Back

Bivalent and Broad Chromatin Domains Regulate Pro-metastatic Drivers in Melanoma

Terranova, C.; Tang, M.; Maitituoheti, M.; Raman, A.; Schulz, J.; Amin, S.; Orouji, E.; Tomczak, K.; Sarkar, S.; Oba, J.; Creasy, C.; Wu, C.-J.; Zhao, D.; Chen, K.; Haydu, L. E.; Wang, W.-L.; Lazar, A. J.; Woodman, S. E.; Bernatchez, C.; Rai, K.

2019-08-01 cancer biology
10.1101/721480 bioRxiv
Show abstract

Chromatin deregulation is an emerging hallmark of cancer. However, the extent of epigenetic aberrations during tumorigenesis and their relationship with genetic aberrations are poorly understood. Using ChIP-sequencing for enhancers (H3K27ac and H3K4me1), promoters (H3K4me3), active transcription (H3K79me2) and polycomb (H3K27me3) or heterochromatin (H3K9me3) repression we generated chromatin state profiles in metastatic melanoma using 46 tumor samples and cell lines. We identified a strong association of NRAS, but not BRAF mutations, with bivalent states harboring H3K4me3 and H3K27me3 marks. Importantly, the loss and gain of bivalent states occurred on important pro-metastasis regulators including master transcription factor drivers of mesenchymal phenotype including ZEB1, TWIST1, SNAI1 and CDH1. Unexpectedly, a subset of these and additional pro-metastatic drivers (e.g. POU3F2, SOX9 and PDGFRA) as well as melanocyte-specific master regulators (e.g. MITF, ZEB2, and TFAP2A) were regulated by exceptionally wide H3K4me3 domains that can span tens of thousands of kilobases suggesting roles of this new epigenetic element in melanoma metastasis. Overall, we find that BRAF, NRAS and WT melanomas may use bivalent states and broad H3K4me3 domains in a specific manner to regulate pro-metastatic drivers. We propose that specific epigenetic traits - such as bivalent and broad domains - get assimilated in the epigenome of pro-metastatic clones to drive evolution of cancer cells to metastasis.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.