Leave the light on: chronic optogenetic tachypacing of human engineered cardiac tissue constructs

This editorial refers to ‘Chronic intermittent tachypacing by an optogenetic approach induces arrhythmia vulnerability in human engineered heart tissue’, by M. Lemme et al., pp. 1487–1499.

Animals models involving chronic tachypacing of the heart have provided a valuable platform for characterizing conditions like heart failure (HF)¹ and atrial fibrillation (AF)² but their translational relevance is inherently limited by differences from human physiology. 3D engineered heart tissue (EHT) constructs comprising cardiomyocyte-like human induced pluripotent stem cells (hiPSC-CM) have emerged as a valuable experimental system in which large numbers of small (∼1 cm-long), spontaneously beating units of human-like myocardium are produced, and the response to various stimuli (electric pacing, drug application, etc.) can be systematically characterized by analysing dozens of samples in parallel.³ However, until now it has not been possible to study EHTs subjected to chronic tachypacing via electric stimulation due to the cytotoxic by-products from electrochemical reactions that occur at the contact site.⁴

Optogenetics is an emergent experimental approach where photosensitive proteins (opsins) are expressed in excitable cells.⁵ This enables the use of light pulses instead of electric stimuli to perturb membrane potential and elicit a regenerative propagating response. Over the past decade, many applications of cardiac optogenetics have been explored, ranging from light-based defibrillation of ventricular arrhythmias⁶ to all-optical experimental systems for high-throughput electrophysiological characterization of cardiac tissue samples.⁷ Rapid progress in this area has in part been catalyzed by the development of a robust framework for representing optogenetic sensitization and light stimulation in multiscale simulations of the heart.⁸

Lemme et al.⁹ use a new model that combines EHT and optogenetics to investigate consequences of chronic tachypacing in human cardiac tissue. The light-sensitive cation channel Channelrhodopsin-2 (ChR2) was expressed via lentiviral transfection in EHTs predominantly composed of ventricular-like hiPSC-CMs. Flow cytometry analysis revealed ChR2 expression in ∼25% of cells, predominantly localized to the sarcolemma. EHTs were mounted in an array with side-by-side light sources (blue/red) to facilitate parallel optogenetic stimulation and contraction characterization.

In their first set of experiments, the authors compared response to illumination, contractile properties, and spontaneous beating rate in ChR2 transduced EHTs (ChR2-EHTs) and non-transfected EHTs. As expected, blue light had no effect on non-transfected tissue but entrained excitation in ChR2-EHTs (up to 5 Hz optogenetic pacing frequency). Baseline contractile properties between the two groups were similar but there were electrophysiological differences (even in the absence of direct optical stimulation). Compared to non-transfected EHTs, onset of autorhythmicity in ChR2-EHTs was delayed by ∼3 days and spontaneous excitation rate was twice as fast.

Next, the authors compared ChR2-EHTs in the presence or absence of chronic tachypacing via intermittent optical stimulation, alternating between 15 s of quiescence and 15 s of optogenetic burst stimulation at 3 Hz. Microelectrode recordings taken in both preparations after 4 weeks showed that, compared to control ChR2-EHTs, chronically tachypaced ChR2-EHTs had accelerated contraction and relaxation kinetics; abbreviated action potential duration (APD₉₀) and effective refractory period (ERP); depolarized take-off potential; decreased upstroke velocity; and reduced action potential amplitude. Patch-clamp experiments in hiPSC-CM isolated from ChR2-EHT had lower cell capacitance and reduced L-type calcium current density. In 64% of tachypaced ChR2-EHTs, burst electrical stimulation induced long-lasting (∼30 min) episodes of tachycardia, in which beating rate accelerated by a factor of ∼3×. In contrast, the tachycardia inducibility rate in control ChR2-EHTs was only 13%. In all inducible EHTs (with or without chronic optogenetic tachypacing), diastolic depolarization rate was increased compared to non-inducible constructs.

To further characterize the pro-arrhythmic substrate induced by chronic tachypacing, the authors measured mRNA and protein expression levels for several potentially relevant subcellular players. Most notably, the abundance of calsequestrin, which plays a critical role in sarcoplasmic reticulum (SR) calcium buffering, was reduced in chronically tachypaced EHTs. This suggests ryanodine receptor (RyR2) leak due to SR overloading as a potential mechanism for increased tachycardia vulnerability. Further support for this explanation is provided by the authors’ subsequent attempt to pharmacologically terminate tachycardia in ChR2-EHTs by applying a variety of conventional and experimental antiarrhythmic drugs. Drugs with a primary (e.g. JTV-519) or off-target (e.g. flecainide) RyR2-stabilizing effect were among the most effective compounds tested for tachycardia termination. Notably, arrhythmias could also be terminated by direct lengthening of APD/ERP via the hERG channel blocker E-4031.

This work by Lemme et al. has several important findings and implications. This is the first systematic examination of the consequences of chronic tachypacing (optogenetic or otherwise) in human myocardium-like tissue constructs. While the cells used in ChR2-EHTs here were generic hiPSC-CM phenotype, the same technology platform could be used with patient-derived EHT to study arrhythmia mechanisms (and potentially develop custom-tailored treatment strategies) in individuals with HF or other progressive conditions. Likewise, it would be fascinating to conduct chronic tachypacing experiments with EHTs made from atrial-like hiPSC-CMs. Analysis of electrophysiological remodelling at multiple timepoints could help decipher mechanisms of progression from paroxysmal to persistent to long-standing forms of AF, which are not yet well understood.¹⁰

Beyond the study of human arrhythmias that occur due to disease processes, this approach could also enrich understanding of engraftment tachycardia following cell therapy to prompt revascularization of infarcted hearts. In animal models, the window of arrhythmia vulnerability is from ∼1 to ∼4 weeks after injection of derived cells and tachycardia rates are far faster than the intrinsic beating rate of the same cells in vitro.¹¹ It is thus noteworthy that chronic tachypacing of ChR2-EHTs induced a similarly dramatic acceleration of spontaneous activity. Although the physiological contexts of the two models are fundamentally different and the relevance of chronic tachypacing to engraftment arrhythmias is completely unclear, the ability to scrutinize hiPSC-CM beating rate acceleration in a controlled experimental system may lead to discovery of novel explanations for this complex emergent behaviour.

Another interesting aspect of this study is the ChR2-EHT model itself, as no prior study has shown the viability of ChR2 expression in human tissue over such a long time period. Also, while it is known that mouse hearts show robust ChR2 1 year after light sensitization via AAV-ChR2 injection,⁶ this study’s finding that ChR2-EHTs can tolerate intermittent optical pacing but not continuous illumination over long time periods will influence any future attempts to study chronic optogenetic pacing in such animals. Interestingly, the observed preferential localization of ChR2 to the sarcolemma in transfected hiPSC-CMs differs from past cardiac optogenetics studies in transgenic mice, which reported robust ChR2 co-localization with the t-tubular system in ventricular myocytes.¹²

Finally, just as computational modelling has played a beneficial complementary role in the realm of cardiac optogenetics,⁸ robust tools have been developed for simulating hiPSC-CMs at the cell^13,14 and organ scale.¹⁵ It is conceivable that these computational approaches could be combined to create a comprehensive model of EHT-ChR2 that could be carefully calibrated and then used in an iterative manner to drive rapid progress in interpretation of complex results observed in vitro.

Conflict of interest: none declared.

Funding

This work was supported by a grant from the National Institutes of Health (NIH) and National Heart, Lung, and Blood Institute (NHLBI) (U01-HL141074 to N.A.T., with subcontract to P.M.B.).

The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.