Pro-arrhythmogenic effects of the S140G KCNQ1 mutation in human atrial fibrillation - insights from modelling

Abstract

Functional analysis has shown that the missense gain-in-function KCNQ1 S140G mutation associated with familial atrial fibrillation produces an increase of the slow delayed rectifier potassium current (I(Ks)). Through computer modelling, this study investigated mechanisms by which the KCNQ1 S140G mutation promotes and perpetuates atrial fibrillation. In simulations, Courtemanche et al.'s model of human atrial cell action potentials (APs) was modified to incorporate experimental data on changes of I(Ks) induced by the KCNQ1 S140G mutation. The cell models for wild type (WT) and mutant type (MT) I(Ks) were incorporated into homogeneous multicellular 2D and 3D tissue models. Effects of the mutation were quantified on AP profile, AP duration (APD) restitution, effective refractory period (ERP) restitution, and conduction velocity (CV) restitution.Temporal and spatial vulnerabilities of atrial tissue to genesis of re-entry were computed. Dynamic behaviours of re-entrant excitation waves (lifespan (LS), tip meandering patterns and dominant frequency) in 2D and 3D models were characterised. It was shown that the KCNQ1 S140G mutation abbreviated atrial APD and ERP and flattened APD and ERP restitution curves. It reduced atrial CV at low excitation rates, but increased it at high excitation rates that facilitated the conduction of high rate atrial excitation waves. Although it increased slightly tissue temporal vulnerability for initiating re-entry, it reduced markedly the minimal substrate size necessary for sustaining re-entry (increasing the tissue spatial vulnerability). In the 2D and 3D models, the mutation also stabilized and accelerated re-entrant excitation waves, leading to rapid and sustained re-entry. In the 3D model, scroll waves under the mutation condition MT conditions also degenerated into persistent and erratic wavelets, leading to fibrillation. In conclusion, increased I(Ks) due to the KCNQ1 S140G mutation increases atrial susceptibility to arrhythmia due to increased tissue vulnerability, shortened ERP and altered atrial conduction velocity, which, in combination, facilitate initiation and maintenance of re-entrant excitation waves.

Figure 1. Experimental and simulated I_Ks under WT and MT conditions, where experimental data are from Chen et al. (2003)

Aa, voltage-clamp protocol. Ab, simulated I_Ks traces under WT conditions. Ac, simulated I_Ks traces under MT conditions. B, experimental I–V relationships for I_Ks for WT (open circles) and S140G mutant I_Ks (filled circles) superimposed upon simulated I–V relationships for WT and MT cases (continuous lines).

Figure 2. Simulated AP profiles and current traces under WT (continuous line), and MT conditions where simulated data for ϕ= 0.1 (dashed line), ϕ= 0.25 (dotted line) and ϕ= 1 (grey line) are shown in panels A–F

A, AP profiles. B, I_Ks profiles during APs. C, I_Kr profiles during APs. D, I_to profiles during APs. E, I_K1 profiles during APs. F, I_CaL profiles during APs. G, changes in APD as the modelling parameter ϕ is increased from 0 (WT) to 1 (MT) condition.

Figure 3. APD and ERP restitution

A, APDr curves under WT and MT conditions. B, maximum slopes of the APDr curves. C, ERP restitution curves under WT and MT conditions.

Figure 4. Simulated CVr and VW under WT and MT conditions

A, CVr curves under WT, ϕ= 0.1, ϕ= 0.25 and ϕ= 1 conditions. B, temporal VW under WT, ϕ= 0.1, ϕ= 0.25 and ϕ= 1 conditions.

Figure 5. Critical length of the minimal substrate size for re-entry (MS) in 2D sheets

Aa and b, Ba and b, illustration of MS required to induce a pair of re-entrant circuits in homogeneous 2D sheets. C, critical length under WT, ϕ= 0.1, ϕ= 0.25 and ϕ= 1 conditions. The critical length under mutant conditions was dramatically shorter than under Control conditions.

Figure 6. Simulation of spiral waves in 2D model of human atrium

Top panels show results from 2D re-entrant wave simulation under WT condition; the second row of panels show results from ϕ= 0.1 conditions; the third row of panels show results from ϕ= 0.25 conditions; and the bottom row show results from ϕ= 1. Frames from the 2D simulation at time t= 1 s (column I), time t= 1.5 s (column II) and time t= 2 s (column III) are shown. Column IV shows the re-entrant wave tip trajectories. Column V shows time traces of localized electrical excitations (top) and dominant frequency (bottom).

Figure 7. Simulation of scroll waves in 3D virtual human atrium

Top panels show representative frames from the simulation under WT condition and bottom panels show frames from the simulation under MT (ϕ= 0.1). Column I shows frames at time t= 1 s, Column II at time t= 2 s, Column III at time t= 3 s and Column IV at time t= 4 s. Column V shows time traces of localized electrical excitations. Column VI shows dominant frequencies.