CLL Subsets with Distinct Stereotyped B Cell Receptors Have Distinct Epigenetic Make-up, Even Beyond IGHV Gene Mutational Status: DNA Methylation Profiling of IGHV-Unmutated CLL Stereotyped Subsets #6 and #8

Abstract 3869

In CLL, subsets of patients with stereotyped B cell receptors (BcRs) account for one-third of the cohort. Increasing evidence suggests that cases assigned to the same subset can share similar biological and clinical features independently of IGHV gene mutational status, at least for selected major subsets. Consequently, the study of BcR stereotypy has important implications for refining patient stratification with the ultimate aim of implementing targeted therapy and, eventually, improving outcome. CLL stereotyped subsets #6 (IGHV1–69/IGKV3–20) and #8 (IGHV4–39/IGKV1(D)-39) both express unmutated BCRs yet exhibit different clinical behavior. In particular, subset #6 is less aggressive compared to subset #8, which exhibits the highest risk for development of Richter's syndrome among all CLL. So far, no explanation is available for this difference, thus prompting investigation into the underlying mechanisms. Aberrant DNA methylation is increasingly recognized as relevant for CLL with strong correlations between promoter methylation and transcriptional silencing for critical genes, often depending on IGHV gene mutational status. With this in mind, we profiled the DNA methylation in CLL stereotyped subsets #6 and #8 and searched for epigenetic modifications that could be linked with their distinct clinicobiological features. DNA from CD19⁺ peripheral blood B cells of 10 subset #6 and 10 subset #8 cases was bisulfite-converted and analyzed with the Infinium HumanMethylation450 BeadChip array which allows interrogation of approximately 485,000 CpG sites. The methylation level of each CpG site was calculated in GenomeStudio Methylation module, while differential methylation analysis was performed using the Genomestudio software. Overall, 118 differentially methylated CpG sites (DMCpGs) were identified, of which 70 were hypomethylated while 48 were hypermethylated in subset #8 in comparison to subset #6. Hierarchical clustering of these DMCpGs enabled the discrimination of the two subsets. Seventy of the 118 DMCpGs were related to 61 genes, being preferentially located in regions of the corresponding locus associated with RNA transcripts, whereas for the remaining 48 DMCpGs no links with annotated transcripts could be identified. Of note, 14/48 of the latter DMCpGs were located in a CpG island on chromosome 14q32 and were hypermethylated in subset #8 versus subset #6, likely as a consequence of the t(14;19)(q32;q13) (BCL3/IGH) chromosomal translocation, which is frequently found in subset #8. In order to search for distinctive gene patterns among differentially methylated genes, we performed separate Gene Ontology (GO) analysis using the WebGestalt bioinformatics tool for (i) genes hypermethylated (n=19) and (ii) genes hypomethylated (n=42) in subset #8 versus subset #6. Interestingly, the most pronounced differences in methylation status concerned 10 genes in the latter category (HOXα5, PAX3, HLA-C, TP63, ZNF300, PRDM16, MGRN1, EMX2, PP1R14A, CCDC140). For genes hypermethylated in subset #8, GO analysis did not reveal any significant biological processes or molecular functions. In contrast, among genes hypomethylated in subset #8, we noted a significant overrepresentation of genes involved in developmental and metabolic biological processes, in particular, transcription factor and transcription regulator activity and DNA binding. In conclusion, we found distinct DNA methylation profiles between two CLL stereotyped subsets, both carrying unmutated BcRs yet showing different clinical behavior. The observed differential DNA methylation may be implicated in differences in gene expression programs that shape the biologic make-up and eventual clinical behavior of each subset.