Abstracts

Rationale: The identification of mechanistically novel drug targets for epilepsy is highly challenging. To address this problem we developed and experimentally validated a new computational approach to drug target identification that combines gene-regulatory information with a causal reasoning framework (“causal reasoning analytical framework for target discovery” – CRAFT). CRAFT aims to map the underlying drivers and regulators of disease states as novel candidate drug targets for disease modification.  Methods: Network-based systems analyses provide powerful new approaches to elucidate molecular processes and pathways underlying disease. The strength of the gene network approach comes from the analysis of multiple genes in functionally enriched pathways, as opposed to single gene approaches that examine only one component of a complex system at a time. Using genome-wide transcriptional profiling in tissues relevant to the disease under investigation, network analysis can identify modules (i.e., sets of co-expressed genes) as candidate regulators and drivers of disease. Network-based drug discovery aims to harness this knowledge to identify drugs capable of restoring the expression of disease modules toward health. At this systems-level framework, therapeutic compounds are judged not by their binding affinity to a particular protein, but by their ability to induce a transcriptional response (i.e., a gene expression profile) that is anti-correlated to the coordinated transcriptional program underpinning the disease state. This systems-level approach to target discovery is orthogonal to traditional concepts of drug discovery. Specifically, CRAFT utilizes known regulatory interactions between membrane receptors, transcription factors and target genes to establish a predictive functional genomics framework for identifying membrane receptors exerting a regulatory influence over disease states.  Results: Starting from genome-wide hippocampal gene expression data from a model of acquired epilepsy (the post-status epilepticus pilocarpine model of temporal lobe epilepsy), CRAFT predicted the tyrosine kinase receptor Csf1R as a novel therapeutic target for epilepsy. Specifically, Csf1R was predicted to regulate an inflammatory module underpinning the acquired epilepsy disease state. Csf1R is a membrane receptor expressed by myeloid lineage cells including microglia. Using a small molecule inhibitor of Csf1R the therapeutic effect of Csf1R blockade in epilepsy was validated in two separate pre-clinical models of acquired epilepsy, including a model of pharmacoresistant epilepsy (kainate model). In addition to Csf1R, CRAFT identified many other candidate regulators of acquired epilepsy that may warrant further investigation as potential novel antiepilepsy drug targets.  Conclusions: These results suggest Csf1R blockade as a novel therapeutic strategy in acquired epilepsy, and highlight CRAFT as a systems-level framework for predicting mechanistically new drugs and targets. Although in this study we applied CRAFT to epilepsy, our method is equally applicable to any disease for which an underlying disease module can be identified. From a clinical perspective, our study highlights Csf1R as a potential novel drug target for acquired epilepsy, and further supports the potential for immunomodulatory therapies as a valid therapeutic approach in acquired epilepsy.  Funding: UCB PharmaEU FP7 EPITARGET Imperial College NIHR BRC