Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - INTEGRAL (Signal Integration by Gene Regulatory Landscapes)

Teaser

Biologists have been fascinated for a long time by the question how a fertilized egg gives rise to a whole organism, which is a fundamental question as humans are curious about their origins. All organs and tissues of a functional organism at birth are formed during...

Summary

Biologists have been fascinated for a long time by the question how a fertilized egg gives rise to a whole organism, which is a fundamental question as humans are curious about their origins. All organs and tissues of a functional organism at birth are formed during embryogenesis. Another fascinating aspect of embryogenesis is that features of an organism’s evolutionary history are recapitulated during development. This is very relevant to a better understanding of how the molecular mechanisms and interactions that control embryonic development were established. By now, the vast majority if not all of the genes that control organ and tissue development are known and their functions have been analyzed. One of the major current challenges is to understand how these genes interact as part of the pathways and gene regulatory networks (GRNs) that regulate embryonic development. There is increasing evidence that these often self-regulatory gene regulatory networks endow the molecular circuits and interactions controlling embryonic development with robustness. The spatio-temporal regulation of gene expression is a key component of the interactions that orchestrate the morphogenesis of organs and tissues in embryos. Recent research has shown that many alterations of gene expression are caused by mutations and/or deletions affecting so-called cis-regulatory modules (CRMs) within the non-coding part of the genome. These cis-regulatory mutations alter or even inactivate gene expression, which will likely result in embryonic lethality and/or birth defects. Likewise, alterations affecting the expression of developmental regulator genes have been proposed as drivers of evolutionary diversification of organ and tissue development. Therefore, the functional analysis of the effects of molecular changes that alter gene expression during development will also provide insight into the cis-regulatory changes that underlie evolutionary diversification. In this ERC-funded research project, we focus our analysis mostly on CRMs that function as transcriptional enhancers as they up-regulate transcription in a tissue-specific and/or spatio-temporally controlled manner. Enhancer activities are controlled by transacting complexes of transcriptional regulators in embryonic cells, which express the target gene. In summary, the activation and spatio-temporal regulation of gene expression is controlled by (1) the interaction of transcription factor complexes with specific enhancers and (2) their interactions with the promoter. This analysis is of general relevance as many congenital malformations and diseases are caused by mutations that alter CRMs and/or chromatin architecture rather than protein coding exons. This study is expected to provide fundamental novel insights into molecular mechanisms that regulate robustness and evolvability of gene expression patterns during embryonic development. As a paradigm for this analysis, we have chosen the cis-regulation of gene expression during limb bud development.
The developing limb of tetrapods (= four limbed animals) is our experimental paradigm to study how signaling impacts gene regulation. To understand how signaling inputs are integrated by cells into the regulation of gene expression we study both the relevant transcription factors complexes and the CRMs that regulate transcription. Previously, we showed that a self-regulatory system of interlinked signaling feedback loops controls both limb bud patterning and outgrowth. The BMP antagonist Gremlin1 (Grem1) is a key node in this self-regulatory signaling system and its spatio-temporal expression in the limb bud mesenchyme is controlled by inputs from at least three major pathways, namely SHH, FGF and BMP signaling. Using loss-of-function genetics in the mouse, we showed that Grem1 is essential for maintaining and propagating this signaling system and normal limb bud development. This analysis also uncovered the large genomic landscape that is require

Work performed

We have identified and analysed the functions of the CRMs in the Grem1 genomic landscape by combining gain-of-function and loss-of-function genetic approaches in mouse embryos. Initially classical LacZ reporter assays were used to identify active enhancers among the CRMs that are active in the limb bud mesenchyme of transgenic mouse embryos This identified at least eight enhancers in the large Grem1 genomic landscape with partially overlapping activities, which pointed to functional cooperativity and redundancies. In particular, this analysis revealed two clusters of enhancers and we used CRISPR/Cas9-mediated genome engineering to assess the functional requirement of both individual and clusters of enhancers for regulating the Grem1 expression dynamics. In addition, the effects of deleting CRMs on the self-regulatory signaling system during limb bud development was also assessed. This ongoing analysis has already revealed some remarkable features: inactivation of individual enhancers has in many cases no effect on Grem1 expression and limb bud development which points to cis-regulatory robustness. Removing either of two enhancer clusters reduces Grem1 expression significantly but limb bud development progresses normally in embryos homozygous for these larger loss-of-function mutations which points to signaling systems robustness. Indeed, our ongoing analysis provides molecular evidence that compensatory changes in the self-regulatory signaling system and gene regulatory networks balance the reduction in Grem1 expression which results in normal progression of limb bud development. One possible molecular explanation of the observed cis-regulatory- and systems level robustness is the fact that the downstream transcriptional mediators of SHH and WNT signaling and HOX transcription factor complexes interact with the majority of all CRMs in the Grem1 genomic landscape. In contrast, SMAD4 transcriptional complexes which mediate BMP signal transduction, interact with specific enhancers in the Grem1 genomic landscape. In addition, the ongoing chromatin architecture analysis is revealing another feature, namely that even large deletions do not prevent the remaining CRMs to establish contacts with the Grem1 promoter. Together this analysis reveals how a gene expression pattern is elaborated in a robust manner by the presence of multiple CRMs/enhancers and by compensatory changes in the associated gene regulatory networks.
When comparing the Grem1 distribution in limb buds from different species, it becomes immediately clear that its spatio-temporal expression is rather plastic and e.g. prefigures changes in the pattern of digits at early limb bud stages. This prompted us to initiate a comparative evolutionary analysis using limb buds of specific species that represent different classes of tetrapods. As the ancestor of modern tetrapods conquered land, this ultimately resulted in a remarkable diversification of limbs allowing for colonization of very different habitats. In particular, mammals are a class of modern tetrapods with very successful radiation and the diversity of their limb structures provides an ideal setting to study the molecular changes underlying evolutionary diversification. Therefore, another major focus of our research is to understand how the Grem1 cis-regulatory landscape changed during evolution of tetrapod limbs. The activities of the two functionally most closely related enhancers among the CRMs identified were comparatively analysed using LacZ reporter assays in mouse embryos. These CRMs were isolated from select species representing different classes and branches of tetrapods (rabbit, pig, cow, chicken and lizard in addition to mouse), which showed that the activity of one CRM is very robust across different species, while the other is very variable. Initially rather unexpected, these two Grem1 enhancers were detected in cartilaginous fish and shown to be active in mouse limb buds, which suggested that the c

Final results

In addition to the results described in the previous section, we have also initiated a systematic and integrative genome-wide bioinformatics analysis to comparatively analyse the molecular regulation of mouse and chicken limb bud development. One major aim of this analysis is to gain insight into the changes in the relevant gene regulatory networks that control limb bud development in the two species. In the long run this approach together with the molecular analysis of the relevant enhancer may pinpoint key molecular alterations underlying diversification of the type we have previously identified for artiodactyl limb bud development.
1. The first major aim of our ongoing genetic studies of the Grem1 genomic landscape is to complete the analysis of the CRMs that control Grem1 expression and gain a comprehensive understanding of how the spatio-temporal expression dynamics are regulated. As part of this analysis, we will also investigate which are the CRMs /enhancers required for BMP/SMAD4-mediated activation and SHH/GLI-mediated upregulation of Grem1 expression using a combination of genetic and loss-of-function analysis with experimental manipulation of pathway activities in cultured limb buds. Similarly, we will analyse the role of WNT/ß-Catenin signaling in regulating Grem1 expression. The ultimate aim of this analysis is to gain a comprehensive understanding of the trans-/cis-regulatory interactions that impact Grem1 expression during progression of limb bud development.
2. As Grem1 is a key node in the self-regulatory feedback signaling system, the second aim will be to understand how changes in Grem1 expression impact this system and/or are compensated by the feedback signaling interactions (see before). This will be done by combining genetics with transcriptome analysis. In addition, our genetic and functional analysis of the WNT/ß-Catenin and BMP/SMAD4 pathways has uncovered additional signaling interactions that may be part of the self-regulatory signaling system. These will be analysed further in the framework of this ERC project by combining genetics with transcriptome analysis. As part of this systems analysis, we plan to perform in silico simulations of wild-type and relevant mutant phenotypes to gain insights into the underlying cis-regulatory logic and crosstalk between different pathways.
3. This analysis will also be of direct relevance to the third aim, which is to gain insight into how the Grem1 cis-regulatory landscape was modified during limb evolution and diversification. In particular, we will analyse the functionally most relevant CRMs in more detail to identify the regions responsible for conserved (i.e. evolutionary constraint) and species-specific (i.e. evolutionary variable) activity patterns. Our analysis already pinpointed an evolutionary novelty in one of the CRMs by which its non-polarized activity in lower species is restricted to the posterior limb bud in mammals. We aim to identify the cis-regulatory changes underlying this differential activity and their potential impact on trans/cis interactions with the relevant transcription factor complexes (see before). Ultimately, we will consider to swop the most relevant evolutionary divergent enhancer into the mouse using genome editing and study the effects the endogenous Grem1 expression and the signaling system in developing mouse limb buds.