Oral Presentation 39th Annual Lorne Genome Conference 2018

DNA replication timing shapes the cancer epigenome (#17)

Qian Du 1 , Saul A Bert 1 , Nicola J Armstrong 2 , Elizabeth Caldon 3 , Clare Stirzaker 1 4 , Susan Clark 1 4
  1. Genomics and Epigenetics, Garvan Institute of Medical Research, Sydney, NSW, Australia
  2. Mathematics and Statistics, School of Engineering and Information Technology, Murdoch University, Murdoch, Western Australia, Australia
  3. Cancer Division, Garvan Institute of Medical Research, Sydney, NSW, Australia
  4. St Vincent's Clinical School, University of NSW, Sydney, New South Wales, Australia

Cancer cells are highly proliferative and undergo successive rounds of DNA replication. Replication of the genome follows a highly organised ‘replication-timing’ program where genomic domains are replicated in a specific temporal order during S-phase, from early to late. Replication timing is known to modulate the faithfulness of DNA replication. As the transmission of epigenetic information during S- phase is intimately associated with the replication fork, replication timing may similarly modulate the faithful transmission of the epigenome. Here, we investigate how the replication timing program organizes the epigenome and contributes to aberrant epigenetic change in normal prostate and cancer cells.

We performed Repli-Seq that utilizes next generation sequencing to map the order of nascent DNA replication across S-phase in normal and prostate cancer cells. We find that, while replication timing is mostly maintained from normal to cancer, domains of replication timing change are accompanied by aberrant changes in gene expression, chromatin marks and DNA methylation. Strikingly, domains of replication timing change correspond to our previously identified domains of Long Range Epigenetic Silencing and Activation found in prostate cancer (LRES, LREA). Importantly, late replication appears to predispose cancer cells to aberrant changes in heterochromatin and DNA methylation, and becomes enriched for chromosomal rearrangements in cancer. Furthermore, we find that different replication times biases towards either cis or trans chromosomal rearrangements. We propose that this mutational bias is related to the differences in cancer epigenetic remodelling that occur in early, compared to late, replication timing. Our findings highlight for the first time that the replication timing program associates with aberrant epigenetic remodelling in cancer, and together, potentially contributes to the mutational pattern in cancer.