A complex CTCF binding code defines TAD boundary structure and function

Abstract

Topologically Associating Domains (TADs) compartmentalize vertebrate genomes into sub-Megabase functional neighbourhoods for gene regulation, DNA replication, recombination and repair. TADs are formed by Cohesin-mediated loop extrusion, which compacts the DNA within the domain, followed by blocking of loop extrusion by the CTCF insulator protein at their boundaries. CTCF blocks loop extrusion in an orientation dependent manner, with both experimental and in-silico studies assuming that a single site of static CTCF binding is sufficient to create a stable TAD boundary. Here, we report that most TAD boundaries in mouse cells are modular entities where CTCF binding clusters within extended genomic intervals. Optimized ChIP-seq analysis reveals that this clustering of CTCF binding does not only occur among peaks but also frequently within those peaks. Using a newly developed multi-contact Nano-C assay, we confirm that individual CTCF binding sites additively contribute to TAD separation. This clustering of CTCF binding may counter against the dynamic DNA-binding kinetics of CTCF, which urges a re-evaluation of current models for the blocking of loop extrusion. Our work thus reveals an unanticipatedly complex code of CTCF binding at TAD boundaries that expands the regulatory potential for TAD structure and function and can help to explain how distant non-coding structural variation influences gene regulation, DNA replication, recombination and repair.

Reconstructing the 3D genome organization of Neanderthals reveals that chromatin folding shaped phenotypic and sequence divergence

“…Thus, we are observing complex sequence patterns underlying 3D genome folding that could not be determined by simply considering sequence divergence or intersecting AH variants with all CTCF sites. This is concordant with previous results which suggest that 3D genome folding is governed by a complex CTCF binding grammar [50, 80, 82, 83].…”

“…Thus, we are observing complex sequence patterns underlying 3D genome folding that could not be determined by simply considering sequence divergence or intersecting AH variants with all CTCF sites. This is concordant with previous results which suggest that 3D genome folding is governed by a complex CTCF binding grammar [50,80,82,83]. (C) To evaluate whether 3D genome organization constrained sequence divergence, we estimate the null distribution of expected 3D divergence based on sequence differences between the Neanderthal (Vindija) and African MH (HG03105) genomes.…”

“…Given the weak relationship between sequence and 3D divergence, we sought to identify some properties of sequence differences that result in large 3D divergence. Based on the importance of CTCF-binding in maintaining 3D genome organization [50, 62, 79, 80], we quantified the effects of AH-MH nucleotide differences overlapping CTCF binding motifs. Disruption of CTCF binding sites is important, but not all disruptions are likely to influence 3D divergence.…”

“…However, the role of 3D genome organization in the divergence between AHs and MHs has never been explored because chromatin contacts cannot be assayed in ancient DNA. 3D genome folding is facilitated by a complex interplay of CTCF binding with cohesin and other architectural factors [50, 62, 79, 80]. Recent deep learning methods have been developed that learn the sequence “grammar” underlying 3d genome folding patterns [81–84].…”

“…3D genome folding is facilitated by a complex interplay of CTCF binding with cohesin and other architectural factors [50,62,79,80]. Recent deep learning methods have been developed that learn the sequence "grammar" underlying 3d genome folding patterns [81][82][83][84].…”