METTL14 aggravates endothelial inflammation and atherosclerosis by increasing FOXO1 N6-methyladeosine modifications

Abstract

Aims: The N6-methyladenosine (m⁶A) modification plays an important role in various biological processes, but its role in atherosclerosis remains unknown. The aim of this study was to investigate the role and mechanism of m⁶A modification in endothelial cell inflammation and its influence on atherosclerosis development. Methods: We constructed a stable TNF-α-induced endothelial cell inflammation model and assessed the changes in the expression of m⁶A modification-related proteins to identify the major factors involved in this process. The m⁶A-modified mRNAs were identified by methylated RNA immunoprecipitation (RIP) sequencing and forkhead box O1 (FOXO1) was selected as a potential target. Through cytological experiments, we verified whether methyltransferase-like 14 (METTL14) regulates FOXO1 expression by regulating m⁶A-dependent mRNA and protein interaction. The effect of METTL14 on atherosclerosis development in vivo was verified using METTL14 knockout mice. Results: These findings confirmed that METTL14 plays major roles in TNF-α-induced endothelial cell inflammation. During endothelial inflammation, m⁶A modification of FOXO1, an important transcription factor, was remarkably increased. Moreover, METTL14 knockdown significantly decreased TNF-α-induced FOXO1 expression. RIP assay confirmed that METTL14 directly binds to FOXO1 mRNA, increases its m⁶A modification, and enhances its translation through subsequent YTH N6-methyladenosine RNA binding protein 1 recognition. Furthermore, METTL14 was shown to interact with FOXO1 and act directly on the promoter regions of VCAM-1 and ICAM-1 to promote their transcription, thus mediating endothelial cell inflammatory response. In vivo experiments showed that METTL14 gene knockout significantly reduced the development of atherosclerotic plaques. Conclusion: METTL14 promotes FOXO1 expression by enhancing its m⁶A modification and inducing endothelial cell inflammatory response as well as atherosclerotic plaque formation. Decreased expression of METTL14 can inhibit endothelial inflammation and atherosclerosis development. Therefore, METTL14 may serve as a potential target for the clinical treatment of atherosclerosis.

Keywords: FOXO1; METTL14; atherosclerosis; endothelial inflammation; m6A modification.

Figure 1

METTL14 is increased in TNF-α-induced endothelial cells and atherosclerotic lesions. (A) QRT-PCR detection of METTL3, METTL14, KIAA1429, WTAP, FTO and ALKBH5 in HUVECs stimulated with 10 ng/ml TNF-α for 6 h or 12 h. (B) Representative western blot results of METTL3, METTL14, KIAA1429, WTAP, FTO and ALKBH5 after TNF-α stimulation in HUVECs. (C and D) Endothelial mononuclear adhesion assays in HUVECs after knockdown of METTL14 (with or without TNF-α stimulation) and the statistics of the relative monocyte adhesion rate. (E) QRT-PCR detection of VCAM-1, ICAM-1 and E-selectin in HUVECs stimulated with 10 ng/ml TNF-α for 12 h with or without METTL14 knockdown. (F) Representative western blot results of VCAM-1 and ICAM-1 in HUVECs stimulated with 10 ng/ml TNF-α for 12 h with or without METTL14 knockdown. (G) QRT-PCR detection of VCAM-1 and ICAM-1 in HUVECs with or without METTL14 knockdown or METTL14 overexpression. (H) Representative western blot results of VCAM-1 and ICAM-1 in HUVECs with or without METTL14 knockdown or METTL14 overexpression. (I and J) METTL14 immunostaining in atherosclerosis samples from APOE^-/- mice and the statistics on the proportion of METTL14-positive cells. Non AS represents the normal vascular area near the atherosclerotic plaques. AS stands for the atherosclerotic plaque area. The white and yellow arrows indicate METTL14-positive endothelial cells in the AS plaque and Non AS area, respectively. (K) QRT-PCR detection of METTL14 mRNA in normal vascular tissues or atherosclerotic lesions collected from APOE knockout mice. (Data are presented as the mean ± SEM. A, D, E, and G, One-way ANOVA with Bonferroni's post hoc test was applied to compare the indicated groups. J and K, Two-tailed unpaired Student's t test was applied to compare the two groups. *P<0.05, **P<0.01, and ***P<0.001).

Figure 2

METTL14 mediates TNF-α-induced m⁶A/A ratio upregulation in endothelial cells. (A) The overall m⁶A/A ratio in polyadenylated RNA is detected using the EpiQuik^TM m6A RNA methylation quantification kit after HUVECs were stimulated with TNF-α (10 ng/ml, 12 h). (Data are presented as the mean ± SEM. Two-tailed unpaired Student's t test was applied to compare the two groups. **P<0.01, compared with the Normal tissue group). (B) Dot blot assay using the anti-m⁶A antibody in HUVECs stimulated with TNF-α (10 ng/ml, 12 h). (C) Overall m⁶A/A ratio in polyadenylated RNA is detected in HUVECs transfected with si-METTL14 after TNF-α stimulation (10 ng/ml, 12 h). (Data are presented as the mean ± SEM. One-way ANOVA with Bonferroni's post hoc test was applied to compare the indicated control groups. **P<0.01, ***P<0.001). (D) Dot blot assay using the anti-m⁶A antibody in HUVECs transfected with si-METTL14 after TNF-α stimulation (10 ng/ml, 12 h). (E and F) RIP analysis of the interaction of METTL14 or m⁶A with VCAM-1 and ICAM-1 mRNA with or without TNF-α (10 ng/ml, 6/12 h) stimulation. The enrichment of VCAM-1, ICAM-1 and NANOG mRNA with antibodies targeted against METTL14 or m⁶A was measured by qRT-PCR and normalized to the input.

Figure 3

FOXO1 mRNA is modified by METTL14 during TNF-α-induced endothelial-monocyte adhesion. (A) The number of overlapping bound genes between the PBS (control) and TNF-α-stimulated groups of HUVECs. (B) Distribution of the m⁶A peaks across the length of the mRNAs between the control and TNF-α-stimulated groups. (C) GO analysis with the upregulated transcripts that were covered by a unique peak. The cutoff parameters for enrichment analysis using Cytoscape software are as follows: p < 0.005, FDR: q < 0.1, and overlap cutoff: > 0.5. (D) Potential mRNAs that were significantly up- or down-regulated with m⁶A modification after TNF-α stimulation. (E) RIP analysis of the interaction of METTL14 or m⁶A with FOXO1 mRNA with or without TNF-α stimulation (10 ng/ml, 6/12 h). The enrichment of FOXO1 mRNA was measured by qRT-PCR with antibodies targeted against METTL14 or m⁶A and normalized to the input. (F) RIP analysis of the interaction of METTL14 or m⁶A with FOXO1 mRNA with or without si-METTL14 transfection and TNF-α stimulation (10 ng/ml, 6/12 h). The enrichment of FOXO1 mRNA m⁶A was measured by qRT-PCR with antibodies against METTL14 or and normalized to the input. (G) RT-PCR detection of FOXO1 mRNA expression with or without si-METTL14 transfection and TNF-α stimulation (10 ng/ml, 12 h). (H) Representative western blot results and its relative expression fold change of FOXO1 in HUVECs stimulated with 10 ng/ml TNF-α for 12 h with or without METTL14 knockdown. (I) Representative western blot results showing the relative fold change of FOXO1 expression in HUVECs transfected with METTL14 overexpression lentivirus (LV-M14-1, MOI=10 and LV-M14-2, MOI=20). (J) Representative western blot results showing the relative fold change of VCAM-1 and ICAM-1 expression in HUVECs stimulated with 10 ng/ml TNF-α for 12 h with or without FOXO1 knockdown. (K) Endothelial-monocyte adhesion assay in HUVECs after knockdown of FOXO1 (with or without TNF-α stimulation) and the relative monocyte adhesion rate. (L) Endothelial-monocyte adhesion assay in HUVECs after METTL14 knockdown or FOXO1 overexpression (with or without TNF-α stimulation) and the relative monocyte adhesion rate. (E-J) Data are presented as the mean ± SEM. One-way ANOVA with Bonferroni's post hoc test was applied to compare the indicated groups. (K and L) Two-tailed unpaired Student's t-test was applied to compare the indicated two groups. *P<0.05, **P<0.01, and ***P<0.001). (K and L) Three independent experiments were performed; magnification, ×100.

Figure 4

YTHDF1 enhances FOXO1 translation by recognizing the m⁶A modification site on the 3ʹ-UTR of FOXO1 mRNA. (A) The expression of FOXO1 mRNA was measured using qRT-PCR after si-METTL14 transfection of HUVECs with or without TNF-α (10 ng/ml, 12 h) stimulation and later treated with ActD (4 µM). (B) The stability of the FOXO1 protein was measured by western blotting after HUVECs were treated with or without TNF-α (10 ng/ml, 12 h). Cells were then treated with CHX (10 µM). (C) RIP analysis of the interaction of YTHDF1/2/3 with FOXO1 mRNA with or without TNF-α stimulation (10 ng/ml, 6/12 h). The enrichment of FOXO1 mRNA was measured by qRT-PCR with antibodies against YTHDF1, YTHDF2, or YTHDF3 and normalized to the input. (D) qRT-PCR analysis of FOXO1 mRNA in HUVECs after TNF-α stimulation (10 ng/ml, 12 h) with or without si-YTHDF1 transfection and fractioned into polysomes (fraction 3 and 14 were selected). (E) RIP analysis of the interaction of m⁶A with different suspected m⁶A regions of FOXO1 mRNA (CDS-1, CDS-2, CDS-3, and 3ʹ-UTR) with or without TNF-α stimulation (10 ng/ml, 6/12 h). The enrichment of the suspected m⁶A regions was measured by qRT-PCR with an antibody against m⁶a and normalized to the input. (F) Luciferase activity of HUVECs transfected with WT or mutant plasmids with or without TNF-α stimulation (10 ng/ml, 12 h) was measured using the dual luciferase reports system. (G) RIP analysis of the interaction of m⁶A with WT or mutant plasmids in HUVECs with or without TNF-α stimulation (10 ng/ml, 12 h). The enrichment of the suspected m⁶A regions was measured by qRT-PCR with an antibody against m⁶A and normalized to the input. (H) RIP analysis of the interaction of YTHDF1 with WT or mutant plasmids in HUVECs with or without TNF-α stimulation (10 ng/ml, 12 h). The enrichment of the suspected m⁶A regions was measured by qRT-PCR with an antibody against m⁶A and normalized to the input. (I) RNA pulldown assay of biotin-tagged WT or mutant plasmids cultured with HUVEC lysates (with or without TNF-α stimulation). Data are presented as the mean ± SEM. (A, C, D, and E) One-way ANOVA with Bonferroni's post hoc test was applied to compare the indicated groups. (F, G, and H) Two-tailed unpaired Student's t-test was applied to compare the indicated two groups. *P<0.05, **P<0.01, and ***P<0.001).

Figure 5

METTL14 cooperates with FOXO1 to promote VCAM-1 and ICAM-1 transcription. (A) HUVECs were lysed and precipitated using anti-METTL14 or anti-FOXO1 antibodies, followed by immunoblot analyses with indicated antibodies. (B) ChIP assay showed that the ability of METTL14 and FOXO1 proteins to bind the VCAM-1 and ICAM-1 promoters was significantly increased in TNF-α-stimulated HUVECs, while there was no change in their ability to bind the E-selectin promoter. (C and D) ChIP assay demonstrated the effect of METTL14 knockdown on the binding of FOXO1 to VCAM-1, ICAM-1, or E-selectin promoter regions, as well as the effect of FOXO1 knockdown on the binding of METTL14 to VCAM-1 and ICAM-1 promoter regions. (E and F) CHIP-re-CHIP results showed that the binding of the METTL14-FOXO1 complex to the promoter regions of VCAM-1 and ICAM-1 was significantly increased after TNF-α stimulation of HUVECs. (B-F) Data are presented as mean ± SEM. One-way ANOVA with Bonferroni's post-hoc test was applied to compare the indicated two groups. *P<0.05 and **P<0.01.

Figure 6

METTL14 knockout can significantly inhibit atherosclerosis development. (A) METTL14^+/-/APOE^-/- and control APOE^-/- mice were fed WD for 12 weeks. Representative images and quantification of the aortic root lesion area stained with oil red O are shown. (n=10-12 for each group). (B) Representative images and quantification of the aorta en face lesion stained with oil red O (n=10-12 for each group). (C) Necrotic core area (H&E) in the aortic root (n=10 per group). (D and E) Immunofluorescence staining showing the expression of FOXO1-positive cells in the atherosclerotic plaque and non-plaque regions of METTL14^+/-/APOE^-/- and APOE^-/- mice (n=10 per group). All representative images are from mice fed WD. (A, B, C, and E) Data are presented as mean ± SEM. (E) Two-tailed unpaired Student's t-test was applied to compare the indicated two groups. (A, B, and C) One-way ANOVA with Bonferroni's post-hoc test was applied to compare the indicated groups. (***P<0.001).

Figure 7

Proposed scheme of the mechanism of METTL14-mediated m⁶A modification affects endothelial inflammation and atherosclerosis by regulating FOXO1. The expression of METTL14 was increased in endothelial cells stimulated by TNF-α, which enhanced the m⁶A modification of FOXO1 mRNA and promoted the expression of FOXO1, and thus promoting the expression of adhesion molecules. At the same time, METTL14 can also interact with FOXO1 to promote the expression of adhesion molecules, which in turn mediates TNF-α-induced endothelial inflammation and atherosclerosis.