Polychlorinated biphenyls alter hepatic m6A mRNA methylation in a mouse model of environmental liver disease

Abstract

Exposure to polychlorinated biphenyls (PCBs) has been associated with liver injury in human cohorts and with nonalcoholic steatohepatitis (NASH) in mice fed a high fat diet (HFD). N (6)-methyladenosine (m6A) modification of mRNA regulates transcript fate, but the contribution of m6A modification on the regulation of transcripts in PCB-induced steatosis and fibrosis is unknown. This study tested the hypothesis that PCB and HFD exposure alters the levels of m6A modification in transcripts that play a role in NASH in vivo. Male C57Bl6/J mice were fed a HFD (12 wks) and administered a single oral dose of Aroclor1260, PCB126, or Aroclor1260 + PCB126. Genome-wide identification of m6A peaks was accomplished by m6A mRNA immunoprecipitation sequencing (m6A-RIP) and the mRNA transcriptome identified by RNA-seq. Exposure of HFD-fed mice to Aroclor1260 decreased the number of m6A peaks and m6A-containing genes relative to PCB vehicle control whereas PCB126 or the combination of Aroclor1260 + PCB126 increased m6A modification frequency. ∼41% of genes had one m6A peak and ∼49% had 2-4 m6A peaks. 117 m6A peaks were common in the four experimental groups. The Aroclor1260 + PCB126 exposure group showed the highest number (52) of m6A-peaks. qRT-PCR confirmed enrichment of m6A-containing fragments of the Apob transcript with PCB exposure. A1cf transcript abundance, m6A peak count, and protein abundance was increased with Aroclor1260 + PCB126 co-exposure. Irrespective of the PCB type, all PCB groups exhibited enriched pathways related to lipid/lipoprotein metabolism and inflammation through the m6A modification. Integrated analysis of m6A-RIP-seq and mRNA-seq identified 242 differentially expressed genes (DEGs) with increased or reduced number of m6A peaks. These data show that PCB exposure in HFD-fed mice alters the m6A landscape offering an additional layer of regulation of gene expression affecting a subset of gene responses in NASH.

Keywords: Epitranscriptome; High fat diet; Liver; PCBs; Readers; Writers; m6A.

Fig. 1.

Effect of PCB exposure in HFD-fed mice on m6A in total mouse liver RNA. Male C57BL6/J mice were exposed to 12 wks of a HFD and either vehicle control or a single oral dose of the indicated PCB (Supplemental Fig. 1). m6A levels were quantified in total RNA from individual livers using a colorimetric kit against a standard curve. The data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons post-hoc test: *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 2.

Summary of m6A peaks in liver genes identified by m6A-RIP-seq in mRNA from HFD-fed mice with control (vehicle, corn oil) or PCB-exposure. A) Heat map of m6A peaks in genes in the 4 mouse liver exposure groups. Data are the avg. of five individual mouse livers/exposure group. B) The distribution of m6A peaks in chromosomes. C) The distribution of m6A peaks in chromosomes normalized to the size of each chromosome (Mbp/chromosome). (Cont. = PCB vehicle control) D) The distribution of mA peaks identified in genes. E) The distribution of m6A peaks per gene in the 4 exposure group.

Fig. 3.

Comparison of m6A-containing genes and integrated analysis of DEGs with Ar1260+ PCB126 co-exposure and m6A peaks in the DEG. A) Number of m6A-peaks in genes in the livers from HFD control, Ar1260, PCB126, and Ar1260+ PCB126 co-exposed mice. The values are the number of genes and the % of total number of m6A-containing genes in that exposure group. B) Integrated analysis of DEGs regulated by Ar1260+ PCB126 co-exposure in HFD-fed mouse liver and differential m6A peaks in the DEGs. The size of the green dots corresponds to reduced m6A peaks/DEG whereas the bigger size of the red dots indicates higher m6A/peaks/DEG. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4.

qRT-PCR for Apob and m6A-RIP qRT-PCR on Apob m6A Peak Specific Amplicons. A) The top panel shows qRT-PCR for Apob in polyA-selected mRNA from all 40 livers (n = 10 from the HFD-fed mice in the PCB vehicle control and in each of the three PCB exposure groups: Ar1260, PCB126, and Ar1260+ PCB126 co-exposed mice. Apob was normalized to Gapdh. For A bottom panel and B-D, m6A-RIP was performed on 5 livers for each experimental group. For A bottom panel, qRT-PCR was performed on these m6A-RIP samples for Apob (same primers used in A top panel). For B, C, and D, primers were designed for three regions of the Apob gene containing m6A peaks identified in m6A-RIP-seq and containing RRACH motifs (Supplementary Table 1). The primers are in brown letters, the RRACH motif is bold and underlined. qRT-PCR was performed to examine the enrichment of the three m6A peak-containing Apob regions in each of the four experimental groups. The data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons post-hoc test: *p < 0.05, **p < 0.01. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5.

Western blots for FZD, A1CF, and BCL9L that showed an increase in m6A peaks with PCB exposures. Representative western blots for FZD (A), A21DCF(B), and BCL9L (C) are shown. Bands were normalized to Ponceau S and to the average of those values within individual blots. The bar graphs show the normalized Western blot protein quantification from the five samples used for m6A RIP-seq. The data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons post-hoc test: **p < 0.01, *p < 0.05.