Menu

Close

Developmentally regulated Shh expression is robust to TAD perturbations

Iain Williamson, Lauren Kane, Paul S. Devenney, Eve Anderson, Fiona Kilanowski, Robert E. Hill, Wendy A. Bickmore, Laura A. Lettice

Preprint posted on April 15, 2019 https://www.biorxiv.org/content/10.1101/609941v1

Article now published in Development at http://dx.doi.org/10.1242/dev.179523

Are TAD borders that important to prevent enhancer rewiring? Tissue-specific enhancers maintain robust control of Shh gene expression after challenging the architectural chromatin context.

Selected by Jesus Victorino

 

Background & Summary

The 3D structure of the DNA is organized in smaller self-interacting areas of approximately 1 Mb (ranging from hundred kbs to several megabases) called Topologically Associating Domains (TADs) where it is common to find the regulatory landscapes of developmental genes. TADs are thought to play a role as functional domains that allow for cis-regulatory elements to interact with promoters shaping the control of gene expression. However, although there are studies showing enhancer rewiring upon TAD perturbation in human diseases [Lupiañez 2015], global CTCF depletion alters chromatin conformation but has limited effects on gene expression [Nora 2017].

In this preprint, Williamson et al. challenged the stability of the Shh TAD with a number of deletions including several CTCF sites located at both borders of the ~1 Mb TAD. The authors generated mouse embryonic stem cells (mESCs) and mouse lines carrying the deletions to analyze their effect on chromatin structure and gene expression. As a result, the Shh-containing locus seemed to be a stable genomic environment in which alterations at the TAD borders did not result in the merging of neighboring TADs. Despite the increase in inter-TAD contacts, the expression patterns of the Shh gene appeared not to be affected by contacts with neighboring enhancers.

 

Key findings

Shh-ZRS interaction is not distance-dependent:

The authors performed a >700 kb deletion containing most of the TAD and including many known enhancers as well as the sub-TAD boundaries. ZRS enhancer-Shh promoter interaction was maintained and the Shh gene kept its domain of expression at the Zone of Polarizing Activity (ZPA) driven by the ZRS enhancer despite now being at <100 kb instead of 850 kb.

Fig.1 – Effect of deletions on the 3-D landscape of the Shh TAD. Heat- maps showing 5C data from cells of the bodies of E11.5 wild-type embryos (A) and embryos homozygous for the 700kb deletion (B), across the 1.7-Mb Shh region. C) Images of representative nuclei from wild type (left) and ∆CTCF1 (right) ESCs showing increased distance for Shh/SBE2/ZRS probes in the mutant as seen by FISH signals . Scale bars = 5 μm. D) Heat maps comparing ∆CTCF1 enrichment (red) with wild type (blue). Green dashed lines indicate TAD boundaries, black dashed lines highlight change in TAD boundary position. Figures taken from this preprint.

 

– Shh-ZRS interaction is enriched in the posterior limb bud:

The strongest long-range interacting spot with the Shh gene was an intronic region of the Limbr1 gene. However, since this region had not been described as an enhancer, this interaction seemed to be non-functional as it also appeared not to change between the active (posterior) and the inactive (anterior) domain in the limb bud.

On the other hand, the interaction between Shh and ZRS, the well-known limb enhancer, is enriched in the cell populations within the posterior limb bud, the domain where the enhancer is functional.

 

Deletion of CTCF sites in mESCs caused border relocation and/or intra-TAD reorganization with enriched interaction within the sub-TADs:

Deletion of the first CTCF site at the 3’ Shh TAD border relocated the boundary by ~40 kb to the 5’ end, approximately where the second CTCF site is located. This caused those 40 kb to lose intra-TAD interactions and gain interactions with the neighbor TAD.

 

– Mutant mice where CTCF sites were deleted had no phenotype:

Deletions of CTCF sites in vivo reduced Shh-ZRS colocalization in the active (posterior) domain of the limb bud to the level of the inactive (anterior) domain. Yet mice had no developmental problems or detectable phenotype and were fertile.

 

– A larger deletion at the boundary resulted in ectopic expression of Mnx1 at the ZPA:

Finally a 35 kb deletion at the 5’ Shh TAD border relocated the boundary by ~40 kb closer to the promoter of Nom1. As a result, ZRS enhancer and Mnx1 are able to interact and the Mnx1 gene starts expressing ectopically in the posterior limb bud.

 

Why I chose this preprint

Regulatory elements control gene expression in time and space. Genomic compartmentalization into functional smaller regions such as A/B compartments (that contributes to distinguishing active from repressed chromatin) or TADs (often containing genes with some degree of shared regulation) may help optimize colocalization of the transcription machinery and bringing enhancers to their target. In fact, deletions or inversions of large pieces of DNA containing TAD borders can alter the expression patterns of the genes in such regions and these mutations can be found in patients with severe phenotypes. Importantly, CTCF at the borders is thought to impede cohesin-mediated loop extrusion, therefore preventing insulating neighbor domains from interacting with each other.

I chose this paper for the systematic approach the authors followed to address the questions of how much CTCF sites account for insulating TADs and how structure would affect the expression of the genes within a specific locus. Somewhat unexpectedly, none of the deletions affecting exclusively the TAD boundaries had a deleterious phenotype, nor was there a significant rewiring of the enhancer repertoire as shown by wild type gene expression patterns in the mutants (except in the particular case of Mnx1).

Dr. Iain Williamson and colleagues also showed that upon large deletions and inversions intra-TAD (deletion of a 700kb region inside the TAD), long-range enhancer-promoter interactions were kept as well as the TAD borders, despite deleting most of the TAD itself.

 

Acknowledgements

I would like to thank preLighter Clarice Hong and Mate Palfy for edits and comments.

 

Question to the authors

– Shh gene expression domains seem to remain uninfluenced by enhancers located in neighbor TADs after challenging CTCF sites at the borders. However, the interaction landscape resulting from these perturbations in the shh TAD shows only a small degree of inter-TAD contacts. Do the authors think that the additive effects of such deletions would increase the degree of inter-TAD crosstalk and, thus, enhancer-mediated ectopic expression?

 

– Deleting a 35 kb region at the ZRS border of the Shh TAD shows increased expression of Mnx1 gene at ZPA domain, probably due to an increased influence of the limb enhancer, showing no phenotype along mutant mice lifespan. However, the expression patterns of Shh seems unaltered in the mutants. Do the authors think that the Shh TAD itself evolutionarily gained extra layers of isolation since subtle ectopic expression of Shh can lead to severe phenotypes as it happens in many cases of polydactyly?

 

– The degree of promoter-enhancer colocalization between Shh and ZRS at the posterior limb bud in the mutant mice with CTCF sites deleted is reduced to the levels of colocalization that occur at the anterior limb bud. This does not affect the expression domain at ZPA as can be seen by in situ hybridization (ISH). Since ISH is rather qualitative, would there be quantitative changes in the expression levels in the posterior limb bud?

 

References:

1. Lupiáñez DG, Kraft K, Heinrich V, Krawitz P, Brancati F, Klopocki E, Horn D, Kayserili H, Opitz JM, Laxova R, Santos-Simarro F, Gilbert-Dussardier B, Wittler L, Borschiwer M, Haas SA, Osterwalder M, Franke M, Timmermann B, Hecht J, Spielmann M, Visel A, Mundlos S. 2015. Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. Cell. 161(5), 1012-1025.

2. Nora EP, Goloborodko A, Valton AL, Gibcus JH, Uebersohn A, Abdennur N, Dekker J, Mirny LA, Bruneau BG. 2017. Targeted Degradation of CTCF Decouples Local Insulation of Chromosome Domains from Genomic Compartmentalization. Cell. 169(5), 930-944.

 

 

Posted on: 21st May 2019 , updated on: 23rd May 2019

Read preprint (No Ratings Yet)




  • Author's response

    Wendy Bickmore shared

    – Shh gene expression domains seem to remain uninfluenced by enhancers located in neighbor TADs after challenging CTCF sites at the borders. However, the interaction landscape resulting from these perturbations in the shh TAD shows only a small degree of inter-TADs contacts. Do the authors think that the additive effects of such deletions would increase the degree of inter-TAD crosstalk and, thus, enhancer-mediated ectopic expression?

     

    We certainly consider it a possibility that combining the CTCF site deletions may increase the degree of inter-TAD interactions and perhaps enhancer adoption. In our future work we will certainly be exploring this question.

     

    – Deleting a 35 kb region at the ZRS border of the Shh TAD shows increased expression of Mnx1 gene at ZPA domain, probably due to an increased influence of the limb enhancer, showing no phenotype along mutant mice lifespan. However, the expression patterns of Shh seems unaltered in the mutants. Do the authors think that the Shh TAD itself evolutionarily gained extra layers of isolation since subtle ectopic expression of Shh can lead to severe phenotypes as it happens in many cases of polydactyly?

     

    There is no evidence. – e.g. from 5C or Hi-C that the boundaries of the Shh are special or more ‘insulated’ than other TADs to protect Shh expression.

     

    – The degree of promoter-enhancer colocalization between Shh and ZRS at the posterior limb bud in the mutant mice with CTCF sites deleted is reduced to the levels of colocalization that occur at the anterior limb bud. This does not affect the expression domain at ZPA as can be seen by in situ hybridization (ISH). Since ISH is rather qualitative, would there be quantitative changes in the expression levels in the posterior limb bud?

     

    This is something we are looking at. Even crossing the CTCF1 deletion mice to a null Shh allele revealed no phenotype, so any reduced levels of expression as a result of the TAD boundary deletion must be relatively small.

    1 comment

    3 months

    steve

    nice work at he “cutting” edge of epigenetics 😉

    Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here
    Close