Computational analysis of arrhythmogenesis in KCNH2 T618I mutation-associated short QT syndrome and the pharmacological effects of quinidine and sotalol

Fig. 1. Simulation results at ion channel and cellular levels. A The fitted current–voltage (I–V) curves in wild-type (black) and T618I (red) conditions using IKr Markov chain models. Experimental data from Sun et al.11 (their Fig. 4B). Error bars represent the standard error of measurement (SEM). B The IKr Markov chain model. C Steady-state (1.25 Hz) action potentials (Ci) and the corresponding IKr (Cii) for EPI cells. Solid and dashed lines represent wild-type and T618I conditions, respectively. D Steady-state action potentials and the corresponding IKr for MIDDLE cells. E Steady-state action potentials and the corresponding IKr for ENDO cells. F The comparison of the peak IKr in wild-type and T618I conditions. G The comparison of transmural APD differences. H The comparison of transmural ERP differences.

Fig. 2. Simulation results of APD and CV restitution curves and rate adaption curve of QT interval. A APD restitution curves obtained from the S1–S2 protocol (Ai) and the dynamic protocol (Aii), where APDs were plotted against PCL and DI, respectively. B CV restitution curves obtained from the S1–S2 protocol (Bi) and the dynamic protocol (Bii). C Rate adaption curve of QT interval.

Fig. 3. Simulation results of the unidirectional conduction block using the transmural 1D strand. A The evoked unidirectional conduction block in the wild-type condition. B The unidirectional conduction block in the T618I mutation condition. C Distributions of VWs in the two conditions. D The comparison of the average VW width in the two conditions.

Fig. 4. Simulation results of the critical length in the idealized 2D sheet. A The schematic diagram of the idealized 2D sheet. B The critical length of S2 for initiating spiral waves in the wild-type tissue. C The critical length in the T618I mutant tissue. D The comparison of critical length in the two situations.

Fig. 5. Simulation results of reentry arrhythmias in the realistic 2D ventricular slice. Spiral waves were evoked by the S1–S2 protocol. The stimulating locations of S2 were indicated by the black arrows. A Spiral waves in the wild-type condition with a lifespan of only 480 ms. Snapshots were recorded at 320, 390, 490, 800 ms. B The spiral wave in the T618I mutation condition. Snapshots were recorded at 260, 410, 1090, 1760, 1940, 2330, 2630, 2930 ms. Spiral waves took the form of functional reentry initially before it transformed into anatomical reentry arrhythmia at about 2000 ms and persisted since then. C The comparison of lifespan of spiral waves in two situations.

Fig. 6. Simulation results of reentry arrhythmias in the realistic 3D bi-ventricle model. The S2 stimulus was applied to the upper part of the LV epicardium to initiate scroll waves. A The scroll wave in the wild-type condition. The lifetime was only 410 ms, and the recorded onset and disappearance time of the scroll wave were separately 270 and 680 ms. B The scroll wave in the T618I mutation condition, which persisted throughout the whole simulation period of 10,000 ms.

Fig. 7. Actions of quinidine at ion channel and cellular levels. A The fitted current–voltage (I–V) curves in wild-type (black), T618I (red), and quinidine (blue) conditions using IKr Markov chain models. Experimental data from Sun et al.11 (their Fig. 7B). Error bars represent the standard error of measurement (SEM). B The IKr Markov chain model with two additional drug-related states. C–E Steady-state (1.25 Hz) action potentials and the corresponding IKr for EPI, MID, and ENDO cells. F The comparison of the peak IKr. G The comparison of transmural APD differences. H The comparison of transmural ERP differences.

Fig. 8. Proportion of states under WT, T618I, and QUIN conditions. A Action potential traces. B The corresponding state proportions under different conditions. The Markov model demonstrated that the enhanced IKr in T618I was due to the loss of inactivation, as evidenced by the significantly dropped proportion of the Inactivation (“I”) state in panels Bi and Bii. Simulation also showed that quinidine exerted its effects by directly deprecating the Open (“O”) state rather than restoring the “I” state (see panels Bii and Biii).

Fig. 9. Simulation results of APD and CV restitution curves and the rate adaption curves of QT interval under wild-type, T618I mutation, and quinidine conditions. A APD restitution curves obtained from the S1–S2 protocol (Ai) and the dynamic protocol (Aii), where APDs were plotted against PCL and DI, respectively. B CV restitution curves obtained from the S1–S2 protocol (Bi) and the dynamic protocol (Bii). C The rate adaption curves of QT interval.

Fig. 10. Simulation results of the reentry arrhythmias in the condition of quinidine. A Spiral waves on the 2D ventricular slice. Spiral waves were evoked by the S1–S2 protocol, and the black arrow indicates the stimulating location of S2. Quinidine prevented the persistent arrhythmias, which was evidenced by the disappearance of spiral waves at about 1000 ms. B Scroll waves in the 3D ventricle. In line with the 2D simulation results, the evoked scroll waves were not sustainable and disappeared at about 900 ms.