Molecular basis of ranolazine block of LQT-3 mutant sodium channels: evidence for site of action

Abstract

1 We studied the effects of ranolazine, an antianginal agent with promise as an antiarrhythmic drug, on wild-type (WT) and long QT syndrome variant 3 (LQT-3) mutant Na(+) channels expressed in human embryonic kidney (HEK) 293 cells and knock-in mouse cardiomyocytes and used site-directed mutagenesis to probe the site of action of the drug. 2 We find preferential ranolazine block of sustained vs peak Na(+) channel current for LQT-3 mutant (DeltaKPQ and Y1795C) channels (IC(50)=15 vs 135 microM) with similar results obtained in HEK 293 cells and knock-in myocytes. 3 Ranolazine block of both peak and sustained Na(+) channel current is significantly reduced by mutation (F1760A) of a single residue previously shown to contribute critically to the binding site for local anesthetic (LA) molecules in the Na(+) channel. 4 Ranolazine significantly decreases action potential duration (APD) at 50 and 90% repolarization by 23+/-5 and 27+/-3%, respectively, in DeltaKPQ mouse ventricular myocytes but has little effect on APD of WT myocytes. 5 Computational modeling of human cardiac myocyte electrical activity that incorporates our voltage-clamp data predicts marked ranolazine-induced APD shortening in cells expressing LQT-3 mutant channels. 6 Our results demonstrate for the first time the utility of ranolazine as a blocker of sustained Na(+) channel activity induced by inherited mutations that cause human disease and further, that these effects are very likely due to interactions of ranolazine with the receptor site for LA molecules in the sodium channel.

Figure 1

Effect of ranolazine on peak sodium current (I_peak) carried by WT and two disease-associated mutant human sodium channels. (a) Averaged TTX-sensitive traces recorded upon a depolarizing step (200 ms at −10 mV, pulse frequency 0.33 Hz) show currents from HEK 293 cells expressing WT and ΔKPQ sodium channel before and after steady-state block of peak Na⁺ current by ranolazine (50 μM). (b) Bar graphs summarize the effect of ranolazine on peak Na⁺ current measured in WT (n=3), Y1795C (n=4) and ΔKPQ channels (n=7). Normalized block was determined as the fraction of the pulse current normalized to control current reduced by the drug. Shown are mean±s.e.m. data. ns, nonsignificantly different from WT.

Figure 2

Effect of ranolazine on LQT-3 mutant human sodium channel sustained current (I_sus). (a) High gain recordings show averaged sustained Na⁺ current, normalized to peak Na⁺ current, and its block by ranolazine (50 μM) for ΔKPQ channels (peak currents are off-scale). (b) Bars summarize the normalized mean block (±s.e.m.) of ranolazine on I_sus and I_peak for two LQT-3 mutant sodium channels: ΔKPQ (n=7) and Y1795C (n=4). ^**P<0.01 for I_peak vs I_sus.

Figure 3

Concentration–response relationships for ranolazine inhibition of peak and sustained sodium current (I_Na) in ΔKPQ murine cardiomyocytes. (a) Averaged current traces recorded during depolarizing pulses (200 ms, −10 mV, 0.33 Hz) in cardiomyocytes isolated from mice expressing ΔKPQ channels shown at low (left) and high (right) gain before and after steady-state block of peak I_Na (I_peak) and sustained I_Na (I_sus) by ranolazine (20 μM). High-gain traces are normalized to peak current. (b) Concentration–response curves of I_peak and I_sus for ΔKPQ myocytes. The averaged data were fitted with the following function: y=A₁+((A₂–A₁)/(1+10^(log₁₀(IC₅₀)–x)^*p)) where A₁ and A₂ are fractional amplitudes of each component, p is the Hill slope and IC₅₀ is the drug concentration that inhibit the response at 50%. IC₅₀ values for I_peak and I_sus are 135 μM and 15 μM, respectively, n=3–5 cells per concentration. ^*P<0.05; ^**P<0.01 for I_peak vs I_sus.

Figure 4

Effects of ranolazine on the AP in WT and ΔKPQ murine cardiomyocytes. Superimposed APs recorded under control conditions and after addition of ranolazine (10 μM) in WT and ΔKPQ mouse cardiomyocytes stimulated at 0.5 Hz.

Figure 5

Ranolazine and its structural homology with local anesthetics. (a) Structural comparison of ranolazine and lidocaine (blue color represents nitrogen and red color represents oxygen). (b) Schematic view of sodium channel pore illustrating two key residues (F1760 and Y1767) in the LA binding site in the S6 helical transmembrane segment of domain IV (red).

Figure 6

Mutation of the local anesthetic receptor residue (F1760A) diminishes tonic and UDB of peak sodium channel current by ranolazine. (a) Shown are averaged peak I_Na recorded at 5 Hz (UDB) in WT and F1760 (see text) channels expressed in HEK 293 cells before and after inhibition of current by ranolazine (50 μM). (b) Bar graphs summarize the fraction of current blocked for tonic block (0.33 Hz) and UDB (5 Hz) measured in WT (n=3 for both tonic block and UDB) and F1760 channels (n=6 for tonic block and n=5 for UDB). Data are expressed as mean±s.e.m. ^**P<0.01 for WT vs F1760A.

Figure 7

Mutation of the local anesthetic receptor binding site residue (F1760A) diminishes ranolazine block of ΔKPQ channel sustained current. (a) Shown are averaged high-gain current traces recorded in response to 200 ms voltage pulses (−10 mV, 0.33 Hz) before and after exposure to ranolazine (50 μM) recorded in HEK 293 cells expressing ΔKPQ channels and ΔKPQ mutant channels in an α subunit construct harboring the F1760A mutation. (b) The bar graph summarizes the normalized mean ranolazine block of I_sus (±s.e.m.) for ΔKPQ (n=7) and ΔKPQ_F1760A (n=6) channels. ^**P<0.01 for ΔKPQ vs ΔKPQ_F1760A.

Figure 8

Simulated effects of ranolazine on APs in human myocytes. Simulated APs obtained in the absence and in the presence of ranolazine (5 μM) in WT and LQT-3 human myocytes. Each AP shown is the 100th AP in a train of APs stimulated at 1 Hz. At this pacing rate, ranolazine decreases the APD by 15 ms in WT cells and 39 ms in LQT-3 mutant cells.