NEW YORK (GenomeWeb) – A UK-led team has uncovered a network of genes that tends to be altered in individuals with epilepsy — a set that it is starting to explore as a source of information on new drug targets.
Using an integrated, systems biology approach, the researchers brought together array-based gene expression data for nine regions in post-mortem brain samples from dozens of healthy individuals. By overlaying epilepsy mutation and genome-wide association study data onto the resulting brain gene co-expression network, they focused in on a network of 360 genes that appears to be altered in individuals with forms of epilepsy involving mutations to one gene or small changes to many genes.
"[O]ur results implicate a convergent role for a synaptic gene network (M30) in rare monogenic and common (presumed polygenic) forms of epilepsy and suggest targeting the expression of this network is [a] potential new therapeutic strategy in epilepsy," co-corresponding authors Michael Johnson of Imperial College London and Enrico Petretto of ICL and Duke-NUS Medical School and their co-authors wrote.
When the team screened some 1,300 known drugs for influence on this network, it found both components from established epilepsy treatments, including valproic acid, as well as new candidates, such as a compound found in Indian ginseng. The study was published in Genome Biology today.
"[W]e are increasingly aware that genes don't work in isolation. Identifying groups of genes that work together, and then targeting these networks of genes may lead to more effective treatments," Johnson said in a statement. "Our proof-of-concept study suggests this network biology approach could help us identify new medications for epilepsy, and the methods can also be applied to other diseases."
Effective treatments remain out of reach for a significant proportion of individuals with epilepsy, he and his co-authors explained, even though dozens of drugs have been discovered and licensed to treat the neurological, seizure-causing condition.
"[D]espite almost 30 different drugs licensed for the condition, a third of people with epilepsy continue to suffer from uncontrolled epileptic seizures," Johnson said, adding that "very little progress in finding more effective drugs for epilepsy has been made in the past 100 years."
For their analysis, the researchers began by analyzing array-based gene expression data for regions in post-mortem brain samples from 88 individuals available through the UK Brain Expression Consortium. There, they identified modules of genes that appeared to be co-expressed across the brain, along with genes with brain region-specific co-expression — results they compared with co-expression data previously reported for the Genotype-Tissue Expression project and other analyses.
By adding in de novo epilepsy variants found from prior exome sequencing studies, the team defined a conserved 320-gene expression module, dubbed M30, that appeared to overlap with changes expected to accompany common, epilepsy-associated variants from past GWAS studies.
With this proposed epilepsy network, the researchers went on to explore everything from protein interaction patterns to drugs with potential effects on the co-expressed genes. In the latter analysis, for example, they used an available Connectivity Map of drug exposure-related expression patterns to narrow in on 15 compounds expected to return differentially expressed genes in the epilepsy network back to control-like expression levels.
The authors noted that their analyses so far suggest that a systemic screen of drug-like molecules based on their measured effect on M30 gene expression "might represent an efficient therapeutic strategy for new drug discovery in epilepsy.