Potent hERG channel inhibition by sarizotan, an investigative treatment for Rett Syndrome

Abstract

Rett Syndrome (RTT) is an X-linked neurodevelopmental disorder associated with respiratory abnormalities and, in up to ~40% of patients, with prolongation of the cardiac QT_c interval. QT_c prolongation calls for cautious use of drugs with a propensity to inhibit hERG channels. The STARS trial has been undertaken to investigate the efficacy of sarizotan, a 5-HT_1A receptor agonist, at correcting RTT respiratory abnormalities. The present study investigated whether sarizotan inhibits hERG potassium channels and prolongs ventricular repolarization. Whole-cell patch-clamp measurements were made at 37 °C from hERG-expressing HEK293 cells. Docking analysis was conducted using a recent cryo-EM structure of hERG. Sarizotan was a potent inhibitor of hERG current (I_hERG; IC₅₀ of 183 nM) and of native ventricular I_Kr from guinea-pig ventricular myocytes. 100 nM and 1 μM sarizotan prolonged ventricular action potential (AP) duration (APD₉₀) by 14.1 ± 3.3% (n = 6) and 29.8 ± 3.1% (n = 5) respectively and promoted AP triangulation. High affinity I_hERG inhibition by sarizotan was contingent upon channel gating and intact inactivation. Mutagenesis experiments and docking analysis implicated F557, S624 and Y652 residues in sarizotan binding, with weaker contribution from F656. In conclusion, sarizotan inhibits I_Kr/I_hERG, accessing key binding residues on channel gating. This action and consequent ventricular AP prolongation occur at concentrations relevant to those proposed to treat breathing dysrhythmia in RTT. Sarizotan should only be used in RTT patients with careful evaluation of risk factors for QT_c prolongation.

Keywords: KCNH2; Long QT syndrome; Rett Syndrome; Sarizotan; hERG.

Fig. 1

(A) Upper traces show I_hERG in control solution and in the presence of 100 nM and 10 μM sarizotan; voltage protocol shown as lower trace. Inset shows recordings from a separate experiment, showing partial reversibility of 1 μM sarizotan. Cross-over of washout and control I_hERG tails is consistent with ‘foot-in-the door’ inhibition. (B) Concentration-response relation for I_hERG inhibition. Mean fractional block of I_hERG is shown plotted against corresponding drug concentration. Replicate numbers at each concentration shown in brackets. IC₅₀ and n_H values given in “Results” text. (C) Representative traces of I_hERG in control (black) and 100 nM sarizotan (grey) at selected potentials during protocol like that in A, but with test potentials between −40 and +40 mV. (D) Mean fractional block of I_hERG tails by 100 nM sarizotan plotted against test voltage (n = 7). Activation relations for control and sarizotan are shown superimposed (V_0.5 and k values in the main text).

Fig. 2

(A) Representative traces of I_hERG in control and 100 nM elicited by ventricular AP clamp (AP shown superimposed). (B) Representative traces of guinea-pig ventricular deactivating I_Kr tails (elicited on repolarization to −40 mV following 500 ms command to +20 mV) in control, 1 μM sarizotan and 10 μM E-4031. Sarizotan was applied in these experiments for 3–5 min and 1 μM E-4031 for 1 min. (C) Ventricular APs (elicited at 0.1 Hz) in control solution, after 100 nM and 1 μM sarizotan and following washout. Mean AP duration at 25% and 90% repolarization (APD₂₅ and APD₉₀) values are given in the main text.

Fig. 3

(A) Upper traces show representative current traces elicited by the envelope of tails protocol shown in as lower traces. Left hand panel shows data in Control, right hand panel shows data in 100 nM sarizotan. (B) Mean (n = 5) I_hERG blocking time-course at +20 mV derived from application of the envelope of tails protocol. Protocol was applied in control, cells then rested in sarizotan for 3–5 min, and protocol was reapplied to ascertain fractional block values. The inset shows the protocol with a high gain insert to show more clearly the steps of short duration early during the protocol. Fractional block data at different time-points were fitted with an exponential to ascertain the time constant value given in the Results text.

Fig. 4

(A) Protocol used to investigate I_hERG ‘availability’. From a − 80 mV, the membrane potential was stepped to +40 mV for 500 ms; this was then followed by 2 ms repolarization steps to potentials ranging from −140 mV to +50 mV, membrane potential was then stepped back to +40 mV for 100 ms (inset shows portion of protocol encompassing the transition to and from repolarizing steps). The amplitude of current transients elicited by the second step to +40 mV was used to assess I_hERG availability. (B) Representative traces in control. Selected traces are shown for clarity. Numbers indicate corresponding voltage of preceding 2 ms repolarization step. (C) Representative traces at 100 nM sarizotan. (D) Plots against voltage of normalized current transient amplitude in control and in 100 nM sarizotan (n = 7 cells); the inactivation V_0.5 and k values were derived from fits with standard Boltzmann function and are given in the Results text.

Fig. 5

(A) Representative traces showing effects of sarizotan on I_hERG tail current for WT, N588K, S624A and Y652A hERG channels (shown as upper sets of traces) with expanded portion of voltage protocol shown as lower traces. (B) Bar charts showing mean fractional block of I_hERG tails by 1 μM and 10 μM sarizotan for WT, N588K, S624A, F557 L, Y652A and F656A hERG channels. Replicate numbers shown in brackets. * and ** are significance at P < .05 and P < .01 respectively (one-way ANOVA with Tukey post-test). (Ci) Low energy score configuration for R sarizotan (yellow space filling representation) docked into the cryo-EM structure of hERG [36]. Subunit A of the hERG pore is colored grey and the side chains shown are identified in Cii. The location of N588 in chain A is also shown. The selectivity filter is occupied by K⁺ ions in the 1 and 3 positions (purple spheres) and waters in the 2 and 4 positions. The S enantiomer of sarizotan (not shown) was similarly able to bind with a hydrophobic aromatic part of the drug projecting into a hydrophobic pocket. PH: pore helix. (Cii) Close up of the bound configuration for R sarizotan as in panel Ci showing side chains in the binding “pocket” [36] of chain A (grey sticks) and other side chains important for binding to hERG blockers and mentioned in the text. (For interpretation of the references to colour in this figure legend, the reader is referred to the online version of this article.)