Bioinformatics guided rotenone adjuvant kindling in mice as a new animal model of drug-resistant epilepsy

Drug-resistant epilepsy results from multiple mechanisms which are difficult to fully acquire in animal models. Technological advances, that allow transformation of big data into novel therapies, are now assisting in identification a disease targets for animal modeling. Our goal was to transform the available genomic and proteomic data related to drug-resistant epilepsy into ubiquitous disease target using system biology and network pharmacology approaches, followed by animal modeling and assess its validity. We used a dataset of 42 antiseizure drugs, 175 drug targets, and 601 epilepsy-gene associations to create interactome of 543 diseased proteins linked to drug-resistant epilepsy. DIAMOnD algorithm and DAVID web-services were used to identify 35 disease pathways whereby mitochondrial complex-I was selected for animal modeling. Albino mice were treated with specific inhibitor of mitochondrial complex-I (i.e., rotenone 2.5 mg/kg, i.p on daily basis) along with chemical and electric kindling stimulus for 35 days and 15 days, respectively. According to our results, the rotenone kindling model with inhibited complex-I activity showed significant (P < 0.001) resistance to lamotrigine (15 mg/kg), levetiracetam (40 mg/kg), carbamazepine (40 mg/kg), zonisamide (100 mg/kg), gabapentin (224 mg/kg), pregabalin (30 mg/kg), phenytoin (35 mg/kg), topiramate (300 mg/kg), valproate (200 mg/kg), and drug combinations at doses that had significantly (P < 0.001) controlled seizure severity in lamotrigine-pentylenetetrazole and corneal kindling models. In conclusion, lamotrigine kindling model is more advantageous than earlier described lamotrigine and corneal kindling models which respond to drug combinations. As a result, pre-clinical drug screening through rotenone kindling may uncover broad spectrum drugs with novel antiseizure mechanisms which is a pressing issue to deal with drug-resistant epilepsy.