Mettl14 Mediates the In ammatory Response of Macrophages in Atherosclerosis Through the NF- κB/IL-6 Signaling Pathway

Yang Zheng Harbin Medical University Second A liated Hospital Department of Cardiology Yunqi Li Harbin Medical University Second A liated Hospital Department of Cardiology Xianwen Ran Harbin Medical University Second A liated Hospital Department of Cardiology Di Wang Harbin Medical University Second A liated Hospital Department of Cardiology Xianghui Zheng Harbin Medical University Second A liated Hospital Department of Cardiology Maomao Zhang Harbin Medical University Second A liated Hospital Department of Cardiology Jian Wu (  wujian780805@163.com ) Second A liated Hospital of Harbin Medical University https://orcid.org/0000-0002-1192-321X Yong Sun Harbin Medical University Institute of Biological Information Science and Technology Bo Yu Harbin Medical University Second A liated Hospital Department of Cardiology


Introduction
Monocytes and macrophages are regarded as the main actors in atherosclerosis because of their ubiquitous involvement during all stages of atherosclerosis [4,8]. Considerable amounts of evidence support that atherosclerosis is a chronic in ammatory disease starting with the entry of monocytes from peripheral blood into the endothelium: rolling, adhesion, activation and migration. Then, monocytes differentiate into macrophages [5,27]. Depending on the markers, production of speci c factors and biological functions of the macrophages, macrophages can be divided into two major subtypes: classically activated macrophages (M1) and alternatively activated macrophages (M2). High levels of the glycoprotein Ly6C (Ly6C high ) monocytes in mice, known as CD14 (CD14 ++ CD16 -) in humans, differentiate into M1 macrophages. However, Ly6C low monocytes (CD14 +/low CD16 + in humans) differentiate into M2 macrophages [12,38]. M1 macrophages initiate and sustain in ammation, producing in ammatory cytokines (TNF-α, IL-1β and IL-6) and leading to foam cell formation. M2 macrophages counteract in ammation and secrete anti-in ammatory cytokines (IL-10 and TGF-β) [5,25]. The biological characteristics of macrophages in atherosclerotic plaques determine the size, composition and stability of the lesion [22]. Recent studies suggest that a substantial proportion of patients retain the progression of the plaque despite achieving guideline-directed therapeutic targets caused by in ammation [31,35].
Regulating the in ammatory state and function of macrophages raises hope for atherosclerosis regression.
Macrophages are associated with epigenetic reprogramming and are modi ed by epigenetic enzymes [20]. N 6 -methyladenosine (m 6 A) RNA methylation is a posttranscriptional epigenetic modi cation of eukaryotic RNAs, including coding and noncoding RNAs [9]. The m 6 A modi cation process is jointly regulated by three types of enzymes: "writers" (methyltransferases), "erasers" (demethylases) and "readers" (m 6 A-related binding proteins) [36]. Previous research has established that m 6 A modi cation is associated with RNA metabolism, including changes in RNA structure, splicing, export and translation [16]. Numerous studies have demonstrated that regulators of m 6 A methylation are involved in multiple physiological and pathological processes, such as immune system development, human cancers and cardiovascular disease [16].
However, our understanding of the regulation of m 6 A modi cation in atherosclerosis is still in its infancy. Guo et al. showed that IFN regulatory factor-1 induced macrophage pyroptosis by Mettl3-mediated m 6 A modi cation of circ_0029589 in atherosclerosis [13]. Zhang et al. demonstrated that Mettl14 increased the m 6 A modi cation of pri-miR-19a to promote endothelial cell proliferation and invasion in vitro [44].
Similar studies have reported that Mettl14 regulates endothelial in ammation and that Mettl14 knockdown reduces the development of atherosclerotic plaques [15]. However, the amount of published data on the m 6 A modi cation of macrophages in atherosclerosis is limited.
Here, we found that the Mettl14 "writer" can regulate the in ammatory state of macrophages in atherosclerosis. Knockdown of Mettl14 promoted M2 polarization and inhibited foam cell formation.
Mechanistically, RNA sequencing (RNA-seq) analysis revealed that Mettl14 regulated the expression of Myd88 through m 6 A modi cation. Myd88 affected the transcription of IL-6 through p65 by regulating the distribution of p65 in nuclei. In vivo, Mettl14 gene knockout in mice signi cantly reduced the development of atherosclerosis by decreasing the in ammatory response of macrophages. Collectively, our study offers important insights into the m 6 A modi cation in atherosclerosis and highlights a potential target for treatment.

Study population and ethics statement
A total of 54 participants were selected for our study from the Second A liated Hospital of Harbin Medical University between December 2020 and June 2021: 9 controls, 18 patients with ST-segment elevation myocardial infarction (STEMI), 11 patients with non-ST-segment elevation myocardial infarction (NSTEMI) and 17 patients with unstable angina (UA). Inclusion and exclusion criteria were previously described in detail [23]. Brie y, patients with UA, STEMI and NSTEMI were enrolled according to the guidelines [39]. All participants or their families provided informed consent for inclusion before participation in the study, conforming to the Declaration of Helsinki. The current study was approved by the Ethics Committee of the Second A liated Hospital of Harbin Medical University, China (KY2020-156).
Mettl14 +/-APOE -/mice were generated by breeding Mettl14 +/mice with APOE -/mice. Eight-to 10-weekold male APOE -/-(WT) mice and Mettl14 +/-APOE -/-(KO) mice were fed a high-cholesterol diet (D12108C, Opensource diets) for 12 weeks. Then, the mice were euthanized for further analysis. All mice were housed under speci c pathogen-free (SPF) conditions with controlled temperature and a 12-hour light/dark cycle at the Second A liated Hospital of Harbin Medical University. All experimental protocols were approved by the Institutional Animal Care and Use Committee at the Second A liated Hospital of Harbin Medical University (sydwgzr2020-095). This study was conducted in accordance with the Guide for the Care of Use of Laboratory Animals (Institute of Laboratory Animal Resources/National Institutes of Health, Bethesda, MD, USA).

Peripheral blood mononuclear cells isolated
Overnight fasting blood samples were collected by venipuncture from all participants when they were hospitalized on the rst day. The peripheral blood of mice was collected when the mice were fed a highcholesterol diet for 12 weeks. Peripheral blood mononuclear cells (PBMCs) of humans and mice were isolated using the density gradient technique (TBD; Tianjin, China). Human PBMCs were lysed with TRIzol reagent (Invitrogen, Carlsbad, USA) for RNA extraction. The PBMCs of mice were further analyzed via ow cytometry.

RNA extraction and qRT-PCR
Total RNA from PBMCs of patients and THP-1 cells was extracted using TRIzol reagent (Invitrogen, Carlsbad, USA) and converted to complementary DNA by a Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland). qRT-PCR was performed with a Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland). The settings were as follows: 40 cycles of 10 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C. All relative mRNA expression levels were analyzed using the 2 -ΔΔCt method. The primers used are listed in Table S1.

Dot-blot assays
Total RNA from PBMCs of patients and THP-1 cells was extracted using TRIzol reagent (Invitrogen, Carlsbad, USA). The RNAs (200, 100 and 50 ng) were double diluted, denatured by heating at 95°C for 5 min and chilled on ice immediately. The RNAs were then spotted onto nitrocellulose membranes (Solarbio, Beijing, China). Then, the membranes were ultraviolet (UV) crosslinked, blocked and incubated with an m6A-speci c antibody (Synaptic Systems, Gottingen, Germany). The other membrane was stained with methylene blue as a loading control.

Culture and transfection of THP-1 cells
The human macrophage cell line THP-1 was purchased from the American Type Culture Collection. The cells were cultured in RPMI 1640 medium (Gibco, Thermo Fisher Scienti c, Waltham, MA, USA) consisting of 10% FBS (Biological Industries, Israel) at 37°C with 5% CO 2 .
The siRNAs for siMettl14 and siMyd88 and the plasmid DNA were purchased from GenePharma (Shanghai, China). All the sequences of the siRNAs are provided in Table S1. The siRNAs were  RNA-seq and bioinformatics analysis RNA from the siM14 and NC groups was isolated using TRIzol reagent. RNA-seq and data analysis were carried out by Sangon Biotech (Shanghai, China). The cDNA library was constructed using a TruSeq PE Cluster Kit v4-cBot-HS (Illumina, USA). Sequencing was performed on a MGISEQ-2000 platform. The RNAseq reads were mapped by the human reference genome (hg19) from the NCBI using HISAT2. The differentially expressed genes (DEGs) were ltered based on P values <0.05 and fold-changes >1.5. GO enrichment and KEGG pathway enrichment analyses of DEGs were performed by using DAVID Bioinformatics Resources 6.8 (https://david.ncifcrf.gov/).

Immuno uorescence
The cells were xed with 4% paraformaldehyde for 20 min. The tissue was cut into 7 μm thin slices and xed in acetone. Then, the cells or cryosections were immersed in Triton X-100 (Biosharp, Hefei, China) for 30 min followed by supplementation with 10% normal goat serum (Solarbio, Beijing, China) for 30 min. The cells or cryosections were incubated overnight with primary antibodies. Secondary antibodies were added to the cells or cryosections and then labeled with DAPI. The images were captured using a confocal laser scanning microscope (Zeiss LSM 700).
mRNA decay analysis THP-1 cells were treated with actinomycin D at a nal concentration of 5 μg/ml for 5 or 10 h. Total RNA was extracted at the indicated time points for reverse transcription and qRT-PCR. The mRNA decay rate was normalized to that at 0 h.

RNA-binding protein immunoprecipitation assays
RIP assays were performed using an RNA Immunoprecipitation Kit (Geneseed Biotech, Guangzhou, China) according to the manufacturer's instructions. In brief, the cells were harvested in RIP buffer on ice for 10 min. Then, 100 μL RIP lysis was used as input. Then, the cell lysates were incubated with 5 μg antim 6 A antibody or control IgG containing protein A/G-agarose beads. After rotation at 4°C for 2 h, the beads were washed. The immunoprecipitated and input RNAs were isolated and subjected to RIP-qPCR analysis. The PCR primers used are listed in Table S1.

Chromatin immunoprecipitation (CHIP) assays
ChIP assays were performed using a SimpleChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology, Danvers, MA, USA). The cells were crosslinked with 1% formaldehyde for 10 min and then sonicated in lysis buffer. Ten microliters of the lysate was used as an input. The remaining lysate was subjected to a ChIP assay using p65 or IgG antibodies. The PCR primers used are provided in Table S1.

Pathological staining
The hearts from the mice were perfused with PBS and immediately embedded in Tissue OCT-Freeze Medium (Tissue-Tek, Sakura, Torrance, CA). Serial cryosections (7 μm) were made using a cryostat. The serial cryosections from the caudal aortic sinus to the proximal aorta were stained with hematoxylin and eosin (HE) to visualize the plaques, oil red O to evaluate the lipid content, and Masson trichrome to analyze lesion size, collagen content, brous cap thickness and necrotic core formation. The HE staining kit, oil red O staining kit and Masson trichrome staining kit were purchased from Solarbio (Beijing, China).

ELISA analysis
Plasma was collected from the mice. Then, the levels of IL-6 were assessed with ELISA kits (Lianke Biotech, Hangzhou, China) according to the manufacturer's instructions. The absorbance was measured using a microplate reader (Tecan In nite M200).

Serum lipid determination
Sera were collected from the mice. Serum lipids, including triglycerides, total cholesterol, LDLs and HDLs, were determined based on commercial kits (Jiancheng, Nanjing, China) using a microplate reader (Tecan In nite M200).

Statistical analysis
The results are presented as the means ± SDs. The statistical analyses were performed using GraphPad Prism version 9.1.0 or SPSS 23.0 software (IBM). Two-group comparisons were performed by unpaired two-tailed Student's t-test, and three or more group comparisons were performed by ordinary one-way analysis of variance (ANOVA). P values <0.05 were considered statistically signi cant. All experiments were performed at least three times.

Results
Mettl14 is upregulated in coronary heart disease patients and macrophages The role of m 6 A in PBMCs of patients with coronary heart disease (CHD) was evaluated using an m 6 A mRNA dot blot assay. We divided patients with CHD into the UA group, STEMI group and NSTEMI group.
The levels of m 6 A modi cation, including UA, STEMI and NSTEMI, were signi cantly increased in CHD (Fig. 1a). Given that the m 6 A modi cation is primarily catalyzed by methyltransferase (writers) and demethylase (erasers), we hypothesized that the abnormal m 6 A modi cation in CHD was caused by the dysregulation of m 6 A writers and erasers. Then, we measured the expression of writers (Mettl3, Mettl14, Mettl16 and WTAP) and erasers (FTO and ALKBH5) in the PBMCs of CHD patients. Interestingly, the expression of Mettl3 and Mettl14 was signi cantly increased in CHD patients, including those in the UA group, STEMI group and NSTEMI group (Fig. 1b,c). The level of WTAP was increased in the UA group and NSTEMI group but did not change in STEMI group (Fig. 1e). FTO was upregulated in the UA group but did not change in the STEMI or NSTEMI group (Fig. 1f). The level of ALKBH5 was decreased in the STEMI and NSTEMI groups but did not change in the UA group (Fig. 1g). In addition, Mettl16 did not change signi cantly in CHD patients (Fig. 1d). Consistent with the results in CHD patients, m 6 A modi cation in LPS-stimulated THP-1 cells (a human monocyte cell line) increased (Fig. 2a). Moreover, the expression of Mettl3, Mettl14, Mettl16 and WTAP was increased in LPS-stimulated THP-1 cells, and ALKBH5 and FTO expression decreased in LPS-stimulated THP-1 cells (Fig. 2b-c, Fig. S1a-e). In particular, the expression level of Mettl14 was highest in LPS-stimulated THP-1 cells. Taken together, these results prompted us to explore the consequences of increased Mettl14 in macrophages caused by atherosclerosis.
Mettl14 knockdown conditioned macrophages toward M2 polarization and suppressed NF-κB signaling The m 6 A writer Mettl14 plays a critical role in the regulation of m 6 A modi cation in macrophages in atherosclerosis. Hence, we knocked down Mettl14 in THP-1 cells through two different siRNAs. qRT-PCR and Western blot analysis showed that both siM14-1 and siM14-2 caused a decrease in the expression of Mettl14 ( Fig. S1f-h). LPS-and IL-4-stimulated THP-1 cells were used as positive controls for M1 and M2 polarization, respectively (Fig. 2d, Fig. S1i-j). As shown in Fig. 2d and Fig. S1i-j, knockdown of Mettl14 signi cantly increased the proportion of M2 macrophages (CD68 + CD163 + cells) and decreased the proportion of M1 macrophages (CD68 + CD86 + cells). Like ow cytometry, qRT-PCR con rmed that the mRNA expression of IL-1β and TNF-α was downregulated and that IL-10 and CD163 were upregulated in siM14-1-and siM14-2-treated THP-1 cells (Fig. 2e-h). Previous studies have demonstrated that the NF-κB pathway plays an important role in macrophage polarization by inducing the expression of in ammatory genes. Then, we investigated the NF-κB pathway using western blotting to evaluate whether the NF-κB pathway was involved in macrophage polarization. Knockdown of Mettl14 markedly reduced the phosphorylation of NF-κB-p65, but the level of p65 was not signi cantly changed (Fig. 2i, Fig. S1k-l). The expression of IκBα increased in M2 phenotype macrophage cells (Fig. 2i, Fig. S1m). Taken together, these data suggested that knockdown of Mettl14 prevents the macrophage in ammatory response by promoting M2 polarization via the NF-κB pathway.

Mettl14 de ciency inhibited foam cell formation and macrophage migration
In arteries, foam cell formation and macrophage migration play important roles in the development of atherosclerosis. We asked whether Mettl14 has a role in foam cell formation and macrophage migration. As shown in Fig. 3a and 3b, the siM14-1 and siM14-2 groups had much less lipid accumulation than the NC group stained with oil red O did. Then, we measured the expression level of lipid metabolism-related proteins. Several studies have shown that ABCA1 and ABCG1 have an essential role in cholesterol e ux during foam cell formation and that PPAR-γ and LXR-α regulate ABCG1/ABCA1 [6]. The expression of ABCA1, ABCG1, PPAR-γ and LXR-α was increased in the Mettl14 knockdown group compared to the NC group ( Fig. 3c-g), indicating that the protective function of Mettl14 in macrophages may be related to the PPARγ-LXRα-ABCA1/ABCG1 pathway. Next, using a scratch test, we con rmed that Mettl14 affected migration. The results showed that knockdown of Mettl14 signi cantly decreased macrophage migration (Fig. 3h-i). Together, the above ndings demonstrated that Mettl14 de ciency inhibited foam cell formation and macrophage migration.
Mettl14 modi es Myd88 mRNA stability by m 6 A modi cation To determine Mettl14-regulated genes in macrophages, we used RNA-seq. For this, we used three samples of Mettl14 knockdown THP-1 cells and three samples of THP-1 cells treated with NC. The RNA-seq data showed that 150 mRNAs were upregulated and that 128 mRNAs were downregulated in the Mettl14 knockdown group compared with the NC group (Fig. 4a, Table S2). Gene Ontology (GO) analysis of these DEGs showed that they were involved mainly in the immune response, in ammatory response, chronic in ammatory response, positive regulation of interleukin-6 production and positive regulation of I-κB kinase/NF-κB signaling (Fig. 4b). The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway was mainly enriched in cytokine-cytokine receptor interaction, rheumatoid arthritis, pertussis, Toll-like receptor signaling pathway and in ammatory bowel disease (Fig. 4c). These data suggested that Mettl14 regulated the in ammatory response of macrophages.
Among the DEGs, Myd88 and interleukin-6 (IL-6) were very interesting because they play an important role in the in ammatory response of macrophages [1,3,7]. First, we detected whether the expression of Myd88 was regulated by Mettl14. qRT-PCR analysis revealed that the mRNA expression levels of Myd88 were markedly lower in Mettl14 knockdown THP-1 cells than in those in the NC group, whereas the mRNA expression of Myd88 was increased in the Mettl14 overexpression group (Fig. 4d). Moreover, using western blots and immuno uorescence staining, we obtained similar results of Myd88 protein expression ( Fig. 4e-f, Fig. S2a). To determine the regulatory mechanism by which Mettl14 regulates Myd88, we examined the mRNA stability of Myd88. Similar to our hypothesis, the stability of Myd88 mRNA was decreased in the Mettl14 knockdown group (Fig. 4i). RNA immunoprecipitation (RIP) analysis revealed that knockdown of Mettl14 reduced the m 6 A modi cation on site 4 of Myd88 mRNA (Fig. 4j). These results suggested that Mettl14 modi es Myd88 mRNA stability by m 6 A modi cation.
Mettl14 regulates the expression of IL-6 by affecting the nuclear distribution of p65 through Myd88 IL-6 is a key in ammatory factor involved in the activation and regulation of macrophages [40]. To validate the RNA-seq results, we performed qRT-PCR to detect the expression of IL-6 mRNA. The level of IL-6 was signi cantly decreased in the Mettl14 knockdown group, and the expression of IL-6 was the opposite in the Mettl14 overexpression group (Fig. 4h). Next, we identi ed the mechanism by which Mettl14 regulates IL-6. Unfortunately, the stability of IL-6 mRNA and translation was not signi cantly affected by Mettl14 (Fig. 4g-k, Fig. S2b). Therefore, Mettl14 does not regulate IL-6 through m 6 A modi cation. Therefore, how does Mettl14 regulate the expression of IL-6? We assumed that Mettl14 regulated IL-6 through the Myd88/NF-κB pathway according to the prediction of p65 binding to the promoter region of IL-6. Mettl14 did not affect the expression level of p65 (Fig. 2i). Immuno uorescence staining showed that knockdown of Mettl14 impeded p65 translocation from the cytosol to the nucleus (Fig. 4l). However, p65 translocation from the cytosol to the nucleus was facilitated in THP-1 cells treated with siM14 and overexpressing Myd88 (Fig. 4l). As expected, ChIP analysis revealed that knockdown of Mettl14 inhibited p65 binding to the IL-6 promoter (Fig. 4m, Fig. S2c). P65 binding to the IL-6 promoter was increased in cotransfected THP-1 cells. These results indicated that Mettl14 regulates its expression through the Myd88/NF-κB pathway.

Knockdown of Mettl14 induces a low in ammatory response in macrophages through Myd88
We showed that Mettl14 knockdown induced a low in ammatory response in macrophages ( Fig. 2 and Fig. 3) and that Myd88 was regulated by Mettl14 (Fig. 4). We next examined whether the functions of Mettl14 are mediated by Myd88 in macrophages. For this, we conducted rescue experiments. We generated a Mettl14 expression vector, which was cotransfected with siMyd88. Overexpression of Mettl14 promoted M1 polarization rather than M2 polarization in THP-1 cells (Fig. 5a-c). As expected, siMyd88 reversed Mettl14 overexpression-induced M1 polarization (Fig. 5a-c). The expression level of p65 protein did not signi cantly change in the OE-M14 group or cotransfection group. The p-p65 protein level increased in Mettl14-overexpressing THP-1 cells, and its expression was decreased by siMyd88 (Fig. 5d,  Fig. S3a-c). We found that the in ammatory factors IL-1β and TNF-α increased and that the antiin ammatory factors IL-10 and CD163 decreased in Mettl14-overexpressing cells (Fig. 5e-h). Moreover, the levels of in ammatory factors and anti-in ammatory factors were reversed in cells cotransfected with siMyd88. Furthermore, Mettl14 overexpression increased foam cell formation and decreased the abundance of lipid metabolism-related proteins (Fig. 5i-k). However, foam cell formation decreased in response to siMyd88, and lipid metabolism-related proteins increased (Fig. 5i-k, Fig. S2d-g). The migration of macrophages had similar results. Overexpression of Mettl14 promoted the migration of macrophages (Fig. 5l-m). However, siMyd88 reversed OE-M14-induced migration. Collectively, these results suggested that Mettl14 regulates the functions of macrophages through Myd88.

Mettl14 de ciency attenuates atherosclerosis progression in vivo
Given that Mettl14 regulates the functions of macrophages in vitro, we further assessed the effect of Mettl14 on atherosclerosis progression in vivo. Previous studies have demonstrated that Mettl14 gene knockout causes embryo death. Mettl14 heterozygous knockout mice were used for research [15]. To do this, age-and weight-matched male APOE -/-(WT) mice and Mettl14 +/-APOE -/-(KO) mice were fed a western diet for three months. After three months, the aortas and peripheral blood of these mice were dissected.
Compared to the WT mice, the KO mice did not exhibit changes in the plasma lipid pro les (Fig. S4a-c). As shown in Fig. 6a, images of aortic arches suggested that KO mice were signi cantly protected from atherosclerosis. HE staining of the aortic roots revealed that the lesion areas were smaller in KO mice than in WT mice (Fig. 6b, Fig. S4d). The extent of proximal aortic atherosclerosis was reduced in KO mice using oil red O (Fig. 6c, Fig.S4e). These results suggested that Mettl14 knockout can signi cantly reduce atherosclerosis.
Next, using Masson trichrome staining, we examined the characteristics of plaque stabilization between WT and KO mice. Compared to the WT mice, the KO mice had an increased collagen content in the proximal aorta (Fig. 6d, Fig. S4f). In addition, the thickness of the brous cap was greater in the lesions of KO mice than in those of WT mice (Fig. 6d, Fig. S4g). Importantly, the necrotic area was decreased in KO mice vs. WT mice (Fig. 6d, Fig. S4h). Taken together, these data showed that Mettl14 de ciency suppressed the characteristics of vulnerable plaque formation.

Mettl14 regulates the function of macrophages via Myd88/IL-6 in vivo
To verify the function and mechanism of macrophages regulated by Mettl14 in vivo, we analyzed the subtypes of monocytes in the peripheral blood of the mice. The subtypes of monocytes include proin ammatory Ly6C high CX3CR1 low monocytes and anti-in ammatory Ly6C low CX3CR1 high monocytes.
We found that the subpopulations of Ly6C high CX3CR1 low monocytes were lower in KO mice than in WT mice and that Ly6C low CX3CR1 high monocytes were higher (Fig. 6e-g). These data indicate that Mettl14 de ciency reprograms monocytes/macrophages to anti-in ammatory effects. Then, we detected the expression of Myd88 in plaque macrophages. Fig. 6i shows that the expression of Myd88 signi cantly decreased in plaque macrophages in KO mice. Surprisingly, we also found that the plasma level of IL-6 was reduced in KO mice (Fig. 6h). In conclusion, the results suggest that Mettl14 can regulate the function of macrophages in atherosclerosis via Myd88/IL-6 in vivo.

Discussion
The presence of macrophages is a common condition that has considerable impact on all stages of atherosclerosis [4,8]. The functions of macrophages are mainly regulated by epigenetic reprogramming [20]. Recent evidence suggests that epigenetic modi cations can be grouped into three categories: epigenetic triad, DNA methylation, and histone modi cation and nucleosome positioning [30]. In recent years, the mechanism of epigenetic modi cations has been updated with the development of speci c methylated RNA immunoprecipitation and next-generation sequencing [28]. M 6 A is the most common epitranscriptomic modi cation of mRNA from yeast, plants, les, humans and other mammals [45]. M 6 A methylation marks are dynamic and reversible. Brie y, m 6 A methylation occurs at the consensus sequence RRACH (R=G or A; H=A, C or U) through methyltransferases (writers) and demethylases (erasers). Then, the m 6 A-related binding proteins (readers) selectively bind the site of m 6 A modi cation. After this regulatory event, RNAs are cleaved, stable, degraded and translated [18,21,34].
Several studies have demonstrated a link between m 6 A modi cation and human diseases [34].
Zhang et al. demonstrated that m 6 A modi cation and Mettl14 were signi cantly increased in the atherosclerotic vascular endothelial cells of patients with carotid stenosis [44]. Guo et al. showed that the expression of Mettl3 was signi cantly elevated in macrophages in patients with acute coronary syndrome [13]. Another study showed opposite results, which may be caused by different detection methods for m 6 A modi cation. The m 6 A levels were signi cantly decreased in peripheral blood mononuclear cells, as detected by colorimetry. These three studies suggest that m 6 A modi cation plays an important role in atherosclerosis. Here, we observed that the levels of m 6 A modi cation and Mettl14 were increased in the peripheral blood mononuclear cells of patients with coronary heart disease and LPS-stimulated THP-1 cells. It is possible, therefore, that Mettl14 has a pivotal role in macrophages in atherosclerosis. Existing studies recognize the critical role played by m 6 A modi cation in the immune response of macrophages. An earlier study showed that the reader YTHDF2 was upregulated after LPS stimulation and that YTHDF2 knockdown promoted the in ammatory response in LPS-stimulated macrophages [43]. Subsequent studies con rmed that m 6 A modi cation regulated M1/M2 polarization and cholesterol e ux [24,29,41,42]. However, the function of Mettl14 in macrophages in atherosclerosis has not been reported. In this study, we rst reported that Mettl14 regulated the in ammatory state of macrophages in atherosclerosis. Knockdown of Mettl14 in macrophages promotes M2 polarization. Moreover, foam cell formation and migration were inhibited in Mettl14 knockdown macrophages. Atherosclerotic plaques and the in ammatory response were signi cantly reduced in Mettl14 knockout mice. These data provide evidence that Mettl14 plays an essential role in the regulation of macrophages in atherosclerosis.
The mechanisms of m 6 A modi cation are diverse, including the fold, stability, degradation and cellular interactions of the modi ed RNA [28]. Jian et al. demonstrated that Mettl14 enhances the transcription factor FOXO1 by increasing its translation, not RNA stabilization, in endothelial in ammation [15]. In bacterial infection, Mettl14 depletion blocked m 6 A methylation of SOCS1, diminishing its RNA stability [10]. In addition, Mettl14 forms a complex with Mettl3, called the Mettl3/Mettl14 complex, modifying nascent transcripts whose translation is enhanced [14]. We performed RNA-seq of Mettl14knockdown macrophages to explore the regulatory mechanism of Mettl14. The DEGs regulated by Mettl14 were enriched in the in ammatory response, indicating that Mettl14 plays an important role in the in ammatory response in macrophages. We found two interesting DEGs, those encoding Myd88 and IL-6.
qRT-PCR further con rmed that the expression of Myd88 and IL-6 was consistent with the RNA-seq results. Most Toll-like receptors (TLRs) and several cytokine receptors signal through Myd88 to initiate a rapid immune response when alarmins stimulate macrophages [7]. In addition, Myd88 knockout mice showed smaller atherosclerotic plaques and less macrophage activation, lipid accumulation and foam cell formation [3,26,37]. Myd88 plays a central role in the in ammatory response of macrophages. IL-6 is a major contributor to the development of atherosclerosis. The aortas of atherosclerotic mice and rats had higher levels of macrophage-attracting IL-6 than did those of the control, and the expression level of IL-6 was higher with aging [2,11]. The mice were given an anti-mouse IL-6 receptor antibody, and the atherosclerotic lesion size and in ammation were reduced [1]. In clinical trials, the rst cytokine inhibition, IL-1β inhibition, was used for atherosclerosis treatment and prevention and achieved results.
The results showed that IL-6 was the central signaling cytokine of IL-1β inhibition, in turn reducing the in ammatory response [33]. In a recent RESCUE trial, the IL-6 ligand monoclonal antibody ziltivekimab was highly effective at reducing the in ammatory response and atherosclerotic biomarkers, suggesting that IL-6 has become a new therapeutic target for atherosclerosis [32]. In our study, IL-6 was the target of Mettl14. The expression of IL-6 was decreased in the Mettl14 knockdown, but the stability and translation of IL-6 mRNA did not signi cantly change in the Mettl14 knockdown.  19]. Therefore, we hypothesized that Mettl14 regulates IL-6 through the Myd88/NF-κB pathway. As expected, Mettl14 regulated the distribution of p65 in nuclei, which regulates the transcription of IL-6, re ecting the mechanism of upstream regulation of IL-6 in macrophages.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.