CE45 - Mathématique, informatique, automatique, traitement du signal pour répondre aux défis de la biologie et de la santé

LoToReg: Local Topology and global Regulation, modeling of a dynamic and multiscale coupling – LoToReg

LoToReg: Local Topology and global Regulation, modelling of a dynamic and multiscale coupling

DNA topology, i.e., torsional deformation experienced by the double-helix, is an ubiquitous feature both in prokaryotes and eukaryotes. In bacteria, it is controlled by conserved topoisomerase enzymes, and is involved in an ancestral double-sided coupling with transcription. Relating the structure and function of the genome, it acts as a basal and global transcription factor that remains poorly understood. Characterizing this mechanism is thus a fundamental problem in bacterial genetics.

Mapping the supercoil distribution along a bacterial genome and modelling its interplay with global transcription

The spatial distribution of supercoils along bacterial genomes is currently not experimentally accessible. We intend to map this distribution with high-throughput sequencing techniques based on the intercalating agent TMP (tri-methyl-psoralen). This analysis will be carried in the model phytopathogen Dickeya dadantii, in different culture conditions relevant to the pathogenic process where DNA topology was shown to play a key role, as well as supercoil variations induced by topoisomerase inhibitors. In particular, we will complement the usual gyrase inhibitor novobiocin with the novel and antagonistic topoisomerase 1 inhibitor seconeolitsin. In all conditions investigated, the transcriptome will be analyzed by RNA-Seq. <br />Based on these new experimental data, we will develop a computational model of the transcription-supercoiling coupling. It will be based on an existing stochastic model transcription developed and tested on small-scale experiments, which will be expanded to the whole genome, and allows inferring the supercoil distribution a given condition-dependent expression pattern. The model predictions and experimental data will then be compared and the model refined in an iterative manner.

- High-throughput mapping of supercoil distributions based on tri-methyl-psoralen intercalation and crosslinking.
- Stress conditions mimicking the dynamics of plant infection
- Gyrase and topoisomerase 1 transient inhibition by antibiotics
- Transcriptome analysis by RNA-Seq
- Stochastic simulations

1. Development of a computational model of the transcription-supercoiling coupling applicable to both in vitro or in vivo experiments on model promoters, but also to whole-genome expression analysis for a wide range of bacterial species
2. Publication of a simulation code for the latter model
3. Characterization of the action of the topoisomerase 1 inhibitor seconeolitsin in the Gram-negative bacterium Dickeya dadantii

Combining the functional (gene expression) and structural (DNA topology) aspects of the genome within a single mechanistic modelling will contribute to a paradigm shift in gene regulation. It will also contribute in bridging the communities of biophysicists (who describe topological domains independent of genes) and systems biologists (who study regulation networks independent of chromatin structure).

1. El Houdaigui, B., & Meyer, S. (2020). TwisTranscripT: stochastic simulation of the transcription-supercoiling coupling. Bioinformatics
2. Martis, B. S., Forquet, R., Reverchon, S., Nasser, W., & Meyer, S. (2019). DNA supercoiling: An ancestral regulator of gene expression in pathogenic bacteria? Computational and Structural Biotechnology Journal, 17, 1047-1055
3. El Houdaigui, B., Forquet, R., Hindré, T., Schneider, D., Nasser, W., Reverchon, S., & Meyer, S. (2019). Bacterial genome architecture shapes global transcriptional regulation by DNA supercoiling. Nucleic acids research, 47(11), 5648-5657.

Accumulating data suggest that bacterial genes are co-regulated along spatial genomic domains, even when they do not share any transcription factor, thus escaping classical regulation models. The LoToReG project aims at explaining these observations through a novel ancestral form of transcriptional regulation based on the activity of RNA Polymerase (RNAPol) on DNA, through the formation of coupled topological-transcriptional domains. DNA supercoiling or topology has been long known as a key factor influencing bacterial transcription, but has been disregarded in genome-wide regulation models. In our working hypothesis in contrast, transcription and local topology affect each other in a dynamic coupling, which results in a basal and complex interaction between adjacent genes in a wide range of organisms.
The objective is to develop a quantitative computational modelling of this basal regulation mode. Our strategy integrates an analysis of novel high-throughput data obtained in the phytopathogenic Gram-negative bacterium Dickeya dadantii, together with a dynamic 1D modelling/simulation of the transcription – local topology coupled process along an entire chromosome. The modelling will be based on the first simultaneous high-resolution in vivo maps of gene expression (by RNA-Seq) and DNA supercoiling distribution (by psoralen intercalation mapping) in bacteria. New information of decisive importance for deciphering the regulation mechanism will be obtained by analyzing (1) the complementary conditions of the DNA gyrase and of topoisomerase I inhibition (the latter by a new antibiotics, never tested in Gram-negative bacteria), (2) nucleoid-associated-protein mutants, and (3) environmental stresses relevant to the infection process.
The dynamic model involves only a few global mechanistic parameters, calibrated on published small-scale experiments involving model systems; condition-dependent input parameters (promoter initiation rates, topological domain boundaries) will be inferred from the data, in order to simulate the in vivo process. The computed supercoiling distribution profiles will be compared to the data for validation, and may lead to model refinements in a back-and-forth strategy. The modelling will provide the quantitative spatial regulatory networks arising from the activity of RNAPol itself, which may account for a substantial part of neighbour gene co-expressions currently unexplained. Being potentially applicable to a wide range of evolutionary distant organisms, it will have a strong impact in the bacterial regulation community, and may contribute in a paradigm shift in gene regulation, from the classical vision centred on transcription factors, towards a multiscale view where the chromatin state also plays an important role. In this shift, our model will be the first to explicitly and quantitatively describe the interaction between the functional (gene expression) and structural (DNA physical state) aspects of an entire chromosome.
This modelling will then be applied to biologically interesting cases. In particular, we will assess the effect of DNA supercoiling in the pathogenicity islands of several bacterial species, where this mechanism could play a crucial role in coordinating the proper regulation of virulence genes during the infection process. By identifying an ancestral regulation mode based on common ingredients, this analysis could open the road to new common strategies in the fight against pathogenic bacteria.

Project coordinator


The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.



Help of the ANR 185,068 euros
Beginning and duration of the scientific project: December 2018 - 48 Months

Useful links

Sign up for the latest news:
Subscribe to our newsletter