Personalized structural biology reveals the molecular mechanisms underlying heterogeneous epileptic phenotypes caused by de novo KCNC2 variants

Abstract

Background. Next-generation whole exome sequencing (WES) is ubiquitous as an early step in the diagnosis of rare diseases and the interpretation of variants of unknown significance (VUS). Developmental and epileptic encephalopathies (DEE) are a group of rare devastating epilepsies, many of which have unknown causes. Increasing WES in the clinic has identified several rare monogenic DEEs caused by ion channel variants. However, WES often fails to provide actionable insight, due to the challenges of proposing functional hypotheses for candidate variants. Here, we describe a "personalized structural biology" (PSB) approach that addresses this challenge by leveraging recent innovations in the determination and analysis of protein 3D structures. Results. We illustrate the power of the PSB approach in an individual from the Undiagnosed Diseases Network (UDN) with DEE symptoms who has a novel de novo VUS in KCNC2 (p.V469L), the gene that encodes the Kv3.2 voltage-gated potassium channel. A nearby KCNC2 variant (p.V471L) was recently suggested to cause DEE-like phenotypes. We find that both variants are located in the conserved hinge region of the S6 helix and likely to affect protein function. However, despite their proximity, computational structural modeling suggests that the V469L variant is likely to sterically block the channel pore, while the V471L variant is likely to stabilize the open state. Biochemical and electrophysiological analyses demonstrate heterogeneous loss-of-function and gain-of-function effects, respectively, as well as differential inhibition in response to 4-aminopyridine (4-AP) treatment. Using computational structural modeling and molecular dynamics simulations, we illustrate that the pore of the V469L variant is more constricted increasing the energetic barrier for K+ permeation, whereas the V471L variant stabilizes the open conformation Conclusions. Our results implicate KCNC2 as a causative gene for DEE and guided the interpretation of a UDN case. They further delineate the molecular basis for the heterogeneous clinical phenotypes resulting from two proximal pathogenic variants. This demonstrates how the PSB approach can provide an analytical framework for individualized hypothesis-driven interpretation of protein-coding VUS suspected to contribute to disease.