HDAC9 represses cholesterol efflux and generation of alternatively activated macrophages in atherosclerosis development

Abstract

Objective

Recent genome-wide association studies revealed that a genetic variant in the loci corresponding to histone deacetylase 9 (HDAC9) is associated with large vessel stroke. HDAC9 expression was upregulated in human atherosclerotic plaques in different arteries. The molecular mechanisms how HDAC9 might increase atherosclerosis is not clear.

Approach and Results

In this study, we show that systemic and bone marrow cell deletion of HDAC9 decreased atherosclerosis in LDLr^−/− mice with minimal effect on plasma lipid concentrations. HDAC9 deletion resulted upregulation of lipid homeostatic genes, downregulation of inflammatory genes, and polarization towards an M2 phenotype via increased accumulation of total acetylated H3 and H3K9 at the promoters of ABCA1, ABCG1, and PPAR-γ in macrophages.

Conclusions

We conclude that macrophage HDAC9 upregulation is atherogenic via suppression of cholesterol efflux and generation of alternatively activated macrophages in atherosclerosis.

Introduction

Recent genome-wide association studies revealed that a genetic variant in the loci corresponding to histone deacetylase 9 (HDAC9) is associated with large vessel stroke¹. Moreover, HDAC9 is also a common genetic variant that share genetic susceptibility to ischemic stroke and coronary artery disease². HDAC9 expression was upregulated in human atherosclerotic plaques in different arteries³. The molecular mechanisms how HDAC9 region might increase the risk of atherosclerosis is unknown.

There are eighteen mammalian histone deacetylases (HDACs), which fall into four classes on the basis of their structural and biochemical characteristics⁴. HDAC9 is expressed in heart, pancreatic islets, neuron, spinal cord, teeth, smooth and skeletal muscles, T lymphocytes, endothelium and adipose tissues⁴. One of the best characterized mechanisms of action of HDAC9 is its ability to partner with MEF-2 and repress MEF-2 activity⁴. HDAC9 deacetylates Lys¹¹⁶ on ataxia telangiectasia group D-complementing (ATDC; also called TRIM29) protein, prevents ATDC from binding p53, and consequently leads to activation of p53-inducible genes⁵. Macrophages consist of at least two subgroups M1 and M2. Whereas M1 macrophages are pro-inflammatory and have a central role in atherosclerosis development and plaque rupture, M2 macrophages are associated with response to anti-inflammatory reactions, tissue remodeling, fibrosis and atherosclerosis regression⁶. Recent studies demonstrate that among HDACs, HDAC3 and HDAC7 have been identified to play a key role in inflammatory gene expression program and alternative activation of macrophages^7–9. However, the role of HDAC9 in these processes in macrophages is unknown. Understanding whether HDAC9 plays a role in macrophage development and the pathogenesis of atherosclerosis will provide the rationale for the development of selective histone deacetylase 9 inhibitors that target macrophages for the treatment and prevention of atherosclerosis because of its significance with GWAS study.

In the present study, we report that HDAC9 is expressed in macrophages and its role in cholesterol homeostasis and inflammation in macrophages and on atherosclerosis development. Our experiments revealed that systemic and bone marrow cell HDAC9 deficiency in LDLr^−/− mice reduced atherosclerosis. HDAC9 deletion resulted upregulation of lipid homeostatic genes, downregulation of inflammatory genes, and polarization of macrophages towards M2-like phenotype via increased accumulation of acetylated H3 at the promoters of ABCA1, ABCG1, and PPAR-γ in macrophages. We conclude that macrophage HDAC9 upregulation is atherogenic via suppression of cholesterol efflux and generation of alternatively macrophages in atherosclerosis.

Materials and Methods

Materials and Methods are available in the Online–only Supplement. Link to the online supplement.

Results

Systemic deletion of HDAC9 reduces atherosclerosis in male and female LDLr^−/− mice

Initially, we utilized animal model of atherosclerosis to understand the underlying molecular pathways of HDAC9 in atherogenesis. To examine whether systemic deficiency of HDAC9 inhibits atherogenesis, male and female LDLr^−/− (SKO) and LDLr^−/− HDAC9^−/− (DKO) mice were generated and fed an atherogenic diet (Supplementary Table 1) for 16 weeks. Female and male SKO and DKO mice had similar body weight and plasma lipids concentrations on a chow diet (Supplementary Fig. 1).

Plasma triglyceride (TG) and free cholesterol (FC) levels were decreased in female DKO mice compared with SKO female mice (Fig. 1A); however, no difference was observed in male mice (Supplementary Fig. 2A). Total cholesterol (TC), cholesterol ester (CE) and phospholipid (PL) concentrations were similar for SKO vs. DKO mice (Fig. 1A, Supplementary Fig. 2A). VLDL cholesterol was decreased, whereas LDL and HDL were increased in DKO female, but not male DKO mice, compared to SKO (Fig. 1B–C, Supplementary Fig. 2B–C).

Figure 1. Systemic HDAC9 deficiency reduces atherosclerosis in female LDLr^−/− mice.

Plasma lipid and atherosclerosis measurements were made after 16 weeks of atherogenic diet consumption by SKO (LDLr^−/−) and DKO (LDLr^−/−HDAC9^−/−) mice.

(A) Plasma lipid concentrations.

(B–C) Cholesterol distribution in VLDL, LDL and HDL by FPLC analysis of pooled plasma from the terminal bleeds.

(D) Representative examples of light microscopic images taken of gross dissected aortas in situ. Atherosclerotic lesions are visualized as white areas, denoted by arrows, inside the arteries.

(E–F) Representative en face image (E) and surface lesion quantification (F) of aortas.

(G–H) Representative cross-sectional image and quantification of aortic root stained with Oil Red O (G–H.

Data are shown as the mean ± SD. * p<0.05; ** p<0.01

Compared to SKO mice, DKO mice had visibly decreased lesions in the unstained aortic arch in situ (Fig. 1D, Supplementary Fig. 2D), by en face analysis (Fig. 1E–F, Supplementary Fig.2E–F), and by morphometric analysis of Oil Red O-stained aortic root sections (Fig. 1G–H, Supplementary Fig. 2G–H). Smooth muscle cells synthesize the interstitial collagens which are responsible for resistance to rupture of atherosclerotic plaques. Pathological studies in human ruptured atherosclerotic plaques have shown a thin fibrous cap poor in smooth muscle cells and collagen^{10, 11}. To determine whether HDAC9 deficiency promotes plaque stability, we measured smooth muscle cell and collagen content in aortic root atherosclerotic lesions by immunohistochemistry. There was a significant increase in smooth muscle cells and an increased trend in collagen deposition in aortic atherosclerotic plaques in DKO compared to SKO mice (Supplementary Fig. 3A–D).

We determined the effect of HDAC9 deletion in genes expression levels in lesions that play key roles in atherogenesis in vivo. Real time PCR analysis was done in pooled samples of aortic roots from the SKO and DKO mice. Aortic roots of DKO mice had increased mRNA expression of ABCA1, ABCG1, and arginase-1, decreased expression of IL-1β and MCP-1, but no change in expression of CD36, SRA, TNF-α and iNOS compared to SKO mice (Supplementary Fig. 4).

Deletion of HDAC9 in bone marrow cells inhibits the formation of atherosclerosis

HDAC9 is expressed in multiple tissues including endothelial cells, T lymphocytes and adipose tissues^{4, 12}. To test the hypothesis whether HDAC9 expressed on hematopoietic cells (e.g., monocytes/MΦ, lymphocytes, and platelets) vs., endothelial cells or all the cell types those promote the development of atherosclerosis lesions, we transplanted male SKO and DKO bone marrow (BM) into female LDLr^−/− recipient mice (Supplementary Fig. 5). Five weeks after transplantation, the recipient mice were fed an atherogenic diet for 16 weeks and sacrificed. Plasma TC, FC, and CE levels were similar, but TG concentrations were reduced, in LDLr^−/− mice reconstituted with DKO vs. SKO bone marrow (Fig. 2A). VLDL cholesterol was reduced, whereas HDL was increased in mice receiving DKO bone marrow (Fig. 2B–C). LDLr^−/− mice transplanted with DKO vs. SKO bone marrow had decreased visible atherosclerotic plaques in the aortic arch (Fig. 2D), aortic surface lesion area (Fig. 2E–F), and aorta root intimal area and lipid staining (Fig. 2G–I). Although BM transplantation experiments suggested the possible deletion of macrophage HDAC9 was sufficient to decrease atherosclerosis, a beneficial role for other BM cell types in this process cannot be excluded.

Figure 2. HDAC9 deficiency in hematopoietic cells reduces atherosclerosis in LDLr^−/− mice.

Plasma lipid and atherosclerosis measurements were made after 16 weeks of atherogenic diet consumption. Irradiated female LDLr^−/− mice were transplanted with bone marrow cells isolated from male SKO (white bar) and DKO (black bar) mice.

(A) Plasma lipid concentrations.

(B–C) Cholesterol distribution in VLDL, LDL and HDL fractions by FPLC analysis of pooled plasma from the terminal bleeds.

(D) Representative examples of light microscopic images taken of gross dissected aortas in situ. Atherosclerotic lesions are visualized as white areas, denoted by arrows.

(E–I) Representative en face image (E), quantification of en face surface lesion area (F), cross-sectional image of aortic root stained with Oil Red O (G), intimal area (H), and Oil Red O positive area (I). For (A) and (C), genotypes of the donor cells are indicated below the figures.

Data are shown as the mean ± SD. * p<0.05; ** p<0.01

HDAC9 expression is increased during macrophage differentiation

To determine the role of HDAC9 biologic and pathologic functions in macrophages, we performed following experiments. We first determined HDAC9 expression in human THP-1 monocytes during in vitro differentiation into macrophages, and mouse bone marrow-derived macrophages (BMDM). HDAC9 mRNA was most highly expressed in differentiated macrophages (Fig. 3A and B). HDAC9 is alternatively spliced to generate different proteins. Among these, two major isoforms (HDAC9- a form containing HDAC domain and a truncated form HDRP/MITR without HDAC domain but acquire deacetylase activity by recruitment of HDAC1 or HDAC3) expressed in tissue specific manner^{13, 14}. These major isoforms were expressed in mouse and human macrophages (Fig 3C–D). HDAC9 mRNA expression is further increased in response to ox-LDL, ac-LDL and TLR (LTA, LPS, flagellin but not CpG DNA) signals, thioglycollate-elicited peritoneal macrophages (TEPMs) (Supplemental Fig. 6).

Figure 3. HDAC9 is overexpressed during macrophage differentiation.

(A) Real-time PCR analysis of Class IIa HDAC mRNAs in THP-1 monocyte cells and differentiated macrophages. THP1 cells were cultured with PMA (20 ng/mL) for 4 days to differentiate to macrophages. Data were normalized using the housekeeping gene GAPDH. mRNA expression was calculated relative to undifferentiated THP1 cells (controls).

(B) HDAC9 mRNA level is increased during differentiation of mouse bone marrow-derived cells to macrophages. Bone marrow-derived cells from C57BL/6J mice were cultured with LCM media and cells were isolated at given time points after differentiation. Data were normalized using the housekeeping gene GAPDH. mRNA expression was calculated relative to differentiated macrophages at day 1 (control).

(C) RT PCR analysis demonstrates two major HDAC9 isoforms (HDAC9 and HDRP) in mouse macrophages. (1) Mouse bone marrow cells; (2) Macrophages differentiated from mouse bone marrow cells; and (3) TEPMs from an LDLr^−/− mouse.

(D) RT PCR analysis demonstrates two major HDAC9 isoforms (HDAC9 and HDRP) in human macrophages. (1) THP1 cells; (2) Macrophages differentiated from THP1 cells; (3) Human monocytes; and (4) Macrophages differentiated from human monocytes.

HDAC9 deficiency in macrophages results increased cholesterol efflux via increased expression of ABCA1 and ABCG1 via accumulation of acetylated histone at H3 and H4 at promoters

Cholesterol efflux capacity from macrophages via ABCA1 and ABCG1 has a strong inverse relation with atherosclerosis development¹⁵. To determine the role of macrophages HDAC9 in cholesterol efflux pathway, we performed following experiments. HDAC9-deficient macrophages demonstrated increased expression of the cholesterol efflux genes ABCA1 and ABCG1 with acLDL and LXR ligand stimulation compared to WT mice (Both HDAC9-deficient and wild type mice were on C57/Bl6 back ground fed on chow diet) in vitro (Supplemental Fig. 7 A&B). ApoA1- and HDL-mediated cholesterol efflux was increased in HDAC9-deficient vs. WT macrophages Supplemental Fig. 7 C&D). ACAT inhibition further increased ApoA1- but, not HDL-mediated cholesterol efflux in HDAC9-deficient (Supplemental Fig. 7E&F). To determine whether cholesterol efflux genes were upregulated in macrophages from atherogenic diet-fed mice, TEPMs were isolated from SKO and DKO mice fed an atherogenic diet for 16 weeks. TEPMs from DKO mice had significantly less CE compared to foam cells from SKO mice (Fig. 4A), consistent with the finding of decreased atherosclerosis in DKO mice. Expression of ABCA1, ABCG1 in DKO macrophages was increased compared to SKO macrophages (Fig. 4B–D). HDAC9 regulates gene expression via multiple mechanisms, including protein-protein interaction, histone and non-histone protein acetylation, but these pathways are not necessarily mutually exclusive^{5, 16, 17}. Western blot analyses revealed no global changes in histone acetylation in HDAC9-deficient macrophages (Supplementary Fig. 8). Control of inducible gene expression is dependent upon signal induced transcriptional elongation. This process is facilitated by deposition of histone acetylation mark associated with transcriptional activation at the promoters¹⁸. Using stable isotope labeling in combination with mass spectrometry analysis in splenocytes from KO and MRL/lprHDAC9^+/+ (WT) mice, we had previously demonstrated that Lysine residues H3K9 and H3K18 were hyperacetylated (~5-fold) in splenocytes from HDAC9-deficient mice¹⁷. To investigate the consequences of HDAC9 deletion in macrophages and these histone acetylation marks, we performed ChIP assay using total H3, H4, site specific H3K9ac and H3K18 ab. Quantitative ChIP assays demonstrated increased levels of total H3, H4 and H3K9 acetylation but not H3K18 acetylation at the promoters of ABCA1 and ABCG1 in HDAC9-deficient macrophages compared to controls fed on atherogenic diet (Fig. 4E–F). We also performed ChIP assays in BMDMs from these two groups of mice loaded with acLDL in vitro. Similar results were observed in ChiP assay as above (Supplemental Fig 9B&C). There was no increase in deposition of these acetylation marks in SRA promoter where transcription was unchanged between HDAC9 vs. wild type macrophages (Supplemental Fig 9 D).

Figure 4. HDAC9 regulates cholesterol homeostasis via alteration of ABCA1 and ABCG1 promoter architecture.

(A) Quantification of TC, CE and FC in thioglycollate-elicit peritoneal macrophages isolated from SKO and DKO mice fed with 16-week atherogenic diet.

(B–D) Thioglycollate-elicit peritoneal macrophages isolated from SKO and DKO mice fed with 16-week atherogenic diet. ABCA1 and ABCG1 mRNA level (B), protein level (C) and Western Blot quantification (D).

(E–F) Quantitative ChIP assays were performed on the promoter regions of ABCA1 and ABCG1 using anti-Ac-H3,Ac-H4, Ac-H3K9, Ac-H3K18 or IgG (control) for immunoprecipitation. Non-immunoprecipitated sample served as an input control.

All assays performed in triplicate and are representative of three independent experiments. Data are presented as mean ± SD. * p<0.05; ** p<0.01

HDAC9 deficiency in macrophages promotes M2 polarization and decrease M1 inflammatory genes via upregulation of PPAR-γ via chromatin remodeling

To determine the role of HDAC9 in generation of M1 and M2 macrophages, we performed following experiments. Ex vivo stimulation with LPS demonstrated that DKO TEPMs produced significantly less pro-inflammatory cytokines (TNF-α, IFN-γ, MCP-1, IL-6, iNOS and IL-1β) compared with SKO macrophages (Fig. 5A–B). TEPMs from HDAC9-deficient mice had a significant increase in subset of genes characteristics of alternatively activated, M2-like, macrophages (MGL-1, arginase-1, and IL-10) in the basal states compared to wild type mice fed on atherogenic diet (Fig. 5C). Compared with wild type mice macrophages, the macrophages from HDAC9-deficient mice fed on chow diet had a significant increase in basal, IL-4 and PPAR-γ agonists (rosiglitazone) stimulated expression of CD206, MGL-1, MGL-2, arginase-1, Chi3l3 and IL-10 mRNA (Supplemental Fig. 10). The upregulation of M2 and downregulation of M1 genes in HDAC9-deficient macrophages are probably through PPAR-γ pathway¹⁹. Expression of PPAR-γ in DKO macrophages was increased compared to SKO macrophages (Fig. 5D). Quantitative ChIP assays demonstrated increased levels of total H3, H4,H3K9, but not H3K18acetylation at the promoters of PPAR-γ (Fig. 5E) but not at the arginase-1 promoter (Supplemental Fig 11).

Figure 5. Macrophage HDAC9 deficiency reduces LPS-mediated inflammatory cytokines secretion and increases M2 polarization through changing PPAR-γ1 promoter architecture.

(A–C) Thioglycollate-elicit SKO and DKO peritoneal macrophages were isolated from hypercholesterolemic mice and stimulated with LPS for 24 hours, M1 marks were detected at mRNA level (A) and protein level (B), and M2 markers were checked at mRNA level without LPS stimulation (C).

(D) PPAR-γ1 mRNA level in thioglycollate-elicit peritoneal macrophages isolated from SKO and DKO mice fed with 16-week atherogenic diet.

(E) Quantitative ChIP assays were performed on the promoter regions of PPAR-γ1 using anti-Ac-H3,Ac-H4, Ac-H3K9, Ac-H3K18 antibodies or IgG (control) for immunoprecipitation. Non-immunoprecipitated sample served as an input control.

All assays performed in triplicate and are representative of three independent experiments. Data are presented as mean ± SD. * p<0.05; ** p<0.01

These in vitro and ex vivo experiments in macrophages support the concept that increased HDAC9 expression in macrophages is atherogenic via suppression of cholesterol efflux and generation of alternatively activated macrophages via decreased accumulation of histone acetylation at ABCA1, ABCG1 and PPAR γ promoters. The consequences of HDAC9 deletion in macrophages on a genome-wide scale in response to different Toll-like receptor agonists or infections need further investigations.

Discussion

In this study, we report several new findings. Among class IIa HDACs, HDAC9 is most abundantly expressed during macrophage differentiation. Systemic and macrophage HDAC9 deficiency reduces atherosclerosis development in different sites in LDLr^−/− mice fed an atherogenic diet. The molecular mechanisms behind the decreased atherosclerosis in HDAC9 deficient mice are likely multi-factorial, including increased macrophage cholesterol efflux by ABCA1 and ABCG1 and phenotypic switching of macrophages from a pro-inflammatory M1 to a less inflammatory M2 state via PPAR-γ.

We have not observed cardiac hypertrophy or polydactyly in DKO mice in contrast to previously reported studies in HDAC9 KO mice^{20, 21}. The difference between the phenotypes observed between ours and others are most likely due to difference in background(C57Bl6 vs mixed 129 and C57Bl6) and age of the mice studied. Interestingly, in contrast to Chatterjee TK et al study¹², the double knockout (LDLr^−/−HDAC9^−/−) mice have increased body, liver, and adipose tissue weight compared to single knockout (LDLr^−/−) mice on the atherogenic diet; these phenotypes were neither observed in chow-fed mice nor during atherogenic diet feeding of LDLr^−/− mice transplanted with HDAC9 knockout vs. wild-type bone marrow (data not shown). The difference between the phenotypes observed between theirs and ours are most likely due to difference in background (C57Bl6 vs LDLr) and diet. The mechanisms behind the phenotypes observed with systemic deficiency of HDAC9 in atherogenic diet-fed LDLr^−/− mice are currently unknown.

Recent genome-wide association studies revealed that a genetic variant in the loci corresponding HDAC9 is associated with large vessel stroke¹. Moreover, HDAC9 is also a common genetic variant that share genetic susceptibility to ischemic stroke and coronary artery disease. HDAC9 expression was upregulated in human atherosclerotic plaques in different arteries³. HDAC9 knockout mice are protected from adipose tissue dysfunction and systemic metabolic disease during high fat feeding¹².

Moreover, HDAC9 deficiency increases regulatory T cells and decrease effector T cells^{17, 22}. Finally, HDAC9 promotes angiogenesis by targeting the antiangiogenic MicroRNA-17-92 cluster in endothelial cells²³. These studies consistent with our results to support the concept of inhibition of HDAC9 may have beneficial effect in atherosclerosis development.

Lund et al demonstrated that epigenetic changes, such as DNA methylation polymorphisms, preceded any histological sign of atherosclerosis in apoE deficient mice, and human THP-1 cells cultured with mixtures of different proportions of lipoproteins (VLDL, LDL and HDL) decreased histone H4 acetylation²⁴. Studies in endothelial cells have demonstrated opposing roles of HDACs in endothelial function²⁵. HDAC3 expression protected endothelial integrity and HDAC3 knockdown increased atherosclerosis in aortic isografts of apoE mice²⁶. Laminar blood flow inhibited HDAC5 expression, resulting in decreased binding of monocytes to endothelial cells²⁷. HDAC2 and HDAC5 played an important role in smooth muscle cell differentiation^{27, 28}. Moreover, short interfering RNA-mediated knockdown of HDAC1, 2 or 3, and pharmacologic inhibition of HDAC by scriptaid prevented smooth muscle cell proliferation and neointima formation²⁹. HDAC3 and HDAC7 promote TLR-4 dependent pro-inflammatory gene expression whereas HDAC3 promotes macrophage alternate activation. Finally, pan-HDAC inhibitors have both anti- and pro-inflammatory effects on macrophages and prevent smooth muscle cells migration in vitro³⁰.Importantly, endogenous HDAC inhibitor β-hydroxybutyrate protects against oxidative stress by upregulating oxidative stress resistance genes Foxo3a and MT2³¹. Based upon these opposing roles of HDACs in different cell types that are involved in atherosclerosis development, it is not surprising that trichostatin A (TSA), which inhibits both class I and class II HDACs, increased atherosclerosis in LDLr^−/− mice³². It is also possible that the high concentration of dietary cholesterol (1.25%) overwhelms a beneficial effect of TSA.

The nuclear receptors (PPARs, LXRs and RXRs) not only influence lipid metabolism at the systemic level, but also regulate lipid homeostasis and inflammation in macrophages³³. There is considerable interest in the development of selective PPAR and LXR agonists for the treatment of atherosclerosis^33–35. Active repression of LXR target genes, such as ABCA1 and ABCG1, is mediated, in part, by interaction with co-repressors NCoR and SMART to repress basal expression of target promoters via recruitment of HDAC3³⁶. Intriguingly, HDAC9 interacts and co-localizes in vivo with a number of transcriptional repressors or co-repressor including NCoR¹³. HDAC9, which represses these nuclear receptors by forming multi-protein complexes, antagonizes the anti-atherogenic effect of natural and synthetic nuclear receptor ligands. HDAC9 deletion relieves inhibition and results in increased transcription of ABCA1, ABCG1, and PPAR-γ in macrophages. Activation of the genes in macrophages decreased foam cell formation and inflammation via increased cholesterol efflux and phenotypic switching of macrophages from a pro-inflammatory M1 to anti-inflammatory M2 phenotype ultimately with decreased atherosclerosis phenotype in LDLr^−/− mice. Our findings are consistent with recent study by Li, etal that macrophage specific deletion of NCoR results anti-inflammatory phenotype and generation of M2 macrophages via increase in histone acetylation at certain promoters³⁷.

Although, it is not clear from our study whether enzyme activity of HDAC9 or protein itself is required for atherogenesis, it is tempting to speculate that HDAC9 inhibition in macrophages may provide a therapeutic benefit in atherosclerosis in humans. HDAC9 is recently identified a common genetic variant that share genetic susceptibility to large artery stroke and coronary artery disease¹. Moreover, HDAC9 expression was upregulated in human atherosclerotic plaques in different arteries³. Our study provides a mechanistic clue how HDAC9 gene variant may cause atherosclerosis. We envision that this study may provide the foundation for development of HDAC9 isoform specific inhibitors that can selectively be delivered to macrophages via chemical conjugates technology may provide therapeutic benefit in individuals with coronary artery disease and stroke with genetically defined HDAC9 population.

Significance.

Recently, genome-wide association studies revealed that a genetic variant in the loci corresponding to histone deacetylase 9 (HDAC9) is associated with large vessel stroke. Moreover, HDAC9 expression was identified in human atherosclerotic plaques in different arteries. However, the molecular mechanism of HDAC9 on atherogenesis is unknown. In this study, we demonstrate that HDAC9 in macrophage is atherogenic. Systemic and bone marrow cell deletion of HDAC9 decreases atherosclerosis in LDLr^−/− mice. Macrophages lacking HDAC9 suppress foam cell formation by increasing cholesterol efflux via increased expression of macrophage ABCA1 and ABCG1 gene expression. Moreover, macrophages lacking HDAC9 produce less inflammatory mediators and polarize towards M2-like macrophages. These important finding points towards development of epigenetic therapy for atherosclerosis by using small molecule inhibitors that targets HDAC9 isoform in macrophages.

Abbreviations

HDACs

Histone Deacetylases

HDAC9

Histone Deacetylase 9

ATDC

Ataxia Telangiectasia group D-complementing

TG

Triglyceride

FC

Free Cholesterol

TC

Total Cholesterol

CE

Cholesterol Ester

PL

Phospholipid

BMDM

Bone Marrow Derived Macrophages

TEPMs

Thioglycollate-Elicited Peritoneal Macrophages

SKO

LDLr−/−

DKO

LDLr−/− HDAC9−/−