The Role of DNA Methylation in Transcriptional Regulation of Pro-Nociceptive Genes in Rat Trigeminal Ganglia

Abstract

Epigenetic modulation by DNA methylation is associated with aberrant gene expression in sensory neurons, which consequently leads to pathological pain responses. In this study, we sought to investigate whether peripheral inflammation alters global DNA methylation in trigeminal ganglia (TG) and results in abnormal expression of pro-nociceptive genes. Our results show that peripheral inflammation remotely reduced the level of global DNA methylation in rat TG with a concurrent reduction in DNMT1 and DNMT3a expression. Using unbiased steps, we selected the following pro-nociceptive candidate genes that are potentially regulated by DNA methylation: TRPV1, TRPA1, P2X3, and PIEZO2. Inhibition of DNMT with 5-Aza-dC in dissociated TG cells produced dose-dependent upregulation of TRPV1, TRPA1, and P2X3. Systemic treatment of animals with 5-Aza-dC significantly increased the expression of TRPV1, TRPA1, and PIEZO2 in TG. Furthermore, the overexpression of DNMT3a, as delivered by a lentiviral vector, significantly downregulated TRPV1 and PIEZO2 expression and also reliably decreased TRPA1 and P2X3 transcripts. MeDIP revealed that this overexpression also significantly enhanced methylation of CGIs associated with TRPV1 and TRPA1. In addition, bisulfite sequencing data indicated that the CGI associated with TRPA1 was methylated in a pattern catalyzed by DNMT3a. Taken together, our results show that all 4 pro-nociceptive genes are subject to epigenetic modulation via DNA methylation, likely via DNMT3a under inflammatory conditions. These findings provide the first evidence for the functional importance of DNA methylation as an epigenetic factor in the transcription of pro-nociceptive genes in TG that are implicated in pathological orofacial pain responses.

Keywords: DNA methylation; inflammation; pain; sensory ganglia.

Figure 1.

Effects of inflammation on DNA methylation and on enzymes that regulate DNA methylation in TG. (A) The extent of gDNA methylation was compared between naïve and CFA-treated rats (n = 5/group). (B) mRNA levels for DNMT1, DNMT3a, DNMT3b, GADD45α, MBD4, TET1, TET2, and TET3 detected by qRT-PCR were compared between naïve and CFA-treated rats on day 3 (CFA, 3d) (n = 4/group).*P < .05.

Figure 2.

Effects of 5-Aza-dC on mRNA level of pro-nociceptive genes in TG. (A) TG cultures treated with vehicle (PBS) or 1 of the 3 doses of 5-Aza-dC for 24 hours. N = 5 to 7. qRT-PCR was conducted as described in methods. *P < .05 compared to the vehicle condition analyzed with one-way ANOVA. (B) 5-Aza-dC or its vehicle (PBS) was administered (i.p.) for 3 consecutive days after which TG was extracted and analyzed. Each group consisted of 5 to 7 rats. *P < .05 between the 2 groups analyzed with Student’s t-test.

Figure 3.

Effects of LV-mediated DNMT3a overexpression on mRNA of pro-nociceptive genes in TG. mRNA fold changes in TG cultures treated with pseudo-viruses (LV-CMVeNR1-EGFP-DNMT3a for EGFP-DNMT3a or LV-CMVeNR1-EGFP for EGFP) were compared to those of untreated TG cultures (naïve). Median values with interquartile range in box plots are presented. N = 9. *P < .05 compared to naive condition.

Figure 4.

Methylation analyses of the TRPA1 gene. (A) Schematic presentation of PCR primers for TRPA1 MeDIP and bisulfite-modified TRPA1 CGI. Primers were designed based on sequences of the rat TRPA1 gene. (B) Representative methylated clones of BS sequencing of the TRPA1 CGI. CpGs are presented as circles aligned in a row for each individual clone. Open circles indicate unmethylated CpGs, and solid circles indicate methylated CpGs. (C) MeDIP results of the TRPA1 CGI. Data are presented in the same format as in Figure 3. Experiments were performed as described in methods.N = 7~8. *P < .05 compared to naive condition.

Figure 5.

Methylation analyses of the TRPV1 gene. (A) Schematic presentation of PCR primers for TRPA1 MeDIP. Primers were designed based on sequences of the rat TRPV1 gene. (B) MeDIP results of the TRPV1 CGI. Others are the same as Figure 4.