Comparison of HERG channel blocking effects of various β‐blockers – implication for clinical strategy

Abstract

1 b-Blockers are widely used in the treatment of cardiovascular diseases. However, their effects on HERG channels at comparable conditions remain to be defined. We investigated the direct acute effects of b-blockers on HERG current and the molecular basis of drug binding to HERG channels with mutations of putative common binding site (Y652A and F656C). 2 b-Blockers were selected based on the receptor subtype. Wild-type, Y652A and F656C mutants of HERG channel were stably expressed in HEK293 cells, and the current was recorded by using wholecell patch-clamp technique (231C). 3 Carvedilol (nonselective), propranolol (nonselective) and ICI 118551 (b 2 -selective) inhibited HERG current in a concentration-dependent manner (IC 50 0.51, 3.9 and 9.2 mM, respectively). The IC 50 value for carvedilol was a clinically relevant concentration. High metoprolol (b 1 -selective) concentrations were required for blockade (IC 50 145 mM), and atenolol (b 1 -selective) did not inhibit the HERG current. 4 Inhibition of HERG current by carvedilol, propranolol and ICI 118551 was partially but significantly attenuated in Y652A and F656C mutant channels. Affinities of metoprolol to Y652A and F656C mutant channels were not different compared with the wild-type. 5 HERG current block by all b-blockers was not frequency-dependent. 6 Drug affinities to HERG channels were different in b-blockers. Our results provide additional strategies for clinical usage of b-blockers. Atenolol and metoprolol may be preferable for patients with type 1 and 2 long QT syndrome. Carvedilol has a class III antiarrhythmic effect, which may provide the rationale for a favourable clinical outcome compared with other b-blockers as suggested in the recent COMET (Carvedilol Or Metoprolol European Trial) substudy.

Prediction of drug-induced QT interval prolongation in telemetered common marmosets

“…The reason why dl-sotalol shows such weak hERG blockade even though it prolongs QT interval via the I K blockade (Antonaccio and Gomoll, 1993) is unclear, but it is considered that the QT-prolonging effect of dl-sotalol might be caused, at least partially, through a mechanism other than direct hERG inhibition. Contrary to the case of dl-sotalol, dl-propranolol showed no QTc interval prolongation even though it has apparently stronger potential than dl-sotalol to inhibit hERG (IC 50 = 3.9 μM) (Kawakami et al, 2006). It was reported that dl-propranolol inhibits not only I Kr /hERG but also I Na (Varro et al, 1990), leading to a shortening of APD in isolated guinea pig papillary muscles at 3 μM and above (Hayashi et al, 2005) or no effect on QTc interval, as in the present study, in conscious dogs up to 30 mg/kg (Toyoshima et al, 2005).…”

“…Nifedipine, which has no QT-prolonging risk in clinic, was reported to cause tachycardia and false QTc interval prolongation in a canine telemetry study using Fridericia's formula (Toyoshima et al, 2005). Verapamil, anoth- Hoppu et al 1991 andSaviuc et al 1993, c Kimura et al 1996, d Zhou et al 1999, e Kirsch et al 2004, f Kawakami et al 2006, g Zhang et al 1999 er calcium channel blocker, also caused tachycardia and false QT-prolonging results in the same canine study (Toyoshima et al, 2005) and in a monkey study using Bazett's formula (Ando et al, 2005). The explanation given for the QTc overestimations was that the heart rate was out of a valid correction range due to hyper-tachycardia (130-170% of baseline value) (Ando et al, 2005;Toyoshima et al, 2005).…”

A Composite Model for hERG Blockade

“…Metoprolol was excluded because it has been shown that, in contrast to all other compounds tested, it does not lose binding affinity when Phe 656 or Tyr 652 are mutated. [35] It must therefore bind in a different mode than the other compounds. For compounds with multiple IC 50 values, the mean IC 50 value was used if pIC 50 max ÀpIC 50 min < 1; otherwise the compounds were excluded.…”

“…[29][30][31][32][33][34] Phe 656 and Tyr 652 have been identified as very important binding partners for almost all compounds tested [29] apart from metoprolol. [35] Additionally, depending on the compound, one or more of the amino acids Ser 624, Thr 623, Val 625, Gly 648, and Val 650 are necessary for binding. In all homology models and docking studies published, [22,23,[36][37][38] the side chains of Phe 656, Tyr 652, Ser 624, Thr 623, and Val 625 point towards the inside of the pore.…”

Cardiomyocytes Derived from Human Induced Pluripotent Stem Cells: An In-Vitro Model to Predict Cardiac Effects of Drugs

“…PPB are either from published data (plain text) or measured in our laboratories (bold italics). [22]; c: [23]; d: [24,25]; e: [26]; f: [27]; g: [28]; h: [19]; i: [29]; j: [30]; k: [31]; l: [32]; m: [33]; n: [34]; o: [35]; p: [36]; q: [37]; r: [38]; s: [39]; t: [40]; u: [41]; v: [42]; w: [43]; x: [44]; y: [45]; z: [46]; aa: [47]; bb: [48]; cc: [49]; dd: [50]; ee: [51]; ff: [52]; gg: [53]; hh: [54]; ii: [55]; jj: [56]; kk: [57]; ll: [58]; mm: [59]; nn: [60]; oo: [61]; pp: [62]; qq: [63]; rr: [64]; ss: [65]; tt: [66]; uu: [67]; vv: [68]; ww: …”