Identification of key regulators for a gene network associated to epilepsy – EpiReg

In WP1, candidate drugs are tested on rats in which epilepsy of the temporal lobe is induced by injection of lithium and pilocarpine. We are studying their electroencephalograms to objectify the anti-epileptic effect of the selected drugs by comparing rats exposed to these drugs and to a control saline solution. We collect the brains of these rats to study the drug-induced expression changes by sequencing genome-wide messenger RNAs (ie by transcriptome or «RNAseq«).

For WP2, the identification of key regulators of the M30 gene network is accomplished using systems genetics and gene regulatory network analysis. The mapping of the regulatory locus of the M30 network expression was done by Bayesian inference. The method can select the best models of multivariates single nucleotide polymorphisms (SNP) associated to the expression of M30 (GUESS software: www.bgx.org.uk/software/guess.html).

Several regulatory loci were identified using data from 129 patients with temporal lobe epilepsy, but none of them was replicated using data from the United Kingdom Brain Expression Consortium (UKBEC, n = 88).
Tests on the effects of candidate drugs on the rat temporal epilepsy model are underway. Meanwhile, the tests performed on zebrafish are promising: they indicate a protective effect of a candidate drug when exposed to a toxic proconvulsant.

The two WP will generate new genome wide expression data associated with functional and phenotypic data in animal models of epilepsy. The in vivo validation of a candidate anti-epileptic drug and the identification of primary genetic regulator of the epilepsy-associated gene network M30 will provide clear directions for novel drug development for this devastating condition.

Réda C, Kaufmann E, and Delahaye-Duriez A. Machine learning applications in drug development. Comput. Struct. Biotechnol. J. 2020 doi:10.1016/j.csbj.2019.12.006. [Epub ahead of print] hal.inria.fr/hal-02533303/

The main objective of the EpiReg project is the identification of genetic drivers for epilepsy by integrating systems genetics approaches and experimental functional in vivo expertise of the INSERM U1141 collaborators.
Epilepsy is a common, serious neurological disease mainly characterized by a tendency to recurrent unprovoked epileptic seizures. Systems biology and network analyses provide powerful approaches to elucidate the molecular pathways underlying epilepsy and can provide new candidate targets for drug discovery or repurposing. During her recent EU Marie Curie Fellowship in London Dr Delahaye-Duriez led a series of computational and bioinformatic analyses culminating in the identification of a candidate gene regulatory network for epilepsy. Specifically, with her collaborators they identified a network of 320 co-expressed genes (termed M30), which is expressed widely throughout the human brain, is enriched for genes that cause epilepsy when mutated, is under tight neurodevelopmental control, and which encodes proteins that physically interact in synaptic processes. This research was reported in Genome Biology (2016) and the application presented here builds on this published foundation.
The main scientific hypothesis of the EpiReg project is that M30 might be targeted as a novel therapeutic strategy in epilepsy. As a proof of concept, they have already demonstrated that among the drugs capable of inducing transcriptional changes in the expression of M30 network in neurons, valproic acid (VPA), a widely used anti-epileptic drug (AED) with a broad spectrum of efficacy, is the one that most significantly restores the expression of M30 in epilepsy toward health. Using this “signature reversion paradigm”, they undertook a computationally-based screening of drug-like molecules based on their measured effect on the expression of M30 which highlighted several other drugs such as withaferin A (WFA) as having the potential to reverse the epilepsy-associated expression changes of M30 toward health, and therefore representing potential novel candidate drug therapies for epilepsy.
The EpiReg project presented here is articulated in two work packages (WP). Both aim to modulate epilepsy by targeting the M30 gene regulatory network.
WP1 is based on the candidate drug prioritization approach that they applied in the preceding published analyses (summarized above). WP1 aims to validate WFA as an AED in a lithium-pilocarpine rat model of epilepsy, and by comparing its effect on seizure activity and on M30 expression to VPA. The purpose is to assess the correlation between restoration of M30 expression toward health and clinical efficacy.
WP2 intends identifying the M30 gene network key regulator(s) using systems genetics approaches. The first objective of WP2 will be to map key regulatory genes of M30 using Bayesian methods that Dr Delahaye acquired in Dr Petretto’ and Pr Johnson’s lab. To this end, we will take advantage of the available transcriptional and genotype data in human brain from two consortiums: the United Kingdom Brain Expression Consortium (UKBEC) and the Gene Tissue Expression project (GTEx). Then, we will perform in vivo functional validation experiments of the identified M30 key regulatory genes exploring the effect of their knock-down on seizure susceptibility induced by pentylenetetrazol (PTZ) in the zebrafish model, and on the level of expression of M30 genes.
The two WP will generate new genome wide expression data associated with functional and phenotypic data in animal models of epilepsy. The in vivo validation of a candidate AED and the identification of primary genetic regulator of the epilepsy-associated gene network M30 will provide clear directions for novel drug development for this devastating condition.