Changes in H3K27ac at Gene Regulatory Regions in Porcine Alveolar Macrophages Following LPS or PolyIC Exposure

Abstract

Changes in chromatin structure, especially in histone modifications (HMs), linked with chromatin accessibility for transcription machinery, are considered to play significant roles in transcriptional regulation. Alveolar macrophages (AM) are important immune cells for protection against pulmonary pathogens, and must readily respond to bacteria and viruses that enter the airways. Mechanism(s) controlling AM innate response to different pathogen-associated molecular patterns (PAMPs) are not well defined in pigs. By combining RNA sequencing (RNA-seq) with chromatin immunoprecipitation and sequencing (ChIP-seq) for four histone marks (H3K4me3, H3K4me1, H3K27ac and H3K27me3), we established a chromatin state map for AM stimulated with two different PAMPs, lipopolysaccharide (LPS) and Poly(I:C), and investigated the potential effect of identified histone modifications on transcription factor binding motif (TFBM) prediction and RNA abundance changes in these AM. The integrative analysis suggests that the differential gene expression between non-stimulated and stimulated AM is significantly associated with changes in the H3K27ac level at active regulatory regions. Although global changes in chromatin states were minor after stimulation, we detected chromatin state changes for differentially expressed genes involved in the TLR4, TLR3 and RIG-I signaling pathways. We found that regions marked by H3K27ac genome-wide were enriched for TFBMs of TF that are involved in the inflammatory response. We further documented that TF whose expression was induced by these stimuli had TFBMs enriched within H3K27ac-marked regions whose chromatin state changed by these same stimuli. Given that the dramatic transcriptomic changes and minor chromatin state changes occurred in response to both stimuli, we conclude that regulatory elements (i.e. active promoters) that contain transcription factor binding motifs were already active/poised in AM for immediate inflammatory response to PAMPs. In summary, our data provides the first chromatin state map of porcine AM in response to bacterial and viral PAMPs, contributing to the Functional Annotation of Animal Genomes (FAANG) project, and demonstrates the role of HMs, especially H3K27ac, in regulating transcription in AM in response to LPS and Poly(I:C).

Keywords: Poly(I:C); alveolar macrophages; chromatin state; histone modifications; lipopolysaccharide; pig; transcriptome.

FIGURE 1

Transcriptional response of alveolar macrophages to LPS or Poly (I:C). (A) QPCR results of inflammatory marker genes in AM stimulated with LPS and Poly (I:C). Data are shown as the mean of Log2 Fold change +/–SD (n = 8). Kruskal-Wallis-one-way ANOVA was used to compare treatments with non-stimulated AM. Significance was set at P < 0.05. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, see Supplementary Table S2 for details. (B) Principal component analysis of transformed RNA-seq reads counts for whole transcriptomes. Axis indicate component scores (C) Heatmap showing DEG between non-stimulated and LPS or Poly(I:C) stimulated AM at 2 and 6 h. Color code is based on Z-score of log₂ transformed CPM across all samples. Genes, treatments and timepoints were hierarchically clustered (row, genes; columns, treatments and timepoints).

FIGURE 2

Dot plots showing GO term enrichment for DEGs. The top 5 most significantly enriched biological processes (A) and KEGG pathways (B) are displayed. The X axis corresponds to downregulated and upregulated genes at 2 and 6 h post stimulation, and Y axis represents the GO terms. The size and color of the dots corresponds to the number of DEGs associated with the GO terms and pathways and the P-values of hypergeometric tests, respectively.

FIGURE 3

Chromatin histone modification analysis of porcine alveolar macrophages. (A) Principal component analysis plot of histone ChIP-seq samples. (B) The average genome wide histone fold enrichment (average log₂ IP/input) near TSS (± 5.0 Kb) was calculated for each individual histone mark. Non-stimulated alveolar macrophages at 2 h were displayed in the figure as representative.

FIGURE 4

The relationship between H3K27ac signal and gene expression. (A) Histogram of the H3K27ac signal intensity and all expressed genes in alveolar macrophages stimulated with LPS treatment at 2 h. Genes were divided into four quartiles, high expression (red), middle high (green), middle low (purple) and low expression (blue). (B) Peak visualization using the Integrative Genomics Viewer (IGV) of a highly expressed gene ACTB and a low expressed gene MYOT in non-stimulated alveolar macrophages.

FIGURE 5

Chromatin states of porcine alveolar macrophages. (A) At left is shown a heatmap of the emission parameters, each row corresponds to a different state, and column for each histone mark. The darker blue color corresponds to a greater probability of observing the histone mark. At right is shown the description of the specific chromatin state (B) The TSS neighborhood heatmap shows the overlap enrichment for each state for each 200-bp bin within 2 kb around a set of TSSs. (C) Heatmap showing the emission parameters of non-stimulated and stimulated alveolar macrophages with LPS or Poly (I:C) at 2 h. The heatmap displays the overlap enrichment of the histone mark on the current pig genome annotation.

FIGURE 6

Changes in chromatin state of the TNF gene in response to LPS and Poly(I:C) at 2 h and 6 h. (A) IGV screenshots showing DHMRs-H3K27ac with chromatin states around 1kb of promoter regions of TNF gene in respond to treatments. Annotation of the chromatin states is shown as legend on the bottom and as (B) summary table. (C) Gene expression values of TNF gene from RNA-seq of stimulated AM.

FIGURE 7

Integrated analysis of H3K27ac modification and RNA expression response to inflammatory stimuli demonstrates enrichment of binding motifs for TF induced by these stimuli. (A) Ontology enrichment clusters of upregulated transcription factors that were induced by the LPS-2 h treatment whose TFBM was enriched in regions with a gain of H3K27ac modification of DEG. The most statistically significant term within similar term clusters (enclosed by dotted lines) was chosen to represent the cluster. Term color is given by cluster ID and the size of the terms is given by –log10 P-value. The stronger the similarity among terms, the thicker the edges between them. (B) PPI network of upregulated transcription factors with enriched H3K27ac motifs of DEG for LPS-2h treatment. A unique color is assigned to each MCODE network.