Figure 2
Figure 2. Mutant WT1 induces DNA hypermethylation in AML. (A) Venn diagram showing cooccurrence between the 3 mutations most linked to DNA hypermethylation (WT1, IDH2, and CEBPA) from 191 sequenced patients in the TCGA cohort. (B) Venn diagram showing overlap of hypermethylated CpG sites identified by Boolean implications for mutation in WT1, IDH2, and CEBPA. (C) Venn diagram showing overlap of all identified protein-coding genes associated with the hypermethylated CpG sites using the annotation file provided by Illumina. (D) Schematic showing the domain structure of the WT1 protein and the location of insertion/deletion mutations for the WT1 mutations in the TCGA AML cohort. Shown subsequently is the mutated form of WT1 used in methylation validation studies truncated at the first zinc finger domain (amino acid position 306) and linked to self-cleaving T2A peptide and GFP in a lentiviral vector. P/Q rich, proline/glutamine rich domain; ZF, zinc finger domain. (E) Principal component analysis of all probes with a dynamic β value range >0.2 assayed by Illumina HumanMethylation450 BeadChip array in THP1 AML cells transduced with either mutant WT1 or mutant IDH2 R172W in triplicate and compared with parental cells after 10 passages in culture. (F) Heatmap showing differential methylation between parental and WT1-mutant–expressing THP1 cells. Increased β values indicate increased methylation. (G) Venn diagram showing the overlap of genes methylated upon expression of mutant WT1 in THP1 cells and genes identified to be methylated in the TCGA data set by Boolean implications; *P = 8.5E-34, Fisher’s exact test, right tailed. A gene could only be included in this analysis if at least 1 of its associated CpG sites appeared in the Boolean implication-based methylome analysis of TCGA patients and in the differential methylome analysis of the THP1-derived cell lines. The total number (universe) of genes used in Fisher’s exact test calculation was 13 018. (H) Volcano plot showing the difference in methylation for WT1 mutated vs WT1 wild-type samples in the TCGA data set. The y-axis shows the negative log of Benjamini-Hochberg corrected P value of a t test for differential methylation, and the x-axis shows the magnitude change in methylation (difference in average β value between WT1-mutant and WT1 wild-type samples). Each dot represents the mean β value difference and the P value of a given CpG site in a t test between WT1-mutated vs WT1 wild-type AML.

Mutant WT1 induces DNA hypermethylation in AML. (A) Venn diagram showing cooccurrence between the 3 mutations most linked to DNA hypermethylation (WT1, IDH2, and CEBPA) from 191 sequenced patients in the TCGA cohort. (B) Venn diagram showing overlap of hypermethylated CpG sites identified by Boolean implications for mutation in WT1, IDH2, and CEBPA. (C) Venn diagram showing overlap of all identified protein-coding genes associated with the hypermethylated CpG sites using the annotation file provided by Illumina. (D) Schematic showing the domain structure of the WT1 protein and the location of insertion/deletion mutations for the WT1 mutations in the TCGA AML cohort. Shown subsequently is the mutated form of WT1 used in methylation validation studies truncated at the first zinc finger domain (amino acid position 306) and linked to self-cleaving T2A peptide and GFP in a lentiviral vector. P/Q rich, proline/glutamine rich domain; ZF, zinc finger domain. (E) Principal component analysis of all probes with a dynamic β value range >0.2 assayed by Illumina HumanMethylation450 BeadChip array in THP1 AML cells transduced with either mutant WT1 or mutant IDH2 R172W in triplicate and compared with parental cells after 10 passages in culture. (F) Heatmap showing differential methylation between parental and WT1-mutant–expressing THP1 cells. Increased β values indicate increased methylation. (G) Venn diagram showing the overlap of genes methylated upon expression of mutant WT1 in THP1 cells and genes identified to be methylated in the TCGA data set by Boolean implications; *P = 8.5E-34, Fisher’s exact test, right tailed. A gene could only be included in this analysis if at least 1 of its associated CpG sites appeared in the Boolean implication-based methylome analysis of TCGA patients and in the differential methylome analysis of the THP1-derived cell lines. The total number (universe) of genes used in Fisher’s exact test calculation was 13 018. (H) Volcano plot showing the difference in methylation for WT1 mutated vs WT1 wild-type samples in the TCGA data set. The y-axis shows the negative log of Benjamini-Hochberg corrected P value of a t test for differential methylation, and the x-axis shows the magnitude change in methylation (difference in average β value between WT1-mutant and WT1 wild-type samples). Each dot represents the mean β value difference and the P value of a given CpG site in a t test between WT1-mutated vs WT1 wild-type AML.

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal