Polymer Simulations of Heteromorphic Chromatin Predict the 3D Folding of Complex Genomic Loci

Adam Buckle,Chris A Brackley,Shelagh Boyle,Davide Marenduzzo,Nick Gilbert

doi:10.1016/j.molcel.2018.09.016

Adam Buckle, Chris A Brackley + Show 3 more

Open Access

https://doi.org/10.1016/j.molcel.2018.09.016

Copy DOI

Journal: Molecular Cell	Publication Date: Oct 18, 2018
Citations: 141	License type: cc-by

Affiliation: MRC Human Genetics Unit, University of Edinburgh

Abstract

SummaryChromatin folded into 3D macromolecular structures is often analyzed by chromosome conformation capture (3C) and fluorescence in situ hybridization (FISH) techniques, but these frequently provide contradictory results. Chromatin can be modeled as a simple polymer composed of a connected chain of units. By embedding data for epigenetic marks (H3K27ac), chromatin accessibility (assay for transposase-accessible chromatin using sequencing [ATAC-seq]), and structural anchors (CCCTC-binding factor [CTCF]), we developed a highly predictive heteromorphic polymer (HiP-HoP) model, where the chromatin fiber varied along its length; combined with diffusing protein bridges and loop extrusion, this model predicted the 3D organization of genomic loci at a population and single-cell level. The model was validated at several gene loci, including the complex Pax6 gene, and was able to determine locus conformations across cell types with varying levels of transcriptional activity and explain different mechanisms of enhancer use. Minimal a priori knowledge of epigenetic marks is sufficient to recapitulate complex genomic loci in 3D and enable predictions of chromatin folding paths.

Highlights

Chromatin fiber folding in cells is dictated by a vast number of interactions between nucleosomes, chromatin-binding proteins, and structural components such as CCCTC-binding factor (CTCF)-cohesin loops, as well as the inherent structure of the underlying fiber
Activity States of Pax6 Show Differential Epigenetic Marks and CTCF Binding In the present work, we set out to develop a universal approach for modeling chromatin fiber folding with limited experimental knowledge, based only on extensive freely available data generated from the ENCODE project
We investigated the folding of 5 Mb around the Pax6 locus using three different immortalized cell lines that expressed Pax6 at different levels (Figure S1), referred to as Pax6-OFF, ON, and HIGH cell lines

Summary

Introduction

Chromatin fiber folding in cells is dictated by a vast number of interactions between nucleosomes, chromatin-binding proteins, and structural components such as CCCTC-binding factor (CTCF)-cohesin loops, as well as the inherent structure of the underlying fiber. The ENCODE project (ENCODE Project Consortium, 2012) comprehensively mapped the distribution of epigenetic and structural features in different human and mouse cell lines Many of these marks are surrogates for transcriptional activity that can impact local chromatin fiber structure. We recently combined the TF model with the popular loop extrusion (LE) model for chromosome organization (Pereira et al, 2018), which explains features of chromatin loops mediated by cohesin and CTCF (Fudenberg et al, 2016; Sanborn et al, 2015) While this strategy successfully predicts large-scale features of genome organization, we show below that it cannot accurately predict the folding of the complex Pax genomic locus at high resolution, which we probed experimentally at different levels using fluorescence in situ hybridization (FISH) imaging and Capture-C. Pax is surrounded by constitutively expressed genes and multiple enhancers, providing a paradigm for complex genetic interactions (Buckle et al, 2018; Lacomme et al, 2018; McBride et al, 2011)

Methods

Results

Discussion

Conclusion