Novel chemical suppressors of long QT syndrome identified by an in vivo functional screen

Abstract

Background: Genetic long QT (LQT) syndrome is a life-threatening disorder caused by mutations that result in prolongation of cardiac repolarization. Recent work has demonstrated that a zebrafish model of LQT syndrome faithfully recapitulates several features of human disease, including prolongation of ventricular action potential duration, spontaneous early afterdepolarizations, and 2:1 atrioventricular block in early stages of development. Because of their transparency, small size, and absorption of small molecules from their environment, zebrafish are amenable to high-throughput chemical screens. We describe a small-molecule screen using the zebrafish KCNH2 mutant breakdance to identify compounds that can rescue the LQT type 2 phenotype.

Methods and results: Zebrafish breakdance embryos were exposed to test compounds at 48 hours of development and scored for rescue of 2:1 atrioventricular block at 72 hours in a 96-well format. Only compounds that suppressed the LQT phenotype in 3 of 3 fish were considered hits. Screen compounds were obtained from commercially available small-molecule libraries (Prestwick and Chembridge). Initial hits were confirmed with dose-response testing and time-course studies. Optical mapping with the voltage-sensitive dye di-4 ANEPPS was performed to measure compound effects on cardiac action potential durations. Screening of 1200 small molecules resulted in the identification of flurandrenolide and 2-methoxy-N-(4-methylphenyl) benzamide (2-MMB) as compounds that reproducibly suppressed the LQT phenotype. Optical mapping confirmed that treatment with each compound caused shortening of ventricular action potential durations. Structure activity studies and steroid receptor knockdown suggest that flurandrenolide functions via the glucocorticoid signaling pathway.

Conclusions: Using a zebrafish model of LQT type 2 syndrome in a high-throughput chemical screen, we have identified 2 compounds, flurandrenolide and the novel compound 2-MMB, as small molecules that rescue the zebrafish LQT type 2 syndrome by shortening the ventricular action potential duration. We provide evidence that flurandrenolide functions via the glucocorticoid receptor-mediated pathway. These 2 molecules and future discoveries from this screen should yield novel tools for the study of cardiac electrophysiology and may lead to novel therapeutics for human LQT patients.

Conflict of interest statement

Conflict of Interest Disclosures

CTJ is a co-founder of Cellular Dynamics International, Inc, a Madison-based stem cell company.

Figures

Figure 1

A chemical screen for suppressors of long QT syndrome resulted in the identification of two structurally unrelated compound classes. (a) Flow chart indicating the small molecule screening strategy and lead compounds. (b) Dose-response curve of the two lead compounds, 2-MMB and flurandrenolide.

Figure 2

Flurandrenolide and 2-MMB treatment of the breakdance mutant partially correct the repolarization defect. (a) Ventricular action potential duration (APD) curves (left) of wildtype (blue) and bkd−/− (red) fish treated with flurandrenolide (dashed lines). Bar graphs (right) indicate the mean APD90 +/− the standard deviation (n=7–8). (b) Ventricular APD curves (left) of wildtype (blue) and bkd−/− (red) fish treated with 2-MMB (dashed lines). Bar graphs (right) indicate the mean APD90 +/− the standard deviation (n=7–16).

Figure 3

zERG I59S mutation leads to a trafficking defect, which is not corrected by flurandrenolide or 2-MMB. Summary data from Western blots (n = 3–4) indicating the amount of 155 kd (upper) cell surface isoform of zERG and zERG-I59S as a percentage of total zERG protein under various conditions is shown above on a representative Western blot. Tubulin is shown as a loading control.

Figure 4

Subcellular localization of zERG and I59S zERG is shown in red and compared to a surface marker in green. The overlay demonstrates surface expression of wildtype in yellow but not I59S zERG. Treatment with flurandrenolide or 2-MMB does not alter surface expression of I59S zERG. Scale bars are 10 μm.

Figure 5

zERG-I59S current density is reduced compared to wildtype. Exemplar traces from from zERG (a) and zERG-I59S (b) transfected cells. (c) The current-voltage relationship for zERG (black circles) and zERG-I59S (white squares). The voltage clamp protocol is shown as an inset. There is a significant reduction in peak current density in I59S zERG compared to wild type (see Table 1).

Figure 6

Flurandrenolide acts via the glucocorticoid signaling pathway. (a) Dose-response curve using steroids known to act through specific signaling pathways: dexamethasone (glucocorticoid pathway), testosterone (androgen pathway), and DOCA (mineralocorticoid pathway). (b) Knockdown of the glucocorticoid receptor blocks rescue of long QT by flurandrenolide and dexamethasone (n= 26–31).