Reconfiguration of nucleosome-depleted regions at distal regulatory elements accompanies DNA methylation of enhancers and insulators in cancer

Abstract

It is well established that cancer-associated epigenetic repression occurs concomitant with CpG island hypermethylation and loss of nucleosomes at promoters, but the role of nucleosome occupancy and epigenetic reprogramming at distal regulatory elements in cancer is still poorly understood. Here, we evaluate the scope of global epigenetic alterations at enhancers and insulator elements in prostate and breast cancer cells using simultaneous genome-wide mapping of DNA methylation and nucleosome occupancy (NOMe-seq). We find that the genomic location of nucleosome-depleted regions (NDRs) is mostly cell type specific and preferentially found at enhancers in normal cells. In cancer cells, however, we observe a global reconfiguration of NDRs at distal regulatory elements coupled with a substantial reorganization of the cancer methylome. Aberrant acquisition of nucleosomes at enhancer-associated NDRs is associated with hypermethylation and epigenetic silencing marks, and conversely, loss of nucleosomes with demethylation and epigenetic activation. Remarkably, we show that nucleosomes remain strongly organized and phased at many facultative distal regulatory elements, even in the absence of a NDR as an anchor. Finally, we find that key transcription factor (TF) binding sites also show extensive peripheral nucleosome phasing, suggesting the potential for TFs to organize NDRs genome-wide and contribute to deregulation of cancer epigenomes. Together, our findings suggest that ''decommissioning'' of NDRs and TFs at distal regulatory elements in cancer cells is accompanied by DNA hypermethylation susceptibility of enhancers and insulator elements, which in turn may contribute to an altered genome-wide architecture and epigenetic deregulation in malignancy.

The role of DNA methylation in directing the functional organization of the cancer epigenome

“…The highly consistent phasing patterns around the C4 CTCF sites (Supplemental Fig. 1D) substantiated our earlier findings that DNA methylation rates are significantly higher within internucleosome linker regions than within nucleosome cores Berman et al 2013;Taberlay et al 2014). DNA methylation loss in DKO1 cells did not affect the strong nucleosome positioning around CTCF, further demonstrating that the effects of DNA methylation on nucleosome organization are limited and context specific.…”

“…This early observation, along with the fact that >60% of promoters are located within CGIs, has driven the focus on CGIs as a model of study for epigenetic regulation (Tazi and Bird 1990;Irizarry et al 2009;Deaton and Bird 2011;Portela et al 2013). Unbiased whole-genome methylation platforms, however, have revealed that methylation changes at other regulatory regions such as CGI shores, non-CGI promoters, and enhancers may also play a role in tumorigenesis (Doi et al 2009;Irizarry et al 2009;Rach et al 2011;Berman et al 2012;Aran and Hellman 2013;Hovestadt et al 2014;Taberlay et al 2014). Epigenomic mapping projects such as the NIH Roadmap Epigenomics Program and the ENCODE Project Consortium (Bernstein et al 2010; The ENCODE Project Consortium 2012) have shown that distinct epigenetic marks are highly correlated and form consistent epigenetic states (Rivera and Ren 2013).…”

Incorporating Functional Annotations for Fine-Mapping Causal Variants in a Bayesian Framework Using Summary Statistics

“…These include predicted chromatin state using ChromHMM (Ernst and Kellis 2012) in breast and prostate normal and cancer cells (Taberlay et al 2014), DHSs from the normal and cancer prostate cells from the ENCODE project (The ENCODE Project Consortium 2012), histone modifications such as histone acetylation, Figure 3 Proportion of causal variants identified by different methods. The top panels correspond to 20 initial loci ($12 causal loci) and the bottom 5panels correspond to 100 initial loci ($65 causal loci).…”

“…These include predicted chromatin state using ChromHMM (Ernst and Kellis 2012) in breast and prostate normal and cancer cells (Taberlay et al 2014), DHSs from the normal and cancer prostate cells from the ENCODE project (The ENCODE Project Consortium 2012), histone modifications such as histone acetylation, …”

ZFX acts as a transcriptional activator in multiple types of human tumors by binding downstream from transcription start sites at the majority of CpG island promoters

“…C4-2B NOMe-seq data were generated as previously described (Rhie et al 2018) and sequenced using an Illumina HiSeq 2000 PE 100 bp to produce FASTQ files. FASTQ files of MCF-7 NOMe-seq data were obtained from GSE57498 (Taberlay et al 2014). Each FASTQ file was aligned to a bisulfite-converted genome (hg19) using BSMAP (Xi and Li 2009) and processed as previously described (Rhie et al 2018).…”