The site specific demethylation in the 5'-regulatory area of NMDA receptor 2B subunit gene associated with CIE-induced up-regulation of transcription

Abstract

Background: The NMDA receptor represents a particularly important site of ethanol action in the CNS. We recently reported that NMDA receptor 2B (NR2B) gene expression was persistently up-regulated following chronic intermittent ethanol (CIE) treatment. Increasing evidence that epigenetic mechanisms are involved in dynamic and long-lasting regulation of gene expression in multiple neuroadaptive processes prompted us to investigate the role of DNA methylation in mediating CIE-induced up-regulation of NR2B gene transcription. To dissect the changes of DNA methylation in the NR2B gene, we have screened a large number of CpG sites within its 5'-regulatory area following CIE treatment.

Methods: Primary cortical cultured neurons were subjected to ethanol treatment in a CIE paradigm. Bisulfite conversion followed by pyrosequencing was used for quantitative measurement and analysis of CpG methylation status within the 5'-regulatory area of the NR2B gene; chromatin immunoprecipitation (ChIP) assay was used to examine DNA levels associated with methylation and transcription factor binding. Electrophoretic mobility shift assay (EMSA) and in vitro DNA methylation assays were performed to determine the direct impact of DNA methylation on the interaction between DNA and transcription factor and promoter activity.

Results: Analysis of individual CpG methylation sites within the NR2B 5'regulatory area revealed three regions with clusters of site-specific CpG demethylation following CIE treatment and withdrawal. This was confirmed by ChIP showing similar decreases of methylated DNA in the same regions. The CIE-induced demethylation is characterized by being located near certain transcription factor binding sequences, AP-1 and CRE, and occurred during treatment as well as after ethanol withdrawal. Furthermore, the increase in vitro of methylated DNA decreased transcription factor binding activity and promoter activity. An additional ChIP assay indicated that the CIE-induced DNA demethylation is accompanied by increased occupation by transcription factors.

Conclusions: These results suggest an important role of DNA demethylation in mediating CIE-induced NR2B gene up-regulation, thus implicating a novel molecular site of alcohol action.

Figure 1. Schematic view of the 5′-regulatory area of the NR2B gene.

A. The structure of 5′-regulatory area includes promoter (from −2000 bp to 0) and 5′-UTR region (from +1 to +1100 bp) of mouse NR2B gene, containing previously identified transcription factor CREB and AP-1 binding sites as indicated. B. The detailed distribution of CpG dinucleotides and their relative positions to AP-1 and CRE in 5′-regulatory area of the NR2B promoter is shown (note that the bars represent one or multi-CpG sites depending on the thickness). C. a to g represent the seven regions chosen in this study being amplified by qChIP PCR using different primer pairs in NR2B regulatory area and relative position in 5′-regulatory area of the NR2B (for the sequences of these primers see Table S1).

Figure 2. 5′AZA induced up-regulation of NR2B gene expression.

The cortical neuronal cultures were treated by ethanol in CIE regimen or 5′AZA (48 hrs), respectively. Total RNA was isolated and real time PCR for NR2B and GAPDH mRNA expression was performed. Results are presented as the mean ratio to control ± SEM; *, p<0.05; **, p<0.01 compared with control levels.

Figure 3. CIE induced DNA demethylation in the specific CpG sites determined by bisulfite-pyrosequencing.

This graph outlines the relative CpG positions, where CIE induced significantly demethylation, to the transcription start site and the binding sites of AP-1 and CREB. Methylation status of individual CpG was determined by bisulfite-pyrosequencing at 116 CpG sites in the 5′-regulatory area of the NR2B gene. All bars represent the average basal methylation level (%) of individual CpG (control cells). The arrows in red only indicate these CpG sites where CIE induced significant demethylation (the detailed alterations are shown in Figure 4). The locations of transcription factor AP-1 and CREB binding sites are indicated by arrow. The bent arrow indicates the location of transcription start site.

Figure 4. Quantitative analysis of DNA demethylation levels in the CpG sites following CIE treatment with or without SAM.

Genomic DNA isolated from control and CIE-treated cortical neuronal cultures was bisulfite-treated and pyrosequencing analyzed. The 15 individual graphs show the detailed changes in percentage of CpG demethylation following CIE treatment and withdrawal with or without SAM in the indicated CpG sites, which are indicated with arrows in Figure 3. The values in negative are defined in this study to represent demethylation. Results are presented as the mean percentage of demethylation verse to the level of control (zero) ± SEM; *, p<0.05; **, p<0.01 compared with respective control levels.

Figure 5. SAM prevents CIE-induced up-regulation of NR2B expression.

The cortical neuronal cultures were treated by CIE regimen with and without the presence of SAM (2 mM). Total RNA was isolated and real time PCR for mRNA expression of NR2B was performed. Results are presented as the mean ratio to control ± SEM; *, p<0.05; **, p<0.01 compared with their corresponding control levels.

Figure 6. CIE reduced the association of MeCP2 with chromatin in specific regions.

The chromatin fragments were prepared by cross-linking using 1% formaldehyde and precipitating with antibody specific to MeCP2. The qChIP PCR was performed to selectively amplify the region a, b, c, d, f and g. Each of the individual graphs represents the CIE-induced changes in region a to g (see Figure1). *, p<0.05; **, p<0.01 compared to the control.

Figure 7. The influence of methylated CpG sites on the binding activities of CREB and AP-1.

The representatives of four independent EMSA assays in each experiment are shown. Left: nuclear extracts from cortical neuronal culture were incubated with end-labeled double-stranded oligonucleotides with nonmethylated (Lane 1) and methylated (Lane 2) CpG 32 within CRE site. Right: nuclear extracts were incubated with probes incorporated with methylated CpG32 in 0%, 25%, 50%, 75% and 100% (Lane 3∼7). 2. The probes of oligonucleotides containing AP-1 binding site from nonmethylated (Lane 1), hemi-methylated in sense (S, Lane 2) and anti-sense (A, Lane 3) and methylated (Lane 4) were used in EMSA. Since we have previously identified CREB and AP-1 (family) proteins binding by antibodies shift assay, the arrows indicate CREB or AP-1 bands.

Figure 8. 5′AZA and CIE cause increased the occupancy of transcription factors with chromatin.

The cultured neuronal cells were treated by 5′AZA (4 days) or CIE. The chromatin fragments were prepared by cross-linking using 1% formaldehyde and precipitating with antibodies, CREB, c-Fos, c-Jun and FosB. The qChIP PCR was performed by using the primers amplifying the regions containing CRE and AP-1 sites, respectively. *, p<0.05 compared to the control.

Figure 9. In vitro methylation inhibits promoter activities.

A. The structure of the luciferase reporter gene constructs of the NR2B gene promoter, containing CREB (0.8 kb) or both CREB and AP-1 (1.3 kb) binding sites. B. Constructs 1.3 and 0.8 kb were in vitro mock-methylated and methylated and transfected into cultured cortical neurons. Cells were collected 48 h after transfection. Solid and open bars represent the luciferase activities from mock-methylated and methylated promoter, respectively. Data (mean ± SEM) shown are from at least three independent experiments. *, p<0.05 compared to the mock-methylated control constructs.

Figure 10. CIE reduced Dnmt1 expression.

The cortical neuronal cultures were treated by CIE regimen. Total RNA was isolated and real time PCR was performed. Results are presented as the mean relative ratio to control ± SEM; *, p<0.05 compared with control level.