Abstracts

Rationale: Mutations in SCN2A (encoding human voltage-gated sodium channel NaV1.2) have been associated with early-onset epileptic encephalopathy (EOEE). Studies suggest that gain-of-function (GOF) is common among EOEE-associated mutations, but loss-of-function (LOF) defects have also been observed. Knowing which variants can be pharmacologically corrected by specific drugs could guide clinical studies to determine the efficacy of genotype-specific treatments. Diverse patterns of NaV1.2 channel dysfunction may underlie differences in disease severity, clinical course, and pharmacological responses. The availability of a robust, rigorous, and standardized in vitro platform for testing drugs on specific ion channel variants will be valuable for achieving the goal of precision medicine in genetic epilepsy and may contribute to future drug approvals. Here we used an automated electrophysiology platform to determine the effects of PRAX-330 and carbamazepine (CBZ) on two SCN2A gain-of-function mutations (R1626Q and M1879T) associated with EOEE.  Methods: The mutants (R1626Q and M1879T) were engineered in a recombinant human NaV1.2 plasmid and expressed heterologously (by electroporation) in HEK-293T cells stably expressing the b1 and b2 subunits. Automated patch clamp recording was performed using the Nanion Syncropatch 768 PE. We determined the biophysical properties of each variant in the absence of compound to establish the functional defects and then assayed the mutations in the absence and presence of PRAX-330 or CBZ.  Results: Whole cell current density recorded at 0 mV from a holding potential of -120 mV was slightly smaller while inactivation rate was substantially slower for R1626Q and M1879T compared to WT. The voltage dependence of inactivation for both mutants was shifted towards more depolarized potentials compared to WT channels (V0.5 values: WT, -51.5 +- 0.7 mV; R1626Q, -48.1 +- 0.6; and M1879T, -43.7 +- 0.9 mV; p < 0.005). The voltage dependence of activation was not different. The overall effect of both mutations was judged to be GOF. PRAX-330 and CBZ evoked a concentration-dependent hyperpolarized shift in the voltage dependence of inactivation exhibited by WT NaV1.2 and both mutants. 3 µM PRAX-330 shifted the V0.5 for WT, R1626Q and M1879T by 10.7 +- 1.2, 13.2 +- 1.0 and 11.1 +- 1.7 mV respectively. 600 µM CBZ shifted the V0.5 for WT, R1626Q and M1879T by 12.2 +- 2.7, 12.8 +- 1.6 and 4.7 +- 2.1 mV respectively. The potency was substantially greater for PRAX-330 than carbamazepine. The voltage dependence of activation and the inactivation rate were not modified by the drugs. PRAX-330 exerted a similar degree of use-dependent block (UDB) of peak current on WT NaV1.2 and both mutant channels at 5, 10 and 20 Hz. The concentration required to inhibit peak current by 50% (IC50) at 10 Hz was 294, 266 and 224 nM for WT, R1626Q and M1879T, respectively. CBZ did not produce significant UDB.  Conclusions: In this study, we demonstrated that an automated electrophysiology platform can be used to determine the effects of drugs on SCN2A mutations associated with EOEE. We observed that PRAX-330 can correct the voltage dependence of inactivation for two different GOF mutants more potently than CBZ, and uniquely exerted a UDB of peak current. Based on this in vitro investigation, we predict that PRAX-330 will exhibit potent anticonvulsant activity in the setting of SCN2A GOF mutations.  Funding: Research grants from Praxis Precision Medicines, Inc., and NIH (NS108874)