Physiological calcium combined with electrical pacing accelerates maturation of human engineered heart tissue

Abstract

Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have wide potential application in basic research, drug discovery, and regenerative medicine, but functional maturation remains challenging. Here, we present a method whereby maturation of hiPSC-CMs can be accelerated by simultaneous application of physiological Ca²⁺ and frequency-ramped electrical pacing in culture. This combination produces positive force-frequency behavior, physiological twitch kinetics, robust β-adrenergic response, improved Ca²⁺ handling, and cardiac troponin I expression within 25 days. This study provides insights into the role of Ca²⁺ in hiPSC-CM maturation and offers a scalable platform for translational and clinical research.

Keywords: calcium; engineered heart tissue; force-frequency relationship; maturation; pacing.

Figure 1

Physiological level Ca²⁺ alone increases contractile force but does not improve overall functional maturation profile (A) Experimental design schematic of HC-NP and LC-NP groups. (B and C), (E and F) both n = 6, (H and I) HC-NP n = 6, LC-NP n = 4. All data came from one differentiation batch. (B) Unnormalized force-frequency relationship (FFR). (C) Normalized FFR. (D) Representative twitches of the two groups at 1 Hz. (E) TTP at 1 Hz. (F) RT50 at 1 Hz. (G) Representative HC-NP post-rest potentiation data sequence. (H) Post-rest potentiation. (I) Diastolic excess fraction. two-way ANOVA with repeated measures (B and C) and unpaired two tailed t-tests (E, F, H, and I) were performed. two-way ANOVA p = n.s. (B and C), and t-tests p = n.s. (E, F, H, and I).

Figure 2

Physiological level Ca2+ and pacing result in functional improvement (A) Experimental design of high-Ca2+ ramp-paced (HC-RP) and low-Ca2+ ramp-paced (LC-RP) groups with 2-week continuous 2-4 Hz ramp pacing. C-J n = 6 for both groups from a single differentiation batch. M-N HC-RP n = 8, LC-RP n = 11 from two batches. O-P n = 6 for both groups from two batches. (B) SOLIDWORKS drawing of the pacing apparatus for individual tissues cultured in a 12-well plate. (C) Representative force-frequency relationship (FFR) for HC-RP (blue) and LC-RP (red) groups respectively from 1–3 Hz with 0.25 Hz increments. (D) Unnormalized FFR. (E) Normalized FFR (p < 0.0001). (F) Time to peak (TTP) at 1–3 Hz (p = 0.0003). (G) Time to 50% relaxation (RT50) at 1–3 Hz (p < 0.0001). (H) Representative twitches at 1 Hz. (I) Time to peak (TTP) at 1 Hz (p = 0.0009). (J) Time to 50% relaxation (RT50) at 1 Hz (p = 0.0028). (K) Representative sequence of HC-RP group to measure post-rest potentiation and diastolic excess fraction as the ratio between 3 Hz and 1Hz diastolic baselines. (L) Representative LC-RP data sequence. (M) Post-rest potentiation (p = 0.0049). (N) Diastolic excess fraction (p = 0.0325). (O) Sample Fura-2 Ca2+ transients at 1 Hz. (P) Fura-2 Ca2+ transient signal intensity at 1 Hz (p = 0.0248). (Q) Ca2+ transient decay time constant from peak to 80% relaxation (τ80, p < 0.0001). Two way ANOVA with repeated measures (D–G) and unpaired two-tailed t tests (I, J, M, N, P, Q) were performed. ∗p < 0.05, ∗∗p < 0.005, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 3

Physiological level Ca²⁺ and pacing result in improved β-adrenergic response to ISO (A and B) Representative normalized HC-RP and LC-RP twitches at 1 Hz before and after 1 μM ISO infusion. (C) Systolic force increases in percentage (both n = 7, p = 0.0012). (D) TTP changes in percentage (both n = 7, p = 0.0033). (E) RT50 change in percentage (both n = 7). (F and G) Representative normalized HC-RP and LC-RP tissue Ca²⁺ transients with Fura-2 before and after ISO infusion. (H) Ca²⁺ transient intensity increases (HC-RP n = 5, LC-RP n = 7, p = 0.0223). (I) Ca²⁺ transient time decay constant from peak to 80% relaxation (τ₈₀, HC-RP n = 5, LC-RP n = 7, p = 0.0035). Unpaired two-tailed t-tests were performed for panels (C, D, H, and I). ^∗p < 0.05, ^∗∗p < 0.005. All data came from one differentiation batch.

Figure 4

Physiological level Ca²⁺ and pacing result in more mature protein and RNA expression All protein and RNA samples were extracted from tissues fresh frozen immediately after pacing. For western data, n = 4 for both groups and all measurement and ratios were first normalized against total protein. For RNA-seq, HC-RP n = 4, LC-RP n = 3, HC-NP n = 3, and LC-NP n = 4. Both western blotting and RNA-seq data came from one separate differentiation batch of the same single cell. (A) TPM1 and SERCA content (p = 0.0003). (B) PLN (p = 0.0477) and p-PLN. (C) cTnI (p < 0.0001) and p-cTnI (p < 0.0001). (D) RNA expression comparisons for Ca²⁺-handling proteins, sarcomere proteins, fatty acid metabolism, ion channels, and maturation markers. (E) HC-RP versus LC-NP volcano plot. (F) HC-RP versus LC-RP. (G) HC-RP versus HC-NP. (H) HC-NP versus LC-NP. Unpaired two-tailed t-tests were performed for (A–C). ^∗p < 0.05, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001.

Figure 5

Functional improvements from high calcium and ramp pacing were observed in a second iPSC line (A) Two-way ANOVA experimental design (high Ca²⁺ vs. low Ca²⁺; ramp vs. no pacing) and spontaneous beating rate after ramp completion (two-way ANOVA p < 0.0001). Mechanical data (B–K) were from three separate differentiation batches (HC-NP n = 14, HC-RP n = 13, LC-NP n = 14, and LC-RP n = 12). FURA data (M, N, and T–W) were from two batches (n = 8, 9, 7, and 8, respectively). ISO data were from three batches (n = 13, 10, 11, and 9, respectively). (B) Raw and normalized FFR (p < 0.0001). (C) TTP from 1 to 3 Hz. (D) RT50. (E) Sample normalized peak force at 1 Hz. (F) TTP at 1 Hz (two-way ANOVA p = 0.0002, HC-RP vs. all others p < 0.0001). (G) RT50 at 1 Hz (two-way ANOVA p < 0.0001, HC-NP vs. LC-NP p = 0.0003, HC-NP vs. LC-RP p < 0.0001, HC-NP vs. HC-RP p < 0.0001, HC-RP vs. LC-NP p < 0.0001, HC-RP vs. LC-RP p = 0.0018). (H and I) Raw post-rest potentiation data for HC-RP and LC-RP. (J) Post-rest potentiation. (K) Diastolic excess fraction (two-way ANOVA p < 0.0019, HC-RP vs. HC-NP p = 0.0001, HC-RP vs. LC-NP p = 0.0009, HC-RP vs. LC-RP p = 0.0003). (L) Ca²⁺ transient signal intensity. (M) Ca²⁺ transient decay (τ₈₀). (N and O) Sample HC-RP and LC-RP force responses to 1 μM ISO. (P) Summary force changes to 1 μM ISO (two-way ANOVA p = 0.004, HC-RP vs. all others p < 0.0001). (Q) TTP changes to ISO. (R) RT50 changes to ISO. (S and T) Example HC-RP and LCRP Ca²⁺ transients in response to 1 μM ISO. (U) Ca²⁺ transient intensity changes to ISO. (V) τ₈₀ changes to ISO (two-way ANOVA p = 0.0058, HC-RP vs. HC-NP p = 0.0003, HC-RP vs. LC-NP p = 0.0053, HC-RP vs. LC-RP p = 0.0025). Three-way ANOVA with repeated measures was performed for (B–D), two-way ANOVA with planned comparisons for (F, G, J–M, P–R, U, and V). ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001.