Decitabine alters the expression of Mecp2 isoforms via dynamic DNA methylation at the Mecp2 regulatory elements in neural stem cells.

Schematics of the Methyl CpG binding protein 2 gene (Mecp2), Mecp2e1/e2 transcripts, and known regulatory elements (REs). (A) Generation of MeCP2 isoforms by alternative splicing; mature Mecp2e1 transcripts comprise of exons 1, 3, and 4. Mature Mecp2e2 transcripts comprise of exons 2, 3, and 4 (adapted from [4,11]). Exons are denoted as Ex. (B) Regulatory elements of the MECP2/Mecp2 gene. The MECP2/Mecp2 gene is reported to be regulated by negative and positive REs within the promoter and a silencer element within the intron 1 (information extracted from [12,13]). For our studies we selected a 500-bp region in the promoter upstream of the exon 1 and a 1-kb region in the intron 1 upstream of the exon 2. Each sequence was divided into three regions, R1 to R3 in the Mecp2 promoter and R4 to R6 in the intron 1. Note that there are no CpG dinucleotides in the mouse genomic sequence between R5 and R6.

Characterizing in vitro neural stem cells to study Methyl CpG binding protein 2 (MeCP2) expression. (A) Schematic representation of in vitro neural stem cell (NSC) differentiation. (B) Detection of (a) NESTIN+ and (b) KI67+ cells in self-renewing neurospheres. Scale bars represent 20 μm. (C) Immunofluorescent detection of different cell-type markers in the day 8 (D8) population (a) TUBULIN III (TUB III): neurons, (b) Glial fibrillary acidic protein (GFAP): astrocytes, (c) S100B: mature astrocytes, (d) 2',3'-Cyclic-nucleotide 3'-phosphodiesterase (CNPase): oligodendrocytes, (e) Myelin basic protein (MBP): oligodendrocytes, (f) Oligodendrocyte lineage transcription factor 2 (OLIG2): early oligodendrocytes and progenitors, and (g) KI67: proliferating cells. Scale bars represent 10 μm. The percentages represent average number of cells from three individual experiments (n = 3 ± standard error of the mean). (D) (a) Immunofluorescent detection of MeCP2 in a sectioned primary neurosphere. Scale bar represents 20 μm. (b) Double labeling of MeCP2 and NESTIN within primary neurosphere cells. Scale bar represents 5 μm. (E) Immunofluorescent detection of MeCP2 in D8 cell types: (a) TUB III, (b) GFAP, (c) S100B, (d) CNPase, (e) MBP, (f) OLIG2, and (g) KI67. Scale bars represent 2 μm.

Methyl CpG binding protein 2 gene (Mecp2) isoform-specific transcript expression and DNA methylation at the Mecp2 regulatory elements during neural stem cell (NSC) differentiation. (A) Analysis of Mecp2e1 and Mecp2e2 transcript levels during NSC differentiation: n = 3 ± standard error of the mean (SEM). Significant differences: ***P <0.001; **P <0.01; *P <0.05. (B) Average percentage methylation over Mecp2 promoter and intron 1 regions at day 0 (D0), day 2 (D2) and day 8 (D8) during NSC differentiation. The regions are promoter regions R1, CpG island contains 13 CpG sites; R2, 4 CpG sites; R3, 2 CpG sites, and intron 1 regions R4, 1 CpG site; R5, 1 CpG site; and R6, 2 CpG sites; n = 3 ± SEM. Significant differences: **P <0.01; *P <0.05. Gapdh, glyceraldehyde-3-phosphate dehydrogenase gene.

Effect of decitabine on global DNA methylation (5mC and 5hmC) and Dnmt genes in differentiating neural stem cells (NSC). (A) Schematic representation of decitabine treatment. Briefly, 2.5 μM of decitabine was added to dissociated neurospheres on day 0 (D0) at the onset of NSC differentiation for 48 h, and the treatment was withdrawn at D2. Cells were kept in culture till D8. Top panel (B-E), after exposure to decitabine at D2. (B) Immunofluorescent detection of DNA methylation using 5-methylcytosine (5mC) antibody. Decitabine caused reduced levels of DNA methylation, note the presence of 4',6-diamidino-2-phenylindole (DAPI) signals in decitabine-treated cells with no 5mC signal. Scale bars represent 5 μm. (C) Detection of overall DNA methylation levels by DNA dot blot with antibodies specific for (a) 5mC, (b) 5-hydroxymethylcytosine (5hmC). (D) Quantification of the 5mC and 5hmC levels after decitabine exposure. (E) Detection of Dnmt transcript levels by qRT-PCR. Bottom panel (F-I), after withdrawal of decitabine at D8. (F) DNA methylation detection by immunofluorescence using 5mC antibody. Scale bars represent 5 μm. (G) Detection of global DNA methylation levels by DNA dot blot, (a) 5mC, (b) 5hmC. (H) Quantification of 5mC and 5hmC levels after withdrawal of decitabine. (I) Detection of Dnmt transcript levels by qRT-PCR. Fold changes are calculated relative to transcript levels at D2 or D8 control; n = 3 ± standard error of the mean. Significant differences from control: **P <0.01; *P <0.05. MB, methylene blue (used for visualizing total DNA).

Effect of decitabine exposure and withdrawal on Methyl CpG binding protein gene (Mecp2/MeCP2) expression. (A-C), After exposure to decitabine at D2. (A) Analysis of Mecp2 (total), Mecp2e1 and Mecp2e2 transcript levels by qRT-PCR. (B) Detection of MeCP2 (total) protein expression levels by western blot; n = 2 ± standard error of the mean (SEM). (C) Detection of MeCP2E1 protein expression levels by western blot; n = 2 ± SEM. (D-F), After withdrawal of decitabine at day 8 (D8). (D) Analysis of total Mecp2, Mecp2e1 and Mecp2e2 transcript levels by qRT-PCR. (E) Detection of MeCP2 (total) protein expression levels by western blot. (F) Detection of MeCP2E1 protein expression levels by western blot. (G-H) Pearson's correlation analysis of the relation between Mecp2 transcript levels and MeCP2 protein levels at D2 (G) and D8 (H); r = Pearson’s correlation coefficient, r2 = coefficient of determination. (I) Transcript detection of cell type-specific markers for neurons (Tub III, NeuN); astrocytes (Gfap, S100b); oligodendrocytes (Cnpase, Mbp) by qRT-PCR in D2 control and decitabine-treated cells. (J) Quantification of neurons, astrocytes and oligodendrocytes using cell type-specific markers by immunofluorescence in D2 control and decitabine-treated cells. (K) Transcript detection of cell type-specific markers for neurons (Tub III, NeuN); astrocytes (Gfap, S100b); oligodendrocytes (Cnpase, Mbp) by qRT-PCR in D8 control and decitabine-treated cells. (L) Quantification of neurons, astrocytes and oligodendrocytes using cell type-specific markers in D8 control and decitabine-treated cells. (M) Comparison of immunofluorescent detection of (a) Glial fibrillary acidic protein (GFAP) and (b) S100B between control and decitabine-treated cells. Images were taken at the same exposure time. Scale bars represent 20 μm. For all the panels, fold changes are calculated relative to expression levels at D2 or D8 controls. Significant differences from controls: ****P <0.0001; ***P <0.001; **P <0.01; *P <0.05, n = 3 ± SEM, unless specifically mentioned.

Bisulfite pyrosequencing analysis of DNA methylation at the Methyl CpG binding protein 2 gene (Mecp2) regulatory elements after decitabine treatment. (A) Effect of decitabine exposure at D2 on the percentage DNA methylation of Mecp2 regulatory regions. The three promoter regions are (a) R1, (b) R2, (c) R3, and the three intron 1 regions are (d) R4, (e) R5, and (f) R6. (B) Effect of decitabine exposure on average methylation over the entire region 2 (R2) (a), and intron 1 (R4 to R6) (b). Significant differences from controls: **P <0.01; *P <0.05; n = 3 ± standard error of the mean. (C) Effect of decitabine withdrawal at D8 on percentage methylation of Mecp2 regulatory regions. The regions are promoter regions (a) R1, (b) R2 and (c) R3, and intron 1 regions (d) R4, (e) R5 and (f) R6. (D) Effect of decitabine withdrawal on average DNA methylation over entire region 2 (R2) (a), and intron 1 (R4 to R6) (b).

Correlation analysis between DNA methylation at the Methyl CpG binding protein 2 gene (Mecp2) regulatory elements and Mecp2 expression after decitabine treatment (day 2) and decitabine withdrawal (day 8). All graphs represent the Pearson's correlation coefficient (r) for Mecp2e1 (black), and Mecp2e2 (pink): statistical significance: ***P <0.001; **P <0.01; *P <0.05; n = 3. (A) Correlation coefficients for the relation between Mecp2 expression and average methylation over entire regions in Mecp2 promoter (region (R)1 to R3) and intron 1 (R4 to R6) after decitabine exposure on day 2 (D2) (a), and after decitabine withdrawal on D8 (b). (B) After decitabine exposure: correlation coefficients for Mecp2e1 (black), and Mecp2e2 (pink) with individual CpG methylation at the promoter regions (a) R1, (b) R2 and (c) R3, and intron 1 regions (d) R4, (e) R5 and (f) R6. (C) After decitabine withdrawal: correlation coefficients for Mecp2e1 (black), and Mecp2e2 (pink) with individual CpG methylation at promoter regions (a) R1, (b) R2 and (c) R3, and intron 1 regions (d) R4, (e) R5 and (f) R6. Statistical significance: *P <0.05; n = 3.

Summary of the correlations between the expression of Methyl CpG binding protein 2 gene (Mecp2) isoforms and DNA methylation at the Mecp2 regulatory elements. (A) Dynamic changes in the expression of Mecp2 isoforms (Mecp2e1 and Mecp2e2) at different time points of neural stem cell (NSC) differentiation at day 0 (D0), D2 and D8. Decitabine caused upregulation of Mecp2e1/MeCP2E1 but not Mecp2e2 at D2. Decitabine effect on MeCP2E2 at the protein levels is unknown. Decitabine withdrawal by D8 downregulated Mecp2e1/MeCP2E1, and Mecp2e2/MeCP2E2 (unknown) to different extents. (B) Schematic representation of the correlation between Mecp2 isoform-specific expression and DNA methylation at the Mecp2 promoter regions (R1 to R3), and intron 1 regions (R4 to R6). The size of the signs, plus (+), and minus (-) represents the relative degree of correlation with either Mecp2 isoform. The statistically significant correlations are represented in red (red-circled plus, red-circled minus). After decitabine exposure Mecp2e1 expression negatively correlated with promoter R1 and R3, and intron 1 R5. Mecp2e2 isoform negatively correlates with promoter R3. In contrast, after decitabine withdrawal, Mecp2e1 expression negatively correlated with promoter R1, R2 and R3, and positively correlated with intron 1 R6. Hence, correlation between Mecp2e1 and DNA methylation at REs changed depending on the stage of NSC differentiation. Mecp2e2 isoform positively correlated with intron 1 R4 and R6.