Bromate-induced Changes in p21 DNA Methylation and Histone Acetylation in Renal Cells

Abstract

Bromate (BrO3-) is a water disinfection byproduct (DBP) previously shown to induce nephrotoxicity in vitro and in vivo. We recently showed that inhibitors of DNA methyltransferase 5-aza-2'-deoxycytidine (5-Aza) and histone deacetylase trichostatin A (TSA) increased BrO3- nephrotoxicity whereas altering the expression of the cyclin-dependent kinase inhibitor p21. Human embryonic kidney cells (HEK293) and normal rat kidney (NRK) cells were sub-chronically exposed to BrO3- or epigenetic inhibitors for 18 days, followed by 9 days of withdrawal. DNA methylation was studied using a modification of bisulfite amplicon sequencing called targeted gene bisulfite sequencing. Basal promoter methylation in the human p21 promoter region was substantially lower than that of the rat DNA. Furthermore, 5-Aza decreased DNA methylation in HEK293 cells at the sis-inducible element at 3 distinct CpG sites located at 691, 855, and 895 bp upstream of transcription start site (TSS). 5-Aza also decreased methylation at the rat p21 promoter about 250 bp upstream of the p21 TSS. In contrast, sub-chronic BrO3- exposure failed to alter methylation in human or rat renal cells. BrO3- exposure altered histone acetylation in NRK cells at the p21 TSS, but not in HEK293 cells. Interestingly, changes in DNA methylation induced by 5-Aza persisted after its removal; however, TSA- and BrO3--induced histone hyperacetylation returned to basal levels after 3 days of withdrawal. These data demonstrate novel sites within the p21 gene that are epigenetically regulated and further show that significant differences exist in the epigenetic landscape between rat and human p21, especially with regards to toxicant-induced changes in histone acetylation.

Keywords: p21; DNA methylation; bromate; chromatin immunoprecipitation; epigenetics; kidney; nephrotoxicity; next-generation sequencing.

Figure 1.

Sub-chronic dosing regimen for ${\text{BrO}}_3^{-}$ and epigenetic inhibitors. Cells were exposed to various concentrations of ${\text{BrO}}_3^{-}$ at log phase (after 24 h of seeding) for 72 h. A portion of the cells was used for reseeding and the rest for sample collection for targeted gene bisulfite sequencing and chromatin immunoprecipitation. This was called passage 1 (P1) and the regimen continued until 18 days (P6). Cells were allowed to recover by discontinuing the exposures for the next 3 passages (Days 21–27 or P7–P9) to obtain withdrawal samples.

Figure 2.

Schematic of p21 gene organization highlighting the loci of interest for (A) DNA methylation analysis: the human p21 promoter region adjacent to the transcription start site (hp21-TSS), the human transcription factor binding site called the sis-inducible element (hp21-SIE1), the rat p21 promoter region starting near the start site (rp21-TSS), and the rat p21 coding region (rp21-coding). (B) Histone acetylation analysis: the human p21 promoter region (hp21-ChIP) about 1000 bp upstream of the transcription start site and rat p21 promoter region (rp21-ChIP) at the TSS.

Figure 3.

Differential methylation analysis. Comparison of methylation between rat (A) and human (B) p21 transcription start sites. DNA methylation data are represented as the percent methylation of each CpG site in the analyzed fragments of the human and rat p21 promoter regions near the respective transcription start sites (rp21-TSS and hp21-TSS). C, Comparison of methylation in different regions of the rat and human p21 gene. Differential methylation data are represented as the percent DNA methylation of the transcription start site, sis-inducible element (SIE-1), and gene coding regions of human and rat p21. Next-generation sequencing data was analyzed using Bismark bisulfite mapper. Data are represented as the mean ± SEM of 3-independent experiments (n = 3).

Figure 4.

Comparison of basal DNA methylation of the p21 promoter region between HEK293 cells and freshly isolated human proximal tubule (hPT) cells. (A) Heat map of the site-specific percent DNA methylation changes as determined by targeted gene bisulfite sequencing in the human p21 promoter region at the transcription start site (hp21-TSS). (B) Comparison of methylation of human p21 promoter at the transcription factor binding site SIE-1 between HEK293 and hPT cells. Data are represented as the mean ± SEM of 3 different passages of HEK293 cells and 3 different pools of freshly isolated hPT cells (n = 3). *p < .05 compared with HEK293.

Figure 5.

Sub-chronic effect of ${\text{BrO}}_3^{-}$ and epigenetic inhibitors on HEK293 cell morphology and number. HEK293 cells were sub-chronically exposed to 0–10 ppm ${\text{BrO}}_3^{-}$ (A–F), vehicle control dimethyl sulfoxide (DMSO) (G), 100 nM TSA (H), or 40 µM 5-Aza (I) for 18 days (P6). The cell number data in (J) are represented as mean ± SEM of 3 separate passages (n = 3). ^#p < .05 compared with DMSO.

Figure 6.

Effect of ${\text{BrO}}_3^{-}$ and epigenetic inhibitors on HEK293 cell morphology and number after withdrawal. HEK293 cells were sub-chronically exposed to ${\text{BrO}}_3^{-}$ for 18 days and then allowed to recover (A–F), vehicle control dimethyl sulfoxide (DMSO) (G), 100 nM TSA (H), or 40 µM 5-Aza (I) for 9 days (P9). The cell number data in (J) are represented as mean ± SEM of 3 separate passages (n = 3). ^#p < .05 compared with DMSO.

Figure 7.

Acute and sub-chronic effects of ${\text{BrO}}_3^{-}$ on the percent DNA methylation of CpGs in the SIE-1 site of the human p21 promoter. HEK293 cells were treated with water (vehicle control for ${\text{BrO}}_3^{-}$), 0.001–100 ppm ${\text{BrO}}_3^{-}$, dimethyl sulfoxide (DMSO) (vehicle control for 5-Aza) or 40 µM 5-Aza (positive control) for 3 days or P1 (A and B), 9 days or P3 (C and D), 18 days or P6 (E and F). The cells were allowed to recover by discontinuing the exposures for 9 days or P9 (G and H). The first row in the heat map represents the position of the cytosine in the CpG dinucleotide context relative to the transcription start site. Data are represented as the average percent DNA methylation of 3 separate passages (n = 3) as determined by targeted gene bisulfite sequencing analysis. The effects of 5-Aza are emphasized in the bar graphs for the respective exposure times. Data are represented as the mean ± SEM of 3 different passages (n = 3). *p < .05 compared with DMSO.

Figure 8.

Sub-chronic effects of ${\text{BrO}}_3^{-}$ on acetylation of lysine 9 and 14 on histone H3 (H3K9/14 Ac) of the rat p21 promoter region. Normal rat kidney cells to were exposed to 0.001–100 ppm ${\text{BrO}}_3^{-}$ for 3 days or P1 (A), 9 days or P3 (B), and 18 days or P6 (C). The cells were allowed to recover by discontinuing exposures for 9 days or P9 (D). Data are represented as the relative fold increase in H3K9/14 Ac levels as normalized to GAPDH as determined using the chromatin immunoprecipitation assay. Data are represented as the mean ± SEM of 3 different passages (n = 3). *p < .05 compared with 0 ppm ${\text{BrO}}_3^{-}$ and ^#p < .05 compared with DMSO.