INCB054329 | Bace Signal

BET bromodomain containing epigenetic reader proteins regulate vascular smooth muscle cell proliferation and neointima formation

Jochen Dutzmann, MD1,2,*, Marco Haertle2,*, Jan-Marcus Daniel, MD1,2, Frederik Kloss2, Robert-Jonathan Musmann2, Katrin Kalies, MSc1, Kai Knöpp, MD1,2, Claudia

Abstract

Aims: Recent studies revealed that the bromodomain and extraterminal (BET) epigenetic reader proteins resemble key regulators in the underlying pathophysiology of cancer, diabetes or cardiovascular disease. However, whether they also regulate vascular remodeling processes by direct effects on vascular cells is unknown. In this study we investigated the effects of the BET proteins on human smooth muscle cell (SMC) function in vitro and neointima formation in response to vascular injury in vivo. Methods and results: Selective inhibition of BETs by the small molecule (+)-JQ1 dose dependently reduced proliferation and migration of SMCs without apoptotic or toxic effects. Flow cytometric analysis revealed a cell cycle arrest in the G0/G1 phase in the presence of (+)-JQ1. Microarray- and pathway-analysis revealed a substantial transcriptional regulation of gene sets controlled by the FOXO1-transcription factor. Silencing of the most significantly regulated FOXO1-dependent gene, CDKN1A, abolished the antiproliferative effects. Immunohistochemical colocalization, coimmunoprecipitation and promoter binding ELISA assay data confirmed that the BET protein BRD4 directly binds to FOXO1 and regulates FOXO1 transactivational capacity. In vivo, local application of (+)-JQ1 significantly attenuated SMC proliferation and neointimal lesion formation following wire-induced injury of the femoral artery in C57BL/6 mice.
Conclusions: Inhibition of the BET containing protein BRD4 after vascular injury by (+)-JQ1 restores FOXO1 transactivational activity, subsequent CDKN1A expression, cell cycle arrest and thus prevents SMC proliferation in vitro and neointima formation in vivo. Inhibition of BET epigenetic reader proteins might thus represent a promising therapeutic strategy to prevent adverse vascular remodeling.

Translational Perspective

Here we demonstrate for the first time that the bromodomain and extraterminal (BET) epigenetic reader proteins are important regulators of smooth muscle cell function. Currently available, highly specific inhibitors of these proteins like (+)-JQ1 potently prevent SMC proliferation after acute vascular injury, limiting neointima formation and vessel re-occlusion following interventional treatment. Interfering with BET-function by using such inhibitors which are already in clinical use for the treatment of different diseases may represent an attractive and approach with high translational potential to support current interventional therapies by minimizing negative vascular remodeling processes.

1 Introduction

The process of neointima formation is common to various forms of vascular diseases such as atherosclerosis, in-stent restenosis, vein bypass graft failure, and transplant vasculopathy. The neointimal layer is formed by activated medial smooth muscle cells (SMCs) that proliferate and migrate into the intima in response to vascular injury.1 A key component in SMC activation and vascular proliferative diseases progression is a disturbed epigenetic regulation and chromatin remodeling.2,3 Epigenetic modifications including DNA methylation and chromatin acetylation, which alter defined transcriptional cellular programs, have extensively been studied during the last decades.4 Covalent epigenetic modifications of the nucleosome are recognized by so-called “epigenetic readers”, that facilitate chromatin remodeling, transcriptional initiation, and elongation.5 One of them, the bromodomain motif, interacts with acetylated lysine-sidechains and is part of numerous chromatinassociated proteins.6 In addition to the bromodomain motif, the bromodomaincontaining proteins (BRD) 2, 3, and 4 as well as the bromodomain testis-specific protein (BRDT) contain an extraterminal molecular interaction site and are thus classified as bromodomain and extraterminal (BET) proteins.7
Available BET inhibitors competitively target the BRD proteins’ binding site for acetylated histones and have recently been shown to exert favorable effects in the prevention and treatment of multiple diseases, e.g. cancer or diabetes.8,9 To understand the role of BETs in the modulation of SMC function and to evaluate the impact of BET inhibition on vascular remodeling processes, we first studied the effects of the potent selective BET inhibitors (+)-JQ1 and I-BET 151 in human coronary SMC in vitro. In response to bromodomain inhibition, we found a Forkhead box O (FOXO)1-dependent upregulation of CDKN1A, which, in agreement with previous reports, caused a cell cycle arrest in human SMCs.10,11 Therefore, we further investigated the effect of BET inhibition in vivo on neointima formation following wire-induced injury of murine femoral arteries.

2 Material and Methods

A detailed methods section is available as an online supplement.

2.1 Cell culture

Human coronary artery SMCs and human coronary artery endothelial cells (ECs) were purchased from Lonza (Switzerland). Cells between passages 3 and 7 were used for all experiments. Cells were treated with the BET-specific bromodomain inhibitors (+)-JQ1 dissolved in DMSO (BPS Bioscience, USA) and I-BET 151 hydrochloride dissolved in DMSO (Tocris, USA) at indicated concentrations.

2.2 Cell proliferation and viability assay

Cell proliferation was determined by BrdU incorporation (Cell Proliferation ELISA Kit, Roche, Switzerland) and colorimetric measurement following manufacturer’s instructions and as previously described.12 Cell viability was determined by cleavage of WST-1 to formazan (Cell Proliferation Reagent WST-1, Roche) and colorimetric measurement following manufacturer’s instructions and as previously described.13 For siRNA-mediated knockdown cells were transfected with siRNA as indicated before serum-starvation followed by growth medium (GM) stimulation (Provitro, Berlin, Germany). For inhibitor studies, indicated concentrations of (+)-JQ1 or I-BET 151 or an equal volume of vehicle control (DMSO; Carl Roth, Germany) were added to the growth medium.

2.3 Migration assay

Cell migration was assessed by a modified Boyden chamber assay as previously described.14 Migration was determined 4 h after stimulation and incubation at a wavelength of 528 nm.

2.4 Cell cycle analysis

Cells were serum starved for 16 h and treated with either (+)-JQ1, I-BET151, or DMSO at indicated concentrations for 24 h. Cell cycle was analyzed using a flow cytometer at 605-635 nm and Cytometry List Mode Data Acquisition & Analysis Software (both: GalliosTM, Beckman Coulter, USA).

2.5 Apoptosis assays

Apoptosis was determined by measurement of caspase 3 and 7 activity using the FLICA® 660 Caspase 3/7 Assay kit (ImmunoChemistry Technologies, LLC, USA) following manufacturer’s instructions. Cells were treated with the respective inhibitors in basal medium (BM; Provitro, Berlin, Germany), 100 mM hydrogen peroxide (Carl Roth), or GM. Fluorescence signal was measured using a flow cytometer at 695/30 nm and Cytometry List Mode Data Acquisition & Analysis Software (both: GalliosTM, Beckman Coulter).

2.6 siRNA-mediated gene knockdown

For siRNA-mediated knockdown of BRD2, BRD3, BRD4, CDKN1A, GADD45A, and FOXO1 cells were transfected with respective siRNAs (3 unique 27mer siRNA duplexes; Origene, USA) according to manufacturer’s instructions. Scrambled negative siRNA served as control.

2.7 RNA isolation and reverse transcription

Isolation of RNA from cells was performed using the RNeasy mini kit (Qiagen, Germany) according to the manufacturer’s instructions. RNA from tissues was isolated after homogenization in 500 µl TrizolTM reagent (Invitrogen). The obtained RNA was reverse transcribed with the High-Capacity RNA-to-cDNATM kit (Applied Biosystems, USA) according to manufacturer’s instructions.

2.8 PCR and qRT-PCR

cDNA was amplified according to manufacturer’s instructions with TaqDNA Polymerase, dNTP Set (both: Qiagen) and the respective primers in Mastercycler® reagent (Eppendorf, Germany). Subsequently, the cDNA was separated and visualized with the FlashGelTM System (Lonza). Quantitative real-time PCR was performed using SYBR green® master mix with the respective primers and the CFX96 TouchTM Real Time PCR Detection System (both: Bio-Rad, USA). All analyses were performed in triplicate and either the DNA template or the reverse transcriptase was omitted for control reactions. Fold change expression levels were quantified and normalized to the geometric mean of at least two reference genes with the highest expression stability by use of the 2-∆∆Ct relative quantification method. Primer sequences and design are mentioned in the online supplementary.

2.9 Protein isolation and determination

Proteins of whole cell lysates were isolated using RIPA buffer (cell signaling, USA) containing PMSF (Sigma-Aldrich) and EDTA-free protease inhibitor cocktail (Roche). For co-immunoprecipitation proteins were previously crosslinked with dithiobis succinimidyl propionate (Thermo FisherTM, USA) according to manufacturer’s instructions. Protein concentration was determined using the Protein Assay Dye Reagent Concentrate (Bio-Rad).

2.10 Immunoprecipitation

To analyze protein-protein association in FOXO1-transfected SMCs we separated BRD4 protein complexes using an anti-BRD4 antibody (sc-48772, Santa Cruz, USA) and the DynabeadsTM Protein G Immunoprecipitation Kit (Invitrogen) following manufacturer’s instructions. Normal rabbit IgG (sc-2027, Santa Cruz) was used as control.

2.11 SDS-PAGE and Western Blotting

Protein isolates were denatured in Roti®-Load 1 buffer (Carl Roth), added to NuPAGETM LDS sample buffer (Invitrogen), H2O, and DTT (Sigma-Aldrich). Proteins were separated on 4-20 % Mini-ProteanTGX precast gels (Bio-Rad) and transferred to the Immun-Blot® PVDF membrane (Bio-Rad). Following primary antibodies were used: BRD2 (sc-393720, Santa Cruz), BRD3 (SAB1412098, Sigma-Aldrich), BRD4 (ab128874, Abcam, UK), FOXO1 (14952, Cell Signaling, USA or ab52857, Abcam), and GAPDH (sc-32233, Santa Cruz).

2.12 Isolation of nuclear proteins and transcription factor activity

Nuclear proteins were obtained utilizing a commercial nuclear extraction kit (Abcam) following manufacturer’s instructions. Activity of the transcription factor FOXO1 was analyzed by means of the TransAMTM FKHR assay (Active Motif, USA) performed following the manufacturer’s instructions.

2.13 Microarray analysis and sample preparation

RNA was isolated using the RNeasy mini kit (Qiagen) according to the manufacturer’s instructions. RNA concentrations were measured using Nanodrop 2000c (Thermo Fisher, USA) and Agilent 2100 Bioanalyzer (Agilent Technologies, USA) according to the manufacturer’s instructions.
Microarray analysis was conducted using the SurePrint G3 Human GE V2 8x60K Microarray Kit (Agilent Technologies, USA). Synthesis of Cy3-labeled cRNA was performed with the ‘Quick Amp Labeling kit, one color’ (Agilent Technologies) according to the manufacturer’s recommendations. Slides were scanned on the Agilent Micro Array Scanner G2565CA (pixel resolution 3 µm, bit depth 20). Data extraction was performed with the ‘Feature Extraction Software V10.7.3.1’ using the extraction protocol file ‘GE1_107_Sep09.xml’. Processed intensity values of the green channel (‘gProcessedSignal’ or ‘gPS’) were normalized by global linear scaling. gPS values of one sample were multiplied by an array-specific scaling factor and calculated by the following formula: normalized gPSArray i = gPSArray i x (1500 / 75th PercentileArray i).
Data were analyzed with the Gene Set Enrichment Analysis (GSEA) software whereas C5 gene ontology gene sets were obtained from MSigDB. A further gene set of 323 FoxO1-dependent genes was created based on a literature review (Supplemental Table). Fold changes to the mean of basal medium probes were calculated and represented in graphs, using GraphPad Prism 7 (GraphPad Software Inc., USA).Microarray data has been deposited in the GEO repository (GEO accession number: GSE138323).

2.14 Chromatin immunoprecipitation

2×107 cells were cultured in growth medium at a density of 13,500 cells/cm2 and treated with (+)-JQ1 for 1 h. Cell samples were fixed, harvested and lysed according to optimized manufacturer’s instructions using the ChIP-IT™ Express Enzymatic Chromatin Immunoprecipitation Kit (Active Motif). Chromatin immunoprecipitation (ChIP) reactions were incubated overnight on an end-to-end rotator using 95 µL of isolated chromatin and either 2 µg of FOXO1 antibody (sc-374427, Santa Cruz) or of normal mouse IgG (sc-2025, Santa Cruz). Samples were washed, eluted, reverse cross-linked and treated with Proteinase K according to manufacturer’s instructions (Active Motif). DNA was analyzed by qRT-PCR as described above. The FOXO1 binding motif GTAAACAA chosen for the analysis was selected using the ConTra v3 web server.15

2.15 Vascular injury models

All animal experiments have been performed according to Directive 2010/63/EU of the European Parliament as well as to local ethical guidelines. All procedures involving animals have been approved by the Lower Saxony’s institutional committee for animal research (LAVES). Adult male C57BL/6 mice were purchased from Charles River (Germany).

2.16 Mouse femoral artery injury model of neointima formation

The dilatation of the femoral artery was performed as previously described.14 In brief, mice were anesthetized using a singular intraperitoneal injection of ketamine hydrochloride (100 mg/kg body weight; Anesketin, Albrecht, Germany) and xylazine (16 mg/kg body weight Rompun® 2%, Bayer Health Care AG, Germany). A straight spring wire (0.38 mm in diameter, Cook Medical, USA) was then inserted through the profunda femoris artery up to 1 cm into the femoral artery and left in place for one minute to achieve an adequate wire-induced vessel injury. After removal of the wire, the profunda femoris artery was ligated (7-0 Prolene, Ethicon, USA) and perfusion of the dilated femoral artery was reestablished. Immediately after dilatation we covered the injured femoral artery with 50 μl of a 25 % thermosensitive Pluronic® F-127 gel containing (+)-JQ1 (10 mM) or vehicle (DMSO). Mice were sacrificed at 10 or 21 days by cervical dislocation. The femoral artery was carefully harvested and embedded in Tissue-Tek OCT embedding medium (Sakura Finetek Europe B.V., The Netherlands).

2.17 Mouse carotid artery model of reendothelialization

Perivascular electric injury of the carotid artery was performed as previously described.16 Mice were anesthetized as described above. After preparation of the left common carotid artery through ventral middle line neck incision, electric deendothelialization was carried out with a bipolar microregulator (ICC50, ERBEElektromedizin GmbH, Germany) at a length of 5 mm below the carotid bifurcation (2 watt for 2 seconds). The treated artery was subsequently coated with 50 μl of a 25 % thermosensitive pluronic F-127 gel containing (+)-JQ1 (10 mM) or vehicle (DMSO). Five days after electric injury of the left carotid artery, mice were sacrificed by cervical dislocation and reendothelialization was assessed following injection 50 µL 5% Evan’s blue solution (Sigma-Aldrich) into the tail vein, careful dissection of the left carotid artery, and en face staining (Eclipse Ni-E microscope, Nikon, Japan). Reendothelialization was calculated as difference between the length of the bluestained area and the initially injured area, using computer-assisted morphometric analysis (ImageJ 1.48 software, National Institutes of Health, USA).

2.18 Morphometry

After harvesting the dilated femoral arteries at the indicated time-points following injury, vessels were sliced in 6 µm serial sections and van Gieson staining (Carl Roth) was performed for 6 cross-sections from regular intervals throughout each artery. ImageJ 1.48 software was used to calculate circumference of external elastic lamina, internal elastic lamina and lumen as well as medial and neointimal area.

2.19 Immunofluorescence

Samples were incubated with antibodies targeting α-SMA (C6198, Sigma-Aldrich), Ki-67 (ab15580, Abcam), BRD2 (sc-393720, Santa Cruz), BRD3 (SAB1412098, Sigma-Aldrich), BRD4 (ab128874, Abcam), or FOXO1 (ab52857, Abcam). After incubation with primary antibodies, samples were marked with Alexa 488- or 546coupled secondary antibodies (LifeTechnologies) and counterstained with nuclear 4.6-diamidino-2-phenylindole (Immunoselect Antifading Mounting Medium DAPI, Dianova, Germany). For α-SMA staining we used monoclonal antibodies which were labelled directly with Cy3. Matching species- and isotype control antibodies were used for negative controls (Santa Cruz). For bright-field and immunofluorescence microscopy an Eclipse NI-E microscope (Nikon Instruments Europe B.V., The Netherlands), adequate filter blocks and image processing software were used (NIS Elements AR 4.20.01, Nikon Instruments Europe B.V.,).

2.20 Statistical analysis

All collected data were stored and analyzed on personal and institutional computers which were equipped with Microsoft Excel 2010 (Microsoft Corporation), Microsoft Word 2010 (Microsoft Corporation) and GraphPad Prism 6.01 (GraphPad Software Inc., USA). For statistical analysis of data among study groups, we used unpaired ttest or one-way ANOVA followed by multiple comparisons correction with the HolmSidak method depending on the number of groups and comparisons. All results were reported as mean ± standard error of the mean (SEM). The probability value was set <0.05 to be considered statistically significant for all statistical analysis.

3 Results

3.1 BET proteins are expressed in SMCs, ECs, and the vascular wall PCR and western blot analysis revealed robust RNA expression levels of BET containing proteins BRD2, BRD3 and BRD4 in SMCs as well as in ECs. On the protein level, the expression of BRD2, BRD3 and BRD4 was more pronounced in SMCs than in ECs, as determined by western blotting and immmunohistochemistry (figure 1a and b). Whereas BRD2 and BRD3 revealed a perinuclear expression pattern, BRD4 appeared to be almost exclusively localized in the nucleus (figure 1 c, scale bar 5 µm). Consistently, immunofluorescence analysis of native murine femoral arteries showed a comparable cellular distribution of all three BETs in vivo with cytoplasmatic and perinuclear expression of BRD2 and BRD3 and a nuclear expression pattern of BRD4 (figure 1 d, scale bar 25 µm). In addition, we detected rather prominent BRD3 and BRD4 signals in SMC of the intimal and medial layer, and a weaker expression in the adventitia, whereas BRD2 was expressed in all vascular layers including the adventitial layer in vivo. Moreover, in vivo, BRD2, BRD3 and BRD4 seem to be expressed predominantly in SMC, since signals appeared to be much weaker in EC (figure 1d).

3.2 BET inhibition induces G0/G1 cell cycle arrest and prevents SMC proliferation

To investigate the impact of BETs on vascular cell function in vitro, SMCs were incubated with the BET inhibitors (+)-JQ1 or I-BET 151. After 24 hours, we detected a dose-dependent reduction of SMC proliferation in response to treatment with either (+)-JQ1 or I-BET 151 as assessed by BrdU-incorporation (GM+vehicle 100±19.37% vs. GM+1000 nM (+)-JQ1 14.75±3.251% vs. GM+1000 nM I-BET 151 3.059±0.510%, ****P<0.0001, figure 2 a). Flow cytometry-based cell cycle analysis of propidium iodide-stained cells revealed BET-dependent G0/G1 arrest to the presence of (+)JQ1 or I-BET 151 (figure 2 b). WST-1 conversion to formazan revealed a dosedependent significant reduction of cellular metabolic activity in SMCs in response to (+)-JQ1, and a trend in the same direction in response to I-BET 151 (GM+vehicle 100±7.691% vs. GM+1000 nM (+)-JQ1 81.97±5.452 % vs. GM+1000 nM I-BET 151 93.48±11.76 %, *P<0.05, figure 2 c).
Beyond effects on proliferation and metabolic activity, growth medium-induced SMC migration was also dose-dependently attenuated in response to (+)-JQ1 or I-BET 151, as evaluated by a modified Boyden chamber assay (GM+vehicle 100±8.51% vs. GM+1000 nM (+)-JQ1 64.46±5.609% vs. GM+1000 nM I-BET 151 75.68±11.42%, **P<0.01, ****P<0.0001, figure 2 d). Additionally, immunofluorescence staining of the SMC marker proteins CALD1 and MYH11 verified a spindle-shaped SMC phenotype pointing out the prevention of a growth stimuli-induced phenotypic switch in response to BET inhibition. Conclusively, gene expression of these SMC marker genes was preserved in response to (+)-JQ1 as determined by qPCR analysis (*P<0.05, **P<0.01, n=6, supplemental figure I). We did not observe any change in SMC apoptosis rates in response to BET inhibition, neither with (+)-JQ1 nor with I-BET 151 (BM+vehicle 8.107±3.632 FLICA+ cells vs. BM+1000 nM (+)-JQ1 7.783±2.919 FLICA+ cells vs. BM+1000 nM I-BET 151 7.140±0.5841 FLICA+ cells, P=n.s., figure 2 e).

3.3 FOXO1-dependent genes are regulated in response to BET inhibition

To investigate the underlying mechanisms of the observed effects of BET inhibition on SMC function we performed mRNA microarray expression analysis in SMC stimulated for 6 hours with or without BET inhibition and non-proliferating SMC in basal medium. Due to the broader effects of (+)-JQ1 in vitro and the available data using structure-related substances in vivo, we focused on the use of the BET inhibitor (+)-JQ1 only.
Microarray analysis revealed a broad variety of regulated genes in response to stimulation with growth medium compared to SMCs cultured with basal medium, or SMCs cultured with growth medium and (+)-JQ1. Gene set enrichment analysis (GSEA) revealed that BET inhibition predominantly alters the transcription of genes crucially involved in cell cycle regulation and thus proliferation (figure 3 a and b). A further analysis based on a literature-based compilation of FOXO1-controlled genes revealed that numerous genes out of the regulated gene sets are under transcriptional control of the transcription factor FOXO1 (red spots, figure 3 c and Supplemental Table). Vice versa, the majority of genes known to be regulated by FOXO1 was found to be regulated in response to (+)-JQ1 (**P<0.01, FDR q-value 0.085, figure 3 d and Supplemental Table).
We further confirmed the upregulation of CDKN1A, one of the most significantly regulated FOXO1-dependent genes of the above-mentioned gene sets, by qPCR (figure 3 e). CDKN1A, a potent inhibitor of cell cycle-dependent kinases and responsible for the control of the G1-S checkpoint, and GADD45A, which is implicated in the control of cell cycle G2-M arrest, were confirmed to be significantly upregulated after treatment with (+)-JQ1 (figure 3 e and f). These effects of (+)-JQ1 could be prevented by siRNA-mediated silencing of FOXO1, thus approving a FOXO1-dependent expression of both genes. Moreover, siRNA-mediated silencing of CDKN1A but not of GADD45A partially reversed the antiproliferative effects of (+)JQ1 (****P<0.0001, n=6, figure 3 g), indicating that the anti-proliferative effects of (+)JQ1 are largely mediated by CDKN1A. ChIP and subsequent qPCR validated a direct interaction of the transcription factor FOXO1 and the genomic region of CDKN1A (*P<0.05, n=3, figure 3 h).

3.4 FOXO1 is expressed in SMCs and is crucial for (+)-JQ1-mediated inhibition of cell proliferation

Based on the pathway analysis and data obtained by GSEA we identified the transcription factor FOXO1 as a potential key regulator of the (+)-JQ1-mediated differential mRNA expression. A further immunfluorescence analysis revealed a robust expression and nuclear localization of FOXO1 in quiescent SMCs in vitro and in native murine femoral arteries in vivo (figure 4 a and b). Growth medium stimulation of SMC reduced the FOXO1 transcriptional activity, but this reduction was completely prevented after treatment with (+)-JQ1 (OD 450 nm, BM + vehicle 0.0679±0.005 vs. GM + vehicle 0.0456±0.002 vs. GM + 1000 nM (+)-JQ1 0.0649±0.003, *P<0.05, n=3, figure 4 c). Treatment with (+)-JQ1 did not alter FOXO1 protein expression in SMCs in vitro, as determined by western blot analysis (figure 4 d). These data suggest, that BET proteins are crucial for the growth medium-induced transactivational inactivation of FOXO1 and that there might exist a potential direct interaction of these proteins. To follow up on this hypothesis, we performed a siRNAmediated knockdown of FOXO1 in SMCs (figure 4 e). Following FOXO1 knockdown, the inhibitory effect of (+)-JQ1 on SMC proliferation was indeed abolished, indicating that the anti-proliferative effect of BET inhibition is in fact largely dependent on the presence of FOXO1 (GM+vehicle 100±26.46% vs. GM+1000 nM (+)-JQ1+scrambled siRNA 32.05±12.71% vs. GM+1000 nM (+)-JQ1+FoxO1a siRNA 64.68±10.69%, n=6, *P<0.05, figure 4 f).

3.5 Protein-protein-interaction of FOXO1 and BRD4 is decisive for SMC proliferation

To identify the specific BRD proteins responsible for the observed effects on FOXO1 activity and SMC function we assessed SMC proliferation following siRNA-mediated knockdown of each individual BRD protein (figure 5 a). Whereas depletion of BRD3 even augmented SMC proliferation, silencing of each, BRD2 and BRD4, effectively inhibited SMC proliferation, comparable to the effect of (+)-JQ1 (GM+vehicle 100±8.063% vs. GM+250 nM (+)-JQ1 55.71±13.69% vs. GM+siBRD2 50.50±7.068% vs. GM+siBRD3 247.069±18.84% vs. GM+siBRD4 59.545±9.672%, n=4, ***P<0.001, ****P<0.0001 compared to GM+vehicle, figure 5 b). Interestingly, the combined knock down of all BRD proteins (BRD2, BRD3, and BRD4) also resulted in a significant reduction of SMC proliferation, comparable to the effects of (+)-JQ1 (53.53±8.92%, n=4, ****P<0.0001 compared to GM+vehicle, figure 5 b). To follow up on a potential direct interaction, the localization of BRD proteins was assessed by immunohistochemistry. Whereas BRD2 and BRD3 were localized mainly in the cytoplasm, only BRD4 was found to be predominantly localized in the nucleus, as was FOXO1 (figure 5 c). Subsequent co-immunoprecipitation experiments confirmed a direct nuclear interaction of BRD4 and FOXO1 but not of other BRD proteins (figure 5 d).

3.6 BET inhibition prevents smooth muscle cell proliferation and neointima formation following vascular injury in mice

To determine the effect of (+)-JQ1-dependent Brd inhibition in vivo, we assessed intimal and medial SMC proliferation 10 days following wire-induced injury of the murine femoral artery of C57BL/6 mice. Local application of (+)-JQ1, released from a self-degrading thermosensitive Pluronic® F-127 gel which was applied around the injured vessel, significantly reduced the amount of proliferating (Ki-67+) cells within both the intimal and the medial vascular layer (33.90±5.478 Ki-67+ neointimal cells in vehicle-treated mice vs. 0.80±0.719 Ki-67+ neointimal cells in (+)-JQ1-treated mice, ***P<0.001, n=6, figure 6 a and b). Consistently, neointima formation was significantly reduced in mice treated with 10 mM (+)-JQ1 21 days after vascular injury (luminal stenosis 78.51±2.119% in vehicle-treated mice vs. 22.04±2.065% in (+)JQ1-treated mice, ****P<0.0001, n=6, figure 6 c-e). Importantly, (+)-JQ1 had no significant effect on the reendothelialization of vessel segments following electric injury and denudation of mouse carotid arteries in vivo (P=n.s., n=4, supplemental figure II).

4 Discussion

With the recent advancements in the understanding of epigenetic processes, intensive research has been performed to decipher the impact of chromatin modifying enzymes on cell signaling and thereby on the development and prevention of specific diseases. Although the majority of studies so far focused on manipulating epigenetic writers and erasers, i.e. histone acetyltransferases and histone deacetylases, only recent studies also characterized the chromatin reading molecules like BETs. For this purpose, highly specific pharmacological BET inhibitors have been developed.17 Whereas most basic science- and clinical studies focused on the therapeutic potential in oncology, only a few recent studies investigated the potential effects of BET inhibition in the context of cardiovascular diseases.
Initially, BET inhibition has been shown to suppress cardiomyocyte hypertrophy in vitro and pathologic cardiac remodeling in vivo in two basic science landmark studies.18, 19 The very recent Australian ASSURE (ApoA-1 Synthesis Stimulation and Intravascular Ultrasound for Coronary Atheroma Regression Evaluation; NCT01067820) trial provided clinical evidence for a reduced atherosclerotic plaque burden by the BET inhibitor apabetalone (RVX-208) as add-on therapy to high potency statins, hypothesizing that this effect was due to an additional BETindependent lipid-modifying effect, increasing HDL levels.20
In the present study, we show for the first time that specific BRD proteins are expressed in vascular cells and that BET inhibition plays a crucial role in SMC cell cycle regulation and proliferation in vitro and in neointimal lesion formation in vivo.
In our hands, solely BRD4 expression was restricted to the nucleus in SMCs, whereas BRD2 and BRD3 surprisingly showed a predominantly cytosolic, perinuclear expression. Fukazawa and Masumi showed that BRD proteins own a common conserved 12-amino acid nuclear localization signal and are therefore suggested to be found in the nucleus.21 In line with their data, our data suggest that nuclear retention might differ between the single BRD proteins. After all, the detailed molecular mechanism of BET protein nuclear retention is not fully understood. Although located outside the nucleus, siRNA-mediated silencing of BRD2 or BRD3 profoundly changed SMC proliferation. BRD proteins might thus exhibit mechanisms of action apart from their direct impact on transcription factor activity. Mitochondrial integrity and function have recently been implicated in the regulation of bromodomain inhibition and vice versa.22–24 This relation has been at least in part attributed to altered expression of MYC. However, the respective publications lacked to show evidence for the hypothesized nuclear localization of BRD proteins as well as for a direct involvement of BRD proteins in the transcriptional regulation of MYC. Hence, one could speculate about a cytosolic localization of BRD proteins as a prerequisite for a potential direct impact on mitochondrial and cellular function.
The here presented anti-proliferative effect of BET inhibition is consistent with previously published effects seen in epithelial cells, B-cells, and HeLa cells.25,26 Whereas the inhibitory capacity of (+)-JQ1 on cellular metabolism could clearly be determined, the effect of I-BET151 did not reach statistical significance. In fact, there is no published systematic biophysical comparison of (+)-JQ1 and I-BET151 we could refer to. At least Baker et al. determined higher IC50 values for I-BET151 compared to (+)-JQ1 in an assay comparable to ours in osteosarcoma cells in coherence with our results.27
The maintenance of a contractile SMC phenotype in response to BET inhibition furthermore might implicate an improvement in atherosclerotic plaque stability. This might provide one rationale for the reported discrepancy between just moderate effects of apabetalone on plaque volume in the ASSURE study and a clear reduction in cardiovascular event rates in a pooled analysis.28 Hence, our data clearly imply BET inhibition as a potential therapeutic approach to directly prevent adverse vascular remodeling processes independent of a lipid modulating effect. In previous studies, rather unspecific BET bromodomain inhibitors were used. I.e. apabetalone, which has been developed and clinically investigated as an ApoA1 modulator, initially not being aware of its BET bromodomain inhibiting effects, which were deciphered only later on.29,30 However, in order to allow a rather selective targeting, and to dissect specific effects of BET inhibition from lipid-modifying effects as seen with apabetalone, the highly selective BET bromodomain inhibitors I-BET 151 and (+)-JQ1 were used in the current study.31,32
Mechanistically, microarray expression data, GSEA, and further pathway analyses revealed a predominant regulation of FOXO1-dependent genes in response to (+)JQ1, of which CDKN1A was identified to be decisive for the observed antiproliferative effects in our study. Besides CDKN1A, we found GADD45A to be robustly upregulated in response to (+)-JQ1. The transcriptional and posttranscriptional regulation of GADD45A is complex and orchestrated by numerous regulators including MYC.33 GADD45A is even differentially regulated across different tumor cell lines.34 It obtains pleiotropic effects in vitro including the ability to induce cell cycle arrest in the G2/M phase.35 In contrast, (+)-JQ1 induced a cell cycle arrest in the G0/G1 but not the G2/M phase that was not rescued after silencing of GADD45A. Upregulation of CDKN1A in response to (+)-JQ1 might thus exceed the effect of GADD45A silencing and retain the cells in a quiescent state.
However, GADD45A might still have additional indirect anti-proliferative effects in the G0/G1 phase. First, GADD45A has been shown to regulate the phosphorylation and thus activation status of the signal transducer and activator of transcription 3 (STAT3), which we could implicate in SMC proliferation and the pathogenesis of neointima formation before.14,36 Second, GADD45A and CDKN1A are hypothesized to interact with each other, even though the detailed interaction requires further investigation.37
Addition of (+)-JQ1 to growth medium indeed preserved the transcriptional capacity of FOXO1. Conclusive with the here investigated FOXO1-dependent transcription of CDKN1A in response to (+)-JQ1, siRNA-mediated silencing of FOXO1 partially reversed the (+)-JQ1 mediated effects on smooth muscle cell proliferation. Notably, the level of BrdU incorporation still remained lower than that of vehicle control indicating alternative signaling pathways. The transcription factor activities of both interferon regulatory factor 4 (IRF4) and Krüppel-like factor 5 (KLF5) have previously been shown to be depend on BRD4.38,39 Both transcription factors have furthermore been implicated in the regulation of smooth muscle cell proliferation in vitro and neointima formation in vivo.40,41
Indeed, we found a co-localization and direct interaction of FOXO1 and the BET bromodomain-containing protein BRD4 in the nucleus (figure 7). FOXO1 acetylation by histone acetyltransferases (HAT) decreases CDKN1A transcription, whereas BRD4 has an intrinsic HAT activity located distal to the two bromodomains (BD1 and BD2) and an ET domain.42,43 Therefore, FOXO1 acetylation by BRD4 seems to be conceivable at first sight.
(+)-JQ1, though, acts competitively at the BRD’s binding site for acetylated histones located within BD1 and BD2. A direct interaction of FOXO1 with one of the two bromodomains seems hence more likely. Nagarajan et al. reported histone-like motifs within the amino acid sequence of FOXO1, which were found to be acetylated in a curated mass spectrometry database.44 In contrast to BD1, BD2 is known not only to interact with acetylated lysines of histones, but with those of other protein partners, too.45,46 The histone-like motifs of FOXO1 and the BD2 domain of BRD4 could thus serve as direct interacting sites for each other.
Inhibition of BRD4 by (+)-JQ1 was found to enhance the transcriptional capacity of FOXO1, indicating that BRD4 functions as an endogenous inactivator of FOXO1. FOXO1 binds to condensed chromatin in a quiescent state, thereby recruiting other transcription factors to maintain transcriptional activity necessary for retaining physiological tissue homeostasis.47 In response to mitogenic stimuli, FOXO1 is subjected to multiple modifications, including phosphorylation and acetylation. These modifications result in a modulation or inhibition of its transactivational capacity, a cytosolic translocation and subsequent degradation.48 Remarkably, current reports suggest that BET proteins can also be recruited to condensed chromatin, thereby orchestrating cell cycle regulation. DNA binding affinity and transcriptional activity of FOXO1 is regulated by lysine acetylation and deacetylation on the one hand and protein-protein interaction on the other hand.49
Given the above findings, we postulate an additional mechanism regulating FOXO1 transactivational capacity, which may take part alternatively or in parallel to the previously well-described mechanisms: The epigenetic reader protein bromodomain containing protein BRD4 represents an important regulator of FOXO1 function, contributing to FOXO1 inactivation under mitogenic and/or inflammatory stimulation. FOXO1-induced gene transcription has been implicated in the development of numerous proliferative diseases, e.g. cancer, but also vascular diseases.10,50 We previously found FOXO1-dependent gene transcription to be decisive for pulmonary artery SMCs proliferation in the pathogenesis of pulmonary artery hypertension.10 In addition, FOXO1 function in vascular and inflammatory cells has also been implicated in atherosclerosis and venous bypass graft failure, but more importantly, also in neointima formation.51–54 There, FOXO1 regulates CDKN1B in SMCs, which conclusively was regulated in our microarray analysis. CDKN1A and CDKN1B share a high structural and functional similarity as inhibitors of G1 cyclin-Cdk protein kinase activity.55,56 Moreover, both are suggested to regulate each other’s expression levels and to regulate different stages of G1 phase progression.57,58 Thus, Foxo1dependent regulation of neointima formation was shown to be dependent on both CDKN1B and CDKN1A expression.
Importantly, (+)-JQ1 did not affect the endothelial recovery after denudation and thus does not impair vascular healing in vivo, most likely due to the low expression levels of Brd proteins in coronary artery ECs as compared to SMCs. However, others have shown a significant impact of (+)-JQ1 on the expression of endothelial adhesion molecules including ICAM-1, VCAM-1, and E-selectin in vitro. This results in an impairment of leukocyte recruitment in vivo, which might also result in reduced vascular inflammation and thus reduced SMC proliferation and neointima formation in our study.59

5 Conclusions

In conclusion, we present evidence that the BET bromodomain containing protein BRD4 directly interacts with and modulates the activity of the transcription factor FOXO1. Following mitogenic stimulation, FOXO1 has to be inactivated to allow cell cycle entry and proliferation of SMC. BRD4 binding seems to be mandatory for FOXO1-inactivation, since BRD4 inhibition by (+)-JQ1 keeps FOXO1 in a transactivationally active state even under mitogenic stimulation. Due to the FOXO1dependent continuous transactivation of CDKN1A, SMC remain in a quiescent phenotype, which prevents SMC proliferation and injury-induced neointima formation. The direct INCB054329 effects of BET inhibitors on SMC function provide a rationale for their therapeutic potential in preventing post-angioplasty restenosis and possibly also related vascular diseases like atherosclerosis.

References

1. Dzau VJ, Braun-Dullaeus RC, Sedding DG. Vascular proliferation and atherosclerosis: new perspectives and therapeutic strategies. 2002;8.
2. Alexander MR, Owens GK. Epigenetic Control of Smooth Muscle Cell Differentiation and Phenotypic Switching in Vascular Development and Disease. 2012;74:13–40.
3. Gibbons GH, Dzau VJ. The Emerging Concept of Vascular Remodeling. N Engl J Med 1994;330:1431–1438.
4. Heerboth S, Lapinska K, Snyder N, Leary M, Rollinson S, Sarkar S. Use of epigenetic drugs in disease: an overview. Genet Epigenet 2014;6:9–19.
5. Dawson MA, Kouzarides T, Huntly BJP. Targeting Epigenetic Readers in Cancer. N Engl J Med 2012;367:647–657.
6. Tolani B, Gopalakrishnan R, Punj V, Matta H, Chaudhary PM. Targeting Myc in KSHV-associated primary effusion lymphoma with BET bromodomain inhibitors. Oncogene 2014;33:2928–2937.
7. Belkina AC, Denis G V. BET domain co-regulators in obesity, inflammation and cancer. Nat. Rev. Cancer. 2012. p. 465–477.
8. Amorim S, Stathis A, Gleeson M, Iyengar S, Magarotto V, Leleu X, Morschhauser F, Karlin L, Broussais F, Rezai K, Herait P, Kahatt C, Lokiec F, Salles G, Facon T, Palumbo A, Cunningham D, Zucca E, Thieblemont C. Bromodomain inhibitor OTX015 in patients with lymphoma or multiple myeloma: A dose-escalation, open-label, pharmacokinetic, phase 1 study. Lancet Haematol 2016;3:e196–e204.
9. Siebel AL, Trinh SK, Formosa MF, Mundra PA, Natoli AK, Reddy-Luthmoodoo M, Huynh K, Khan AA, Carey AL, Hall G Van, Cobelli C, Dalla-Man C, Otvos JD, Rye KA, Johansson J, Gordon A, Wong NCW, Sviridov D, Barter P, Duffy SJ, Meikle PJ, Kingwell BA. Effects of the BET-inhibitor, RVX-208 on the HDL lipidome and glucose metabolism in individuals with prediabetes: A randomized controlled trial. Metabolism 2016;65:904–914.
10. Savai R, Al-Tamari HM, Sedding D, Kojonazarov B, Muecke C, Teske R, Capecchi MR, Weissmann N, Grimminger F, Seeger W, Schermuly RT he., Pullamsetti SS ava. Pro-proliferative and inflammatory signaling converge on FoxO1 transcription factor in pulmonary hypertension. Nat Med 2014;20:1289– 1300.
11. Shi F, Li T, Liu Z, Qu K, Shi C, Li Y, Qin Q, Cheng L, Jin X, Yu T, Di W, Que J, Xia H, She J. FOXO1: Another avenue for treating digestive malignancy? Seminars in Cancer Biology 2017;
12. Daniel J-M, Dutzmann J, Brunsch H, Bauersachs J, Braun-Dullaeus R, Sedding DG. Systemic application of sirolimus prevents neointima formation not via a direct anti-proliferative effect but via its anti-inflammatory properties. Int J Cardiol 2017;238.
13. Dutzmann J, Koch A, Weisheit S, Sonnenschein K, Korte L, Haertlé M, Thum T, Bauersachs J, Sedding DG, Daniel J-M. Sonic hedgehog-dependent activation of adventitial fibroblasts promotes neointima formation. Cardiovasc Res 2017;113.
14. Daniel J-M, Dutzmann J, Bielenberg W, Widmer-Teske R, Gündüz D, Hamm CW, Sedding DG. Inhibition of STAT3 signaling prevents vascular smooth muscle cell proliferation and neointima formation. Basic Res Cardiol 2012;107.
15. Kreft L, Soete A, Hulpiau P, Botzki A, Saeys Y, Bleser P De. ConTra v3: a tool to identify transcription factor binding sites across species, update 2017. Nucleic Acids Res 2017;45:W490–W494.
16. Carmeliet P, Moons L, Stassen JM, Mol M De, Bouché A, Oord JJ van den, Kockx M, Collen D. Vascular wound healing and neointima formation induced by perivascular electric injury in mice. Am J Pathol 1997;150:761–776.
17. Hewings DS, Rooney TPC, Jennings LE, Hay DA, Schofield CJ, Brennan PE, Knapp S, Conway SJ. Progress in the Development and Application of Small Molecule Inhibitors of Bromodomain–Acetyl-lysine Interactions. J Med Chem 2012;55:9393–9413.
18. Anand P, Brown JD, Lin CY, Qi J, Zhang R, Artero PC, Alaiti MA, Bullard J, Alazem K, Margulies KB, Cappola TP, Lemieux M, Plutzky J, Bradner JE, Haldar SM. BET Bromodomains Mediate Transcriptional Pause Release in Heart Failure. Cell 2013;154:569–582.
19. Spiltoir JI, Stratton MS, Cavasin MA, Demos-Davies K, Reid BG, Qi J, Bradner JE, McKinsey TA. BET acetyl-lysine binding proteins control pathological cardiac hypertrophy. J Mol Cell Cardiol 2013;63:175–179.
20. Shishikura D, Kataoka Y, Honda S, Takata K, Kim SW, Andrews J, Psaltis PJ, Sweeney M, Kulikowski E, Johansson J, Wong NCW, Nicholls SJ. The Effect of Bromodomain and Extra-Terminal Inhibitor Apabetalone on Attenuated Coronary Atherosclerotic Plaque: Insights from the ASSURE Trial. Am J Cardiovasc Drugs 2018;
21. Fukazawa H, Masumi A. The conserved 12-amino acid stretch in the interbromodomain region of BET family proteins functions as a nuclear localization signal. Biol Pharm Bull 2012;35:2064–2068.
22. Lasorsa E, Smonksey M, Kirk JS, Rosario S, Hernandez-Ilizaliturri FJ, Ellis L. Mitochondrial protection impairs BET bromodomain inhibitor-mediated cell death and provides rationale for combination therapeutic strategies. Cell Death Dis 2015;6:e2014.
23. Ishida CT, Shu C, Halatsch M-E, Westhoff M-A, Altieri DC, Karpel-Massler G, Siegelin MD. Mitochondrial matrix chaperone and c-myc inhibition causes enhanced lethality in glioblastoma. Oncotarget 2017;8:37140–37153.
24. Piquereau J, Boet A, Péchoux C, Antigny F, Lambert M, Gressette M, Ranchoux B, Gambaryan N, Domergue V, Mumby S, Montani D, Adcock IM, Humbert M, Garnier A, Rucker-Martin C, Perros F. The BET Bromodomain Inhibitor I-BET-151 Induces Structural and Functional Alterations of the Heart Mitochondria in Healthy Male Mice and Rats. Int J Mol Sci 2019;20.
25. Yang Z, He N, Zhou Q. Brd4 Recruits P-TEFb to Chromosomes at Late Mitosis To Promote G1 Gene Expression and Cell Cycle Progression. Mol Cell Biol 2008;28:967–976.
26. Ottinger M, Christalla T, Nathan K, Brinkmann MM, Viejo-Borbolla A, Schulz TF. Kaposi’s Sarcoma-Associated Herpesvirus LANA-1 Interacts with the Short Variant of BRD4 and Releases Cells from a BRD4- and BRD2/RING3-Induced G1 Cell Cycle Arrest. J Virol 2006;
27. Baker EK, Taylor S, Gupte A, Sharp PP, Walia M, Walsh NC, Zannettino AC, Chalk AM, Burns CJ, Walkley CR. BET inhibitors induce apoptosis through a MYC independent mechanism and synergise with CDK inhibitors to kill osteosarcoma cells. Sci Rep 2015;5:10120.
28. Nicholls SJ, Kausik •, Ray K, Johansson JO, Gordon A, Sweeney • Michael, Halliday C, Kulikowski E, Wong N, Kim SW, Schwartz GG. Selective BET Protein Inhibition with Apabetalone and Cardiovascular Events: A Pooled Analysis of Trials in Patients with Coronary Artery Disease. Am J Cardiovasc Drugs 2018;18:109–115.
29. Picaud S, Wells C, Felletar I, Brotherton D, Martin S, Savitsky P, Diez-Dacal B, Philpott M, Bountra C, Lingard H, Fedorov O, Müller S, Brennan PE, Knapp S, Filippakopoulos P. RVX-208, an inhibitor of BET transcriptional regulators with selectivity for the second bromodomain. Proc Natl Acad Sci U S A 2013;110:19754–19759.
30. Mclure KG, Gesner EM, Tsujikawa L, Kharenko OA, Attwell S. RVX-208, an Inducer of ApoA-I in Humans, Is a BET Bromodomain Antagonist. PLoS One 2013;8:83190.
31. Kempen herman J, Bellus daniel, Fedorov O, nicklisch silke, Filippakopoulos panagis, picaud sarah, Knapp stefan. Lipid Insights Stimulation of Hepatic Apolipoprotein A-I Production by Novel Thieno-Triazolodiazepines: Roles of the Classical Benzodiazepine Receptor, PAF Receptor, and Bromodomain Binding. Lipid Insights 2013;6–47.
32. Ghosh GC, Bhadra R, Ghosh RK, Banerjee K, Gupta A. RVX 208: A novel BET protein inhibitor, role as an inducer of apo A-I/HDL and beyond. Cardiovasc Ther 2017;35:e12265.
33. Lal A, Gorospe M. Egad, More Forms of Gene Regulation: The gadd45a Story. Cell Cycle 2006;5:1422–1425.
34. Hollander MC, Fornace AJ. Genomic instability, centrosome amplification, cell cycle checkpoints and Gadd45a. Oncogene 2002;21:6228–6233.
35. Zhan Q, Antinore MJ, Wang XW, Carrier F, Smith ML, Harris CC, Fornace AJ. Association with Cdc2 and inhibition of Cdc2/Cyclin B1 kinase activity by the p53-regulated protein Gadd45. Oncogene 1999;18:2892–2900.
36. Yang F, Zhang W, Li D, Zhan Q. Gadd45a Suppresses Tumor Angiogenesis via Inhibition of the mTOR/STAT3 Protein Pathway. J Biol Chem 2013;288:6552–6560.
37. Zhao H, Jin S, Antinore MJ, Lung F-DT, Fan F, Blanck P, Roller P, Fornace AJ, Zhan Q. The Central Region of Gadd45 Is Required for Its Interaction with p21/WAF1. Exp Cell Res 2000;258:92–100.
38. Chen C-H, Yang N, Zhang Y, Ding J, Zhang W, Liu R, Liu W, Chen C. Inhibition of super enhancer downregulates the expression of KLF5 in basallike breast cancers. Int J Biol Sci 2019;15:1733–1742.
39. Gopalakrishnan R, Matta H, Tolani B, Triche Jr T, Chaudhary PM. Immunomodulatory drugs target IKZF1-IRF4-MYC axis in primary effusion lymphoma in a cereblon-dependent manner and display synergistic cytotoxicity with BRD4 inhibitors. Oncogene 2016;35:1797–1810.
40. Fujiu K, Manabe I, Ishihara A, Oishi Y, Iwata H, Nishimura G, Shindo T, Maemura K, Kagechika H, Shudo K, Nagai R. Synthetic Retinoid Am80 Suppresses Smooth Muscle Phenotypic Modulation and In-Stent Neointima Formation by Inhibiting KLF5. Circ Res 2005;97:1132–1141.
41. Cheng W-L, She Z-G, Qin J-J, Guo J-H, Gong F-H, Zhang P, Fang C, Tian S, Zhu X-Y, Gong J, Wang Z-H, Huang Z, Li H. Interferon Regulatory Factor 4 Inhibits Neointima Formation by Engaging Krüppel-Like Factor 4 Signaling. Circulation 2017;136:1412–1433.
42. Chae Y-C, Kim K-B, Kang J-Y, Kim S-R, Jung H-S, Seo S-B. Inhibition of FoxO1 acetylation by INHAT subunit SET/TAF-Iβ induces p21 transcription. FEBS Lett 2014;588:2867–2873.
43. Devaiah BN, Case-Borden C, Gegonne A, Hsu CH, Chen Q, Meerzaman D, Dey A, Ozato K, Singer DS. BRD4 is a histone acetyltransferase that evicts nucleosomes from chromatin. Nat Struct Mol Biol 2016;23:540–548.
44. Nagarajan S, Bedi U, Budida A, Hamdan FH, Mishra VK, Najafova Z, Xie W, Alawi M, Indenbirken D, Knapp S, Chiang C-M, Grundhoff A, Kari V, Scheel CH, Wegwitz F, Johnsen SA. BRD4 promotes p63 and GRHL3 expression downstream of FOXO in mammary epithelial cells. Nucleic Acids Res 2017;45:3130–3145.
45. Huang B, Yang X-D, Zhou M-M, Ozato K, Chen L-F. Brd4 Coactivates Transcriptional Activation of NF- B via Specific Binding to Acetylated RelA. Mol Cell Biol 2009;29:1375–1387.
46. Jang MK, Mochizuki K, Zhou M, Jeong H-S, Brady JN, Ozato K. The Bromodomain Protein Brd4 Is a Positive Regulatory Component of P-TEFb and Stimulates RNA Polymerase II-Dependent Transcription. Mol Cell 2005;19:523–534.
47. Zaret KS, Carroll JS. Pioneer transcription factors: establishing competence for gene expression. Genes Dev 2011;25:2227–2241.
48. Wang Y, Zhou Y, Graves DT. FOXO transcription factors: their clinical significance and regulation. Biomed Res Int 2014;2014:925350.
49. Daitoku H, Sakamaki J ichi, Fukamizu A. Regulation of FoxO transcription factors by acetylation and protein-protein interactions. Biochim. Biophys. Acta – Mol. Cell Res. 2011. p. 1954–1960.
50. Kode A, Mosialou I, Manavalan SJ, Rathinam C V., Friedman RA, TeruyaFeldstein J, Bhagat G, Berman E, Kousteni S. FoxO1-dependent induction of acute myeloid leukemia by osteoblasts in mice. Leukemia 2016;30:1–13.
51. Mahajan S, Fender A, Meyer-Kirchrath J, Kurt M, Barth M, Sagban T, Fischer J, Schrör K, Hohlfeld T, Rauch B. A novel function of FoxO transcription factors in thrombin-stimulated vascular smooth muscle cell proliferation. Thromb Haemost 2012;108:148–158.
52. Tsuchiya K, Westerterp M, Murphy AJ, Subramanian V, Ferrante AW, Tall AR, Accili D. Expanded Granulocyte/Monocyte Compartment in Myeloid-Specific Triple FoxO Knockout Increases Oxidative Stress and Accelerates Atherosclerosis in Mice. Circ Res 2013;112:992–1003.
53. Qiang L, Tsuchiya K, Kim-Muller J-Y, Lin H V., Welch C, Accili D. Increased Atherosclerosis and Endothelial Dysfunction in Mice Bearing Constitutively Deacetylated Alleles of Foxo1 Gene. J Biol Chem 2012;287:13944–13951.
54. Abid MR, Yano K, Guo S, Patel VI, Shrikhande G, Spokes KC, Ferran C, Aird WC. Forkhead transcription factors inhibit vascular smooth muscle cell proliferation and neointimal hyperplasia. J Biol Chem 2005;280:29864–29873.
55. Toyoshima H, Hunter T. p27, a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell 1994;78:67–74.
56. He G, Kuang J, Huang Z, Koomen J, Kobayashi R, Khokhar AR, Siddik ZH. Upregulation of p27 and its inhibition of CDK2/cyclin E activity following DNA damage by a novel platinum agent are dependent on the expression of p21. Br J Cancer 2006;95:1514–1524.
57. Orlando S, Gallastegui E, Besson A, Abril G, Aligú R, Pujol MJ, Bachs O. p27 Kip1 and p21 Cip1 collaborate in the regulation of transcription by recruiting cyclin-Cdk complexes on the promoters of target genes. Nucleic Acids Res 2015;43:6860–6873.
58. Kawamata S, Sakaida H, Hori T, Maeda M, Uchiyama T, Broxmeyer H, Katz A, Vainchenker W, Debili N. The upregulation of p27Kip1 by rapamycin results in G1 arrest in exponentially growing T-cell lines. Blood 1998;91:561–569.
59. Huang M, Zeng S, Zou Y, Shi M, Qiu Q, Xiao Y, Chen G, Yang X, Liang L, Xu H. The suppression of bromodomain and extra-terminal domain inhibits vascular inflammation by blocking NF-κB and MAPK activation. Br J Pharmacol 2017;174:101–115.