Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 May 2;11(1):23.
doi: 10.1186/s13073-019-0635-9.

TCF21 and AP-1 interact through epigenetic modifications to regulate coronary artery disease gene expression

Affiliations

TCF21 and AP-1 interact through epigenetic modifications to regulate coronary artery disease gene expression

Quanyi Zhao et al. Genome Med. .

Abstract

Background: Genome-wide association studies have identified over 160 loci that are associated with coronary artery disease. As with other complex human diseases, risk in coronary disease loci is determined primarily by altered expression of the causal gene, due to variation in binding of transcription factors and chromatin-modifying proteins that directly regulate the transcriptional apparatus. We have previously identified a coronary disease network downstream of the disease-associated transcription factor TCF21, and in work reported here extends these studies to investigate the mechanisms by which it interacts with the AP-1 transcription complex to regulate local epigenetic effects in these downstream coronary disease loci.

Methods: Genomic studies, including chromatin immunoprecipitation sequencing, RNA sequencing, and protein-protein interaction studies, were performed in human coronary artery smooth muscle cells.

Results: We show here that TCF21 and JUN regulate expression of two presumptive causal coronary disease genes, SMAD3 and CDKN2B-AS1, in part by interactions with histone deacetylases and acetyltransferases. Genome-wide TCF21 and JUN binding is jointly localized and particularly enriched in coronary disease loci where they broadly modulate H3K27Ac and chromatin state changes linked to disease-related processes in vascular cells. Heterozygosity at coronary disease causal variation, or genome editing of these variants, is associated with decreased binding of both JUN and TCF21 and loss of expression in cis, supporting a transcriptional mechanism for disease risk.

Conclusions: These data show that the known chromatin remodeling and pioneer functions of AP-1 are a pervasive aspect of epigenetic control of transcription, and thus, the risk in coronary disease-associated loci, and that interaction of AP-1 with TCF21 to control epigenetic features, contributes to the genetic risk in loci where they co-localize.

Keywords: AP-1; Deacetylase; Epigenomics; Histone acetyltransferase; TCF21; Transcription.

PubMed Disclaimer

Conflict of interest statement

Ethics approval and consent to participate

HCASMC provided by commercial vendors Lonza, PromoCell, and Cell Applications were derived from deceased individuals. Because the donors were deceased, and only deidentified information was provided to the investigators; this work was not considered human research.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1
Fig. 1
JUN and TCF21 regulate expression levels of CAD genes SMAD3 and CDKN2BAS/CDKN2B in HCASMC. a HCASMC were transfected with JUN (JUN-KD), TCF21 (TCF21-KD), scrambled (Ctrl) siRNA molecules, or transduced with empty pWPI vector (Ctrl) or a human TCF21 cDNA clone (TCF21-OE) virus. The mRNA expression level of CDKN2BAS was evaluated by qPCR with ACTB normalization. b, c CDKN2B and SMAD3 mRNA expression levels were quantified under identical conditions in HCASMC (mean ± SD, n = 3). d SMAD3, JUN, and TCF21 protein levels were evaluated by western blot, with GAPDH as loading control
Fig. 2
Fig. 2
JUN recruits p300, promotes H3K27ac histone modification, and increases chromatin accessibility and TCF21 binding. af ChIP-qPCR at SMAD3-1, SMAD3-2, and CDKN2BAS locus regions with input normalization in HCASMC. a JUN binding was evaluated under conditions of JUN (JUN-KD) and scrambled (Ctrl) siRNA transfections. b Validation of overexpressed TCF21 mRNA level under conditions of JUN-KD and Ctrl siRNA transfections by RT-qPCR. c pWI lentivirus expressed TCF21 binding was evaluated under conditions of JUN-KD or Ctrl siRNA transfection. d H3K27ac level was evaluated under conditions of JUN-KD or Ctrl siRNA transfection. e Chromatin accessibility assessed by ATAC-qPCR was evaluated under conditions of JUN-KD or Ctrl siRNA transfection. f p300 binding at SMAD3-1/SMAD32 and CDKN2BAS loci was evaluated under conditions of JUN-KD or Ctrl siRNA transfection. g Serial ChIP-qPCR with JUN first IP followed by p300 second IP. h Serial ChIP-qPCR with p300 first IP followed by JUN second IP (mean ± SD, n = 3)
Fig. 3
Fig. 3
TCF21 recruits HDACs 1 and 2, promotes deacetylation at H3K27ac, and decreases chromatin accessibility. All panels show ChIP-qPCR at SMAD3-1, SMAD3-2, and CDKN2BAS locus regions with input normalization in HCASMC. a TCF21 binding, b H3K27ac level, and c chromatin accessibility (ATAC-PCR) were evaluated under the conditions of TCF21 (TCF21-KD) or scrambled (Ctrl) siRNA transfections or lentivirus overexpression of TCF21 (TCF21-OE). d HDAC1 and e HDAC2 binding were evaluated by ChIP-qPCR with TCF21-KD. f Serial ChIP-qPCR with TCF21 first IP followed by HDAC1 or HDAC2 second IP. g Serial ChIP-qPCR with HDAC1 first IP followed by TCF21 second IP. h Serial ChIP-qPCR with HDAC2 first IP followed by TCF21 second IP (mean ± SD, n = 3)
Fig. 4
Fig. 4
JUN and TCF21 co-occupy chromatin at SMAD3 and CDKN2BAS loci. a Serial ChIP-qPCR with JUN first IP followed by TCF21 second IP. b Serial ChIP-qPCR with TCF21 first IP followed by JUN second IP (mean ± SD, n = 3). c Interaction between TCF21 and JUN on chromatin was evaluated by ChIP-western blotting, IP with myc-TCF21 (top) followed by JUN western, or with reverse conditions (bottom). Myc-tagged TCF21 and non-tagged JUN expression constructs were transfected into HEK293 cells. d Co-immunoprecipitation of TCF21 and JUN with IP of myc-TCF21 followed by JUN western (top) and reverse conditions (bottom). Myc-tagged TCF21 and non-tagged JUN were transfected into HEK293 cells
Fig. 5
Fig. 5
JUN and TCF21 co-localize genome-wide and regulate H3K27ac chromatin modification. a JUN ChIPseq peaks and b TCF21 peaks were extended +/− 2 kb from summit, and a density plot created for RPM normalized enrichment levels of JUN, TCF21, and H3K27ac ChIPseq in HACSMC, along with control transcription factor HNF1A. c The number of TCF21 peaks overlapped with JUN binding for distances between peaks less than 1 kb, 5 kb, and 10 kb. d Heatmap distribution of ChIPseq data for JUN, TCF21, and H3K27ac with control (Ctrl) or JUN knockdown (JUN-KD) centered on JUN peaks, or e TCF21 knockdown (TCF21-KD) centered on TCF21 peaks within a 4-kb window. f Peak length distribution of Ctrl or JUN-KD, and g Ctrl or TCF21-KD H3K27ac ChIPseq in HCASMC (box plots with Wilcoxon tests, P < 2.2e−16). h H3K27ac ChIPseq enrichment level of Ctrl or JUN-KD, and i Ctrl or TCF21-KD, centered on all H3K27ac peaks with +/− 2 kb extension, normalized by RPM
Fig. 6
Fig. 6
JUN and TCF21 regulate chromatin accessibility at CAD loci. a Venn diagram showing the number of overlaps between TCF21-regulated (intersection of TCF21 knockdown upregulated and overexpression downregulated ATAC peaks, q < 0.01) and JUN-regulated (intersection of two JUN knockdown downregulated ATAC peaks, q < 1e−10 and fold change < 0.5) open chromatin regions in HCASMC. b Venn diagram showing the number of overlaps between JUN plus TCF21-regulated open chromatin regions (from a) and TCF21 or JUN peaks (q < 0.01) in HCASMC. g Heatmap distribution of ATACseq for control (Ctrl) or two JUN knockdowns (KD1 and KD2), ATACseq of control (Ctrl) or TCF21 knockdown (TCF21-KD) or TCF21 overexpression (TCF21-OE), and ChIPseq of JUN and TCF21, all centered on JUN plus TCF21-regulated open chromatin regions within a 4-kb window in HCASMC. HNF1A serves as a control transcription factor. d Heatmap distribution of ATACseq peaks for Ctrl, JUN-KD1, or JUN-KD2, and ATACseq of Ctrl, TCF21-KD, or TCF21-OE on genes within 10 kb upstream and 5 kb downstream of the transcription start site, in loci located in JUN plus TCF21-regulated open chromatin regions. e Pattern of ATACseq mapping of open chromatin at the human SMAD3 and CDKN2BAS loci, with JUN knockdown (JUN-KD1, -KD2), TCF21 knockdown (TCF21-KD), or TCF21 overexpression (TCF21-OE). ENCODE-layered H3K27ac data are also shown. f Biological processes and g GAD disease enrichment from DAVID Gene Ontology analysis of genes located in JUN plus TCF21-regulated open chromatin regions. Genes were assigned by GREAT with “single nearest” mode
Fig. 7
Fig. 7
Allele-specific binding of JUN, TCF21, and H3K27ac enrichment at the SMAD3 locus. a HaploChIP of JUN, TCF21, H3K27ac, and ATAC at rs17293632 was evaluated by qPCR with allele-specific TaqMan probes in heterozygous HCASMCs. Enrichment was normalized with input, and data was shown as C/T ratio. b SMAD3 expression levels in wild type (WT), deletion (DEL), or converted (ALT) HEK293 CRISPR/Cas9 edited cell lines as detected by qPCR. The genomic sequence of CRISPR edits are shown for corresponding cell lines. c ChIP-qPCR of JUN, d TCF21, e H3K27ac, and f ATAC-qPCR showing the differential transcription factor binding, H3K27ac modification, or ATACseq open chromatin at SMAD3-1, SMAD3-2, and CDKN2BAS locus regions for WT or CRISPR HEK293 cell lines. Dominant negative A-FOS was expressed as a positive control of JUN inhibition, and all data normalized versus input (mean ± SD, n = 3)
Fig. 8
Fig. 8
Mechanism of TCF21 and AP-1 epigenetic interactions in the context of CAD-associated genetic loci. TCF21 is a bHLH transcription factor associated with CAD as depicted here due to allelic variation in causal variants (Y), and its transcriptional regulatory function accounts for the attributable genetic risk at the 6q23.2 locus (brown oval). TCF21 binding is enriched in other CAD loci, where it interacts with AP-1 factors (JUN) that co-localize at these sites (gray oval). JUN promotes recruitment of HAT p300 to promote H3K27ac histone acetylation and open chromatin to recruit TFs, including TCF21, which in turn recruits HDACs 1 and 2 that function to oppose AP-1 effects. These epigenetic effects contribute in cis to the regulation of expression of the causal gene through alteration in the binding of the causal TF through CAD-associated variant Z (green circle) or other mechanisms of disease at this locus. Such interactions likely contribute to attributable genetic risk at both the TCF21 and downstream loci

References

    1. Howson JMM, et al. Fifteen new risk loci for coronary artery disease highlight arterial-wall-specific mechanisms. Nat Genet. 2017;49:1113–1119. doi: 10.1038/ng.3874. - DOI - PMC - PubMed
    1. Klarin D, et al. Genetic analysis in UK Biobank links insulin resistance and transendothelial migration pathways to coronary artery disease. Nat Genet. 2017;49:1392–1397. doi: 10.1038/ng.3914. - DOI - PMC - PubMed
    1. Nelson CP, et al. Association analyses based on false discovery rate implicate new loci for coronary artery disease. Nat Genet. 2017;49:1385–1391. doi: 10.1038/ng.3913. - DOI - PubMed
    1. van der Harst P, Verweij N. The identification of 64 novel genetic loci provides an expanded view on the genetic architecture of coronary artery disease. Circ Res. 2017;122:433–443. doi: 10.1161/CIRCRESAHA.117.312086. - DOI - PMC - PubMed
    1. Maurano MT, et al. Systematic localization of common disease-associated variation in regulatory DNA. Science. 2012;337:1190–1195. doi: 10.1126/science.1222794. - DOI - PMC - PubMed

Publication types

MeSH terms