Introduction
Cardiac action potential (AP) onset and propagation across the heart are physiological processes regulated by several key dynamic factors such as collective ion channel recovery as well as intracellular Ca
2+-cycling [
34,
35]. Pathological conditions alter these properties, often resulting in beat-to-beat changes of AP properties, a phenomenon called “alternans” [
42], which is linked to an increased risk of life-threatening re-entrant cardiac arrhythmias. Experimental methods for manipulating AP dynamics are limited: pharmacological approaches [
18,
19,
41] lack spatial and temporal specificity and in some cases are only partially reversible. In this work, we propose an approach that uses optogenetics to manipulate cardiac AP dynamics, enabling the exploration of the role that cardiac alternans plays in sustaining or terminating re-entrant cardiac arrhythmias.
Optogenetics combines the use of light-sensitive proteins, genetically expressed in cells/tissues of interest, with high-resolution optical tools for contactless control and monitoring of electrical activity in excitable cells. Expression of light-sensitive depolarizing ion channels, such as channelrhodopsin-2 (ChR2) in excitable cells, enables the optical induction of APs [
10,
26]. Following the tremendous advances achieved in neuroscience, optogenetics has been successfully extended to cardiac research [
13]. Optogenetics-based strategies have been proposed as alternatives to wired electrical stimulation for cardiac pacing [
6,
30] and cardioversion [
4,
5,
9,
11,
16,
31,
36,
38]. Moreover, optogenetics recently emerged as a robust tool for investigating wave dynamics in cardiac tissue, studying the mechanisms underlying the induction, maintenance and control of cardiac arrhythmias [
7,
12,
14,
40].
Importantly, optogenetic interventions have so far been mostly used for generating transient and intense depolarizing currents for APs triggering or cardioversion. However, the utility of ChR2 for imposing a continuous depolarizing current with amplitudes that are too low to elicit APs (sub-threshold illumination), but are sufficient to fine-tune AP electrical dynamics has not been fully investigated. Recent studies performed primarily
in silico revealed novel insights about the use of sub-threshold illumination to destabilize and terminate spiral waves in a two-dimensional (2D) model of adult mouse ventricle [
20]. Moreover, sub-threshold stimulation has been used to manipulate the shape of cardiac APs in human atrial models at different spatial scales [
21]. In our present work, we explore the effects of sub-threshold illumination both in ex vivo and
in silico experiments. We first characterized the electrophysiological response to sub-threshold illumination of single cardiomyocytes isolated from mice expressing ChR2 (under the control of a cardiac-specific α-myosin heavy chain promoter) using patch clamp techniques. Next, using an optical mapping system operating in the near-infrared range, in combination with a stimulator generating customisable patterns of blue light, we characterized AP properties in isolated mouse hearts expressing ChR2. We found that sub-threshold illumination alters the shape of APs, allowing us to create spatiotemporal heterogeneities in AP properties across the heart in a completely reversible manner. More importantly, we also found that sub-threshold illumination promotes changes in the dynamics of cardiac electrical activity: sub-threshold illumination increases cardiac alternans, which we studied in the context of self-termination of ventricular tachycardia (VT).
Materials and methods
Mouse model generation
Transgenic mice (ChR2-mhc6-cre +) with cardiomyocyte-specific expression of ChR2 (H134R variant) and control (CTRL) mice (ChR2-wtwt-cre +) were generated [
46] and employed in this study. All animal handling and procedures were performed in accordance with the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes. The experimental protocol was approved by the Italian Ministry of Health (protocol number 944/2018-PR).
Cell isolation and patch clamp recording
Ventricular cardiomyocytes from CTRL and ChR2 mice were isolated by enzymatic dissociation as previously described [
39]. Briefly, mice (6 months old) were heparinized (0.1 mL at 5000 units/mL) and anesthetized by inhaled isoflurane (5%). The excised heart was immediately bathed in cell isolation buffer and the proximal aorta was cannulated for perfusion in a Langendorff system. The buffer solution contained (in mM): 120 NaCl, 1.2 MgCl
2, 10 KCl, 1.2 KH
2PO
4, 10 glucose, 10 HEPES, 20 taurine, 5 pyruvate, pH 7.4 (adjusted with NaOH), oxygenated with oxygen. After perfusion at 36 °C for 15 min with a constant flow of 3 mL/min, the solution was then switched to a recirculating enzyme solution, made from the same buffer with the addition of 0.1 mg/mL Liberase
™ (Roche Applied Sciences, Penzberg, Germany). After 8 min, the ventricles were excised and cut into small pieces in buffer solution added with 1 mg/mL of bovine serum albumin (BSA). Gentle stirring was used to further facilitate dissociation of myocytes. The cell suspension was left to settle, and the cell pellet was resuspended in Tyrode buffer, containing (in mM): 133 NaCl, 4.8 KCl, 1.2 MgCl
2, 10 glucose and 10 HEPES, pH 7.4 (adjusted with NaOH). The calcium concentration of the cell suspension was gradually increased to 0.6 mM by adding 15 μL of CaCl
2 (0.1 M). Finally, cardiomyocytes were superfused with Tyrode buffer containing 1.8 mM CaCl
2 during patch-clamp experiments. Patch-clamp data recordings and analysis were performed as previously described [
8] using a Multiclamp700B amplifier in conjunction with pClamp10.0 and a DigiData 1440A AD/DA interface (Molecular Devices, San Jose, CA, USA). For resting membrane potential (V
rest) and AP recordings, the pipette solution contained (in mM): 130 potassium aspartate, 0.1 Na-GTP, 5 Na
2-AT, 11 EGTA, 5 CaCl
2, 2 MgCl
2, 10 HEPES (pH 7.2 with KOH). Intracellular access was obtained via whole-cell ruptured patch. All experiments were performed at 36 ± 0.5 °C. APs were electrically elicited by inward current injection (3 ms current square pulses, 500–1000 pA) at a stimulation frequency of 1 Hz. To assess cell excitability, we increased the inward current pulse (3 ms duration) gradually (50 pA per step) until an AP was induced. Membrane resistance (
Rm) was measured in the voltage clamp mode (− 80 mV) applying a double step of ± 10 mV. For sub-threshold ChR2 activation, the cells were globally illuminated using a light emitting diode (LED) operating at a wavelength centered at 470 nm (SPECTRA X light engine, Lumencor, Beaverton, OR, USA) and a × 20 objective (NA; 0.5, HCX PL FLUOTAR, Leica Microsystems, Wetzlar, Germany). Light intensities (LIs) were measured at the sample site using a photodiode sensor (PD300-3 W, Ophir Optronics, Jerusalem, Israel).
Isolated and perfused mouse hearts
The excised heart was immediately bathed in Krebs–Henseleit (KH) solution and cannulated through the aorta. The KH buffer contained (in mM): 120 NaCl, 5 KCl, 2 MgS
2 O
4—7H
2O, 20 NaHCO
3, 1.2 NaH
2PO
4—H
2O, 1.8 CaCl
2 and 10 glucose, pH 7.4 when equilibrated with carbogen. Cardiac contraction was inhibited during the entire experiment with 5 μM blebbistatin (Enzo Life Sciences, Farmingdale, NY, USA) in the solution. The cannulated heart was perfused through the aorta (using a horizontal-Langendorff perfusion system) with the KH solution and then transferred to a custom-built optical mapping chamber at a constant flow of 2.5 mL/min at (36 ± 0.5) °C. Two platinum electrodes were placed below the heart for monitoring cardiac electrical activity via electrocardiogram (ECG). 1 mL of perfusion solution containing the voltage sensitive dye (VSD; di-4-ANBDQPQ, 6 μg/mL, University of Connecticut Health Center, Farmington, CT, USA) [
24] was bolus injected into the aorta. All the experiments were performed at (36 ± 0.5) °C within 1 h after dye loading to avoid potential re-distribution of the dye and accumulation of phototoxic by-products.
All-optical imaging and manipulation platform
Optical mapping and control were performed using a custom-made mesoscope. The whole mouse heart was illuminated in a wide-field configuration using a 2 × objective (TL2x-SAP, Thorlabs, Newton, NJ, USA) and a LED operating at a wavelength centered at 625 nm (M625L3, Thorlabs, Newton, NJ, USA), followed by an excitation band-pass filter at 640/40 nm (FF01- 640/40–25, Semrock, Rochester, NY, USA). The heart was illuminated with a maximum intensity of 1 mW/mm
2. A dichroic beam splitter (FF685- Di02-25 × 36, Semrock) followed by a band-pass filter at 775/140 nm (FF01-775/140–25, Semrock) was used for collecting the VSD-emitted fluorescent signal. A × 20 objective (LD Plan-Neofluar × 20/0.4 M27, Carl Zeiss Microscopy, Oberkochen, Germany) was used to focus the fluorescent signal on a central portion (128 × 128 pixels) of the sensor of a sCMOS camera (OrcaFLASH 4.0, Hamamatsu Photonics, Shizuoka, Japan) operating at a frame rate of 1 kHz (1 ms actual exposure time). The detection path allows a field of view (at the object space) of 10.1 × 10.1 mm
2 sampled with a pixel size of 80 μm. To manipulate cardiac electrical activity, a Lightcrafter 4500 projector (Texas Instruments, Dallas, TX, USA), operating at a wavelength of 470 nm, was used for projection of user-defined light patterns onto the heart surface. LIs were measured at sample site using a photodiode sensor (PD300-3 W, Ophir Optronics, Jerusalem, Israel). The system was used to optically probe AP propagation in mouse hearts during sub-threshold illumination using user-defined illumination patterns (whole heart, right half, and left half of the heart). Hearts were electrically paced at the apex with a bipolar electrode using an isolated constant voltage stimulator (DS2A, Digitimer, Welwyn Garden City, Hertfordshire, UK). As described before [
17], the optical platform was implemented with a custom LabVIEW software program (LabVIEW 2015, Version 15.0 64-bit, National Instruments, Austin, TX, USA), allowing it to mimic re-entrant VT during sub-threshold optogenetic illumination. Briefly, the apex of the heart was electrically paced, and the induced excitation wave propagated toward the base of the heart. Once an AP was optically detected in a region of interest (ROI) placed at the base of the heart, a new trigger was generated at the apex after a user-defined fixed delay, thus restarting the cycle. For each delay time (DT), re-entrant VT was established for 10 s.
Data and image analysis
All programs for data acquisition and analysis were developed with LabVIEW (National Instruments). For optical recordings, ΔF/F0 imaging of cardiac electrical activity was performed by processing raw data: for each frame, the mean baseline was subtracted, and the frame was subsequently normalized to the mean baseline, yielding a percentage change in fluorescence over time. For each heart, AP kinetics parameters were measured, trace by trace, to get the mean values after averaging five to ten subsequent trials. AP maximum rising slope (APRS), AP duration (APD) at 50% of repolarization (APD50), APD at 70% of repolarization (APD70), and APD at 90% of repolarization (APD90) were measured in a selected region of interest (ROI) of 10 × 10 pixels (≈ 1 mm2). APD was determined relative to the time of maximum depolarization. During slow pacing, APDs were measured considering the diastolic potential preceding the beat, while during fast pacing bursts and VTs (where a full repolarization could not be measured), AP parameters were measured relative to fluorescence baseline before and after the stimulation burst. In VTs, conduction time (CT) was calculated as cycle length (CL)-DT. APRS, APD50, APD70 and CT alternans were calculated using the following formula: \((\sum_{i=1}^{n-1}\left|X\text{i+1}-X{\text{i}}\right|)/(n-\) 1), where X is the parameter of interest. APRS and APD90 were also analyzed across the whole ventricle after a spatial binning of 4 × 4 pixels, generating maps containing APRS and APD90 values across the ventricle. In addition, spatial dispersion of these parameters were assessed by calculating the standard deviation (SD) of values across all pixels. Conduction velocity (CV) was calculated after a spatial binning of 4 × 4 pixels using a multi-vector approach: a seed reference pixel was arbitrarily chosen, and the cross-correlation of the fluorescence trace was calculated pixel by pixel, to estimate the temporal shift among every pixel (activation map). Next, local velocity maps were generated by calculating the delay between adjacent pixels divided by the pixel size. Since the local direction of AP wave-front is represented by a vector for each pixel, the mean CV was calculated by averaging local CVs. Graphic representation of data was obtained using OriginPro 2018, version 9.5 64-bit (OriginLab Corporation, Northampton, MA USA).
Numerical study
The effect of sub-threshold illumination was also investigated using an ionic mathematical model of the optogenetically modified adult mouse ventricular monolayer. Electrical activity in a single cardiac cell is modeled according to Eq.
1:
$$dV/dt \, = \, - \, \left( {I_{ion} + \, I_{stim} } \right)/C_{m} ,$$
(1)
where
V is the transmembrane voltage that arises from ionic gradients that develop across the cell membranes,
Cm is the membrane capacitance of each cell and
Istim is the electrical stimulation current. The total ionic current
Iion flowing across the membrane of a single cell is mathematically described using the electrophysiological model of an adult mouse ventricular cardiomyocyte, introduced by Bondarenko, including the model improvements in Petkova-Kirova [
3,
32]. The 15 total currents flowing through the cell membrane are: the fast Na
+ current (
INa), the L-type Ca
2+ current (
ICa,L), the Ca
2+ pump current (
IpCa), the rapidly recovering transient outward K
+ current (
Ito,f), the slowly recovering transient outward K
+ current (
Ito,s), the rapid delayed rectifier K
+ current (
IKr), the ultrarapid delayed rectifier K
+ current (
IKur), the non-inactivating steady-state voltage-activated K
+ current (
Iss), the time-independent inwardly rectifying K
+ current (
IK1), the slow delayed rectifier K
+ current (I
Ks), the Na
+/Ca
2+ exchange current (
INa/Ca), the Na
+/K
+ pump current (
INa/K), the Ca
2+ activated Cl
− current (
ICl(Ca)), the background Ca
2+ current (
ICa,b) and the background Na
+ current (
INa,b).
In a ventricular monolayer (2-dimentional (2D) domain), cardiac cells communicate with each other through intercellular coupling. The membrane voltage is then modeled using a reaction–diffusion equation (Eq.
2):
$$dV/dt \, = \nabla .(D\nabla V) \, - \, \left( {I_{ion} + \, I_{stim} } \right)/C_{m} ,$$
(2)
The first term on the right-hand side of the equation controls the intercellular coupling. D is the diffusion tensor, assumed here to be a scalar and set to the value 0.0014 cm2/ms to obtain isotropic plane wave propagation with a velocity of 42 cm/s. In this 2D simulation domain, spatial and temporal resolution are considered with values of 0.025 cm and 10–4 ms, respectively. We also applied a no-flux boundary condition at the unexcitable borders of this 2D region.
To make this model light responsive, we added the mathematical model of channelrhodopsin-2 (ChR2) [
44] to this ionic cell model of ventricular mouse heart. This photo-cycle model describes the dynamics of a non-selective cation channel ChR2 that responds to blue light with a wavelength of 470 nm. The inward ChR2 current (I
ChR2) is mathematically described by the following equation (Eq.
3):
$$I_{chR2} = \, g_{ChR2} G\left( V \right)\left( {O_{1} + \, \gamma O_{2} } \right)\left( {V \, - \, E_{ChR2} } \right).$$
(3)
Here,
gChR2 is the conductance with value of 0.4 mS/cm
2, G(V) is the voltage rectification function, O
1 and O
2 are the open state probabilities of the ChR2, γ is the ratio of conductance of O
2/O
1 with value of 0.1, and E
ChR2 is the reversal potential of this channel with value of 0 mV. The detailed description and values of other parameters can be found in [
44]. By including the mathematical model of ChR2 kinetics in the monolayer model of ventricular mouse heart, it is possible to simulate the effects of light on the ChR2 expressing monolayer at the 2D mono-domain level. To investigate the effect of sub-threshold illumination on the velocity of a propagating planar wave, a 2D domain with size of 2.5 × 2.5 cm
2 was continuously illuminated globally at different LIs (0, 0.005, 0.010, 0.0153, 0.020, 0.025, 0.030 mW/mm
2). Then we measured the CV of the planar wave by measuring the time the wave travels through two spatially distinct points with coordinates of
X: 0.625 cm,
Y: 1.25 cm and
X: 1.875 cm,
Y: 1.25 cm. Global and structured illumination patterns were used to study the morphology of an excitation wave AP under sub-threshold illumination. In both cases, planar waves were triggered by a sequence of electrical pulses with a strength of 80 pA/pF, a pulse length of 0.5 ms, and a stimulation frequency of 5 Hz on the left side of the domain. Then, for the case of a global illumination pattern, we measured the APD90 and APA for an ROI selected in the center of the domain with a coordinate of (
X: 1.25 cm,
Y: 1.25 cm) while the planar wave passes through this single point. To visualize the difference in the CV of a planar wave propagating in illuminated and non-illuminated regions, we used a structured illumination pattern. To do this, we illuminated half of the area where the planar wave propagates perpendicular to the intersection of illuminated and non-illuminated regions.
Statistics
For each experimental condition, data from one cell (in patch clamp measurements) or one heart (optical mapping measurements) was averaged, and this average was used for comparison and statistical analysis. Two-way repeated measures (RM) analysis of variance (ANOVA) tests were used to compare electrophysiological features between CTRL and ChR2 mice at different LIs. This method not only assessed the main effect of each categorical independent variable but also determines if there is any interaction between them, since the effects on the outcome of the change in one factor may depend on the magnitude of the other factor. For the comparison of means at specific LIs, the Tukey’s post hoc analysis was used. To investigate the general influence of illumination on AP features in CTRL and ChR2 mice, a regression test was applied: an ANOVA test was used to assess if the fitting function (linear or exponential) is significantly better than a constant function. In addition, the unpaired Student’s t test was used to compare two experimental groups, without another variable. A p value of < 0.05 was considered as indicative of a statistically significant difference between means (NS: p > 0.05; *p < 0.05; **p < 0.01, ***p < 0.001, ****p < 0.0001). Statistical analysis was performed using OriginPro 2018, version 9.5 64-bit and GraphPad Prism, version 8.4.3.
Discussion
In the present work, sub-threshold optogenetic stimulation was used to destabilize re-entrant arrhythmias in an experimental model system. ChR2 activation elicits an inward current proportional to the level of irradiance. Our experiments and simulations investigated the consequences of this effect at different scales, ranging from the single cardiomyocyte to the whole heart, by using transgenic mice constitutively expressing ChR2.
Illumination enabled manipulation of several electrophysiological parameters in ChR2 expressing cardiomyocytes. Light-induced effects were totally and immediately reversible. It’s important to stress that the light-induced effects were exclusively related to ChR2 activation; indeed, illumination did not impact electrical activity in cells from CTRL mice. Both at rest and in the presence of electrically stimulated APs, we observed gradual membrane depolarization, whose magnitude was a function of irradiance. Stable V
rest values were immediately established when constant illumination was applied, which is probably due to the gating properties of ChR2: at the onset of the light stimulus the channel opens rapidly, and then desensitizes to a smaller steady-state conductance during continued illumination [
28]. Optogenetic illumination allowed us to obtain graded “sub-threshold” (i.e., that failed to trigger an AP) depolarization over a rather wide range of potentials (Fig.
1). This is likely due to the gradual increase of V
rest afforded by the protocol, which may inactivate Na
+ channels before they enter an auto-regenerative loop with membrane potential. Membrane resistance (
Rm) at a diastolic potential did not vary substantially during sub-threshold illumination, albeit it tended to decrease when light was stitched on. At rest potential, membrane conductivity is too large to see any significant variations caused by the opening of a few ChR2 channels. A significant decrease of
Rm by ChR2 opening could potentially be observed at more positive potentials, when inward rectified K
+ current (I
K1) is inactivated. However, AP plateau is very short in mouse cardiomyocytes, and measuring resistance during AP repolarization is therefore technically challenging. The light-activated
IChR2 also allowed AP manipulation in terms of amplitude and duration, in an irradiance-dependent manner. Indeed, upon increasing the irradiance, we observed a gradual reduction of APA and APRS. These effects can be explained by partial inactivation of Na
+ channels, secondary to
Vrest depolarization. Accordingly, a refractoriness prolongation was also observed. Sub-threshold illumination also caused an overall APD prolongation during both continuous illumination and during time-specific illumination, where the sub-threshold stimulus is applied exclusively during the repolarization phase of AP. This response is consistent with the I
ChR2 reversal potential [
27]. Mouse APs display a fast initial repolarization which rapidly brings the membrane potential below 0 mV, causing most of the repolarization phase to occur at membrane potentials below 0 mV. As
IChR2 is inward below 0 mV (i.e., during the repolarization phase of AP in mouse heart), it opposes the repolarizing K
+ currents, thereby slowing down repolarization kinetics.
The effects of illumination in intact hearts expressing ChR2 were consistent with those observed in isolated myocytes. We found a gradual decrease of the APRS as a function of increased irradiance, which is consistent with the depression of the fast depolarization rate observed in isolated cardiomyocytes (Fig.
1E). Likewise, prolongation of APD was found in a light intensity-dependent manner, which is consistent with the single cell experiments as well as the computational study. Upon increasing irradiance, in intact hearts we also found a reduction in CV, likely the consequence of partial I
Na inactivation by
IChR2-induced
Vrest depolarization [
23]. Of note, in CTRL mice, a similar value of APD90 was found in patch-clamp and optical mapping recording, suggesting that the VSD and blebbistatin employed in Langendorff experiments do not significantly perturb the electrophysiology of the heart.
We also investigated how changes in cardiac electrical activity promoted by I
ChR2 could be affected by the pacing rate. We found a strong rate dependency of APD50. In fact, during fast repolarization (AP phase1), higher pacing rates results in more time above the reversal potential of the ChR2 channel (≈ 0 mV) where I
ChR2 is outward rather than inward, thus giving rise to a repolarizing current which counteracts AP prolongation [
13]. More importantly, we found that sub-threshold illumination also affects the dynamics of cardiac electrical activity. The physiological beat-to-beat oscillations in APRS and APD, which usually occurs at fast beat rates, were increased by sub-threshold illumination and this effect was more pronounced at high pacing rates. This result may be related to the optogenetic manipulation of the electrical restitution curve (supplemental figure S1). Indeed, the introduction of the depolarizing current I
ChR2 may result in a slower ion channel recovery from inactivation, eventually leading to an increase in the slope of the restitution curve which would cause an increase in cardiac alternans amplitude [
29].
The pro-arrhythmic effect of electrical alternans has been extensively demonstrated in various cardiac preparations [
43]. Paradoxically, electrical alternans has also been observed before spontaneous termination of re-entrant rhythms [
15]. In a recent study, Biasci and coauthors developed a simplified mathematical model capable of reproducing the electrical dynamics occurring in re-entrant rhythms and demonstrated how alternans are involved in generating non-sustained bursting rhythms [
1]. Briefly, when CT and APD oscillate between beats, a stimulus delivered after a beat with a short CT and long APD will encounter a much shorter recovery time than the preceding stimulus. Consequently, the termination of re-entry based arrhythmias occurs preferentially at the stimulus site following a beat with a short CT and a long duration.
Based on this theoretical study, here we employed sub-threshold illumination as a tool to increase cardiac alternans during an ongoing VT and we experimentally dissected the role of alternans in the context of self-termination of VT. In this respect, we exploited the capability of our optical platform to electrically generate re-entrant VTs in ChR2 mouse hearts using real-time feedback-control. Our platform allowed us to test the effects of sub-threshold illumination during an ongoing VT. In general, several approaches can be used to experimentally induced re-entrant VTs, including ischemia and reperfusion, catecholamine infusion, sodium pentobarbital or caffeine. However, these methods do not allow direct control of the re-entrant circuit length and sometimes are not sufficient to induce re-entrant arrhythmias due to the small size of the mouse heart [
25]. In contrast, our strategy aims to generate user-defined re-entrant circuits with different cycle lengths (CLs) depending on the delay time (DT) of the electrical stimulus. We found a greater tendency for VTs to spontaneously terminate when sub-threshold illumination was applied, suggesting the presence of one or more light-induced mechanisms leading to spontaneous termination. We found that DTs at which illumination promotes VT self-termination are associated with larger light-induced enhancements of APD and CT alternans. The larger oscillations lead to a higher probability of VT termination, which consistently occurred following a beat with a short CT and long AP. More importantly, the relationship between electrical alternans magnitude and the probability of block associated with the combination of a short CT and a long AP was present regardless of illumination, suggesting that alternans may play a role in terminating re-entrant rhythms. Although the current work deals with VTs modeled by mono-dimensional re-entrant circuits, the same mechanism can in principle also occur in other much more complex systems (occurring in larger hearts, e.g., human) where other termination mechanisms (such as collision of multiple spiral waves) can concomitantly occur [
2]. Future investigations should be focused on exploring the role of alternans in more complex geometries. In this respect, panoramic [
33] or volumetric [
37] imaging could provide a more comprehensive view of cardiac dynamics across the whole heart surface as well as within ventricle walls, expanding the epicardial observations reported in this study. Furthermore, investigating the involvement of calcium oscillations during the light-mediated increase of APD alternans is essential, especially considering that altered calcium homeostasis often underlies the development of AP alternans in (patho)physiological conditions [
45].
In conclusion, our results support the idea that electrical alternans is the main mechanism for self-termination of re-entry related tachycardias. Pharmacological interventions aimed at increasing the likelihood of electrical alternans at high pacing rates, such as the use of class 1a Na
+-channel blockers which typically reduce conduction velocity and increase APD and refractoriness in the human heart [
22], may mimic the effects of sub-threshold illumination, thus explaining their efficacy in reducing the risk of sustained arrhythmias in patients.