Design and discovery of new tool compounds for studying the role of Slack potassium channels in malignant migrating partial seizure of infancy

Purpose: Slack channels are sodium-activated potassium channels encoded by the KCNT1 gene, and are key regulators of electrical activity in the central nervous system. Slack channels belong to the Slo family of potassium channels (Slo2.2). The development of malignant migrating partial seizure of infancy (MMPSI), a type of severe infantile epilepsy, has been linked to KCNT1 gain of function mutations. The aim of this project is to design and synthesize Slack channel inhibitors for use as in vivo probes via an iterative hit optimization approach. Methods: 110K-member library was screened in a cell-based assay against wild-type Slack and three MMPSI-associated KCNT1 mutants. The hit compound VU0531245 (VU245) was selected for the development of structure-activity relationship (SAR) studies in order to optimize its potency and drug metabolism and pharmacokinetic (DMPK) properties. A thallium flux assay in HEK-293 cell lines stably expressing Slack channels was utilized to evaluate Slack inhibitory activity of the resulting compounds. Results: Compound libraries were designed around VU245 through the systematic scanning of the chemical space and incorporating various bioisosteric replacements. Our data suggest that modifications at the phenyl ring A lead to mode switching from inhibition to activation. However, fluorinated, alicyclic, and deuterated alkoxy groups at 2-position of the phenyl ring maintained Slack inhibitory activity; moreover, they improved some DMPK properties. We have demonstrated that piperidine replacement at ring B was tolerated. In addition, an ethylene linker in place of the 1,2,4-oxadiazole ring at position C maintained the Slack activity. A 4-fluorophenyl ring at position D was tolerated, improved the metabolic stability, and was used as the basis for further SAR exploration. Finally, a sulfonamide linker was optimal for Slack inhibitory activity. Collectively, the tested analogs within this series demonstrated good passive permeability with no evidence of P-gp-mediated efflux; however, high protein binding (fu ~ 0.01 - 0.04) was observed. Conclusions: SAR for Slack activity and DMPK properties was identified around VU245. Further optimization is required to develop suitable Slack in vivo probes with improved potency and DMPK properties.