Chromatin accessibility maps of chronic lymphocytic leukemia identify subtype- specific epigenome signatures and associated transcription regulatory networks

Chronic lymphocytic leukemia (CLL) is characterized by substantial clinical heterogeneity, despite relatively few genetic alterations. To provide a basis for studying epigenome deregulation in CLL, we established genome-wide chromatin accessibility maps for 88 CLL samples from 55 patients using the ATAC-seq assay. These data were further complemented by ChIPmentation and RNA-seq profiling in ten samples. Based on this dataset, we devised and applied a bioinformatic method that links chromatin profiles to clinical annotations. Our analysis identified sample-specific variation on top of a shared core of CLL regulatory regions. IGHV mutation status – which distinguishes the two major subtypes of CLL – was accurately predicted by the chromatin profiles, and gene regulatory networks inferred for IGHV-mutated vs. IGHV-unmutated samples identified characteristic regulatory differences between these two disease subtypes. In summary, we found widespread heterogeneity in the CLL chromatin landscape, established a community resource for studying epigenome deregulation in leukemia, and demonstrated the feasibility of chromatin accessibility mapping in cancer cohorts and clinical research. Introduction Chronic lymphocytic leukemia (CLL) is the most common type of leukemia in the Western world. It is characterized by remarkable clinical heterogeneity, with some patients pursuing an indolent course while others progress rapidly and require early treatment. The diverse clinical course of CLL patients, particularly those that initially present with low disease burden, fuels interest in prognostic biomarkers and personalized therapies. Current clinical biomarkers for CLL include mutational status of the IGHV genes , IGHV gene family usage, stereotyped B cell receptors , serum markers , chromosomal aberrations , and somatic mutations 13, . Most notably, IGHV mutation status distinguishes between a less aggressive form of CLL with mutated IGHV genes (mCLL) and a more aggressive form with unmutated IGHV genes (uCLL). Several surrogate biomarkers of IGHV mutation status have been described. For example, high levels of ZAP70 expression appear to be associated with uCLL. In addition to these focused biomarkers, transcriptome profiling has been used to define broader molecular signatures that may improve disease stratification independent of IGHV mutation status.