Myeloid TM6SF2 Deficiency Inhibits Atherosclerosis

Genetic variants in transmembrane 6 superfamily member 2 (TM6SF2), such as E167K, are associated with atherosclerotic cardiovascular disease (ASCVD). Chronic inflammation and lipid-laden macrophage foam cell formation are the central pathogeneses in the development of atherosclerosis. This study was undertaken to illustrate the biological function of TM6SF2 in macrophages and its role during atherosclerosis development. We generated myeloid cell-specific Tm6sf2 knockout mice on ApoE-deficient background (LysM Cre+/Tm6sf2fl/fl/ApoE−/−, TM6 mKO) with littermate LysM Cre−/Tm6sf2fl/fl/ApoE−/− (Control) mice as controls. Mice were fed a Western diet for 12 weeks to induce atherosclerosis. Myeloid Tm6sf2 deficiency inhibited atherosclerosis and decreased foam cells in the plaques without changing the plasma lipid profile. RNA sequencing of bone marrow-derived macrophages (BMDMs) from TM6 mKO mice demonstrated the downregulation of genes associated with inflammation, cholesterol uptake, and endoplasmic reticulum (ER) stress. TM6SF2 was upregulated by oxidized low-density lipoprotein (oxLDL) in macrophages. Silencing TM6SF2 in THP-1-derived macrophages and Tm6sf2 deficiency in BMDMs reduced inflammatory responses and ER stress and attenuated cholesterol uptake and foam cell formation, while the overexpression of TM6SF2 showed opposite effects. In conclusion, myeloid TM6SF2 deficiency inhibits atherosclerosis development and is a potential therapeutic target for the treatment of atherogenesis.


Introduction
Atherosclerosis is a major cause of cardiovascular disease (CVD) and has been associated with increased levels of total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and triglycerides (TG) and decreased high-density lipoprotein cholesterol (HDL-C) levels [1,2]. Atherosclerosis is a chronic inflammatory vascular disease in which the infiltration of macrophages and other immune cells and their interaction with arterial tissue play an important role in atherosclerosis progression [3]. Monocyte-derived macrophage uptake cholesterol resulting in foam cell formation and polarization to proinflammatory macrophages, contributing to the atherosclerotic plaque initiation and progression [4,5].
Transmembrane 6 superfamily member 2 (TM6SF2) was first identified in the EU-ROIMAGE consortium sequencing project in 2000 [6]. The human TM6SF2 gene is located at chromosome 19p12 and is highly conserved in Homo sapiens and Mus musculus. Exomewide and genome-wide association studies have identified that a variant in the TM6SF2 gene (rs58542926, encoding p.Glu167Lys, E167K) is associated with reduced myocardial infarction risk and decreased plasma LDL-C levels but increased liver fat levels [7][8][9], demonstrating the paradoxical effect of TM6SF2 variant in cardiovascular and liver diseases. The E167K mutation in TM6SF2 yields a misfolded protein, causing accelerated protein degradation and reduced protein levels in hepatocytes [7]. TM6SF2 facilitates cholesterol synthesis and very-low-density lipoprotein (VLDL) assembly, and knockdown of TM6SF2 reduces TG synthesis in human hepatocytes [10][11][12]. Consistent with human genetic findings, liver-specific Tm6sf2-deficient mice show significantly reduced TC and TG levels and VLDL particle size [10,11] while exhibiting spontaneous hepatic steatosis, fibrosis, and accelerated development of hepatocellular cancer [13]. Better understanding of the biological functions of TM6SF2 in cells affected by atherosclerosis will elucidate the effects of TM6SF2 in atherosclerosis and liver diseases, which may aid in developing new approaches for treating cardiovascular diseases.
Considering that inflammatory responses and lipid accumulation in macrophages are characteristic features of atherosclerotic plaques, in the current study, we sought to determine the contribution of macrophage TM6SF2 to atherosclerosis. We found that TM6SF2 deficiency decreased inflammatory responses, inhibited foam cell formation, and reduced ER stress in macrophages. We demonstrated that the genetic ablation of Tm6sf2 in myeloid cells of the ApoE-deficient (ApoE −/− ) mice significantly attenuated atherosclerotic plaque formation without affecting circulating TC and TG levels. Our findings reveal that the suppression of macrophage TM6SF2 may become an appropriate therapeutic strategy for atherosclerosis treatment.

Animal Procedures
All animal experiments in this study were approved by the Institutional Animal Care and Use Committee (IACUC) and met the animal care standards and use established by the Animal Welfare Act and the NIH Guide for the Care and Use of Laboratory Animals. Tm6sf2 floxed embryonic stem (ES) cell clone (EPD0097_3_E02, ES cell strain: C57BL/6N) were obtained from KOMP Repository at UC Davis. The exon 2 of the Tm6sf2 gene was flanked by loxP sites. The ES clone was microinjected into blastocyst stage embryos to generate chimeric mice. The frt-flanked selection marker was removed by breeding with FlpO mice [14], yielding the floxed Tm6sf2 mice (Tm6sf2 fl/fl ). Germline transmission black offspring were crossed with LysM-Cre transgenic mice (From the Jackson Laboratory, Strain #004781) to obtain myeloid cell-specific Tm6sf2 knockout (mKO) mice (LysM Cre+/Tm6sf2 fl/fl ) that were then crossed with apolipoprotein E knockout (ApoE −/− ) mice (from the Jackson Laboratory, Strain #002052) to generate LysM Cre+/Tm6sf2 fl/fl ApoE −/− (Tm6sf2 mKO) and littermate LysM Cre−/Tm6sf2 fl/fl /ApoE −/− mice (Control) mice for the current study. In the atherosclerosis experiment, both male and female Tm6sf2 mKO and littermate control mice were placed on a Western diet (WD) for 12 weeks (TD. 88137, ENVIGO, 42% of calories from fat and 0.2% cholesterol).
At the endpoint of experiment, animals were euthanized with CO 2 overdose, and blood was collected for lipid profile at the Chemistry Laboratory of the Michigan Diabetes Research and Training Center (University of Michigan). Mice were perfused via the left cardiac ventricle with saline followed by 10% formalin. The whole aorta tree from the ascending aorta to the bifurcation of the common iliac arteries was isolated and stained with Oil Red O [15,16]. Sections from aortic roots (5 µm) were used for the quantification of plaque area, necrotic core, and macrophage content (Mac2) by a person who was blinded to the genotypes [15,16].

RNA Sequencing and Data Analysis
BMDMs were isolated from the control and mKO mice after a 12-week WD challenge. After differentiation, macrophages were treated with 100 µg/mL oxLDL (360-31, by Lowry protein assay, Lee Biosolutions, Inc., Maryland Heights, MO, USA) for 48 h. Each sample was collected from an individual animal (n = 4 for each group). RNA sample preparation by using the Qiagen RNeasy ® mini kit (#74106, Qiagen, Houston, TX, USA) and by following the manufacturer's instructions. Samples were then sent to the Advanced Genomics Core of U-M Biomedical Research Core Facilities for next-generation sequencing and quality control analysis. The data analysis follows a previously published paper [17].
DESeq2 package (1.32.0) was used for differential gene expression analysis in R version 4.2.0. Genes with adjusted p values (p-adj) < 0.05 were considered significantly. For the signaling pathway and functional enrichment analysis of the differentially expressed genes (DEGs), KEGG enrichment, and gene ontology (GO) enrichment, Metascape was used for this study [18]. The heatmaps were provided by heatmap package (Raivo Kolde, version 1.0.12, View on GitHub) in R.
Gene set enrichment analysis (GSEA) [19] was used for determining the statistical significance between control and mKO samples, and figures were generated based on the differential expression rank. GSEA version 4.2.3 was used for the data analysis and followed the GSEA user guide. p < 0.05 were considered significant.

Cholesterol Content Measurement
THP-1-derived macrophages and BMDMs were treated with 100 µg/mL oxLDL for 24 h. After washing with PBS, cells were harvested for cholesterol/lipid extraction. Samples were resuspended by 200 µL Chloroform: Isopropanol: NP-40 (7:11:0.1) mix solution, spun at 15,000× g for 10 min, then all liquid was transferred to EP tube to dry at 37 • C overnight. Cholesterol was measured using a cholesterol fluorometric assay kit (Cayman Chemical, #10007640, Ann Arbor, MI, USA).

Flowcytometry
THP-1-derived macrophages were infected with non-targeting siRNA (NT siRNA) or siTM6SF2 (TM6SF2 siRNA) at 20 nM for 48 h. Then cells were loaded with 5 µg/mL Dil-oxLDL (#770262-9, Kalen Biomedical, Montgomery Village, MD, USA) at 37 • C for 40 min and then collected for flow cytometry to determine cholesterol uptake by Dil intensity. Data were analyzed using FlowJo_v10.8.1 at UMICH Flow Cytometry Core. Blank samples without Dil fluorescence were used as negative controls.

Statistics
The data were analyzed using GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA). RNA seq data were analyzed with R 4.1.3 (https://www.r-project.org/, accessed on 10 March 2022). A two-tailed unpaired Student's t-test was used to compare two independent groups. One-way ANOVA or two-way ANOVA was used to compare more than two groups. Individual values were presented as interleaved scatter with bars plot mean ± SEM. p < 0.05 was considered statistically significant.

Tm6sf2 Knockout in Myeloid Cells Inhibits Atherosclerosis in Mice
To study tissue-specific TM6SF2 functions, we generated Tm6sf2 fl/fl mice that contained loxP sites flanking exons 2 of Tm6sf2 ( Figure S1). To investigate the effects of macrophage-specific Tm6sf2 deficiency on atherosclerosis, we obtained myeloid-specific Tm6sf2-deficient mice on ApoE −/− background by crossbreeding with mice expressing the LysM Cre transgene ( Figure 1A,B). Tm6sf2 depletion was confirmed in BMDMs from Tm6sf2 mKO mice (LysM Cre+/Tm6sf2 fl/fl ApoE −/− ) and littermate controls (LysM Cre−/Tm6sf2 fl/fl ApoE −/− ) ( Figure 1C). After 12 weeks of WD challenge, male myeloid-specific Tm6sf2 deficient mice significantly slowed down the development of atherosclerosis compared to littermate control, as evidenced by the reduced plaque size in whole aortas by en face Oil Red O staining ( Figure 1D). Frozen sections of aortic roots were subjected to hematoxylin and eosin (H&E) staining, Oil Red O staining, and macrophage immunofluorescent staining using an antibody against Mac2/galectin 3, a widely used macrophage marker in mouse studies [16]. Compared to littermate control mice, aortic sinus sections from Tm6sf2 mKO mice demonstrated significantly decreased plaque and necrotic core areas ( Figure 1E). Oil Red O staining showed less lipid accumulation in the atherosclerotic plaques in Tm6sf2 mKO mice ( Figure 1F). Along with the decreased lipid accumulation, the Mac2-positive area in atherosclerotic plaques in Tm6sf2 mKO mice was significantly lower than that in littermate controls ( Figure 1G). There were no significant differences in body weight ( Figure 1H), liver weight/body weight ratio ( Figure 1J), blood glucose ( Figure S2A), and insulin ( Figure S2B) levels between control and mKO mice. Myeloid-specific Tm6sf2 knockout did not affect serum TG, TC, LDL-C, and HDL-levels ( Figure 1J). Samiliar results were found in female mKO and control mice ( Figure S3). These in vivo results indicate that myeloid Tm6sf2 knockout can inhibit atherosclerosis development without changing the lipid metabolism and glucose metabolism. marker in mouse studies [16]. Compared to littermate control mice, aortic sinus sections from Tm6sf2 mKO mice demonstrated significantly decreased plaque and necrotic core areas ( Figure 1E). Oil Red O staining showed less lipid accumulation in the atherosclerotic plaques in Tm6sf2 mKO mice ( Figure 1F). Along with the decreased lipid accumulation, the Mac2-positive area in atherosclerotic plaques in Tm6sf2 mKO mice was significantly lower than that in littermate controls ( Figure 1G). There were no significant differences in body weight ( Figure 1H), liver weight/body weight ratio ( Figure 1J), blood glucose ( Figure  S2A), and insulin ( Figure S2B) levels between control and mKO mice. Myeloid-specific Tm6sf2 knockout did not affect serum TG, TC, LDL-C, and HDL-levels ( Figure 1J). Samiliar results were found in female mKO and control mice ( Figure S3). These in vivo results indicate that myeloid Tm6sf2 knockout can inhibit atherosclerosis development without changing the lipid metabolism and glucose metabolism. weight/body weight ratio, (J) Circulating TG, TC, LDL-C, and HDL-C levels, n = 9 for control, n = 10 for mKO. All data are presented as mean ± SEM, * p < 0.05, ** p < 0.01, *** p < 0.001 as indicated. The lesion area is quantified using Image J. n = 9 for control, n = 10 for mKO. E-G, Representative aortic sinus tissue-section images H&E staining (E), ORO staining (F), and Mac-2 immunofluorescent staining (G). Scale bar = 200 µm. n = 6 for control, n = 7 for mKO. (H) Bodyweight and (I) liver weight/body weight ratio, (J) Circulating TG, TC, LDL-C, and HDL-C levels, n = 9 for control, n = 10 for mKO. All data are presented as mean ± SEM, * p < 0.05, ** p < 0.01, *** p < 0.001 as indicated.

Differential Expression Analysis of BMDMs Isolated from Control and Tm6sf2 mKO Mice
Atherogenesis is a complex disease, and many factors contribute to the progression of atherogenesis. Cholesterol metabolism and local inflammation are two important factors for forming atherosclerotic plaques [20]. Macrophages have been considered a fundamental ele-Cells 2022, 11, 2877 6 of 15 ment in the development of atherosclerosis [21]. To unbiasedly understand the mechanisms by which myeloid-specific TM6SF2 contributes to atherosclerosis development, BMDMs from control and Tm6sf2 mKO mice were used for whole-transcriptome analysis (RNA-Seq). After the analyses, more than 24 million reads were generated from each sample. Compared with the control group, there were 3681 genes downregulated and 5487 genes upregulated (padj < 0.05) in the mKO group, shown in the heatmap (Figure 2A). Next, we used GO and KEGG enrichment to analyze the involved signaling pathways. Among the top 20 significant overrepresented pathways ( Figure 2B), inflammation-associated pathways were the most prominent, such as regulation of cytokine production (GO:0001817), inflammatory response (GO:0006954), cytokine-mediated signaling pathway (GO:0019221), and cytokine signaling in the immune system (R-MMU-1280215). As GSEA (Gene Set Enrichment Analysis) shows ( Figure 2C,D), 163 genes in HALLMARK_INFLAMMATORY_RESPONSE pathway were positively correlated with control, and 31 genes were negatively correlated with mKO ( Figure 2C). All 24 genes in REACTOME_CHOLESTEROL_BIOSYNTHESIS pathway were positively correlated with the control group ( Figure 2D), consistent with the findings using GO enrichment analysis (data showed in Supplementary Table S2). In the HALLMARK_TNFA_SIGNALING_VIA_NFKB pathway, 137 genes were positively correlated with control and 58 genes were negatively correlated with mKO ( Figure 2E). Furthermore, we found that the Hspa5 (Bip), Xbp1, and Nfkb1 (NF-κB) genes were all downregulated in Tm6sf2 KO BMDMs ( Figure 2F). The inflammatory genes such as Il6, Il1b, and Ccl2 (Mcp-1) were also down-regulated in mKO BMDMs. The Bip-IRE-1α-Xbp1 signaling pathway regulates ER stress [22], which is associated with various pathological changes. This pathway also activates the NF-κB pathway, a classical inflammation transcriptional signaling pathway [23,24]. Our data suggest that Tm6sf2 deficiency reduces the inflammatory responses in BMDMs and may lead to the alleviation of atherosclerosis in mice.

TM6SF2 Regulates Inflammatory Responses in Macrophages
We then determined whether macrophage TM6SF2 contributes to inflammatory responses in pro-atherosclerotic conditions. Ox-LDL, a well-established atherosclerosis risk factor, was used to induce inflammatory responses in human and mouse macrophages. oxLDL treatment upregulated the TM6SF2 expression in THP-1-derived macrophages and in BMDMs ( Figure S4). THP-1-derived macrophages were transfected with siTM6SF2 (TM6SF2 siRNA) or non-Targeting siRNA (NT siRNA), and the knockdown efficiency was confirmed as shown in Figure S5A. We found that knockdown of TM6SF2 abolished oxLDL-induced inflammatory responses, indicated by the reduced expressions of TNF, IL1B, and CCR2 ( Figure 3A), while the overexpression of TM6SF2 dramatically enhanced oxLDL-induced inflammatory responses in THP-1-derived macrophages ( Figure 3B). Using BMDMs isolated from control and mKO mice, we confirmed that deficiency of Tm6sf2 showed reduced inflammatory responses ( Figure 3C). We further demonstrated that the overexpression of TM6SF2 in wild-type BMDMs exacerbated ox-LDL-induced inflammation ( Figure 3D). These findings indicate that TM6SF2 plays a role in the inflammatory responses in macrophages.

TM6SF2 Contributes to Foam Cell Formation
Our RNA-seq data suggest that TM6SF2 plays a role in cholesterol handling in macrophages. oxLDL induces the transformation of THP-1-derived macrophages and BMDMs to cholesterol-rich foam cells, evidenced by cholesterol uptake assay, Oil Red O staining, and intracellular cholesterol contents measurement (Figure 4). Knockdown of TM6SF2 in THP-1-derived macrophages significantly reduced cholesterol uptake and foam cell formation ( Figure 4A,B), while the overexpression of TM6SF2 facilitated oxLDLinduced cholesterol uptake and foam cell formation ( Figure 4C,D). Similar findings were confirmed in BMDMs isolated from control and Tm6sf2 mKO mice, as well as in TM6SF2 overexpressing BMDMs ( Figure 4E-H). These data suggest that TM6SF2 contributes to macrophage cholesterol handling and foam cell formation.

TM6SF2 Regulates Inflammatory Responses in Macrophages
We then determined whether macrophage TM6SF2 contributes to inflammatory responses in pro-atherosclerotic conditions. Ox-LDL, a well-established atherosclerosis risk factor, was used to induce inflammatory responses in human and mouse macrophages. oxLDL treatment upregulated the TM6SF2 expression in THP-1-derived macrophages ing BMDMs isolated from control and mKO mice, we confirmed that deficiency of Tm6s showed reduced inflammatory responses ( Figure 3C). We further demonstrated that t overexpression of TM6SF2 in wild-type BMDMs exacerbated ox-LDL-induced inflamm tion ( Figure 3D). These findings indicate that TM6SF2 plays a role in the inflammato responses in macrophages.   (G,H) BMDMs were isolated from wild-type C57BL/6J mice and were transfected with AdLacZ or AdTM6SF2 (100 MOI) for 48 h and treated with oxLDL (100 µg/mL) overnight. n = 3 for flow cytometry, n = 5 for cholesterol contents detection. All data are presented as mean ± SEM, statistics by ordinary one-way ANOVA. * p < 0.05, ** p < 0.01, *** p < 0.001 as indicated. (E,F) BMDMs were isolated from control and mKO mice and treated with oxLDL (100 µg/mL) overnight. (G,H) BMDMs were isolated from wild-type C57BL/6J mice and were transfected with AdLacZ or AdTM6SF2 (100 MOI) for 48 h and treated with oxLDL (100 µg/mL) overnight. n = 3 for flow cytometry, n = 5 for cholesterol contents detection. All data are presented as mean ± SEM, statistics by ordinary one-way ANOVA. * p < 0.05, ** p < 0.01, *** p < 0.001 as indicated.

TM6SF2 Positively Regulates ER Stress Pathway in Macrophages
As TM6SF2 is an ER membrane protein, our previous study also demonstrated that TM6SF2 interacts with the Bip-associated ER stress sensor inositol-requiring enzyme (IRE1α) and affects its downstream XBP1 in the liver [25]. In macrophages, our RNA-seq data also indicated that TM6SF2 modulated the ER stress pathway. oxLDL induced ER stress in macrophages ( Figure 5). We showed that the overexpression of TM6SF2 in THP-1-derived macrophages and BMDMs from C57BL/6J mice upregulated the expression of ER stress markers such as Bip (Hspa5), IRE1α (Ern1), JNK (Mapk8), ASK1 (Map3k5), and XBP1 (Xbp1) (Figure 5B,D). However, the expression of these ER stress markers was downregulated in TM6SF2-deficient macrophages ( Figure 5A,C), indicating that TM6SF2 positively regulates ER stress pathway in macrophages.

TM6SF2 Activates NF-κB in Macrophages
NF-κB pathway can be activated by many atherosclerotic risk factors as well as ER stress, intensifying inflammatory gene expression and inflammation cascade. To further investigate the effects of TM6SF2 on inflammatory responses and ER stress through regulating the NF-κB signaling pathway in macrophages, we determined the phosphorylation of p65 at S536, which contributes to nuclear translocation and the induction of target gene expression. As shown in Figure 6, oxLDL induced phosphorylation of p65. Knockdown of TM6SF2 abolished the effect of oxLDL on p65 activation as well as suppressed the expression of p65 ( Figure 6A), indicating that TM6SF2 contributes to the activation of NF-κB pathway in macrophages. Indeed, the overexpression of TM6SF2 enhanced the activation of p65 in THP-1-derived macrophages ( Figure 6B), consistent with the findings of TM6SF2 in inflammatory responses and ER stress. NF-κB pathway can be activated by many atherosclerotic risk factors as well as ER stress, intensifying inflammatory gene expression and inflammation cascade. To further investigate the effects of TM6SF2 on inflammatory responses and ER stress through regulating the NF-κB signaling pathway in macrophages, we determined the phosphorylation of p65 at S536, which contributes to nuclear translocation and the induction of target gene expression. As shown in Figure 6, oxLDL induced phosphorylation of p65. Knockdown of TM6SF2 abolished the effect of oxLDL on p65 activation as well as suppressed the expression of p65 ( Figure 6A), indicating that TM6SF2 contributes to the activation of NF-κB pathway in macrophages. Indeed, the overexpression of TM6SF2 enhanced the activation of p65 in THP-1-derived macrophages ( Figure 6B), consistent with the findings of TM6SF2 in inflammatory responses and ER stress.

Discussion
Exome-wide association studies and genome-wide association studies have identified that the TM6SF2 variant (encoding p. Glu167Lys, E167K) is associated with decreased myocardial infarction risk and increased fatty liver disease [26,27]. Here, our data demonstrate that myeloid cell-specific Tm6sf2 deficiency inhibits atherosclerosis development independent of the circulating lipid levels. Regarding the mechanism, we discovered that TM6SF2 contributes to inflammatory responses, foam cell formation, and ER stress in macrophages. Therefore, this study determined the role of the macrophage TM6SF2 in atherosclerosis and put forward a new target for treating this disease.
Cardiovascular disease is one of the leading causes of death worldwide [28,29]. The success of genetic studies has significantly advanced our understanding of the etiology of ASCVD and provided new therapeutic strategies, particularly via identifying drug targets

Discussion
Exome-wide association studies and genome-wide association studies have identified that the TM6SF2 variant (encoding p. Glu167Lys, E167K) is associated with decreased myocardial infarction risk and increased fatty liver disease [26,27]. Here, our data demonstrate that myeloid cell-specific Tm6sf2 deficiency inhibits atherosclerosis development independent of the circulating lipid levels. Regarding the mechanism, we discovered that TM6SF2 contributes to inflammatory responses, foam cell formation, and ER stress in macrophages. Therefore, this study determined the role of the macrophage TM6SF2 in atherosclerosis and put forward a new target for treating this disease.
Cardiovascular disease is one of the leading causes of death worldwide [28,29]. The success of genetic studies has significantly advanced our understanding of the etiology of ASCVD and provided new therapeutic strategies, particularly via identifying drug targets for ASCVD treatment. One typical representative is proprotein convertase subtilisin/kexin type 9 serine protease (PCSK9), emerging as an effective cholesterol-lowering target [30,31]. Human genetic studies have documented that the TM6SF2 variant E167K is associated with decreased TC and TG levels and reduced cardiovascular arterial disease risk [7][8][9], presenting cardiovascular protective properties. However, the same variant in TM6SF2 is also associated with increased risks of nonalcoholic fatty liver disease (NAFLD) and diabetes. The E167K mutation in TM6SF2 yields decreased protein levels due to increased degradation [7]. The overexpression of loss-of-function TM6SF2 reduces plasma lipid levels and atherosclerosis development [32]. Knockdown or inactivation of TM6SF2 attenuates apolipoprotein B lipidation and VLDL secretion from hepatocytes, decreasing circulating TC and TG levels while causing lipid accumulation in the liver [13,32,33]. The effects of TM6SF2 in the liver are confirmed in TM6SF2-E167K knock-in mice [25]. Liver-specific Tm6sf2-deficient mice spontaneously exhibit hepatic steatosis, fibrosis, and accelerated development of hepatocellular cancer [11,13,25]. Therefore, the increased NAFLD risk by TM6SF2 inactivation limits its application for cardiovascular arterial disease treatment via regulating lipid metabolism in the liver. Unlike the liver-specific or systemic studies, our data demonstrate that myeloid-specific Tm6sf2 deficiency attenuates atherosclerosis development independent of circulating lipid levels. As expected, we did not observe any changes in body weight, blood glucose, liver weight, or body weight, indicating that myeloid cell Tm6sf2 deficiency does not affect the lipid and glucose metabolism in the liver.
Lipoproteins accumulated in the subendothelial area recruit monocyte/macrophages, T-cells, mast cells, and other immune cells to the lesion sites, where monocyte-differentiated macrophages take up cholesterol and become foam cells, further promoting atherosclerotic plaque progression [21,34,35]. We observed reduced atherosclerotic plaque area and Mac2positive area in myeloid cell-specific Tm6sf2-deficient mice, suggesting that a deficiency of Tm6sf2 prevents macrophage infiltration into the aortic wall. Vascular chronic inflammation further recruits monocytes and macrophages into atherosclerotic lesions. Macrophages uptake cholesterol to form foam cells, which is the key element during the plaque formation and progression of atherosclerosis [4,36]. Our data show that Tm6sf2-deficiency affects the inflammation-associated signaling pathways, such as via the regulation of cytokine production, inflammatory response, and cytokine-mediated signaling pathway, which are all involved in macrophage movement and infiltration. oxLDL, a major risk factor for the initiation and development of atherosclerosis, upregulated TM6SF2 expression in macrophages and increased TM6SF2 expression enhances oxLDL-induced inflammatory responses and foam cell formation, showing a vicious circle of positive feedback. TM6SF2-deficiency blocks this feedback and reduces inflammatory responses and foam cell formation in macrophages, presenting potential protective mechanisms underlying the alleviation of atherosclerosis in Tm6sf2 mKO mice. In endothelial cells, disturbed flow activates crucial proatherogenic pathways [37]. Single-cell RNA sequencing and scATACseq study have demonstrated that TM6SF2 expression is upregulated during disturbed flow-induced endothelial cell phenotype change from atheroprotective phenotypes to proinflammatory cells [38]. These results suggest that TM6SF2 in non-hepatocytes, such as macrophages and endothelial cells, plays a role in atherosclerosis development.
In atherosclerotic lesions, macrophages are the prominent cells undergoing ER stress. Lipoprotein and inflammatory cytokines induce ER stress in macrophages and foam cells, causing apoptosis and recruiting more macrophages to the atherosclerotic sites to accelerate progression of atherosclerosis [39]. TM6SF2 is localized in the ER membrane and ERGIC (ER-Golgi intermediate compartment) [40]. TM6SF2 deletion or loss-of-function mutation affects ER ultrastructure and induces ER stress markers in zebrafish and human livers due to lipid accumulation under physiological conditions [33]. However, loss-of-function of TM6SF2 in E167K knock-in mice are resistant to high-fat-diet-induced ER stress in the liver [25]. Consistent with the findings from the high-fat-diet challenge study [25], after a 12-week WD challenge, the expression of ER stress markers are downregulated in BMDMs from Tm6sf2 mKO mice compared with that from littermate controls. We also confirmed that oxLDL upregulates the expression of ER stress markers, which can be enhanced by overexpression of TM6SF2 in macrophages, while TM6SF2 deficiency largely abolishes its effects, consistent with the protective effects during atherosclerosis development. These findings show that TM6SF2 may function differently in regulating ER stress under adaptive processes and pathological conditions.
In response to pro-atherosclerotic and stress stimuli, NF-κB members are activated, and the NF-κB signaling pathway is vital for inflammatory responses and ER stress [23,41]. ER stress also activates the NF-κB signaling pathway and causes the translocation to the nucleus to activate pro-inflammatory gene expression [24,42,43]. Our transcriptome analysis revealed that TM6SF2 contributes to the activation of the NF-κB signaling pathway in macrophages. The phosphorylation of Ser536 in the cytosolic p65 promotes its nuclear translocation and facilitates p65 binding to specific promoter sequences [44]. We found that oxLDL activates NF-κB, reflected in the increased phosphorylation of the NF-κB p65 subunit, in a TM6SF2-dependent manner. Furthermore, we found that knockdown of TM6SF2 decreased the total p65 level in macrophages. As TM6SF2 is an ER membrane protein, it may regulate the p65 level through posttranslational modifications, such as degradation, ubiquitination, or acetylation. Further investigation of the role of TM6SF2 in p65 expression and activation is warranted in our follow-up studies.
It should be noted that oxLDL induces the expression of TM6SF2 in macrophages. Further examination of the intracellular mechanisms underlying the regulation of TM6SF2 expression will be necessary for clarifying the function of TM6SF2 in atherosclerosis development fully. Whether TM6SF2 in other vascular cells, such as endothelial cells and vascular smooth muscle cells, plays a role in atherosclerosis is unknown. Our future studies will explore the functions of TM6SF2 in atherosclerosis using endothelial cell or vascular smooth muscle-specific Tm6sf2 knockout mouse models, as well as loss of function of TM6SF2 in E167K knock-in mice.

Conclusions
In conclusion, our findings demonstrate that myeloid cell-specific Tm6sf2 deficiency inhibits atherosclerosis development in mice without affecting lipid and glucose metabolism. We discovered that a deficiency of TM6SF2 in macrophages inhibits inflammatory responses, foam cell formation, and ER stress upon oxLDL stimulation. Therefore, the repression of TM6SF2 in macrophages could be a new therapeutic target for atherosclerosis treatment.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cells11182877/s1, Figure S1: Strategy for the conditional disruption of the Tm6sf2 gene 2; Figure S2: Blood Glucose (A) and Insulin (B) Levels in Myeloid-Specific Tm6sf2 Knockout Mice; Figure S3: Tm6sf2 Knockout in Myeloid Cells Inhibits Atherosclerosis Development in Female Mice; Figure S4: TM6SF2 levels upon oxLDL Treatment; Figure S5: Expression levels of TM6SF2 in siRNA-induced knockdown or AdTM6SF2-mediated overexpression in macrophages; Table S1: Sequences of the primers used in the study;  Funding: This work was supported in part by the National Institutes of Health (grants HL137214, HL134569, and HL159871 to Y.E.C.; HL138139 to J.Z.; and HL151524 to L.C.).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Not applicable.