Abstract
During the last years, the study of molecular alterations associated with multiple myeloma (MM) has been mostly focused on the analysis of the genome, transcriptome and DNA methylome. These analyses are showing that (epi)genetic heterogeneity and extensive perturbation of the transcriptional landscape are hallmarks of MM. Our previous analysis of the whole DNA methylome of MM revealed that this epigenetic mark globally shows a poor correlation with gene expression, and therefore did not allow us to better understand gene deregulation in MM. In contrast, the chromatin structure and histone modifications are emerging as essential epigenetic layers to understand the mechanisms underlying gene expression changes in cancer, but remain widely unexplored in MM.
We have now performed a deep ChIP-seq profiling of CD138+ sorted cells from bone marrow samples obtained from four MM patients and three biological replicates of normal plasma cells (NPCs) using antibodies against H3K4me3, H3K4me1, H3K27ac, H3K36me3, H3K27me3 and H3K9me3. Different combinations of these marks allow us to segment the MM and NPC genome into functional chromatin states, including active, weak or poised promoters, active or weak enhancers, transcriptional elongation, polycomb-repressed regions and heterochromatic regions.
The initial unsupervised exploration of the data showed that the chromatin landscape of MM is widely altered as compared to NPCs. A supervised analysis of chromatin states revealed that MM globally shows a more active chromatin structure than NPCs. From over 40,000 regions identified with differential chromatin structure between MM and NPCs, 88% were de novo activated in neoplastic plasma cells. Analyzing the chromatin of individual genes, we observed that there were roughly ten times more genes gaining activity upon neoplastic transformation than those acquiring repressed chromatin marks. Interestingly, the genes showing more activate chromatin were enriched with biosynthesis and metabolic processes, while genes with repressed chromatin were related to gene ontology terms related to B cell signaling. Among those genes gaining de novo activity in MM, we selected several candidates and we are currently performing functional in vitro assays to explore their implication in MM pathogenesis. Furthermore, as extensive chromatin activation is a hallmark of MM, we are currently analyzing additional 15 MM cases and NPCs by ChIP-seq for H3K27ac (marking active promoters and enhancers) and ATAC-seq (marking active regulatory regions) to validate our initial findings and explore chromatin heterogeneity in MM.
Collectively, our initial exploration of histone modification profiles in MM has revealed that MM cells acquire a more active chromatin landscape, with thousands of regions gaining activation as compared to NPCs. Reversing this global activation by epigenetic drugs, such as BET inhibitors, may represent an attractive therapeutic option for MM.
During the meeting, updated information will be presented, including data from all 19 MM patients studied as well as functional data from new candidate genes involved in MM pathogenesis.
Disclosures
Paiva: Celgene: Honoraria, Research Funding; Janssen: Honoraria; Takeda: Honoraria, Research Funding; Sanofi: Consultancy, Research Funding; EngMab: Research Funding; Amgen: Honoraria; Binding Site: Research Funding.
The salt dependence of chicken and yeast chromatin structure. Effects on internucleosomal organization and relation to active chromatin.
