Long-term effects of empagliflozin on excitation-contraction-coupling in human induced pluripotent stem cell cardiomyocytes

Abstract

The SGLT2 inhibitor empagliflozin improved cardiovascular outcomes in patients with diabetes. As the cardiac mechanisms remain elusive, we investigated the long-term effects (up to 2 months) of empagliflozin on excitation-contraction (EC)-coupling in human cardiomyocytes derived from induced pluripotent stem cells (iPSC-CM) in a blinded manner. IPSC from 3 donors, differentiated into pure iPSC-CM (4 differentiations), were treated with a clinically relevant concentration of empagliflozin (0.5 μmol/l) or vehicle control. Treatment, data acquisition, and analysis were conducted externally blinded. Epifluorescence microscopy measurements in iPSC-CM showed that empagliflozin has neutral effects on Ca²⁺ transient amplitude, diastolic Ca²⁺ levels, Ca²⁺ transient kinetics, or sarcoplasmic Ca²⁺ load after 2 weeks or 8 weeks of treatment. Confocal microscopy determining possible effects on proarrhythmogenic diastolic Ca²⁺ release events showed that in iPSC-CM, Ca²⁺ spark frequency and leak was not altered after chronic treatment with empagliflozin. Finally, in patch-clamp experiments, empagliflozin did not change action potential duration, amplitude, or resting membrane potential compared with vehicle control after long-term treatment. Next-generation RNA sequencing (NGS) and mapped transcriptome profiles of iPSC-CMs untreated and treated with empagliflozin for 8 weeks showed no differentially expressed EC-coupling genes. In line with NGS data, Western blots indicate that empagliflozin has negligible effects on key EC-coupling proteins. In this blinded study, direct treatment of iPSC-CM with empagliflozin for a clinically relevant duration of 2 months did not influence cardiomyocyte EC-coupling and electrophysiology. Therefore, it is likely that other mechanisms independent of cardiomyocyte EC-coupling are responsible for the beneficial treatment effect of empagliflozin. KEY MESSAGES: This blinded study investigated the clinically relevant long-term effects (up to 2 months) of empagliflozin on cardiomyocyte excitation-contraction (EC)-coupling. Human cardiomyocytes derived from induced pluripotent stem cells (iPSC-CM) were used to study a human model including a high repetition number of experiments. Empagliflozin has neutral effects on cardiomyocyte Ca²⁺ transients, sarcoplasmic Ca²⁺ load, and diastolic sarcoplasmic Ca²⁺ leak. In patch-clamp experiments, empagliflozin did not change the action potential. Next-generation RNA sequencing, mapped transcriptome profiles, and Western blots of iPSC-CM untreated and treated with empagliflozin showed no differentially expressed EC-coupling candidates.

Keywords: EC-coupling; Electrophysiology; Empagliflozin; iPSC-CM.

Fig. 1

Systolic Ca²⁺ transients and SR Ca²⁺ load (epifluorescence microscopy, Fura-2 AM 5 μM) of human-induced pluripotent stem cell cardiomyocytes (iPSC-CM) after 2 weeks of treatment with either vehicle control (control) or 0.5 μmol/l empagliflozin (EMPA). (a–b) Original representative stimulated Ca²⁺-transient recordings at 0.25 Hz after 2 weeks of treatment and (c–d) Ca²⁺ transients recorded during application of 10 mM caffeine indicating SR Ca²⁺ load. (e) Mean data during increasing stimulation frequencies (0.25, 0.5, and 1 Hz) for systolic Ca²⁺-transient amplitude (f), diastolic Ca²⁺ levels, and (g) exponential decay time of Ca²⁺ transients (τ). (h) Mean data for caffeine-transient amplitude. The sample sizes of iPSC-CM are depicted below each column. Groups were statistically compared using two-way ANOVA with Sidak’s test for multiple comparisons or Student’s t test (h)

Fig. 2

Systolic Ca²⁺ transients and SR Ca²⁺ load (epifluorescence microscopy, Fura-2 AM 5 μM) of human-induced pluripotent stem cell cardiomyocytes (iPSC-CM) after 8 weeks of treatment with either vehicle control (control) or 0.5 μmol/l empagliflozin (EMPA). (a–b) Original representative stimulated Ca²⁺-transient recordings at 0.25 Hz of 8 weeks treated iPSC-CMs and (c–d) caffeine-induced transients indicating SR Ca²⁺ content. (e) Mean data during increasing stimulation frequencies (0.25, 0.5, and 1 Hz) for systolic Ca²⁺-transient amplitude (f), diastolic Ca²⁺ levels, and (g) exponential decay time of Ca²⁺ transients (τ). (h) Mean data for caffeine-transient amplitude. The sample sizes of iPSC-CM are provided below the respective column. For statistical analysis, two-way ANOVA with Sidak’s test for multiple comparisons or Student’s t test (h) was used

Fig. 3

Diastolic sarcoplasmic reticulum Ca²⁺ release (confocal microscopy, Fluo-4 AM 10 μM) of human-induced pluripotent stem cell cardiomyocytes (iPSC-CM) after 2 and 8 weeks of treatment with either vehicle control (control) or 0.5 μmol/l empagliflozin (EMPA). (a) Representative original confocal line scans showing diastolic Ca²⁺ sparks in iPSC-CM after 2 and (d) after 8 weeks treatment with empagliflozin compared with vehicle control. (b) Mean data of diastolic Ca²⁺ spark frequency and (c) the total calculated diastolic Ca²⁺ leak (normalized to vehicle control) after 2 weeks and (e–f) 8 weeks of empagliflozin treatment. The sample sizes of iPSC-CM are presented below the respective column. Groups were statistically tested using Student’s t test

Fig. 4

Action potentials (patch-clamp technique) of human-induced pluripotent stem cell cardiomyocytes (iPSC-CM) after treatment with either vehicle control (Control) or 0.5 μmol/l empagliflozin (EMPA). (a) Representative action potential recordings (0.25 Hz) of human iPSC-CM treated for 2 weeks with vehicle control or (b) empagliflozin. (c) Effects of 2 weeks treatment with control or empagliflozin on action potential duration at 80% (APD₈₀), (d) resting membrane potential (RMP) and (e) action potential amplitude (APA). (f) Original action potential recordings (0.25 Hz) of 8 weeks treated human iPSC-CM with either control or (g) empagliflozin. (h) Mean data for APD₈₀, (i) RMP, and (j) APA of iPSC-CM after 8 weeks of treatment with control or empagliflozin. The sample sizes of iPSC-CM are provided below the respective column. For statistical testing, two-way ANOVA with Sidak’s test for multiple comparisons (c and h) or Student’s t test was used

Fig. 5

Modulation of gene expression in human-induced pluripotent stem cell cardiomyocytes (iPSC-CM) after chronic (8 weeks) treatment with empagliflozin (0.5 μmol/l, n = 4 differentiations). (a) Empagliflozin-affected normalized counts of genes as ryanodine-receptor type 2 (RYR2), sodium-calcium exchanger (NCX1), sarcoplasmic reticulum Ca²⁺ ATPase (SERCA), Ca²⁺-/calmodulin-dependent protein kinase II (CaMKII), phospholamban (PLN), (b) calsequestrin 2 (CASQ2), calmodulin 1 (CALM1), calmodulin 2 (CALM2), (C) Na⁺ voltage-gated channel alpha subunit 5 (SCN5A), Ca²⁺ voltage-gated channel subunit alpha1 C (CACNA1C), K⁺ voltage-gated channel subfamily A member 5 (KCN5A), (d) ATPase plasma membrane Ca²⁺ transporting 1 (ATP2B1). For statistical analysis, Student’s t test was applied

Fig. 6

Western blot of EC-coupling proteins in human-induced pluripotent stem cell cardiomyocytes (iPSC-CM). (a) Representative original Western blots after treatment with empagliflozin (EMPA) or vehicle control (control) for 2 or 8 weeks from 4 differentiation experiments (Diff.) from 2 healthy donors. GAPDH was used as loading control. (b) Mean protein expression levels (normalized to the respective control group at 2 or 8 weeks) in iPSC-CM (n = 4 differentiations, matched groups are displayed with matched individual symbols) and effects of 2 and 8 weeks treatment with empagliflozin (EMPA) on ryanodine-receptor type 2 (RyR2), (c) sodium-calcium exchanger (NCX), (d) sarcoplasmic reticulum Ca²⁺ ATPase (SERCA), (e) phosphorylated Ca²⁺-/calmodulin-dependent protein kinase II (pCaMKII), (f) Ca²⁺-/calmodulin-dependent protein kinase II (CaMKII), and (g) phospholamban (PLB). Groups were statistically analyzed using two-way ANOVA with Sidak’s test for multiple comparisons