In silico investigation of the short QT syndrome, using human ventricle models incorporating electromechanical coupling

Abstract

Introduction: Genetic forms of the Short QT Syndrome (SQTS) arise due to cardiac ion channel mutations leading to accelerated ventricular repolarization, arrhythmias and sudden cardiac death. Results from experimental and simulation studies suggest that changes to refractoriness and tissue vulnerability produce a substrate favorable to re-entry. Potential electromechanical consequences of the SQTS are less well-understood. The aim of this study was to utilize electromechanically coupled human ventricle models to explore electromechanical consequences of the SQTS.

Methods and results: The Rice et al. mechanical model was coupled to the ten Tusscher et al. ventricular cell model. Previously validated K(+) channel formulations for SQT variants 1 and 3 were incorporated. Functional effects of the SQTS mutations on [Ca(2+)] i transients, sarcomere length shortening and contractile force at the single cell level were evaluated with and without the consideration of stretch-activated channel current (I sac). Without I sac, at a stimulation frequency of 1Hz, the SQTS mutations produced dramatic reductions in the amplitude of [Ca(2+)] i transients, sarcomere length shortening and contractile force. When I sac was incorporated, there was a considerable attenuation of the effects of SQTS-associated action potential shortening on Ca(2+) transients, sarcomere shortening and contractile force. Single cell models were then incorporated into 3D human ventricular tissue models. The timing of maximum deformation was delayed in the SQTS setting compared to control.

Conclusion: The incorporation of I sac appears to be an important consideration in modeling functional effects of SQT 1 and 3 mutations on cardiac electro-mechanical coupling. Whilst there is little evidence of profoundly impaired cardiac contractile function in SQTS patients, our 3D simulations correlate qualitatively with reported evidence for dissociation between ventricular repolarization and the end of mechanical systole.

Keywords: 3D model; human ventricles; mechanical contraction; short QT syndrome; stretch-activated channel.

Figure 1

Simulation of ventricular action potentials and the time course of I_Kr and I_K1. (Ai–Ci) Steady state (at a stimulation frequency of 1 Hz) action potentials for EPI (Ai), MCELL (Bi), and ENDO (Ci) cells under wild-type (WT; black), SQT1 (blue) and SQT3 (green) conditions. (Aii–Cii) Corresponding I_Kr current profiles for EPI (Aii), MCELL (Bii), and ENDO (Cii) cells under the WT (black) and SQT1 (blue) conditions. (Aiii–Ciii) Corresponding I_K1 current profiles for EPI (Aiii), MCELL (Biii), and ENDO (Ciii) cells under the WT (black) and SQT1 (blue) conditions.

Figure 2

Force-frequency relationship. Plot of steady state normalized active force vs. heart rate using the EPI cell model. Black continuous line represents the WT electromechanics model while symbols represent experimental data from non-failing control preprarations of human myocardium. Experimental data from Mulieri et al. (1992).

Figure 3

Single cell effects of the SQT1 and SQT3 mutations (without I_sac). (Ai–Ci) WT (black), SQT1 (blue) and SQT3 (green) action potentials in the EPI (Ai), MCELL (Bi), and ENDO (Ci) cell models. (Aii–Cii) WT (black), SQT1 (blue) and SQT3 (green) intracellular calcium concentration and Ca²⁺ transients in the EPI (Aii), MCELL (Bii), and ENDO (Cii) cell models. (Aiii–Ciii) WT (black), SQT1 (blue) and SQT3 (green) sarcomere length (SL) in the EPI (Aiii), MCELL (Biii), and ENDO (Ciii) cell models. (Aiv–Civ) WT (black), SQT1 (blue) and SQT3 (green) active force in the EPI (Aiv), MCELL (Biv), and ENDO (Civ) cell models. Values are normalized to WT maximum active force for each cell type.

Figure 4

Simulated AP clamp using the WT electromechanics model without I_sac. (A) The normal (black) and shortened (red) AP waveforms applied as voltage clamp commands to the WT electromechanics model. (B) I_CaL elicited by the two AP waveforms in (A). (C) Normalized [Ca²⁺]_i elicited by the two AP waveforms in (A). Values are normalized to maximum [Ca²⁺]_i for the “normal” AP waveform. (D) Normalized SL shortening elicited by the two AP waveforms in (A). Values are normalized to maximum SL shortening for the normal AP waveform. (E) Normalized contractile force elicited by the two AP waveforms in (A). Values are normalized to maximum active force for the normal AP waveform.

Figure 5

Changes in steady state SR content induced by conditioning trains of action potentials of differing duration. (A) Protocol used to determine steady state SR content, comprised of train of 100 normal (black) and shortened (red) AP clamp commands followed by a 300 ms square command voltage pulse to +10 mV. (B) I_CaL elicited by the +10 mV voltage command. Peak I_CaL is equal with the two AP waveforms in (A). (C) Normalized [Ca²⁺]_i elicited by the by the +10 mV voltage command. Values are normalized to maximum [Ca²⁺]_i for the “normal” AP waveform. (D) Normalized active force elicited by the +10 mV voltage command. Values are normalized to maximum active force for the normal AP waveform. (E) Normalized maximum SR Ca²⁺ content prior to the application of +10 mV voltage command in (A). Values are normalized to maximum SR Ca²⁺ for the normal AP waveform.

Figure 6

Effects of I_sac on the WT, SQT1, and SQT3 electromechanics model. (Ai–Ci) WT (black), SQT1 (blue) and SQT3 (green) action potentials in the EPI (Ai), MCELL (Bi), and ENDO (Ci) cell models. (Aii–Cii) WT (black), SQT1 (blue) and SQT3 (green) intracellular calcium concentration and Ca²⁺ transients in the EPI (Aii), MCELL (Bii), and ENDO (Cii) cell models. (Aiii–Ciii) WT (black), SQT1 (blue) and SQT3 (green) sarcomere length (SL) in the EPI (Aiii), MCELL (Biii), and ENDO (Ciii) cell models. (Aiv–Civ) WT (black), SQT1 (blue) and SQT3 (green). SAC in these simulations were permeable to Na⁺, K⁺, and Ca²⁺ in the ratio 1:1:0.

Figure 7

Effects of I_sac on the WT, SQT1, and SQT3 electromechanics model. (Ai–Ci) WT (black), SQT1 (blue) and SQT3 (green) action potentials in the EPI (Ai), MCELL (Bi), and ENDO (Ci) cell models. (Aii–Cii) WT (black), SQT1 (blue) and SQT3 (green) intracellular calcium concentration and Ca²⁺ transients in the EPI (Aii), MCELL (Bii), and ENDO (Cii) cell models. (Aiii–Ciii) WT (black), SQT1 (blue) and SQT3 (green) sarcomere length (SL) in the EPI (Aiii), MCELL (Biii), and ENDO (Ciii) cell models. (Aiv–Civ) WT (black), SQT1 (blue) and SQT3 (green). SAC in these simulations were permeable to Na⁺, K⁺, and Ca²⁺ in the ratio 1:1:1.

Figure 8

Summary of effects of SQTS mutations on APD₉₀, [Ca²⁺]_i, SL shortening and active force under different simulation conditions. (Ai–Ci) Changes in APD₉₀under the WT (black), SQT1 (gray) and SQT3 (white) in the EPI, MCELL and ENDO cell types without I_sac (Ai), with I_sac at a permeability ratio of 1:1:0 (Bi) and with I_sac at a permeability ratio of 1:1:1 (Ci). (Aii–Cii) Percentage changes in [Ca²⁺]_i under the WT (black), SQT1 (gray) and SQT3 (white) in the EPI, MCELL and ENDO cell types without I_sac (Aii), with I_sac at a permeability ratio of 1:1:0 (Bii) and with I_sac at a permeability ratio of 1:1:1 (Cii). (Aiii–Ciii) Percentage sarcomere length shortening under the WT (black), SQT1 (gray) and SQT3 (white) in the EPI, MCELL, and ENDO cell types without I_sac (Aiii), with I_sac at a permeability ratio of 1:1:0 (Biii) and with I_sac at a permeability ratio of 1:1:1 (Ciii). Values are relative to WT without I_sac (Aiii). (Aiv–Civ) Normalized active force under the WT (black), SQT1 (gray) and SQT3 (white) in the EPI, MCELL and ENDO cell types without I_sac (Aiv), with I_sac at a permeability ratio of 1:1:0 (Biv) and with I_sac at a permeability ratio of 1:1:1 (Civ).

Figure 9

Reverse mode operation of NCX with the incorporation of SAC. (Ai–Ci) Action potentials of WT (Ai), SQT1 (Bi), and SQT3 (Ci) without stretch (black) and with stretch (red). (Aii–Cii) [Na⁺]_i for WT (Aii), SQT1 (Bii), and SQT3 (Cii) without stretch (black) and with stretch (red). (Aiii–Ciii) Action potentials of WT (Aiii), SQT1 (Biii), and SQT3 (Ciii) without stretch (black) and with stretch (red). As the SAC is permeable to Na⁺, it is of higher amplitude under stretch conditions. (Aiv–Civ) [Ca²⁺]_i for WT (Aiv), SQT1 (Biv), and SQT3 (Civ) without stretch (black) and with stretch (red). (Av–Cv) SR Ca²⁺ release under WT (Av), SQT1 (Bv) and SQT3 (Cv) without stretch (black) and with stretch (red). The SR is refilled to a greater level prior to AP initiation under stretch conditions. (Avi–Cvi) I_NaCa of WT (Avi), SQT1 (Bvi) and SQT3 (Cvi) without stretch (black) and with stretch (red).

Figure 10

Electromechanical coupling in 3D ventricle model under the SQT1 and SQT3 mutations with I_sac. (A) Resting position of the ventricles prior to electrical stimulation. (B) Snapshot of maximum deformation occurring at 230 ms under the WT condition just at the onset of repolarization (black AP). (C) Snapshot of maximum deformation occurring at 200 ms under the SQT1 condition. Repolarization is already significantly advanced (blue AP). (D) Snapshot of maximum deformation occurring at 210 ms under the SQT3 condition. Repolarization is already in progress (green AP). Vertical lines show a comparison of the degree of contraction of the ventricles between the different conditions. Color bar represents the membrane potentials of cells in the ventricles ranging from –86 to 42 mV. The APs shown in the inset are from a left ventricular cell under the WT, SQT1 and SQT3 conditions. Arrows indicate the snapshot time shown in the main figure for each condition corresponding to the repolarization time at which maximal deformation occurs.