Cohesin disrupts polycomb-dependent chromosome interactions

Abstract

16How chromosome organisation is related to genome function remains poorly 17 understood. Cohesin, loop-extrusion, and CTCF have been proposed to create 18 structures called topologically associating domains (TADs) to regulate gene 19 expression. Here, we examine chromosome conformation in embryonic stem cells 20 lacking cohesin and find as in other cell types that cohesin is required to create TADs 21 and regulate A/B compartmentalisation. However, in the absence of cohesin we 22 identify a series of long-range chromosomal interactions that persist. These 23 correspond to regions of the genome occupied by the polycomb repressive system, 24 depend on PRC1, and we discover that cohesin counteracts these interactions. This 25 disruptive activity is independent of CTCF and TADs, and regulates gene repression 26 by the polycomb system. Therefore, in contrast to the proposal that cohesin creates 27 structure in chromosomes, we discover a new role for cohesin in disrupting polycomb-28 dependent chromosome interactions to regulate gene expression. 29

A Central Role for Canonical PRC1 in Shaping the 3D Nuclear Landscape

“…4F-I). By comparing chromatin contacts in the presence and absence of RING1B or CTCF, we conclude that PRC1-mediated interactions are independent of CTCF, a finding consistent with other unpublished observations (Rhodes et al 2019). PRC1…”

“…Associations between RING1B bound CGIs were unaffected by the loss of CTCF ('auxin'; Supplemental Fig. 4E), confirming that the formation of PRC1-mediated interactions is mechanistically distinct from that required for loop extrusion consistent with other un-published observations (Rhodes et al 2019).…”

Coolpup.py:versatile pile-up analysis of Hi-C data

“…The latter analysis not only replicated published data on CTCF-mediated looping changes across the cell cycles, but also revealed novel cell cycle dynamics of polycomb-associated interactions with highest contact enrichment around the time of mitosis. We note that these observations are generally consistent with the dilution and slow recovery of the H3K27me3 mark after the replication fork (Alabert et al, 2015;Reverón-Gómez et al, 2018), as well as an antagonistic relationship between cohesinmediated loop extrusion and looping between RING1B target sites, reported previously (Rhodes et al, 2019). These observations also pose a question of whether polycomb-associated interactions persist in metaphase chromosomes -a possibility since components of CBX2-containing PRC1 remain associated with metaphase chromosomes (Zhen et al, 2014).…”

PRC1 catalytic activity is central to Polycomb system function

“…The control TIR1-only and AID-RING1B lines were generated from E14 ESCs as described previously (Rhodes et al, 2019). Briefly, Cas9 engineering was used to insert the coding sequence for Oryza sativa TIR1 into the Rosa26 locus, thereby generating the TIR1-only control line.…”

“…The high-throughput data reported in this study have been deposited in GEO under the accession number GSEXXXXXX. Published data used in this study include BioCAP-seq (GSE43512 (Long et al, 2013)); MAX ChIP-seq (GSE48175; (Krepelova et al, 2014); H2AK119ub1, H3K27me3 and RING1B cChIP-seq together with corresponding Inputs in PRC1 CKO ESCs (GSE119618; ); cRNA-seq in Pcgf4 -/-4;Pcgf2 fl/fl (GSE119619; ); CaptureC in RING1B deg and control cell lines (ArrayExpress E-XXXX-XXXX (Rhodes et al, 2019)). All R and Perl scripts used for data analysis in this study are available upon request.…”