Optical Investigation of Action Potential and Calcium Handling Maturation of hiPSC-Cardiomyocytes on Biomimetic Substrates

Abstract

Cardiomyocytes from human induced pluripotent stem cells (hiPSC-CMs) are the most promising human source with preserved genetic background of healthy individuals or patients. This study aimed to establish a systematic procedure for exploring development of hiPSC-CM functional output to predict genetic cardiomyopathy outcomes and identify molecular targets for therapy. Biomimetic substrates with microtopography and physiological stiffness can overcome the immaturity of hiPSC-CM function. We have developed a custom-made apparatus for simultaneous optical measurements of hiPSC-CM action potential and calcium transients to correlate these parameters at specific time points (day 60, 75 and 90 post differentiation) and under inotropic interventions. In later-stages, single hiPSC-CMs revealed prolonged action potential duration, increased calcium transient amplitude and shorter duration that closely resembled those of human adult cardiomyocytes from fresh ventricular tissue of patients. Thus, the major contribution of sarcoplasmic reticulum and positive inotropic response to β-adrenergic stimulation are time-dependent events underlying excitation contraction coupling (ECC) maturation of hiPSC-CM; biomimetic substrates can promote calcium-handling regulation towards adult-like kinetics. Simultaneous optical recordings of long-term cultured hiPSC-CMs on biomimetic substrates favor high-throughput electrophysiological analysis aimed at testing (mechanistic hypothesis on) disease progression and pharmacological interventions in patient-derived hiPSC-CMs.

Keywords: action potential; calcium handling; cardiomyocytes; fluorescence; human induced pluripotent stem cells; hydrogels; long-term culture; maturation.

Figure 1

Experimental procedure. Human iPSCs are differentiated into cardiomyocytes (hiPSC-CMs) with monolayer-directed differentiation protocol. (A) At day 20 post differentiation, single hiPSC-CMs are seeded onto hydrogel-based micropatterned surface and cultured until experimental day 60, 75 and 90. (B) Left: fabrication of micropatterned hydrogel with PDMS mold and PEG-DA hydrogel synthesis by soft lithography (scale bars equal to 10 and 60 μm respectively). hiPSC-CM preferential spreading along thepattern direction (indicated by yellow line). (C) Simultaneous recording of action potential and calcium transients using Fluovolt (Ex/em 522/535 nm) and Cal630 (Ex/Em 608/626 nm), respectively.

Figure 2

Dual recording of action potential and calcium transient in later-stages hiPSC-CMs. Single hiPSC-CMs matured on hydrogel-based micropatterned surfaces were subjected to simultaneous optical measurements of action potentials and calcium transients under electrical pacing (1 and 2 Hz) at 37 °C at and external [Ca²⁺] = 1.8 mM. (A) Superimposed action potential (AP) traces of day 75 (N = 2; n = 186) vs. day 90 (N = 2; n = 119) recorded by FluoVolt. AP profile of hiPSC-CMs was recorded both at 1 and 2 Hz to evaluate action potential duration (APD50, ms) and the response to frequency changes at both day 75 and 90. (B) Superimposed normalized traces of calcium transients recorded by Cal630 at day 60 (N = 3; n = 336), 75 (N = 5; n = 251) and 90 (N = 3; n = 165): average calcium transient (CaT) rise (time to peak TTP, ms) and CaT decay (difference of 50% of CaT decay and TTP, RT50, ms) are reported, during pacing at 1 and 2 Hz (C) Representative CaT profiles at day 60 and 90 and average CaT amplitude (in arbitrary fluorescence units, A.U.) at day 60,75 and 90. (D) Representative simultaneous recordings of action potential and intracellular calcium transient from adult ventricular cardiomyocytes, elicited with short current pulses in current-clamp mode at 1 Hz. Average time to peak (Ca TTP) and time from peak to 50% decay (Ca RT50) of Ca transients, and time from stimulus to 50% repolarization (APD50) of action potentials at 1 Hz. Means ± SEM from 27 myocytes (nine control patients). Data are reported as means ± SEM; one-way analysis of variance (ANOVA) with a Tukey post-hoc test with statistical significance set at * p < 0.05 and ** p < 0.01; NS not significant. Supporting information given in Table S2. N = number of differentiations; n = cells.

Figure 3

Correlative analysis of action potential and calcium transient parameters. Pearson correlation coefficient (r²) estimated by linear regression (red line) to correlate APD50 (ms) against CaT amplitude (NS, not significant) and CaT duration (RT50, ms, p < 0.05) of (A) day 90 hiPSC-CMs and (B) (human) hAdult CMs from donor ventricular tissue (p < 0.05).

Figure 4

Sarcoplasmic reticulum contribution during hiPSC-CM maturation. Sarcoplasmic reticulum (SR) contribution in calcium handling maturation was tested by a post rest potentiation protocol and caffeine-induced CaTs elicited in hiPSC-CMs at multiple maturation time-points. (A) The post-rest potentiation of CaT amplitude was estimated after a resting pause of 5 s, inserted in a regular train of stimulation at 2 Hz. The potentiation is expressed as the % increase of CaT amplitude at the first post-rest beat from that of the last calcium transient before the pause (%). Post rest potentiation is estimated at day 60 and day 90. (B) Caffeine-induced CaTs (quick exposure to 10 μM caffeine) after a series of 2 Hz paced CaTs. Average of caffeine transient amplitude was normalized by the amplitude of steady-state calcium transients at 2 Hz prior to caffeine exposure (N = 2; n = 83). Caffeine transient CaT amplitude (CaT_{A CAFF}/CaT_{A 2Hz} ratio) and decay (τ, s⁻¹) of hiPSC-CMs were calculated and compared with caffeine-CaT recorded in hAdult-CMs (N = 5; n = 14). (C) Simultaneously recorded APs and CaTs during the pause protocol. (D) APs and CaTs from the same cells were compared to show Pearson’s correlation (r²) between post rest AP duration (APD50, ms) and post rest CaT decay (RT50, ms, p < 0.05). (E) Variations of post rest APD50 and (F) post rest RT50 were measured both at day 75 (AP: N = 2, n = 119; CaT: N = 5, n = 251) and day 90 (AP: N = 2, n = 119; CaT: N = 3, n = 165). One-way analysis of variance (ANOVA) with a Tukey post-hoc test with statistical significance set at * p < 0.05 and ** p < 0.01; NS not significant. Supporting information is reported in Table S2. N = number of differentiations or patients; n = cells.

Figure 5

Positive β-adrenergic response of later stage hiPSC-CMs. Day 90 hiPSC-CMs were exposed to isoproterenol (ISO, 1 μM) and forskolin (FSK, 1 μM). (A) Representative traces of hiPSC-CM AP recorded before and under ISO stimulation. APD50 reduction under ISO was 12 ± 4%, p > 0.05. (B) Representative traces of ISO effect on the APD50 of human adult cardiomyocytes (15 ± 3%, p < 0.05)(hAdult-CMs: N = 5 patients, n = 12 cells). (C,D) Relative positive inotropic and lusitropic effects of both ISO and FSK (%) in later stages hiPSC-CMs (ISO: N = 2, n = 7; FSK: N = 2, n = 21). Supporting information is reported in Table S2.

Figure 6

Action potential recording by patch clamp in earlier stages hiPSC-CMs. Earlier-stages hiPSC-CMs during patch clamp, elicited with short current pulses in current-clamp mode at day 20, 30 and 60. (A) Representative traces at 0.5, 1 and 2 Hz stimulation rates. (B) Average of resting membrane potential (mV), AP amplitude (mV), time from stimulus to 50% repolarization (APD50, ms) at 1 Hz and to 90% of repolarization (APD90, ms) with frequency variation. (C) Spontaneous beating frequency of action potential from day 20 to day 60. Data are reported as means ± SEM; One-way analysis of variance (ANOVA) with a Tukey post-hoc test with statistical significance set at * p < 0.05 and ** p < 0.01; NS not significant.