis (left panel) and assess cellular glycolysis and glycolytic capacity (right panel, n = 6 independent biological replicates). (D) Detection of acetyl-CoA levels in Ras-induced hVSMCs following TRAP1 deficiency (n = 6 independent biological replicates). (E) Detection of citrate levels in Ras-induced hVSMCs following TRAP1 deficiency (n = 6 independent biological replicates). (F) Western blot analysis of PFK1, HK2, and PKM2 protein levels in Ras-induced hVSMCs following TRAP1 deficiency (n = 5 independent biological replicates). (G) Quantification of protein levels of the blots shown in (F) (n= 5 independent biological replicates). (H) Western blot analysis of PFK1, HK2, and PKM2 protein levels in Ras-induced hVSMCs treated with 1 μM G-TPP for 4 h (n = 5 independent biological replicates). (I) Quantification of protein levels of the blots shown in (H) (n = 5 independent biological replicates). (J) Co-immunoprecipitation analysis of PFK1–TRAP1 binding in Ras-induced hVSMCs (n = 5 independent biological replicates). (K) Using the seahorse system, the OCR of TRAP1-overexpressing hVSMCs was monitored in real-time following citrate (1 mM) treatment for 6 h (left panel), assessing basal respiration, maximal respiration, and ATP-linked OCR (right panel, n = 6 independent biological replicates). (L) ECAR of TRAP1-overexpressing hVSMCs was monitored in real-time using seahorse technology following citrate (1 mM) treatment for 6 h (left panel), assessing cellular glycolysis and glycolytic capacity (right panel, n = 6 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. One-way ANOVA was performed in (A–E, G, I, K, and L)

Figure 3

TRAP1-mediated lactate accumulation promotes VSMC senescence. (A) Detection of lactate levels in Ras-induced hVSMCs with TRAP1 deficiency (n = 6 independent biological replicates). (B) Western blot analysis of senescence markers levels in Ras-induced hVSMCs treated with siTRAP1, followed by treatment with or without exogenous lactate (5 mM) for 24 h (n = 5 independent biological replicates). (C) Proliferation was measured using BrdU assay in Ras-induced hVSMCs treated with siTRAP1, followed by treatment with or without exogenous lactate (scale bar = 20 μm, n = 5 independent biological replicates). (D) Representative SA-β-Gal staining images in Ras-induced hVSMCs treated with siTRAP1, followed by treatment with or without exogenous lactate (scale bar = 50 μm, n = 5 independent biological replicates). (E) RT–qPCR analysis of SASP expression in Ras-induced hVSMCs treated with siTRAP1, followed by treatment with or without exogenous lactate, presented as a heatmap (n = 6 independent biological replicates). (F–H) Senescence markers were detected in Ras-induced hVSMCs following 2-DG (F), siLDHA (G), and oxamate (H) treatment via western blot analysis (n = 5 independent biological replicates). (I) Representative SA-β-Gal staining images in Ras-induced hVSMCs treated with siLDHA (scale bar = 50 μm, n = 5 independent biological replicates). (J) Proliferation measured using the BrdU assay in Ras-induced hVSMCs treated with siLDHA (scale bar = 20 μm, n = 5 independent biological replicates). (K) RT–qPCR was used to detect SASP expression in Ras-induced senescent hVSMCs treated with 2-DG, oxamate, and siLDHA (n = 6 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. One-way ANOVA was performed in (A, B, D, F–I, and K)

Figure 4

TRAP1-mediated H4K12la promotes SASP activation in senescent VSMCs. (A) Western blot analysis of Pan Kla protein levels in Ras-induced hVSMCs following TRAP1 deficiency (n = 5 independent biological replicates). (B) Western blot analysis of Pan Kla, H4K12la, H4K5la, H4K8la, and H3K18la levels in Ras-induced hVSMCs following TRAP1 deficiency (n = 5 independent biological replicates). (C) Expression levels of H4K12la in the nucleus of Ras-induced hVSMCs following TRAP1 deficiency, assessed using super-resolution fluorescence imaging (upper). Volume information of H4K12la hotspots identified in the nucleus of Ras-induced hVSMCs following TRAP1 deficiency (bottom) (scale bar = 5 μm, n = 5 independent biological replicates). (D) Representative images of H4K12la co-stained with DAPI in Ras-induced hVSMCs treated with siTRAP1 or supplemented with exogenous lactate (5 mM) for 24 h (scale bar = 20 µm, n = 5 independent biological replicates). (E) Genome browser tracks of the ChIP-seq signals from GSE188765 at representative SASP loci. (F) The heatmap displayed the enrichment of H4K5la, H4K12la, H3K27ac, and H3K4me1 around the promoter region of SASP, ranging from −1 kb to +1 kb. The degree of binding of H3 to SASP was used as the positive control (n = 6 independent biological replicates). (G) Enrichment of H3K4me1, H4K12la, H3K27ac, and H4K5la at the interleukin-6 (IL-6) promoter analysed via ChIP-qPCR from the results of the heatmap in (F). Data are presented in the form of a column chart (n = 6 independent biological replicates). (H) Global nascent transcripts of SASP were measured using nuclear run-on experiments coupled with RT–qPCR in isolated nuclei from Ras-induced hVSMCs following siTRAP1 treatment (n = 6 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. One-way ANOVA was performed in (C, D, G, and H)

Figure 5

Up-regulation of H4K12la by blocking HDAC3 promotes senescence in VSMCs. (A) Western blot analysis of p300 and HDAC1–3 levels in Ras-induced hVSMCs (n = 5 independent biological replicates). (B) Western blot analysis of HDAC3 levels in hVSMCs treated with lactate (n = 5 independent biological replicates). (C) Western blot analysis of HDAC3 levels in Ras-induced hVSMCs. Ras-induced hVSMCs were treated with siTRAP1, followed by treatment with or without 1 mM oxamate (n = 5 independent biological replicates). (D) Western blot analysis of HDAC3 levels in Ras-induced hVSMCs following TRAP1 deficiency with or without lactate treatment (n = 5 independent biological replicates). (E) Western blot analysis of lactylation modification and senescence markers in HDAC3-overexpressing senescent hVSMCs (n = 5 independent biological replicates). (F) Expression levels of nuclear H4K12la in Ras-induced hVSMCs following HDAC3 overexpression, assessed using super-resolution fluorescence imaging (upper). Volume information of H4K12la hotspots identified in the nucleus of HDAC3-overexpressing senescent hVSMCs (bottom) (scale bar = 5 μm, n = 5 independent biological replicates). (G) Representative images of H4K12la (red) co-stained with DAPI (blue) in Ras-induced hVSMCs. Ras-induced hVSMCs were treated with HDAC3 overexpression or ITSA-1 (150 μM) for 24 h (scale bar = 20 μm, n = 5 independent biological replicates). (H) Representative SA-β-Gal staining images of Ras-induced hVSMCs with HDAC3 overexpression or ITSA-1 treatment, with quantification of SA-β-Gal-positive cells (right panel, scale bar = 50 μm, n = 5 independent biological replicates). (I) Proliferation measured using the BrdU assay in Ras-induced hVSMCs following HDAC3 overexpression or ITSA-1 treatment (scale bar = 20 μm, n = 5 independent biological replicates). (J) RT–qPCR analysis of SASP in Ras-induced hVSMCs after HDAC3 overexpression or ITSA-1 treatment, presented as a heatmap (n = 6 independent biological replicates). (K) Transcriptional regulation of SASP in Ras-induced hVSMCs following HDAC3 overexpression or ITSA-1 treatment was analysed using the run-on assay (n = 6 independent biological replicates). (L) ChIP detection of the binding sites of H4K12la, H3K27ac, and H3K4me1 at the SASP promoter region (n = 6 independent biological replicates). (M) Re-ChIP was performed in Ras-induced hVSMCs with H4K12la antibody in the first round and a pull-down with HDAC3 antibody in the second round (n = 6 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (M). One-way ANOVA was performed in (H, K, and L)

Figure 6

SMC-specific Trap1 knockout ameliorates atherosclerosis. (A) Representative images of aortas stained with Oil Red O from Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice fed with an NC or HFD (n = 6 independent biological replicates). Mice were fed with an NC or HFD for 16 weeks, from 8 weeks old. (B–D) Oil Red O (B), Masson and Sirius red (C), and haematoxylin and eosin (H&E) (D) staining of aortic roots (scale bar = 100 μm, n = 6 independent biological replicates). (E) Representative immunofluorescence staining of the SMC (α-SMA, green) and senescence (P21, red) in aortas from Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice fed with an NC or HFD (scale bar = 50 μm, n = 6 independent biological replicates). (F) Representative immunofluorescence staining of the α-SMA (green) and H4K12la (red) in aortas from Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice fed with an NC or HFD (scale bar = 50 μm, n = 6 independent biological replicates). (G) Western blot analysis of the expression levels of P53, P21, P16, H4K12la, H4K8la, and H4K5la in MOVAS isolated from Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice fed with an NC or HFD (n = 6 independent biological replicates). (H and I) Seahorse analysis of the OCR (H) and ECAR (I) of MOVAS isolated from HFD-fed Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice (n = 6 independent biological replicates). (J) Lactate levels in MOVAS isolated from HFD-fed Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice (n = 10 independent biological replicates). (K) RT–qPCR analysis of SASP expression levels in MOVAS isolated from Apoe^KOTrap1^SMCKO and Apoe^KOTrap1^WT mice fed with an NC or HFD (n = 10 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (H–J). One-way ANOVA was performed in (A–D, G, and K)

Figure 7

Pharmacological inhibition of TRAP1 suppresses atherosclerosis development in vivo. (A) The HFD-fed Apoe^KO mice were intraperitoneally administered with G-TPP or vehicle (Vehicle control) for 12 weeks. The plasma levels of LDL-cholesterol (LDL-C), HDL-cholesterol (HDL-C), total cholesterol (T-CHO), and triglyceride (TG) in mice were measured (n = 6 independent biological replicates). (B) En face aortas stained with Oil Red O from HFD-fed Apoe^KO mice with or without G-TPP treatment (n = 6 independent biological replicates). (C) The Oil Red O, H&E, Masson, and Sirius red staining of aortic roots from HFD-fed Apoe^KO mice with or without G-TPP treatment (scale bar = 100 μm, n = 6 independent biological replicates). (D) Representative immunostaining images of H4K12la (red) and P21 (red) in aortas from HFD-fed Apoe^KO mice with or without G-TPP treatment (scale bar = 50 μm, n = 6 independent biological replicates). (E) Protein levels of P53, P21, P16, H4K12la, H4K5la, and H4K8la in MOVAS from HFD-fed Apoe^KO mice with or without G-TPP treatment were measured by western blot (n = 6 independent biological replicates). (F) The expression levels of SASP in MOVAS from HFD-fed Apoe^KO mice with or without G-TPP treatment were examined using RT–qPCR (n = 10 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (A–C, E, and F)

Figure 8

Degradation of TRAP1 via PROTAC as a novel strategy for atherosclerosis treatment. (A) Western blot analysis of the expression of TRAP1 in hVSMCs. hVSMCs were treated with BP3 (6 h) at different concentrations (n = 5 independent biological replicates). (B) Western blot analysis of TRAP1 expression in hVSMCs. hVSMCs were treated with BP3 (0.5 μM) at different time points (n = 5 independent biological replicates). (C) Western blot analysis of TRAP1 expression in hVSMCs. hVSMCs were treated with BP3 (0.5 μM) for 6 h, followed by treatment with or without 20 μM MG132 for 6 h (n = 5 independent biological replicates). (D) Plasma levels of LDL-C, HDL-C, T-CHO, and TG in HFD-fed Apoe^KO mice. Mice were intraperitoneally administered with BP3 or vehicle (Vehicle control) for 12 weeks (n = 6 independent biological replicates). (E) En face aortas, stained with Oil Red O, from HFD-fed Apoe^KO mice with or without BP3 treatment (left panel). Quantification of the plaque area as a percentage of the total aortic area (right panel, n = 6 independent biological replicates). (F) The Oil Red O, H&E, Masson, and Sirius red staining of aortic roots from HFD-fed Apoe^KO mice with or without BP3 treatment (scale bar = 100 μm, n = 6 independent biological replicates). (G) Protein levels of P53, P21, P16, and H4K12la in MOVAS from HFD-fed Apoe^KO mice with or without BP3 treatment, measured using western blot analysis (n = 6 independent biological replicates). (H) The expression levels of SASP in MOVAS from HFD-fed Apoe^KO mice with or without BP3 treatment, examined via RT–qPCR (n = 10 independent biological replicates). *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (D–H). One-way ANOVA was performed in (A–C)

Figure 9

TRAP1 and H4K12la may be key factors involved in clinical SMC senescence and atherosclerosis. (A) Western blot analysis of H4K12la, H4K5la, and H4K8la in patients with atherosclerosis and healthy individuals (n = 5 independent biological replicates). (B) Linear regression analysis of H4K12la protein expression with P53, P21, and P16 protein expression in aortic tissues from patients with atherosclerosis. (C) Linear regression analysis of TRAP1 mRNA with senescence markers and SASP in aortic tissues from patients with atherosclerosis. *P < .05, **P < .01, ***P < .001. Data are presented as the mean ± SD. Unpaired t-test was used for comparison in (A). Correlation analysis was performed in (B and C)