Back

Loss of heterochromatin at endogenous retroviruses creates competition for transcription factor binding

O'Hara, R.; Banaszynski, L. A.

2022-04-29 genomics
10.1101/2022.04.28.489907 bioRxiv
Show abstract

The mammalian genome is partitioned into active and inactive regions, broadly termed euchromatin and heterochromatin, respectively. The majority of heterochromatin consists of repetitive elements, including endogenous retroviruses (ERVs). ERVs are enriched in regulatory elements containing transcription factor (TF) binding sites with individual families containing hundreds to thousands of distinct copies scattered throughout the genome. We hypothesized that epigenetic derepression of ERVs (such as that observed during early development) may alter the stoichiometry between TFs and their euchromatic target sites, with ERVs effectively competing for these factors. To test this, we modeled acute heterochromatin loss using inducible deletion of the co-repressor KAP1 in mouse embryonic stem cells (ESCs). Upon KAP1 deletion, we observe clear reductions in chromatin accessibility, histone acetylation, and TF binding at euchromatic regions. To directly test the concept of global binding site competition, we designed exogenous binding site arrays (EBSAs) to introduce upwards of 1500 copies of the OCT4 TF binding motif into ESCs. OCT4 EBSAs specifically reduce chromatin accessibility at POU family motifs and result in reduced transcription of the pluripotency machinery with subsequent differentiation. Overall, these data support a model in which heterochromatin at ERVs promotes euchromatic TF binding and transcriptional homoeostasis. We propose that regulated ERV derepression during pre-implantation may serve as a developmental siphon to weaken the robustness of ongoing transcription programs in favor of the plasticity required for cell fate specification.

Matching journals

The top 5 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.