Summary The Polycomb system modifies chromatin and plays an essential role in repressing gene expression to control normal mammalian development. However, the components and mechanisms that define how Polycomb protein complexes achieve this remain enigmatic. Here, we use combinatorial genetic perturbation coupled with quantitative genomics to discover the central determinants of Polycomb-mediated gene repression in mouse embryonic stem cells. We demonstrate that canonical Polycomb repressive complex 1 (PRC1), which mediates higher-order chromatin structures, contributes little to gene repression. Instead, we uncover an unexpectedly high degree of synergy between variant PRC1 complexes, which is fundamental to gene repression. We further demonstrate that variant PRC1 complexes are responsible for distinct pools of H2A monoubiquitylation that are associated with repression of Polycomb target genes and silencing during X chromosome inactivation. Together, these discoveries reveal a new variant PRC1-dependent logic for Polycomb-mediated gene repression.

Highlights d PRC1 catalysis drives PRC2 occupancy and H3K27me3 deposition at target sites d cPRC1 binding and PRC1-mediated chromatin interactions require PRC1 catalysis d DNA-binding vPRC1 complexes occupy target sites independently of PRC1 catalysis d PRC1 catalytic activity is essential for Polycomb-mediated gene repression

Highlights d Interaction between polycomb target genes in ESCs occurs independently of cohesin d Loop extrusion by cohesin disrupts interactions between polycomb target genes d Cohesin removal enhances repression at polycomb target genes with increased interactions

Polycomb group (PcG) proteins function as chromatin-based transcriptional repressors that are essential for normal gene regulation during development. However, how these systems function to achieve transcriptional regulation remains very poorly understood. Here, we discover that the histone H2AK119 E3 ubiquitin ligase activity of Polycomb repressive complex 1 (PRC1) is defined by the composition of its catalytic subunits and is highly regulated by RYBP/YAF2-dependent stimulation. In mouse embryonic stem cells, RYBP plays a central role in shaping H2AK119 mono-ubiquitylation at PcG targets and underpins an activity-based communication between PRC1 and Polycomb repressive complex 2 (PRC2) which is required for normal histone H3 lysine 27 trimethylation (H3K27me3). Without normal histone modification-dependent communication between PRC1 and PRC2, repressive Polycomb chromatin domains can erode, rendering target genes susceptible to inappropriate gene expression signals. This suggests that activity-based communication and histone modification-dependent thresholds create a localized form of epigenetic memory required for normal PcG chromatin domain function in gene regulation.DOI: http://dx.doi.org/10.7554/eLife.18591.001

The Polycomb repressive system is an essential chromatin-based regulator of gene expression. Despite being extensively studied, how its target genes are selected and whether its histone modifying activities are required for transcriptional repression remains controversial. Here, we directly test the requirement for PRC1 catalytic activity in Polycomb system function. We demonstrate that a mutation widely used to disrupt PRC1 catalysis is hypomorphic, complicating the interpretation of previous studies. To overcome this, we develop a new inducible mutation system in embryonic stem cells that completely ablates PRC1 catalytic activity, revealing that catalysis by PRC1 drives Polycomb chromatin domain formation and higher-order chromatin interactions. In the absence of catalysis, we uncover the primary DNA-based targeting determinants that direct Polycomb target site selection. Finally, we discover that Polycomb-mediated gene repression requires PRC1 catalytic activity. Together these discoveries provide compelling new evidence supporting a PRC1-initiated pathway for Polycomb system function in gene regulation.