Loss of Imprinting at the 14q32 Domain Leads to MicroRNA Overexpression in Acute Promyelocytic Leukemia

Abstract

Acute promyelocytic leukemia (APL) is characterized by the chromosomal translocation t(15;17) which results in the expression of the chimeric protein PML- RARα. Compared to the wild type retinoic acid receptor α (RARα), the fusion protein acquires dominant oncogenic properties and the chromosomal rearrangement is identified as the trigger of APL. However the pathogenesis of APL cannot be explained by the sole failure of RARα regulation and additional genetic and epigenetic alterations are required. We and others have shown that the microRNAs (miRNAs) clustered in the chromosome 14q32 imprinted domain and epigenetically regulated by the upstream differentially methylated regions (DMRs) are overexpressed only in APL (Dixon-McIver et al., 2008; Li et al., 2008; Valleron et al., 2012). Here, using high-throughput amplicon bisulfite sequencing (Roche 454), we characterized the DNA methylation profile of the DMRs in bone marrow/peripheral blood samples from patients with APL, other subclasses of acute myeloid leukemia (AML) and from healthy donors. Sequence reads were quality filtered and a total of 923,981 used to determine the methylation status of 202 CpGs. We identified an APL-specific hypermethylation signature (Fig. 1) at the DMR that spans the promoter of the MEG3 gene (MEG3-DMR) and partially overlaps the miRNA cluster. Hypermethylation encompassed the binding site motifs for the enhancer blocking protein CTCF. Consistent with the CTCF insulating activity, CpG methylation at the CTCF binding sites positively correlated with the expression of miRNAs (Fig. 2). Notably, no significant DNA methylation changes were detected at the intergenic imprinting control region (IG-DMR). Indeed, consistent with a scenario whereby only the genes regulated by the MEG3-DMR would be affected by the aberrant methylation, the gene expression profile performed on a cohort of 97 AML patients showed that among the imprinted genes of the domain only MEG3 was distinctively up-regulated in APL. Taking advantage of the long sequence reads obtained, we performed the haplotype analysis of the DNA methylation changes in diagnostic/remission sample pairs and demonstrated that hypermethylation arises in a mono-allelic manner in APL (Fig. 3). As the expression of the 14q32 miRNAs in the adult is normally restricted to the brain, we propose a model in which loss of imprinting (LOI) at 14q32 leads to aberrant expression of the miRNAs in APL cells. This study provides novel insights into the epigenetic characterization of APL and the mechanism underlying the deregulation of a specific cluster of miRNAs in this subtype of leukemia. The 14q32 miRNAs include species with oncogene and tumor-suppressor activity and their up-regulation may play a role in the APL pathogenesis. Further investigations are required to determine whether LOI is involved in the cancer initiation or it occurs at a later stage, possibly in association with the expression of the chimeric protein PML-RARα.Figure 1Unsupervised hierarchical cluster analysis of the CpG methylation levels. Each row represents a CpG site and each column a sample. The percentage of CpG methylation is depicted using color scales of red (CpG methylation > 50%) and green (CpG methylation < 50%). Sample group labels are indicated (APL; Control; Remission; AMLs).Figure 1. Unsupervised hierarchical cluster analysis of the CpG methylation levels. Each row represents a CpG site and each column a sample. The percentage of CpG methylation is depicted using color scales of red (CpG methylation > 50%) and green (CpG methylation < 50%). Sample group labels are indicated (APL; Control; Remission; AMLs).Figure 2Correlation between DNA methylation levels and miRNAs expression. The expression of 14q32 miRNAs was correlated with the DNA methylation level at the DMRs. The position of CpG islands, CTCF binding sites (A-B-C-D-E-F-G) and amplicons is labeled with green, red and blue horizontal bars, respectively. Amplicons 1-9 reside in the MEG3-DMR. The correlation is represented with red and green vertical bars indicating positive and negative values, respectively.Figure 2. Correlation between DNA methylation levels and miRNAs expression. The expression of 14q32 miRNAs was correlated with the DNA methylation level at the DMRs. The position of CpG islands, CTCF binding sites (A-B-C-D-E-F-G) and amplicons is labeled with green, red and blue horizontal bars, respectively. Amplicons 1-9 reside in the MEG3-DMR. The correlation is represented with red and green vertical bars indicating positive and negative values, respectively.Figure 3Haplotype analysis. The sequenced amplicons were interrogated using the available SNPs (UCSC database). When a heterozygous SNP was observed, sequence reads were separated accordingly to the SNP genotype and unsupervised cluster analysis performed on the allelic CpG methylation pattern. Allelic DNA methylation profiles at the MEG3-DMR: the diagnostic (A) and complete remission (B) stages of a patient with APL; (C) healthy donor. D) Allelic DNA-methylation profile at the IG-DMR, healthy donor. Each column represents a CpG site and each row the methylation pattern of a single sequence read. Blue, methylated; yellow not methylated.Figure 3. Haplotype analysis. The sequenced amplicons were interrogated using the available SNPs (UCSC database). When a heterozygous SNP was observed, sequence reads were separated accordingly to the SNP genotype and unsupervised cluster analysis performed on the allelic CpG methylation pattern. Allelic DNA methylation profiles at the MEG3-DMR: the diagnostic (A) and complete remission (B) stages of a patient with APL; (C) healthy donor. D) Allelic DNA-methylation profile at the IG-DMR, healthy donor. Each column represents a CpG site and each row the methylation pattern of a single sequence read. Blue, methylated; yellow not methylated. Disclosures: No relevant conflicts of interest to declare.