Dog left ventricular midmyocardial myocytes for assessment of drug-induced delayed repolarization: short-term variability and proarrhythmic potential

Abstract

Background and purpose: Evaluation of the potential for delayed ventricular repolarization and proarrhythmia by new drugs is essential. We investigated if dog left ventricular midmyocardial myocytes (LVMMs) that can be used as a preclinical model to assess drug effects on action potential duration (APD) and whether in these cells, short-term variability (STV) or triangulation could predict proarrhythmic potential.

Experimental approach: Beagle LVMMs and Purkinje fibres (PFs) were used to record APs. Effects of six reference drugs were assessed on APD at 50% (APD(50)) and 90% (APD(90)) of repolarization, STV(APD), triangulation (ratio APD(90)/APD(50)) and incidence of early afterdepolarizations (EADs) at 1 and 0.5 Hz.

Key results: LVMMs provided stable recordings of AP, which were not affected by four sequential additions of dimethyl sulphoxide. Effects of dofetilide, d-sotalol, cisapride, pinacidil and diltiazem, but not of terfenadine, on APD in LVMMs were found to be comparable with those recorded in PFs. LVMMs, but not PFs, exhibited a proarrhythmic response to I(Kr) blockers. Incidence of EADs was not related to differences in AP prolongation or triangulation, but corresponded to beat-to-beat variability of repolarization, here quantified as STV of APD.

Conclusions and implications: LVMMs provide a suitable preclinical model to assess the effects of new drugs on APD and also yield additional information about putative indicators of proarrhythmia that add value to an integrated QT/TdP risk assessment. Our findings support the concept that increased STV(APD) may predict drug-induced proarrhythmia.

Figure 1

Stability of AP recordings over time in beagle LVMMs. (A) APD₉₀ was measured in standard myocyte Tyrode solution from APs stimulated at a pacing frequency of 1 Hz for 32 min. (B, C and D) Grouped data show that APD, STV(APD) and triangulation (ratio APD₉₀/APD₅₀) were not changed over a period of 32 min [n= 4–10 cells (five dogs); P > 0.05 vs. values from 4 min period]. Effects of four additions of vehicle (0.1% DMSO; to mimic an experiment with an active drug) on STV(APD₉₀) and triangulation as a function of change in APD₉₀ in LVMMs [n= 8–9 cells (two dogs) ] and PFs [n= 8 fibres (eight dogs) ] at pacing frequencies of 1 (E) and 0.5 Hz (F). Note that the effects of vehicle on STV(APD₉₀) and triangulation are plotted on a separate y axis. AP, action potential; APD, action potential duration; APD₅₀ and APD₉₀, duration of the AP at 50% and 90% repolarization; DMSO, dimethylsulphoxide; LVMMs, left ventricular midmyocardial myocytes; PFs, Purkinje fibres; STV, short-term variability.

Figure 3

(A) Fifteen consecutive APs recorded from a PF at a pacing frequency of 0.5 Hz before (red line) and under the influence of 1 µM dofetilide. Note the absence of temporal dispersion of APD. (B) APD₉₀ increases at 1 Hz pacing frequency after exposure to dofetilide (1 µM) in 14 LVMM cells (four dogs). Note the effects on APD₉₀ are largely independent of isolations in LVMMs. (C and D) Fifteen consecutive APs recorded from two LVMMs representing cells with (D) and without (C) EADs. Pacing frequency is 0.5 Hz before (red line) and under the influence of 1 µM dofetilide. Note the presence of temporal dispersion of APD. (E and F) Composite data grouped according to presence of EADs at I_Kr block. STV(APD₉₀) (open bars) and triangulation (open squares) were measured at baseline and during block of I_Kr before first EAD (right) or once steady state was achieved (left) during a pacing frequency of either 1 Hz (E) or 0.5 Hz (F). Note the change in APD₉₀ under the influence of 1 µM dofetilide (black bars). n= 11 cells (three dogs). *P < 0.05 versus vehicle, ^$P < 0.01 versus vehicle (– EADs) and ^#P < 0.05 dofetilide (– EADs). AP, action potential; APD, action potential duration; APD₅₀ and APD₉₀, duration of the AP at 50% and 90% repolarization; EADs, early afterdepolarizations; LVMMs, left ventricular midmyocardial myocytes; STV, short-term variability.

Figure 2

Effect of terfenadine on STV(APD) during the transition to the steady-state decrease in APD. (A) Typical large variations in successive APDs (black traces 198 to 227) were seen in a left ventricular midmyocardial myocyte (LVMM), after exposure to terfenadine (1 µM), at pacing frequency of 1 Hz. This is an example of temporal dispersion of repolarization within a single cell during the transition to the steady-state decrease in APD. (B) Shows the time course of the same cell as in Figure 2A during the perfusion of terfenadine (1 µM) (dashed rectangles illustrate the periods of transition to steady state and steady state; red filled and unfilled triangles illustrate the changes from 1 to 0.5 Hz and back to 1 Hz, respectively; coloured filled circles illustrate the time at which coloured AP traces of Figure 2A were selected). (C) Shows a representative example of a Poincaré plot of APD₉₀ from a LVMM (same cell as in A) under vehicle (0.1% DMSO) and in the presence of 1 µM terfenadine (during transition to steady state and at steady state). Note the large complex polygons during the transition to the steady-state decrease in APD. (D) Increasing STV(APD₉₀) during the transition to the steady-state decrease in APD₉₀[n= 6 cells (two dogs) ]. *P < 0.05 versus vehicle and ^#P < 0.05 versus steady state. APD, action potential duration; DMSO, dimethylsulphoxide; LVMM, left ventricular midmyocardial myocyte; STV, short-term variability.