Modulating cardiac physiology in engineered heart tissue with the bidirectional optogenetic tool BiPOLES

Optogenetic actuators are rapidly advancing tools used to control physiology in excitable cells, such as neurons and cardiomyocytes. In neuroscience, these tools have been used to either excite or inhibit neuronal activity. Cell type-targeted actuators have allowed to study the function of distinct cell populations. Whereas the first described cation channelrhodopsins allowed to excite specific neuronal cell populations, anion channelrhodopsins were used to inhibit neuronal activity. To allow for simultaneous excitation and inhibition, opsin combinations with low spectral overlap were introduced. BiPOLES (Bidirectional Pair of Opsins for Light-induced Excitation and Silencing) is a bidirectional optogenetic tool consisting of the anion channel Guillardia theta anion-conducting channelrhodopsin 2 (GtACR2 with a blue excitation spectrum and the red-shifted cation channel Chrimson. Here, we studied the effects of BiPOLES activation in cardiomyocytes. For this, we knocked in BiPOLES into the adeno-associated virus integration site 1 (AAVS1) locus of human-induced pluripotent stem cells (hiPSC), subjected these to cardiac differentiation, and generated BiPOLES expressing engineered heart tissue (EHT) for physiological characterization. Continuous light application activating either GtACR2 or Chrimson resulted in cardiomyocyte depolarization and thus stopped EHT contractility. In contrast, short light pulses, with red as well as with blue light, triggered action potentials (AP) up to a rate of 240 bpm. In summary, we demonstrate that cation, as well as anion channelrhodopsins, can be used to activate stem cell-derived cardiomyocytes with pulsed photostimulation but also to silence cardiac contractility with prolonged photostimulation.

An arrhythmogenic metabolite in atrial fibrillation

BACKGROUND

Long-chain acyl-carnitines (ACs) are potential arrhythmogenic metabolites. Their role in atrial fibrillation (AF) remains incompletely understood. Using a systems medicine approach, we assessed the contribution of C18:1AC to AF by analysing its in vitro effects on cardiac electrophysiology and metabolism, and translated our findings into the human setting.

METHODS AND RESULTS

Human iPSC-derived engineered heart tissue was exposed to C18:1AC. A biphasic effect on contractile force was observed: short exposure enhanced contractile force, but elicited spontaneous contractions and impaired Ca2+ handling. Continuous exposure provoked an impairment of contractile force. In human atrial mitochondria from AF individuals, C18:1AC inhibited respiration. In a population-based cohort as well as a cohort of patients, high C18:1AC serum concentrations were associated with the incidence and prevalence of AF.

CONCLUSION

Our data provide evidence for an arrhythmogenic potential of the metabolite C18:1AC. The metabolite interferes with mitochondrial metabolism, thereby contributing to contractile dysfunction and shows predictive potential as novel circulating biomarker for risk of AF.

Piezo2 is not an indispensable mechanosensor in murine cardiomyocytes

A short-term increase in ventricular filling leads to an immediate (Frank-Starling mechanism) and a slower (Anrep effect) rise in cardiac contractility, while long-term increased cardiac load (e.g., in arterial hypertension) decreases contractility. Whether these answers to mechanical tension are mediated by specific sensors in cardiomyocytes remains elusive. In this study, the piezo2 protein was evaluated as a potential mechanosensor. Piezo2 was found to be upregulated in various rat and mouse cardiac tissues upon mechanical or pharmacological stress. To investigate its function, C57BL/6J mice with homozygous cardiomyocyte-specific piezo2 knockout [Piezo2-KO] were created. To this end, α-MHC-Cre mice were crossed with homozygous "floxed" piezo2 mice. α-MHC-Cre mice crossed with wildtype mice served as controls [WT-Cre⁺]. In cardiomyocytes of Piezo2-KO mice, piezo2 mRNA was reduced by > 90% and piezo2 protein was not detectable. Piezo2-KO mice displayed no morphological abnormalities or altered cardiac function under nonstressed conditions. In a subsequent step, hearts of Piezo2-KO or WT-Cre⁺-mice were stressed by either three weeks of increased afterload (angiotensin II, 2.5 mg/kg/day) or one week of hypercontractility (isoprenaline, 30 mg/kg/day). As expected, angiotensin II treatment in WT-Cre⁺-mice resulted in higher heart and lung weight (per body weight, + 38%, + 42%), lower ejection fraction and cardiac output (- 30%, - 39%) and higher left ventricular anterior and posterior wall thickness (+ 34%, + 37%), while isoprenaline led to higher heart weight (per body weight, + 25%) and higher heart rate and cardiac output (+ 24%, + 54%). The Piezo2-KO mice reacted similarly with the exception that the angiotensin II-induced increases in wall thickness were blunted and the isoprenaline-induced increase in cardiac output was slightly less pronounced. As cardiac function was neither severely affected under basal nor under stressed conditions in Piezo2-KO mice, we conclude that piezo2 is not an indispensable mechanosensor in cardiomyocytes.

Human-Engineered Atrial Tissue for Studying Atrial Fibrillation

This chapter details the generation of atrial fibrin-based engineered heart tissue (EHT) in standard 24-well format as a 3D model for the human atrium. Compared to 2D cultivation, human-induced pluripotent stem cells (hiPSCs)-derived atrial cardiomyocytes demonstrated a higher degree of maturation in 3D format. Furthermore, we have demonstrated in previous work that the model displayed atrial characteristics in terms of contraction and gene expression patterns, electrophysiology, and pharmacological response. Here, we describe how to embed atrial cardiomyocytes differentiated from hiPSCs in a fibrin hydrogel to form atrial EHT attached to elastic silicone posts, allowing auxotonic contraction. In addition, we describe how force and other contractility parameters can be derived from these beating atrial EHTs by video-optical monitoring. The presented atrial EHT model is suitable to study chamber-specific mechanisms, drug effects and to serve for disease modeling.

Hypertrophic signaling compensates for contractile and metabolic consequences of DNA methyltransferase 3A loss in human cardiomyocytes

The role of DNA methylation in cardiomyocyte physiology and cardiac disease remains a matter of controversy. We have recently provided evidence for an important role of DNMT3A in human cardiomyocyte cell homeostasis and metabolism, using engineered heart tissue (EHT) generated from human induced pluripotent stem cell (hiPSC)-derived cardiomyocytes carrying a knockout of the de novo DNA methyltransferase DNMT3A. Unlike isogenic control EHT, knockout EHT displayed morphological abnormalities such as lipid accumulations inside cardiomyocytes associated with impaired mitochondrial metabolism, as well as functional defects and impaired glucose metabolism. Here, we analyzed the role of DNMT3A in the setting of cardiac hypertrophy. We induced hypertrophic signaling by treatment with 50 nM endothelin-1 and 20 μM phenylephrine for one week and assessed EHT contractility, morphology, DNA methylation, and gene expression. While both knockout EHTs and isogenic controls showed the expected activation of the hypertrophic gene program, knockout EHTs were protected from hypertrophy-related functional impairment. Conversely, hypertrophic treatment prevented the metabolic consequences of a loss of DNMT3A, i.e. abolished lipid accumulation in cardiomyocytes likely by partial normalization of mitochondrial metabolism and restored glucose metabolism and metabolism-related gene expression of knockout EHT. Together, these data suggest an important role of DNA methylation not only for cardiomyocyte physiology, but also in the setting of cardiac disease.

An Important Role for DNMT3A-Mediated DNA Methylation in Cardiomyocyte Metabolism and Contractility

Supplemental Digital Content is available in the text.

Background:

DNA methylation acts as a mechanism of gene transcription regulation. It has recently gained attention as a possible therapeutic target in cardiac hypertrophy and heart failure. However, its exact role in cardiomyocytes remains controversial. Thus, we knocked out the main de novo DNA methyltransferase in cardiomyocytes, DNMT3A, in human induced pluripotent stem cells. Functional consequences of DNA methylation-deficiency under control and stress conditions were then assessed in human engineered heart tissue from knockout human induced pluripotent stem cell–derived cardiomyocytes.

Methods:

DNMT3A was knocked out in human induced pluripotent stem cells by CRISPR/Cas9gene editing. Fibrin-based engineered heart tissue was generated from knockout and control human induced pluripotent stem cell–derived cardiomyocytes. Development and baseline contractility were analyzed by video-optical recording. Engineered heart tissue was subjected to different stress protocols, including serum starvation, serum variation, and restrictive feeding. Molecular, histological, and ultrastructural analyses were performed afterward.

Results:

Knockout of DNMT3A in human cardiomyocytes had three main consequences for cardiomyocyte morphology and function: (1) Gene expression changes of contractile proteins such as higher atrial gene expression and lower MYH7/MYH6 ratio correlated with different contraction kinetics in knockout versus wild-type; (2) Aberrant activation of the glucose/lipid metabolism regulator peroxisome proliferator-activated receptor gamma was associated with accumulation of lipid vacuoles within knockout cardiomyocytes; (3) Hypoxia-inducible factor 1α protein instability was associated with impaired glucose metabolism and lower glycolytic enzyme expression, rendering knockout-engineered heart tissue sensitive to metabolic stress such as serum withdrawal and restrictive feeding.

Conclusion:

The results suggest an important role of DNA methylation in the normal homeostasis of cardiomyocytes and during cardiac stress, which could make it an interesting target for cardiac therapy.

Magnetic Adjustment of Afterload in Engineered Heart Tissues

Afterload is known to drive the development of both physiological and pathological cardiac states. As such, studying the outcomes of altered afterload states could yield important insights into the mechanisms controlling these critical processes. However, an experimental technique for precisely fine-tuning afterload in heart tissue over time is currently lacking. Here, a newly developed magnetics-based technique for achieving this control in engineered heart tissues (EHTs) is described. In order to produce magnetically responsive EHTs (MR-EHTs), the tissues are mounted on hollow silicone posts, some of which contain small permanent magnets. A second set of permanent magnets is press-fit into an acrylic plate such that they are oriented with the same polarity and are axially-aligned with the post magnets. To adjust afterload, this plate of magnets is translated toward (higher afterload) or away (lower afterload) from the post magnets using a piezoelectric stage fitted with an encoder. The motion control software used to adjust stage positioning allows for the development of user-defined afterload regimens while the encoder ensures that the stage corrects for any inconsistencies in its location. This work describes the fabrication, calibration, and implementation of this system to enable the development of similar platforms in other labs around the world. Representative results from two separate experiments are included to exemplify the range of different studies that can be performed using this system.

Blockade of miR-140-3p prevents functional deterioration in afterload-enhanced engineered heart tissue

Afterload enhancement (AE) of rat engineered heart tissue (EHT) in vitro leads to a multitude of changes that in vivo are referred to as pathological cardiac hypertrophy: e.g., cardiomyocyte hypertrophy, contractile dysfunction, reactivation of fetal genes and fibrotic changes. Moreover AE induced the upregulation of 22 abundantly expressed microRNAs. Here, we aimed at evaluating the functional effect of inhibiting 7 promising microRNAs (miR-21-5p, miR-146b-5p, miR-31a-5p, miR-322-5p, miR-450a-5p, miR-140-3p and miR-132-3p) in a small-range screen. Singular transfection of locked nucleic acid (LNA)-based anti-miRs at 100 nM (before the one week AE-procedure) led to a powerful reduction of the targeted microRNAs. Pretreatment with anti-miR-146b-5p, anti-miR-322-5p or anti-miR-450a-5p did not alter the AE-induced contractile decline, while anti-miR-31a-5p-pretreatment even worsened it. Anti-miR-21-5p and anti-miR-132-3p partially attenuated the AE-effect, confirming previous reports. LNA-anti-miR against miR-140-3p, a microRNA recently identified as a prognostic biomarker of cardiovascular disease, also attenuated the AE-effect. To simplify future in vitro experiments and to create an inhibitor for in vivo applications, we designed shorter miR-140-3p-inhibitors and encountered variable efficiency. Only the inhibitor that effectively repressed miR-140-3p was also protective against the AE-induced contractile decline. In summary, in a small-range functional screen, miR-140-3p evolved as a possible new target for the attenuation of afterload-induced pathological cardiac hypertrophy.

Analysis of fibrosis in control or pressure overloaded rat hearts after mechanical unloading by heterotopic heart transplantation

Mechanical unloading (MU) by implantation of left ventricular assist devices (LVAD) has become clinical routine. This procedure has been shown to reverse cardiac pathological remodeling, with the underlying molecular mechanisms incompletely understood. Most studies thus far were performed in non-standardized human specimens or MU of healthy animal hearts. Our study investigates cardiac remodeling processes in sham-operated healthy rat hearts and in hearts subjected to standardized pathological pressure overload by transverse aortic constriction (TAC) prior to MU by heterotopic heart transplantation (hHTx/MU). Rats underwent sham or TAC surgery. Disease progression was monitored by echocardiography prior to MU by hHTx/MU. Hearts after TAC or TAC combined with hHTx/MU were removed and analyzed by histology, western immunoblot and gene expression analysis. TAC surgery resulted in cardiac hypertrophy and impaired cardiac function. TAC hearts revealed significantly increased cardiac myocyte diameter and mild fibrosis. Expression of hypertrophy associated genes after TAC was higher compared to hearts after hHTx/MU. While cardiac myocyte cell diameter regressed to the level of sham-operated controls in all hearts subjected to hHTx/MU, fibrotic remodeling was significantly exacerbated. Transcription of pro-fibrotic and apoptosis-related genes was markedly augmented in all hearts after hHTx/MU. Sarcomeric proteins involved in excitation-contraction coupling displayed significantly lower phosphorylation levels after TAC and significantly reduced total protein levels after hHTx/MU. Development of myocardial fibrosis, cardiac myocyte atrophy and loss of sarcomeric proteins was observed in all hearts that underwent hHTX/MU regardless of the disease state. These results may help to explain the clinical experience with low rates of LVAD removal due to lack of myocardial recovery.

Novel DNA Methylation Sites Influence GPR15 Expression in Relation to Smoking

Smoking is a major risk factor for cardiovascular diseases and has been implicated in the regulation of the G protein-coupled receptor 15 (GPR15) by affecting CpG methylation. The G protein-coupled receptor 15 is involved in angiogenesis and inflammation. An effect on GPR15 gene regulation has been shown for the CpG site CpG3.98251294. We aimed to analyze the effect of smoking on GPR15 expression and methylation sites spanning the GPR15 locus. DNA methylation of nine GPR15 CpG sites was measured in leukocytes from 1291 population-based individuals using the EpiTYPER. Monocytic GPR15 expression was measured by qPCR at baseline and five-years follow up. GPR15 gene expression was upregulated in smokers (beta (ß) = −2.699, p-value (p) = 1.02 × 10⁻⁷⁷) and strongly correlated with smoking exposure (ß = −0.063, p = 2.95 × 10⁻³⁴). Smoking cessation within five years reduced GPR15 expression about 19% (p = 9.65 × 10⁻⁵) with decreasing GPR15 expression over time (ß = 0.031, p = 3.81 × 10⁻⁶). Additionally, three novel CpG sites within GPR15 affected by smoking were identified. For CpG3.98251047, DNA methylation increased steadily after smoking cessation (ß = 0.123, p = 1.67 × 10⁻³) and strongly correlated with changes in GPR15 expression (ß = 0.036, p = 4.86 × 10⁻⁵). Three novel GPR15 CpG sites were identified in relation to smoking and GPR15 expression. Our results provide novel insights in the regulation of GPR15, which possibly linked smoking to inflammation and disease progression.

Clonal dynamics studied in cultured induced pluripotent stem cells reveal major growth imbalances within a few weeks

Background

Human induced pluripotent stem (iPS) cells have revolutionised research and spark hopes for future tissue replacement therapies. To obtain high cell numbers, iPS cells can be expanded indefinitely. However, as long-term expansion can compromise cell integrity and quality, we set out to assess potential reduction of clonal diversity by inherent growth imbalances.

Methods

Using red, green, blue marking as a lentiviral multi-colour clonal cell tracking technology, we marked three different iPS cell lines as well as three other cell lines, assigning a unique fluorescent colour to each cell at one point in culture. Subsequently, we followed the sub-clonal distribution over time by flow cytometry and fluorescence microscopy analysis in regular intervals.

Results

In three human iPS cell lines as well as primary human fibroblasts and two widely used human cell lines as controls (K562 and HEK 293 T), we observed a marked reduction in sub-clonal diversity over time of culture (weeks). After 38 passages, all iPS cultures consisted of less than 10 residual clones. Karyotype and function, the latter assessed by cardiomyocyte differentiation and tissue engineering, did not reveal obvious differences.

Conclusions

Our results argue for a quick selection of sub-clones with a growth advantage and flag a normally invisible and potentially undesired behaviour of cultured iPS cells, especially when using long-term cultured iPS cells for experiments or even in-vivo applications.

Electronic supplementary material

The online version of this article (10.1186/s13287-018-0893-2) contains supplementary material, which is available to authorized users.

Targeting Chondroitin Sulfate Glycosaminoglycans to Treat Cardiac Fibrosis in Pathological Remodeling

Background:

Heart failure is a leading cause of mortality and morbidity, and the search for novel therapeutic approaches continues. In the monogenic disease mucopolysaccharidosis VI, loss-of-function mutations in arylsulfatase B lead to myocardial accumulation of chondroitin sulfate (CS) glycosaminoglycans, manifesting as myriad cardiac symptoms. Here, we studied changes in myocardial CS in nonmucopolysaccharidosis failing hearts and assessed its generic role in pathological cardiac remodeling.

Methods:

Healthy and diseased human and rat left ventricles were subjected to histological and immunostaining methods to analyze glycosaminoglycan distribution. Glycosaminoglycans were extracted and analyzed for quantitative and compositional changes with Alcian blue assay and liquid chromatography–mass spectrometry. Expression changes in 20 CS-related genes were studied in 3 primary human cardiac cell types and THP-1–derived macrophages under each of 9 in vitro stimulatory conditions. In 2 rat models of pathological remodeling induced by transverse aortic constriction or isoprenaline infusion, recombinant human arylsulfatase B (rhASB), clinically used as enzyme replacement therapy in mucopolysaccharidosis VI, was administered intravenously for 7 or 5 weeks, respectively. Cardiac function, myocardial fibrosis, and inflammation were assessed by echocardiography and histology. CS-interacting molecules were assessed with surface plasmon resonance, and a mechanism of action was verified in vitro.

Results:

Failing human hearts displayed significant perivascular and interstitial CS accumulation, particularly in regions of intense fibrosis. Relative composition of CS disaccharides remained unchanged. Transforming growth factor–β induced CS upregulation in cardiac fibroblasts. CS accumulation was also observed in both the pressure-overload and the isoprenaline models of pathological remodeling in rats. Early treatment with rhASB in the transverse aortic constriction model and delayed treatment in the isoprenaline model proved rhASB to be effective at preventing cardiac deterioration and augmenting functional recovery. Functional improvement was accompanied by reduced myocardial inflammation and overall fibrosis. Tumor necrosis factor–α was identified as a direct binding partner of CS glycosaminoglycan chains, and rhASB reduced tumor necrosis factor–α—induced inflammatory gene activation in vitro in endothelial cells and macrophages.

Conclusions:

CS glycosaminoglycans accumulate during cardiac pathological remodeling and mediate myocardial inflammation and fibrosis. rhASB targets CS effectively as a novel therapeutic approach for the treatment of heart failure.

Analysis of Tyrosine Kinase Inhibitor-Mediated Decline in Contractile Force in Rat Engineered Heart Tissue

Introduction

Left ventricular dysfunction is a frequent and potentially severe side effect of many tyrosine kinase inhibitors (TKI). The mode of toxicity is not identified, but may include impairment of mitochondrial or sarcomeric function, autophagy or angiogenesis, either as an on-target or off-target mechanism.

Methods and Results

We studied concentration-response curves and time courses for nine TKIs in three-dimensional, force generating engineered heart tissue (EHT) from neonatal rat heart cells. We detected a concentration- and time-dependent decline in contractile force for gefitinib, lapatinib, sunitinib, imatinib, sorafenib, vandetanib and lestaurtinib and no decline in contractile force for erlotinib and dasatinib after 96 hours of incubation. The decline in contractile force was associated with an impairment of autophagy (LC3 Western blot) and appearance of autophagolysosomes (transmission electron microscopy).

Conclusion

This study demonstrates the feasibility to study TKI-mediated force effects in EHTs and identifies an association between a decline in contractility and inhibition of autophagic flux.

Pharmacokinetics of the Experimental Non-Nucleosidic DNA Methyl Transferase Inhibitor N-Phthalyl-L-Tryptophan (RG 108) in Rats

DNA methyl transferase (DNMT) inhibitors can re-establish the expression of tumour suppressor genes in malignant diseases, but might also be useful in other diseases. Inhibitors in clinical use are nucleosidic cytotoxic agents that need to be integrated into the DNA of dividing cells. Here, we assessed the in vivo kinetics of a non-nucleosidic inhibitor that is potentially free of cytotoxic effects and does not require cell division. The non-specific DNMT inhibitor N-phthalyl-L-tryptophan (RG 108) was injected subcutaneously in rats. Blood was drawn 0, 0.5, 1, 2, 4, 6, 8 and 24 hr after injection and RG 108 in plasma was measured by high-performance liquid chromatography coupled to mass spectrometry. Trough levels and area under the curve (AUC) were significantly higher with multiple-dose administration and cytochrome inhibition. In this group, time to maximal plasma concentration (tmax , mean ± S.D.) was 37.5 ± 15 min., terminal plasma half-life was approximately 3.7 h (60% CI: 2.1-15.6 h), maximal plasma concentration (Cmax) was 61.3 ± 7.6 μM, and AUC was 200 ± 54 μmol·h/l. RG 108 peak levels were not influenced by cytochrome inhibition or multiple-dose administration regimens. Maximal tissue levels (Cmax in μmol/kg) were 6.9 ± 6.7, 1.6 ± 0.4 and 3.4 ± 1.1 in liver, skeletal and heart muscle, respectively. We conclude that despite its high lipophilicity, RG 108 can be used for in vivo experiments, appears safe and yields plasma and tissue levels in the range of the described 50% inhibitory concentration of around 1 to 5 μM. RG 108 can therefore be a useful tool for in vivo DNMT inhibition.

Deciphering the microRNA signature of pathological cardiac hypertrophy by engineered heart tissue- and sequencing-technology

Pathological cardiac hypertrophy and fibrosis are modulated by a set of microRNAs, most of which have been detected in biologically complex animal models of hypertrophy by arrays with moderate sensitivity and disregard of passenger strand (previously "star") microRNAs. Here, we aimed at precisely analyzing the microRNA signature of cardiac hypertrophy and fibrosis by RNA sequencing in a standardized in vitro hypertrophy model based on engineered heart tissue (EHT). Spontaneously beating, force-generating fibrin EHTs from neonatal rat heart cells were subjected to afterload enhancement for 7days (AE-EHT), and EHTs without intervention served as controls. AE resulted in reduced contractile force and relaxation velocity, fibrotic changes and reactivation of the fetal gene program. Small RNAs were extracted from control and AE-EHTs and sequencing yielded almost 750 different mature microRNAs, many of which have never been described before in rats. The detection of both arms of the precursor stem-loop (pre-miRNA), namely -3p and -5p miRs, was frequent. 22 abundantly sequenced microRNAs were >1.3× upregulated and 15 abundantly sequenced microRNAs downregulated to <0.77×. Among the upregulated microRNAs were 3 pairs of guide and passenger strand microRNAs (miR-21-5p/-3p, miR-322-5p/-3p, miR-210-3p/-5p) and one single passenger strand microRNA (miR-140-3p). Among downregulated microRNAs were 3 pairs (miR-133a-3p/-5p, miR-30e-5p/3p, miR-30c-5p/-3p). Preincubating EHTs with anti-miR-21-5p markedly attenuated the AE-induced contractile failure, cardiomyocyte hypertrophy and fibrotic response, recapitulating prior results in whole animals. Taken together, AE-induced pathological hypertrophy in EHTs is associated with 37 differentially regulated microRNAs, including many passenger strands. Antagonizing miR-21-5p ameliorates dysfunction in this model.

Functional improvement and maturation of rat and human engineered heart tissue by chronic electrical stimulation

Spontaneously beating engineered heart tissue (EHT) represents an advanced in vitro model for drug testing and disease modeling, but cardiomyocytes in EHTs are less mature and generate lower forces than in the adult heart. We devised a novel pacing system integrated in a setup for videooptical recording of EHT contractile function over time and investigated whether sustained electrical field stimulation improved EHT properties. EHTs were generated from neonatal rat heart cells (rEHT, n=96) or human induced pluripotent stem cell (hiPSC)-derived cardiomyocytes (hEHT, n=19). Pacing with biphasic pulses was initiated on day 4 of culture. REHT continuously paced for 16-18 days at 0.5Hz developed 2.2× higher forces than nonstimulated rEHT. This was reflected by higher cardiomyocyte density in the center of EHTs, increased connexin-43 abundance as investigated by two-photon microscopy and remarkably improved sarcomere ultrastructure including regular M-bands. Further signs of tissue maturation include a rightward shift (to more physiological values) of the Ca(2+)-response curve, increased force response to isoprenaline and decreased spontaneous beating activity. Human EHTs stimulated at 2Hz in the first week and 1.5Hz thereafter developed 1.5× higher forces than nonstimulated hEHT on day 14, an ameliorated muscular network of longitudinally oriented cardiomyocytes and a higher cytoplasm-to-nucleus ratio. Taken together, continuous pacing improved structural and functional properties of rEHTs and hEHTs to an unprecedented level. Electrical stimulation appears to be an important step toward the generation of fully mature EHT.