Role of DNA methylation in expression control of the IKZF3-GSDMA region in human epithelial cells
Fig 1
5-aza-dC treatment enhances gene expression.
(A) 5-aza-dC treated NuLi-1 cells show reduced proliferation and apoptosis four days after treatment. Arrowheads point to dying/dead cells. (B) Changes in expression levels of 17q12-q21 genes after 5-aza-dC treatment. The y-axis shows fold change in 5-aza-dC treated cells compared to controls. Error bars show standard deviation. Asterisks indicate statistically significant change in expression in 5-aza-dC treated cells compared to controls (* p < 0.05). (C) Allelic expression in 17q12-q21 genes after 5-aza-dC treatment. Arrows show positions of transcribed SNPs in those genes where allelic expression changed post 5-aza-dC treatment. In ZPBP2, 5-aza-dC treatment causes reactivation of the HapA allele. In ORMDL3 it causes a switch in allelic preference. (D) Positions of 51 CGs in the ZPBP2 promoter region that were assayed using the sodium bisulfite sequencing assay are shown in the context of the UCSC browser. The red box indicates the position of the 11 CGs assayed using the pyrosequencing methylation assay. (E) DNA methylation profiles of the ZPBP2 promoter region in control (DMSO) and 5-aza-dC treated cells. Filled circles represent methylated CGs, open circles represent unmethylated CGs. Each row represents a clone. Data are divided by haplotype; allelic percent methylation is shown below the diagram. Type of treatment and average methylation levels are shown on top. Arrow points to CG31 (CG6 in pyrosequencing assays) that has one of the most pronounced allelic differences in methylation.