Enhancer–promoter contact formation requires RNAPII and antagonizes loop extrusion

Competition between transcription and loop extrusion modulates promoter and enhancer dynamics

“…To explore these properties further, we modeled the 800 kb of ESC chromatin around the Sox2 locus as a self-avoiding polymer chain of 1 kb beads, classed as either "neutral", binding PolII (Sox2 promoter and enhancers, determined by peaks of the active histone mark, acetylation of lysine-27 of histone H3 (H3K27ac)), transcribed genic regions (Sox2 gene body), or binding CTCF (determined by ChIP-seq peaks and with motif orientation information included). As we previously did for a theoretical locus to explain chromatin conformation changes on acute loss of PolII 56 , we performed molecular dynamics simulations using Langevin equations to model thermal motion of the chromatin and its binding factors, with the following additions to account for biological processes: PolII molecules have affinity for promoters and enhancers, and traverse the gene body to simulate transcription; PolII molecules also have affinity to each other to simulate condensate formation; cohesin complexes can bind anywhere on the polymer and extrude loops, but has preference to load at promoters and enhancers, and is unable to process past a CTCF-bound site when the motif is oriented towards the extruding loop (see Methods). The ensemble of chromatin conformations resulting from these simulations generated a contact map closely resembling that of the experimental Hi-C results In accordance, application of GPTool to the simulated trajectories gave significantly smaller α app measurements at the promoter and enhancer than control regions (q < 5x10 -22 ; Wilcoxon rank sum test with Benjamini-Hochberg correction) (Fig 3e).…”

“…Dispersed transcription factor binding sites at the Sox2 and other loci have been shown to cluster via protein-protein interactions in ESCs 12 , and such homotypic interactions are believed to be a basis for chromosome compartment formation 63 . Various studies suggest that loop extrusion and chromosome compartmentation are antagonistic processes organizing population-averaged chromatin architectures 47,48,56,62,64 . It is therefore possible that processes responsible for compartmentation are also generally antagonistic to cohesin-mediated loop extrusion.…”

“…Molecular dynamics simulations were performed via the multipurpose EspressoMD package 81 , as in ref 56. Individual proteins are represented by "beads" interacting via phenomenological force fields and move according to the Langevin equation, and the chromatin fiber is represented as a chain of beads connected by bonds.…”

Biallelic non-productive enhancer-promoter interaction precedes imprinted expression ofKcnk9during mouse neural commitment

“…On the maternal allele, the sole pre-existing interaction between the enhancer and Kcnk9 promoter induce its maternal expression. Importantly, the associated recruitment of RNAPolII, which can promote enhancer-promoter interaction ( Zhang et al, 2023 ), will in turn affect chromatin structure by strengthening this interaction, which becomes stronger on the maternal allele as expression increases. This model, which remains to be validated, explains the allelic specificity of the enhancer despite biallelic interactions with Kcnk9 and provides an alternative to the canonical isolation model.…”

May the force be with you: Nuclear condensates function beyond transcription control

“…[57] Of importance, the proteins belonging to the PcG-condensates favor long-range genomic interactions, shaping the genome in hubs to ensure PcG repressive function. [58,59] Recent studies also strengthen transcriptional condensate members' role in genome folding: RNA Pol II sustains the E-P loops, [60] and UTX controls chromatin looping in a condensation-dependent manner. [9] In the same direction, several TFs and coactivators have been described as instrumental in mediating regulatory elements' proximity.…”