Local Control Model of a Human Ventricular Myocyte: An Exploration of Frequency-Dependent Changes and Calcium Sparks

Figure A1. Membrane potential (Vm), myoplasmic Ca2+ ([Ca2+]myo), and SR Ca2+ ([Ca2+]nsr). Several 20 s simulations were conducted in most experiments presented in this study. [Ca2+]nsr recovery achieved steady state around 3–5 s.

Figure A2. IKr and IKs current densities. Similar behavior was observed with IKr in O’Hara–Rudy model from experimental data measured in isolated undiseased human ventricular cells at 37 °C. On the other hand, several isolated myocyte patch clamp experiments underestimate IKs due to enzymatic degradation (i.e., disaggregation of myocytes under electrophysiological recording conditions) [127]. Additionally, the importance of IKs is highlighted by multiple studies, such that clinical long Q-T syndrome type 1 is caused by IKs loss of function [128], and IKs assists in AP repolarization in human ventricle experiments even when the IKr current is blocked [129]. Moreover, in O’Hara–Rudy model, selective IKs block prolonged APD, but only very minimally under basal conditions and no β-adrenergic stimulation. Thus, for further use of the model presented here, IKs was patterned with the Tuscher–Panfilov model for better representation.

Figure A3. IKr and IKs currents. A 20 s simulation was conducted in most experiments presented in this study.

Figure A4. IK1. At various pacing rates, there were no differences observed in IK1 amplitude.

Figure A5. ILCC and INCX amplitude changes. At faster heart rates, ILCC amplitude decreases while INCX increases in amplitude as described in Figure 8J,K, respectively. Moreover, the depolarizing current (indicated by the lowest point at the tail current) is becoming more negative (i.e., the Ca2+ extrusion is greater, which contributes to APD shortening).

Figure A6. Membrane potential (Vm) at 7Hz pacing. Top: 20 s simulation at 7 Hz shows irregular repolarizations. Bottom: 5 s snapshot reveals the presence of alternans. As described, beat-to-beat alternations in intracellular calcium concentration and electrophysiological behavior were observed in human heart experiments at very fast pacing rates [16].

Figure A7. Cytoplasmic and SR Ca2+ loading. Steady state and increased loading were observed in a 20 s simulation. Top: Rate-dependence of calcium transient ([Ca2+]i) shows gradual increase in amplitude (~1–1.5 µM) as described in newly reformulated O’Hara–Rudy model by Tomek et al. [14]. Bottom: SR Ca2+ loading increased from 1000 to 1100 μM in Ca2+ concentrations. Generally, the increase in cytoplasmic Ca2+ is expected to be accompanied by an increase in cytoplasmic Ca2+ content.

Figure A8. Varied LCC Ca2+-dependent inactivation rate. Top: 1 Hz simulations. (A) Because of the reduced LCC inactivation rate, LCC duration is generally prolonged which contributes to longer AP duration. (B) In contrast, increased LCC inactivation rate leads to shorter APD. Bottom: 4 Hz simulations. Similar prolongation (C) and shortening (D) are observed. However, it is important to note that the C4 state (inactivated state) shows a higher amplitude which is reflected in the decreased ILCC peak in Figure 9C.

Figure A9. Effects of varied LCC Ca2+-dependent inactivation rate to RyR open fraction. Top: 1 Hz simulations. (A) Because of the reduced LCC inactivation rate, LCC duration is generally prolonged, and more calcium enters during CICR. As previously discussed, RyR clusters can spontaneously close at increased Ca2+ concentrations, hence its lower open probability under the condition of reduced LCC inactivation. (B) In contrast, higher open probability is observed in an increased inactivation rate. Bottom: 4 Hz simulations. (C,D) Due to the frequent stimulus, the amount of calcium is increased. Therefore, in a similar fashion, RyR open fraction is reduced and goes further into the adapted state as discussed in Figure 5D.

Figure A10. APD90 comparisons of multiple species and Ito block at low diastolic intervals. Recent experimental data from Lovas et al. using human, dog, rabbit, and guinea pig ventricle APDs along with simulation results. The increasing APDs show dynamic restitution curves in response to S1S2 pacing protocol at low diastolic intervals.

Figure A11. l-type calcium channel activation rate during S1S2 protocol. Top: shorter diastolic interval (0.02 s). (A) reduced Ca2+-dependent activation of LCC shows shorter duration of LCC (<0.2 s), while (B) a small increment of 10% to activation rate evidently prolongs LCC activity (>0.2 s). Bottom: longer diastolic interval (0.42 s). Similar shortening (C) and prolongation (D) are observed.

Figure 1. Schematic diagram of a ventricular cell containing 1 transverse tubule (T-tubule) branch with a portion of the 20,000 stochastically gating CRUs. Abbreviations: LCC–L-type calcium channel; RyR2–ryanodine receptor type 2; INCX–sodium-calcium exchange; INaK–sodium-potassium ATPase; IPMCA–plasmalemmal calcium ATPase; CRU–calcium release units; SERCA–sarcoplasmic and endoplasmic reticulum calcium ATPase; SR–sarcoplasmic reticulum; InsCa–nonspecific calcium-activated current; INa–sodium current; IK1–inward rectifier potassium current; IKr–rapid delayed rectifying potassium current; IKs–slow delayed rectifying potassium current; Ito–transient outward potassium current; IbCa–background calcium current; IbNa–background sodium current; IbK–background potassium current. Schematic diagram was created through BioRender.com.