Limits...
Control of VEGF-A transcriptional programs by pausing and genomic compartmentalization.

Kaikkonen MU, Niskanen H, Romanoski CE, Kansanen E, Kivelä AM, Laitalainen J, Heinz S, Benner C, Glass CK, Ylä-Herttuala S - Nucleic Acids Res. (2014)

Bottom Line: We demonstrate that sites of active transcription are more likely to engage in chromatin looping and cell type-specific transcriptional activity reflects the boundaries of chromatin interactions.We provide evidence that these compartments are enriched for clusters of regulatory regions such as super-enhancers and for disease-associated single nucleotide polymorphisms (SNPs).Collectively, these findings provide new insights into mechanisms behind VEGF-A-regulated transcriptional programs in endothelial cells.

View Article: PubMed Central - PubMed

Affiliation: A.I. Virtanen Institute for Molecular Sciences, Department of Biotechnology and Molecular Medicine, University of Eastern Finland, PO Box 1627, 70211 Kuopio, Finland minna.kaikkonen@uef.fi.

Show MeSH

Related in: MedlinePlus

VEGF-regulated compartments are enriched for clusters of enhancers. (A) Sequence motifs associated with the regulatory regions of chromatin compartments containing VEGF-regulated genes when compared to randomly selected genomic loci. Regulatory regions were defined by a significant enrichment of H3K4me1, H3K4me2 or H3K27ac. (B) Number of ETS1 peaks and H3K27ac regions associated with VEGF-regulated compartments or non-regulated compartments. The average number per compartment is indicated in parentheses. (C) Box-and-whisker plots of the expression level (RPKM) of genes interacting with super-enhancers (blue) or normal enhancers (orange) within VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes. Whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar and the mean by a red line. P-value based on one-tailed t-test is shown. (D) Box-and-whisker plots of the fold change in expression of genes interacting with super-enhancers or normal enhancers within the VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes and whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar. For 81 VEGF-regulated compartments, the genes were divided into upregulated (pink boxes 1–2; fold change > 1) and downregulated genes (green boxes 3–4; fold change < 1) whereas for 16 VEGF-upregulated (pink boxes 5–6) and 9 VEGF-downregulated (green boxes 7–8) compartments all active genes are shown. P-values based on one-tailed t-test are indicated. (E) Preferential interactions of VEGF-regulated gene promoters (2 kB around TSS) with endothelial-specific lineage-determining transcription factor peaks based on ChIP-Seq data from (3,37). Heat map represents enrichment ratios, i.e. observed association frequency relative to expected frequency (association strength; red) for each comparison exhibiting significant P-values < 0.005. Non-significant enrichment pairs are represented by gray boxes. Results from the non-regulated genes are represented as median of four randomly selected sets of 108 non-regulated genes to allow comparison to 108 VEGF-regulated genes. (F and G) Overlap of SNPs and disease-associated genes in the NHGRI Catalog of Published GWAS and DisGeNET database (June 2014), respectively, with HUVEC chromatin compartments. Average (F) SNP or (G) gene count is shown for all (2022), VEGF-regulated (81), VEGF-upregulated (16), and VEGF-downregulated (9) compartments. *P-value < 0.007, ** < 0.0002, *** < 2.5E-11, hypergeometric test. (H) Significant interactions at CXCL-loci (chr4:74,433,700-75,000,000). Regulatory regions defined by H3K27ac, gene annotations and the GWAS SNPs are shown. Of the SNPs, only rs1371799 is located within a HUVEC enhancer.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4227755&req=5

Figure 6: VEGF-regulated compartments are enriched for clusters of enhancers. (A) Sequence motifs associated with the regulatory regions of chromatin compartments containing VEGF-regulated genes when compared to randomly selected genomic loci. Regulatory regions were defined by a significant enrichment of H3K4me1, H3K4me2 or H3K27ac. (B) Number of ETS1 peaks and H3K27ac regions associated with VEGF-regulated compartments or non-regulated compartments. The average number per compartment is indicated in parentheses. (C) Box-and-whisker plots of the expression level (RPKM) of genes interacting with super-enhancers (blue) or normal enhancers (orange) within VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes. Whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar and the mean by a red line. P-value based on one-tailed t-test is shown. (D) Box-and-whisker plots of the fold change in expression of genes interacting with super-enhancers or normal enhancers within the VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes and whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar. For 81 VEGF-regulated compartments, the genes were divided into upregulated (pink boxes 1–2; fold change > 1) and downregulated genes (green boxes 3–4; fold change < 1) whereas for 16 VEGF-upregulated (pink boxes 5–6) and 9 VEGF-downregulated (green boxes 7–8) compartments all active genes are shown. P-values based on one-tailed t-test are indicated. (E) Preferential interactions of VEGF-regulated gene promoters (2 kB around TSS) with endothelial-specific lineage-determining transcription factor peaks based on ChIP-Seq data from (3,37). Heat map represents enrichment ratios, i.e. observed association frequency relative to expected frequency (association strength; red) for each comparison exhibiting significant P-values < 0.005. Non-significant enrichment pairs are represented by gray boxes. Results from the non-regulated genes are represented as median of four randomly selected sets of 108 non-regulated genes to allow comparison to 108 VEGF-regulated genes. (F and G) Overlap of SNPs and disease-associated genes in the NHGRI Catalog of Published GWAS and DisGeNET database (June 2014), respectively, with HUVEC chromatin compartments. Average (F) SNP or (G) gene count is shown for all (2022), VEGF-regulated (81), VEGF-upregulated (16), and VEGF-downregulated (9) compartments. *P-value < 0.007, ** < 0.0002, *** < 2.5E-11, hypergeometric test. (H) Significant interactions at CXCL-loci (chr4:74,433,700-75,000,000). Regulatory regions defined by H3K27ac, gene annotations and the GWAS SNPs are shown. Of the SNPs, only rs1371799 is located within a HUVEC enhancer.

Mentions: To identify factors that might be responsible for selecting the compartments for VEGF-mediated regulation, we further studied the regulatory elements located within the compartments. The VEGF-regulated compartments were enriched for the binding motifs of ETS1, AP-1, FOXP1 and NF1 at their regulatory regions (Figure 6A). To see if the amount of ETS1 LDTF was higher in the VEGF-regulated compartments, we took advantage of the public ChIP-Seq data (3). To our surprise, we noticed that VEGF-regulated compartments had significantly more ETS1 peaks than the non-regulated compartments either measured by the number of peaks per compartment or RPKM (Figure 6B, Supplementary Figure S6). This correlated with a higher number of H3K27ac regions in VEGF-regulated compartments, suggesting that VEGF-regulated compartments are enriched for clusters of regulatory regions (Figure 6B, Supplementary Figure S6).


Control of VEGF-A transcriptional programs by pausing and genomic compartmentalization.

Kaikkonen MU, Niskanen H, Romanoski CE, Kansanen E, Kivelä AM, Laitalainen J, Heinz S, Benner C, Glass CK, Ylä-Herttuala S - Nucleic Acids Res. (2014)

VEGF-regulated compartments are enriched for clusters of enhancers. (A) Sequence motifs associated with the regulatory regions of chromatin compartments containing VEGF-regulated genes when compared to randomly selected genomic loci. Regulatory regions were defined by a significant enrichment of H3K4me1, H3K4me2 or H3K27ac. (B) Number of ETS1 peaks and H3K27ac regions associated with VEGF-regulated compartments or non-regulated compartments. The average number per compartment is indicated in parentheses. (C) Box-and-whisker plots of the expression level (RPKM) of genes interacting with super-enhancers (blue) or normal enhancers (orange) within VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes. Whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar and the mean by a red line. P-value based on one-tailed t-test is shown. (D) Box-and-whisker plots of the fold change in expression of genes interacting with super-enhancers or normal enhancers within the VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes and whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar. For 81 VEGF-regulated compartments, the genes were divided into upregulated (pink boxes 1–2; fold change > 1) and downregulated genes (green boxes 3–4; fold change < 1) whereas for 16 VEGF-upregulated (pink boxes 5–6) and 9 VEGF-downregulated (green boxes 7–8) compartments all active genes are shown. P-values based on one-tailed t-test are indicated. (E) Preferential interactions of VEGF-regulated gene promoters (2 kB around TSS) with endothelial-specific lineage-determining transcription factor peaks based on ChIP-Seq data from (3,37). Heat map represents enrichment ratios, i.e. observed association frequency relative to expected frequency (association strength; red) for each comparison exhibiting significant P-values < 0.005. Non-significant enrichment pairs are represented by gray boxes. Results from the non-regulated genes are represented as median of four randomly selected sets of 108 non-regulated genes to allow comparison to 108 VEGF-regulated genes. (F and G) Overlap of SNPs and disease-associated genes in the NHGRI Catalog of Published GWAS and DisGeNET database (June 2014), respectively, with HUVEC chromatin compartments. Average (F) SNP or (G) gene count is shown for all (2022), VEGF-regulated (81), VEGF-upregulated (16), and VEGF-downregulated (9) compartments. *P-value < 0.007, ** < 0.0002, *** < 2.5E-11, hypergeometric test. (H) Significant interactions at CXCL-loci (chr4:74,433,700-75,000,000). Regulatory regions defined by H3K27ac, gene annotations and the GWAS SNPs are shown. Of the SNPs, only rs1371799 is located within a HUVEC enhancer.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4227755&req=5

Figure 6: VEGF-regulated compartments are enriched for clusters of enhancers. (A) Sequence motifs associated with the regulatory regions of chromatin compartments containing VEGF-regulated genes when compared to randomly selected genomic loci. Regulatory regions were defined by a significant enrichment of H3K4me1, H3K4me2 or H3K27ac. (B) Number of ETS1 peaks and H3K27ac regions associated with VEGF-regulated compartments or non-regulated compartments. The average number per compartment is indicated in parentheses. (C) Box-and-whisker plots of the expression level (RPKM) of genes interacting with super-enhancers (blue) or normal enhancers (orange) within VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes. Whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar and the mean by a red line. P-value based on one-tailed t-test is shown. (D) Box-and-whisker plots of the fold change in expression of genes interacting with super-enhancers or normal enhancers within the VEGF-regulated chromatin compartments. Boxes encompass the 25th to 75th percentile changes and whiskers extend to 10th and 90th percentiles. The median fold change is indicated by the central horizontal bar. For 81 VEGF-regulated compartments, the genes were divided into upregulated (pink boxes 1–2; fold change > 1) and downregulated genes (green boxes 3–4; fold change < 1) whereas for 16 VEGF-upregulated (pink boxes 5–6) and 9 VEGF-downregulated (green boxes 7–8) compartments all active genes are shown. P-values based on one-tailed t-test are indicated. (E) Preferential interactions of VEGF-regulated gene promoters (2 kB around TSS) with endothelial-specific lineage-determining transcription factor peaks based on ChIP-Seq data from (3,37). Heat map represents enrichment ratios, i.e. observed association frequency relative to expected frequency (association strength; red) for each comparison exhibiting significant P-values < 0.005. Non-significant enrichment pairs are represented by gray boxes. Results from the non-regulated genes are represented as median of four randomly selected sets of 108 non-regulated genes to allow comparison to 108 VEGF-regulated genes. (F and G) Overlap of SNPs and disease-associated genes in the NHGRI Catalog of Published GWAS and DisGeNET database (June 2014), respectively, with HUVEC chromatin compartments. Average (F) SNP or (G) gene count is shown for all (2022), VEGF-regulated (81), VEGF-upregulated (16), and VEGF-downregulated (9) compartments. *P-value < 0.007, ** < 0.0002, *** < 2.5E-11, hypergeometric test. (H) Significant interactions at CXCL-loci (chr4:74,433,700-75,000,000). Regulatory regions defined by H3K27ac, gene annotations and the GWAS SNPs are shown. Of the SNPs, only rs1371799 is located within a HUVEC enhancer.
Mentions: To identify factors that might be responsible for selecting the compartments for VEGF-mediated regulation, we further studied the regulatory elements located within the compartments. The VEGF-regulated compartments were enriched for the binding motifs of ETS1, AP-1, FOXP1 and NF1 at their regulatory regions (Figure 6A). To see if the amount of ETS1 LDTF was higher in the VEGF-regulated compartments, we took advantage of the public ChIP-Seq data (3). To our surprise, we noticed that VEGF-regulated compartments had significantly more ETS1 peaks than the non-regulated compartments either measured by the number of peaks per compartment or RPKM (Figure 6B, Supplementary Figure S6). This correlated with a higher number of H3K27ac regions in VEGF-regulated compartments, suggesting that VEGF-regulated compartments are enriched for clusters of regulatory regions (Figure 6B, Supplementary Figure S6).

Bottom Line: We demonstrate that sites of active transcription are more likely to engage in chromatin looping and cell type-specific transcriptional activity reflects the boundaries of chromatin interactions.We provide evidence that these compartments are enriched for clusters of regulatory regions such as super-enhancers and for disease-associated single nucleotide polymorphisms (SNPs).Collectively, these findings provide new insights into mechanisms behind VEGF-A-regulated transcriptional programs in endothelial cells.

View Article: PubMed Central - PubMed

Affiliation: A.I. Virtanen Institute for Molecular Sciences, Department of Biotechnology and Molecular Medicine, University of Eastern Finland, PO Box 1627, 70211 Kuopio, Finland minna.kaikkonen@uef.fi.

Show MeSH
Related in: MedlinePlus