Impact of genome architecture on the functional activation and repression of Hox regulatory landscapes

Abstract

Background The spatial organization of the mammalian genome relies upon the formation of chromatin domains of various scales. At the level of gene regulation in cis , collections of enhancer sequences define large regulatory landscapes that usually match with the presence of topologically associating domains (TADs). These domains often contain ranges of enhancers displaying similar or related tissue specificity, suggesting that in some cases, such domains may act as coherent regulatory units, with a global on or off state. By using the HoxD gene cluster, which specifies the topology of the developing limbs via highly orchestrated regulation of gene expression, as a paradigm, we investigated how the arrangement of regulatory domains determines their activity and function. Results Proximal and distal cells in the developing limb express different levels of Hoxd genes, regulated by flanking 3′ and 5′ TADs, respectively. We characterized the effect of large genomic rearrangements affecting these two TADs, including their fusion into a single chromatin domain. We show that, within a single hybrid TAD, the activation of both proximal and distal limb enhancers globally occurred as when both TADs are intact. However, the activity of the 3′ TAD in distal cells is generally increased in the fused TAD, when compared to wild type where it is silenced. Also, target gene activity in distal cells depends on whether or not these genes had previously responded to proximal enhancers, which determines the presence or absence of H3K27me3 marks. We also show that the polycomb repressive complex 2 is mainly recruited at the Hox gene cluster and can extend its coverage to far- cis regulatory sequences as long as confined to the neighboring TAD structure. Conclusions We conclude that antagonistic limb proximal and distal enhancers can exert their specific effects when positioned into the same TAD and in the absence of their genuine target genes. We also conclude that removing these target genes reduced the coverage of a regulatory landscape by chromatin marks associated with silencing, which correlates with its prolonged activity in time. Electronic supplementary material The online version of this article (10.1186/s12915-019-0677-x) contains supplementary material, which is available to authorized users.

Polycomb Repressive Complexes in Hox Gene Regulation: Silencing and Beyond

“…It was thus proposed that PRC2 recruitment at HoxD genes is required to initiate the spreading of H3K27me3 over the proximal limb regulatory elements, to ensure the inactivation of the proximal regulatory landscape during the transition from proximal to distal limb fate. [ 94 ] Accordingly, studies of de novo PRC2 recruitment in ES cells provided evidence that H3K27me3 spreading from nucleation sites occurs both locally as well as in a long‐range manner, consistent with the spatial interaction of PRC‐bound loci within the nucleus. [ 83 ] Based on these data, it was proposed that the numerous H3K27me2/3 domains stem from a limiting number of nucleation sites, [ 83 ] which in turn further emphasizes the importance of the mechanisms underlying the dynamics of PRC occupancy.…”

“…Consistent with the evidence that H3K27me3 spreads from nucleation sites, [ 83 ] targeted replacement of the HoxD cluster with the Hoxd9lacZ transgene affects H3K27me3 occupancy over the regulatory elements controlling HoxD expression. [ 94 ] Indeed, while the 3′TAD, containing the 3′ Hoxd genes and the proximal limb enhancers, becomes normally covered by H3K27me3 in wild type distal limb cells, H3K27me3 occupancy is largely lost upon deletion of the HoxD cluster. It was thus proposed that PRC2 recruitment at HoxD genes is required to initiate the spreading of H3K27me3 over the proximal limb regulatory elements, to ensure the inactivation of the proximal regulatory landscape during the transition from proximal to distal limb fate.…”

Chromatin topology and the timing of enhancer function at thehoxdlocus

“…For E12 ChIPs, the four entire limbs coming from one embryo were used in each experiment. All ChIP experiments were conducted following the ChIPmentation method (Schmidl et al, 2015) as in (Rodríguez-Carballo et al, 2019 and proteinase K. DNA was then purified using Qiagen MiniElute kit and a qPCR was performed to determine the number of cycles to be applied during library amplification.…”

“…DNA libraries were purified, size selected with CleanNGS magnetic beads (CleanNA) and sequenced in a HiSeq 4000 machine as 50 bp or 100 bp reads. The sequencing output was mapped and processed as in (Rodríguez-Carballo et al, 2019) without a normalization step. The track profiles were obtained using pygenometracks (Ramírez et al, 2018).…”

Chromatin topology and the timing of enhancer function at the HoxD locus

“…in ref. 58. Briefly, samples were Polytron minced and homogenized by douncing in Prep Buffer (ChIP-IT High Sensitivity Kit; Active Motif).…”