Comparative Analysis of Promoters and Enhancers in the Pituitary Glands of the Bama Xiang and Large White Pigs

Abstract

The epigenetic regulation of gene expression is implicated in complex diseases in humans and various phenotypes in other species. There has been little exploration of regulatory elements in the pig. Here, we performed chromatin immunoprecipitation coupled with high-throughput sequencing (ChIP-Seq) to profile histone H3 lysine 4 trimethylation (H3K4me3) and histone H3 lysine 27 acetylation (H3K27ac) in the pituitary gland of adult Bama Xiang and Large White pigs, which have divergent evolutionary histories and large phenotypic differences. We identified a total of 65,044 non-redundant regulatory regions, including 23,680 H3K4me3 peaks and 61,791 H3K27ac peaks (12,318 proximal and 49,473 distal), augmenting the catalog of pituitary regulatory elements in pigs. We found 793 H3K4me3 and 3,602 H3K27ac peaks that show differential activity between the two breeds, overlapping with genes involved in the Notch signaling pathway, response to growth hormone (GH), thyroid hormone signaling pathway, and immune system, and enriched for binding motifs of transcription factors (TFs), including JunB, ATF3, FRA1, and BATF. We further identified 2,025 non-redundant super enhancers from H3K27ac ChIP-seq data, among which 302 were shared in all samples of cover genes enriched for biological processes related to pituitary function. This study generated a valuable dataset of H3K4me3 and H3K27ac regions in porcine pituitary glands and revealed H3K4me3 and H3K27ac peaks with differential activity between Bama Xiang and Large White pigs.

Keywords: ChIP-Seq; H3K27ac; H3K4me3; differential peak activity; pigs; pituitary gland; super enhancer.

FIGURE 1

The intersection and conservation of H3K27ac and H3K4me3 peaks. (A) The intersection of H3K27ac and H3K4me3 peaks. (B) The correlation of 16,099 overlapped peaks between two histone modifications in four individuals. (C,D) Boxplots comparing the GC contents and GERP scores of H3K27ac and H3K4me3 peaks.

FIGURE 2

Enrichment of H3K4me3 and H3K27ac peaks across genomic features and transcription start sites. (A) Distribution of H3K4me3 and H3K27ac peaks across genome. (B) Heatmaps depicting normalized ChIP-seq signal (H3K27ac and H3K4me3) at 3 kb near the TSS, sorted by signal intensity.

FIGURE 3

Differential activity analysis on the H3K4me3 and H3K27ac peaks between two breeds and representative validation of differential peaks exhibiting distinct activity of the H3K27ac signal. (A,B) Volcano plot showing the differential activity analysis of H3K4me3 and H3K27ac peaks between BMX and LW breeds. (C,D) Unsupervised hierarchical clustering of the top differential H3K4me3 and H3K27ac peaks between BMX and LW samples. (E,F) The activity track of H3K27ac regions of (Chr13: 75916467–75920943) and (Chr2: 25338342–25347879) with increased hyper-acetylation in the BMX pituitary was related to genes of EPHB1 and FJX1 respectively. (G,H) The activity track of H3K27ac regions of (Chr7: 34554890–34556263) and (Chr3: 17096021–17100322) with higher H3K27ac enrichment in LW was related with KCNK5 and TGFB1I1, respectively.

FIGURE 4

Analysis of super enhancers. (A) The distribution of super enhancers in the four individuals. (B) Enriched Gene Ontology terms for genes covered by the 302 super enhancers shared cross all samples. (C) Volcano plot showing the differential activity analysis of super enhancer between two breeds. (D,E) The tracks of H3K27ac activity and gene in the two representative super enhancers that show differential activity between two breeds for reference (Chr15:109448708–109526513 and Chr13:143955853–144005651).