Introduction: The congenital long QT syndrome (cLQTS) increases the risk of arrhythmia and is an important cause of sudden death in the young. The ion channel subunits KCNQ1 (Kv7.1) and KCNE1 assemble in cardiac myocytes to produce the repolarising slow delayed rectifier potassium current IKs. Loss-of-function (LOF) mutations in the KCNQ1 gene cause cLQTS type 1 (LQTS1) (Wang et al., 1996; Sanguinetti et al., 1996). Missense LQTS1 mutations cause LOF by altering channel gating (Class III), promoting defective channel trafficking (Class II) or through a combination of mechanisms. Recent studies highlight that Class II mechanisms underlie disease pathogenesis for a substantial proportion of LQTS1 mutations (Huang et al., 2018) but current therapeutic managements for LQTS1 patients do not target this defect. Therefore, the aim of this study was to identify small molecules that rescue defective mutant KCNQ1 channel trafficking.
Methods: Trafficking assays: LI-COR-based On/In-Cell Western assays were used to quantify channel trafficking as described in Royal et al., (2017). The ‘On-Cell’ assay quantifies cell surface expression (CSE). Cell lines: Two HEK-293 cell lines were generated which stably express the trafficking deficient KCNQ1 mutant channel G325R (VSV-KCNE1-G325R) and the wild-type channel (VSV-KCNE1-KCNQ1). Compound Screening: 26 compounds were tested in three phases. Compounds were applied for 24 hours at 37 °C unless otherwise indicated. DMSO was the vehicle control. Data are presented as fold-matched control (fold) or normalised arbitrary fluorescent units (NAFUs) ±SEM. Statistical analysis was performed using one-way ANOVA with Dunnett’s multiple comparison test or multiple unpaired t-test.
Results: In Phase 1, clinically approved CFTR channel modulators were screened because they may exhibit cross-channel activity (Mehta et al., 2018). Four modulators (including VX-809 and VX-661) were tested but none altered the CSE of VSV-KCNE1-G325R (P>0.05, n=3). In Phase 2, seven compounds that act as proteostasis regulators were screened. Of these, the proteasome inhibitor MG-132 (1 µM) and Thapsigargin (10 µM) promoted increases in VSV-KCNE1-G325R CSE (3.24±0.26 fold (P<0.0001, n=12) and 1.42±0.04 fold (P<0.01, n=3), respectively). However, both compounds exhibited cell toxicity. In Phase 3, 15 KCNQ1 channel modulators (blockers and activators) were screened. Of these, the activator R-L3 (L-364,373) (100 µM) induced a small but significant increase in CSE (1.44±0.15 fold, P<0.05, n=3) and the activator Docosahexaenoic acid (DHA) at 100 µM significantly increased VSV-KCNE1-G325R CSE by 2.27±0.26 fold (P<0.01, n=3). Furthermore, upon extended treatment (for 48 hours), 5 and 10 µM DHA promoted large increases in VSV-KCNE1-G325R CSE (5.57±0.94 and 8.24±0.66 fold; NAFUs: 0.0768±0.0152 and 0.1123±0.0118 vs. DMSO 0.0135±0.0006, both P<0.001, n=5). By contrast, 10 µM DHA for 48 hours did not alter wild-type (VSV-KCNE1-KCNQ1) channel trafficking (P>0.05, n=5).
Conclusions: These data highlight that the channel activators DHA and R-L3 may be able to promote trafficking rescue. However, the underlying mechanisms are unknown and whether these compounds can correct other Class II mutants within a cardiomyocyte cellular setting warrants investigation.