Utilizing Human-Induced Pluripotent Stem Cells to Study Cardiac Electroporation Pulsed-Field Ablation

BACKGROUND: Electroporation is a promising nonthermal ablation method for cardiac arrhythmia treatment. Although initial clinical studies found electroporation pulsed-field ablation (PFA) both safe and efficacious, there are significant knowledge gaps concerning the mechanistic nature and electrophysiological consequences of cardiomyocyte electroporation, contributed by the paucity of suitable human in vitro models. Here, we aimed to establish and characterize a functional in vitro model based on human-induced pluripotent stem cells (hiPSCs)-derived cardiac tissue, and to study the fundamentals of cardiac PFA. METHODS: hiPSC-derived cardiomyocytes were seeded as circular cell sheets and subjected to different PFA protocols. Detailed optical mapping, cellular, and molecular characterizations were performed to study PFA mechanisms and electrophysiological outcomes. RESULTS: PFA generated electrically silenced lesions within the hiPSC-derived cardiac circular cell sheets, resulting in areas of conduction block. Both reversible and irreversible electroporation components were identified. Significant electroporation reversibility was documented within 5 to 15-minutes post-PFA. Irreversibly electroporated regions persisted at 24-hours post-PFA. Per single pulse, high-frequency PFA was less efficacious than standard (monophasic) PFA, whereas increasing pulse-number augmented lesion size and diminished reversible electroporation. PFA augmentation could also be achieved by increasing extracellular Ca2+ levels. Flow-cytometry experiments revealed that regulated cell death played an important role following PFA. Assessing for PFA antiarrhythmic properties, sustainable lines of conduction block could be generated using PFA, which could either terminate or isolate arrhythmic activity in the hiPSC-derived cardiac circular cell sheets. CONCLUSIONS: Cardiac electroporation may be studied using hiPSC-derived cardiac tissue, providing novel insights into PFA temporal and electrophysiological characteristics, facilitating electroporation protocol optimization, screening for potential PFA-sensitizers, and investigating the mechanistic nature of PFA antiarrhythmic properties.

and complex phenomenon.It may be transient, with the cellular membrane subsequently resealing (reversible electroporation), which may be used for drug or gene delivery while maintaining cell viability.[10][11][12][13][14] Although preclinical work and recent clinical studies found PFA to be both safe and efficacious, [8][9][10][15][16][17] there are significant knowledge gaps concerning cardiac PFA, partially due to paucity of suitable human in vitro cardiac tissue models. Beyon the fact that currently used PFA protocols are not fully disclosed by the industry and the emerging concerns regarding coronary spasm/stenosis 18,19 or phrenic nerve injury, 16 there are many additional aspects requiring further study, including: PFA delivery protocols and equipment optimization, electroporation reversibility thresholds and time constants, tissue selectivity, and possible PFA modifiers or sensitizers, PFA lesion characteristic and their long-term sustainability, the characterization of PFA-mediated cellular damage, and the mechanistic nature of PFA antiarrhythmic properties.Addressing these issues is crucial for further expansion of PFA in the cardiac clinical setup.
The human-induced pluripotent stem cell (hiPSC) technology allows reprogramming of patient-specific somatic cells into pluripotent stem cells, that can be coaxed to differentiate into cardiomyocytes. 20,21][24][25][26] Moving from the cellular level to more complex 2-and 3-dimensional models, the use of advanced optical-mapping techniques [27][28][29][30] allowed the detailed study of cardiac electrophysiological properties using hiPSC-CMs.][29][30] With the advancements made in the differentiation and maturation of hiPSC-CMs, it has been consistently shown that hiPSCs may be used to model primary human cardiomyocytes owing to their similar electrophysiological, molecular, structural, and mechanical properties. 31lthough of clinical and scientific relevance, it is noteworthy that nonhuman or heterologous expression system models may differ significantly from human cardiac tissue in their expression profile of ion channels, calciumhandling proteins, and other structural and functional proteins. 32Recently, hiPSC-based cellular models were WHAT IS KNOWN?
• Pulsed-field ablation (PFA) is emerging as a nonthermal ablation modality for arrhythmia treatment.• Although initial clinical studies found PFA both safe and efficacious, there are significant knowledge gaps concerning the mechanistic nature and electrophysiological consequences of cardiomyocyte electroporation, contributed by the paucity of suitable human in vitro models WHAT THE STUDY ADDS

HF high frequency hiPSC
human-induced pluripotent stem cell hiPSC-CCS human-induced pluripotent stem cellderived cardiac circular cell sheet hiPSC-CM human-induced pluripotent stem cellderived cardiomyocyte PFA pulsed-field ablation PI propidium iodine used to study PFA cardioselectivity and irreversibility thresholds using cell-viability markers, not addressing however electrophysiological parameters. 13,33In the current study, we aimed to harness the advancements made in using hiPSC-derived cardiac tissue models in electrophysiology research to study the fundamentals of cardiomyocyte electroporation, with a specific focus on the impact of PFA on the functional and electrophysiological properties of these models.

METHODS
The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Cardiomyocyte Differentiation and Generation of hiPSC-Derived Cardiac Cell Sheets
Cardiomyocyte differentiation of a previously established healthy-control hiPSC line 22 was achieved using the monolayerdirected differentiation system as previously described. 27,29The generation and use of the hiPSC lines and their differentiated cardiomyocytes was approved by the institutional review board (Helsinki) committee of the Rambam Medical Center (0170-019-RMB), and subjects gave their informed consent for hiPSC generation and use.To generate hiPSC-derived cardiac circular cell sheets (hiPSC-CCSs), differentiating beating monolayers (14-60 days) were enzymatically dissociated and seeded as circular cell sheets (≈0.5-1.5 cm diameter) on Matrigel-coated culture plates, at a seeding density of 0.5 to 1.5×10 6 cells in 150 μL drops, similar to previous reports. 27,28Tissues were cultured in RPMI/B27 containing 1% penicillin/streptomycin and blebbistatin (5 μmol/L).

PFA Experiments
Pulsed fields were delivered using a standard electroporation pulse generator (BTX-ECM830, Holliston, MA) or a custommade generator allowing the delivery of high-frequency, highvoltage protocols. 34Pulses were delivered via 2 needle-shaped electrodes (BTX-450168, 0.5 cm inter-electrode distance), by direct contact with the hiPSC-CCSs.The detailed PFA protocols are described in Figure 1.
PFA was delivered as a train of 20, 40, 50, 60, or 120 sequential pulses, depending upon the specific experiment.The waveform used was either monophasic or high-frequency biphasic, as specified in each experiment.Each single pulse duration was 100 μsec, and pulses were delivered at 1 Hz repetition frequency (Figure 1).In experiments incorporating point by point ablation, pulse trains were delivered in consecutive locations across the culture midline.[29]

Optical Mapping
8][29] The health of the preparations was assessed functionally by documenting typical electric Standard (monophasic) and high-frequency (HF) biphasic PFA (HF-PFA) protocols were used and compared.Monophasic pulses were delivered as square pulses of 550 V, and 100 µsec duration per single pulse.Trains of 20 pulses were delivered sequentially at 1 Hz frequency (1 pulse delivered each second), via 2 needle-shaped electrodes with 0.5 cm inter-electrode spacing.HF biphasic pulses were delivered as biphasic bursts of ±550 V, and 100 µsec duration per single pulse, with 150 kHz biphasic frequency, and no isoelectric delay between phases (hence each single pulse consisted of 15 biphasic bursts, with 3.33 µsec positive and negative phases occurring interchangeably).Depending upon the experiment, trains of 20, 50, 60, or 120 pulses were delivered sequentially at 1 Hz frequency (1 pulse delivered each second) via 2 needle-shaped electrodes with 0.5 cm inter-electrode spacing.

Finite Element Modeling
To calculate the electric field distribution within the hiPSC-CCSs, a 2-dimensional finite elements model 36 was constructed using COMSOL Multiphysics 5.3a (COMSOL Multiphysics; Stockholm, Sweden).The 2-dimensional geometry was designed to match the dimensions of the experimental setup.The well was modeled as a 2×2 cm 2 to match the macroscope's field of view.The hiPSC-CCS was modeled as a circle (0.8 cm radius) at the center of the well and the stainless steel electrodes were modeled as 2 circles (0.05 cm radius) positioned 0.5 cm apart.The medium electric conductivity 13 was set at 2.3 S/m and the electrodes conductivity was set to 1.7×10 6 S/m.The electric field was described by the Laplace equation for electric potential distribution in a volume conductor: (1) ∇ • (σ∇ϕ) = 0, where σ is the electric conductivity of the culture, and ϕ is the potential.Dirichlet boundary condition was applied to the surface of the electrode: (2) ϕ = ϕ 0 and to the ground (3) ϕ = 0, where ϕ 0 is the applied potential on the active electrode.The boundaries where the analyzed domain was not in contact with an electrode were treated as electrically isolative and Neumann boundary condition was set to zero on the outer border of the model: (4) ∂ϕ ∂n = 0, where n denotes the normal to the boundary and ϕ is the potential.
To calculate the electric field thresholds, the lesion areas as derived from optical mapping images were correlated with the electric field contours using custom-written MATLAB software (MathWorks, Inc, MA).

Flow-Cytometry Experiments
MEBCYTO apoptosis kit (MBL) was used to evaluate cell death.Propidium iodine (PI) is a cell-death marker that binds DNA while penetrating dead or dying cells.Annexin-V binds to phosphatidylserine, which is translocated to the outer membrane portion of cells undergoing programmed/regulateddeath. Hence, combining Annexin-V and PI staining may allow differentiating between accidental/necrotic and programmed/ regulated cell-death mechanisms.Untreated control hiPSC-CCSs, and hiPSC-CCSs immediately (T 0 ), and 24 hours following PFA were enzymatically dissociated, washed with PBS, resuspended in binding buffer, and loaded with PI, and Annexin-V-FITC (15 minutes).Over 50 000 cells were analyzed in each sample, using the LSR Fortessa flow cytometer (BD-Biosciences).Data were analyzed using FlowJo software.

Statistical Analysis
PFA lesion areas are reported as mean±SD.Comparisons of cell-death markers in flow-cytometry experiments and of field intensity thresholds as derived from numerical modeling were performed using 1-way ANOVA, with Tukey's post hoc (GraphPad, Prism9).To analyze lesion recovery over time, a repeated-measures linear mixed effect model with an unstructured covariance matrix was applied.The model included fixed factors for group (electroporation protocol), time, group-by-time interaction, and a random effect for the individual experiments to accommodate culture-specific deviations in intercept.As lesion-recovery slope changes were not constant, we applied a 2-piecewise random coefficients model allowing separation of an early rapid recovery period from a later component, each with different slopes.This knot value was selected by fitting the model over a range of potential knots and selecting the model that had the lowest Akaike Information Criteria.P<0.05 was considered statistically significant.Mixed modeling was performed using Stata version 17.0 (College Station, TX).

Standard (Monophasic) PFA in hiPSC-CCSs
We evaluated the electrophysiological effects of monophasic PFA (Figure 1) in the hiPSC-CCSs.Compared with baseline recordings (Figure 2A; Video S1), we found that PFA instantly generated electrically silenced lesions within the hiPSC-CCSs (Figure 2B; Video S2), functioning as conduction barriers.Interestingly, beyond the lack of action-potential generation, PFA-affected areas were also characterized by a relatively depolarized resting membrane potential as compared with nonaffected areas, reflecting the disruption of membrane integrity and its normal electric function.Importantly, while conduction was blocked in the affected region, the surrounding tissue was electrically functional allowing for signal propagation around the lesion (Figures 2B  through 2D).
To correlate the experimental results regarding PFA impact with the electric field generated in the hiPSC-CCSs between the 2 needle-shaped electrodes, we used a numeric model.As can be appreciated in Figure 2E, the resulting electric field is of elliptical shape, with the strongest field located around each electrode, while tapering down with distance.Consistent with previous reports, 36 the electric field intensity and the culture area exposed to that field intensity could be fitted to an exponential function (Figure 2F, R 2 =0.99;P<0.001).
Since electroporation may be reversible, we hypothesized that cells subjected to lower field intensities could potentially recover, regaining normal electric function.We, therefore, evaluated PFA-lesion dimensions over time.Sequential optical mapping images (Figure 3A) demonstrate dynamic, time-dependent recovery of cardiomyocyte electric activity within parts of the PFA lesions, reflecting reversible electroporation.The elliptical lesion shape was concordant with the field intensity distribution as was predicted by computational modeling (Figure 2E).Reversible electroporation and recovery of cardiomyocyte electric activity occurred where field intensities were lower.Conversely, cardiomyocytes located in the region surrounding each electrode, or in the inter-electrode interface, remained electrically silent over time, reflecting irreversible electroporation (Figure 3A).Cardiomyocyte recovery was characterized by 2 phases.Most of the recovery occurred during the initial phase, within 5 minutes following monophasic PFA, whereas the remaining functional recovery occurred at a significantly slower rate (Figure 3B).
Lesion recovery kinetics also differed between the protocols.Although monophasic PFA generated the largest initial functional lesions (0.59±0.04 versus 0.42±0.06cm 2 for HF-PFA 120 pulses and 0.32±0.04cm 2 for HF-PFA 20 pulses; Figure 3B), the rapid cardiomyocyte recovery phase was also most prominent with this protocol.Hence, the slope of the rapid decrease in functional lesion area (within 5-minutes postablation) was significantly steeper with monophasic PFA

Lesion Sustainability at 24 Hours and Mechanisms of PFA-Mediated Cell Death
Considering the possibility of ongoing cardiomyocyte recovery and reversible electroporation, we evaluated the hiPSC-CCSs conduction properties at 24-hour post-PFA.The presence of continuous electric silencing and persistent conduction blocks was documented, reflecting irreversible electroporation (Figure 4A).Comparing optical maps obtained immediately post-PFA (left-panels, T 0 ) and 24 hours later (right panels), it is evident that further cardiomyocyte recovery had occurred over the period of hours.Hence, while cardiomyocytes in regions exposed to the highest field intensities (around each electrode and most of the interelectrode plane) were invariably subjected to irreversible electroporation, cardiomyocytes located further away exhibited reversible electroporation.Overall, lesion dimensions decreased significantly form 0.66±0.07cm 2 immediately following PFA to 0.15±0.04cm 2 at 24-hour post-PFA (n=10 different hiPSC-CCSs; P<0.01).
Numerical modeling revealed a 0.9±0.14kV/cm electroporation irreversibility threshold at 24-hour post-PFA.Sustainable and complete functional lesions were documented at 24 hours within the inter-electrode interface in 7/10 (70%) of the cultures following monophasic PFA (20 pulses; Figure 4A, top panel).Interestingly, field intensity at the central portion of the inter-electrode plane ranged between 0.8 and 1 kV/cm based on the numerical modeling (Figure 2E).This allowed for functional recovery with a midline breach of the conduction block in the remaining 30% of the specimens (Figure 4A; bottom panel).
We next aimed to characterize the mechanisms of PFA-mediated cell death.It was previously suggested that regulated cell death plays an important role in irreversible electroporation. 1,7 low cytometry was used to evaluate the proportion of PI and Annexin-V staining in dispersed cardiomyocytes derived from untreated control hiPSC-CCSs as compared with hiPSC-CCSs immediately (T 0 ), and 24-hour post-PFA.Accidental necrosis (PI positive/ Annexin-V negative staining) did not play a significant role neither immediately nor 24 hours following PFA (Figure 4B, left panel).On the contrary, late-phase regulated cell-death markers (PI positive/Annexin-V positive staining) significantly increased (rising >2-fold) at 24-hour post-PFA as compared with T 0 (Figure 4B; right panel).Importantly, there were no significant differences in Annexin-V and PI positive staining immediately following PFA (T 0 ) as compared with untreated control hiPSC-CCSs.Ca 2+ and Electroporation Sensitization As noted above, PFA lesions were relatively depolarized and electrically silent.Since Ca 2+ plays an important role in cardiomyocyte electrophysiology, and since membrane integrity is crucial for such normal function, we performed Ca 2+ imaging of Fluo-4 loaded hiPSC-CCSs following PFA and found increased Ca 2+ levels within PFA lesions (Figure 5A), consistent with recent reports. 39t was previously suggested that increasing cytoplasmatic Ca 2+ plays an important role in PFA-mediated cytolysis, [5][6][7] and that intracellular Ca 2+ levels significantly increase before electroporation induced cell death. 39We hypothesized that electroporation sensitization and augmentation could be achieved by increasing cardiomyocyte Ca 2+ entry and concentration following electroporation, through elevation of extracellular Ca 2+ levels.To test this hypothesis, we compared PFA lesion dimensions under standard conditions and in a Ca 2+rich environment (Tyrode containing 4 mM CaCl 2 ).Interestingly, increasing Ca 2+ levels mostly impacted later stages of cellular recovery, significantly augmenting lesion size by almost 30% (which was comparable to doubling the pulse number with standard Ca 2+ concentrations, Figure 5B).

Arrhythmia Targeting With PFA
Different forms of reentry underlie most atrial and ventricular arrhythmias.8][29][30] To evaluate arrhythmia targeting by PFA, we used hiPSC-CCSs plated at lower cell-density either focally or diffusely, facilitating the spontaneous emergence of single or multiple rotors (Figure 6A through 6J).PFA was delivered in a linear manner (using a point-by-point approach) to generate a continuous line of conduction block, spanning the entire diameter of the hiPSC-CCSs.The resulting PFA lesions eliminated arrhythmic rotors (or any electric activity) within the ablated area (Figure 6B through 6E, 6H, and 6I) and electrically isolated the opposing portions of the hiPSC-CCSs, creating a bi-directional block (Figure 6B through 6E, and 6H through 6J).In the case of a single spiral wave using the entire culture area (Video S3), such ablation was sufficient to terminate the arrhythmic activity (Figure 6B and 6E, Video S4).In the case of multiple rotors, ablation terminated arrhythmogenicity locally, and allowed for the isolation of the remaining arrhythmic activity to confined regions (Figures 6H through 6J).

DISCUSSION
In this work, we established and characterized a novel functional hiPSC-based in vitro model to study electroporation mechanisms and impact in human cardiomyocytes, while focusing on tissue electrophysiological characteristics.Using this model, we were able to evaluate multiple fundamental aspects of cardiac PFA, gaining insights into its mechanism of action.A, Ca 2+ -imaging of Fluo4-loaded human-induced pluripotent stem cell (hiPSC)-derived cardiac circular cell sheets (hiPSC-CCS) following PFA showing increased intracellular Ca 2+ -levels within the PFA lesion (asterisk).B, Lesion-area measurements over time following high-frequency (HF)-PFA, comparing standard and increased extracellular Ca 2+ concentrations (tyrode's containing 1.8 vs 4 mM CaCl 2 , respectively).Ca 2+ enriched HF-PFA (60 pulses) augmented lesion size at 60 minutes, an effect comparable to doubling the pulse number under standard Ca 2+ levels (*P<0.01 for HF-PFA-60 pulses [4 mM Ca 2+ ] vs HF-PFA-60 pulses, and P=0.7 vs HF-PFA-120 pulses).This effect was related to attenuated late-stage recovery, manifested by a significant decrease in recovery rate at 10 to 60 minUTES ( ‡P<0.01 for trace slope comparing HF-PFA-60 pulses-[4 mM Ca 2+ ] vs HF-PFA-60 pulses, and P=0.53 vs. HF-PFA-120 pulses).
Our main findings include: (1) PFA resulted in the formation of highly localized lesions, exhibiting both reversible and irreversible electroporation; (2) depending on the protocol, reversible electroporation could involve up to 67% of the initial lesion area at 60-minutes post-PFA; (3) significant reversible electroporation was documented within the first 5 to 15-minutes post-PFA; (4) per single pulse, HF-PFA was less efficacious than monophasic PFA, generating smaller lesions; (5) increasing pulse-number augmented lesion area and decreased reversible electroporation; (6) prolonged lesion sustainability (irreversible electroporation) was documented at 24-hours post-PFA, with an irreversibility threshold of ≈0.9 kV/cm; (7) regulated cell death plays an important role following PFA; (8) electroporation sensitization was achieved by increasing extracellular Ca 2+ levels during PFA; (9) PFA was effective in terminating arrhythmic activity and isolating arrhythmogenic substrate within the hiPSC-CCSs by generating sustainable lines of conduction block; (10) PFA effects stem from a combination of cellular electrophysiological changes (including membrane depolarization and increased intracellular Ca 2+ ), cellular damage, and cell death.
We were able to study reversible and irreversible electroporation temporal kinetics and characterize how they were influenced by protocol modifications.Significant cardiomyocyte recovery was observed following PFA, with resumption of electric activity in 40% to 67% of the initial functional lesion within 1-hour post-PFA.Consistent with previous reports, 33,34,38 protocol modifications could influence the extent of reversible electroporation, with a tight spatial correlation between the expected distribution of the electric field intensity and the eventual occurrence of irreversible electroporation.These observations have important clinical consequences, as the immediate development of localized electric silencing and functional conduction blocks does not necessarily mean long-term persistence of these lesions.Our findings also offer time frames that might be potentially relevant for the clinical setup with regards to reassessing the electric activity following initial PFA delivery.
Traditional electroporation protocols incorporated square monophasic pulses, typically of 50 to 100 μsec duration.Such impulses may also be associated with remote nerve stimulation and muscle twitching, limiting their clinical use 37 due to patient movements during ablation, patient-discomfort, or the requirement for general anesthesia and muscle relaxation. 8,10One approach to overcome this obstacle is using modified highfrequency electroporation protocols involving ultra-short bursts of biphasic pulses (<10 μsec/phase). 8,34,37,38ngoing efforts are being directed at comparing such different protocols in vitro and in vivo. 34Here, by comparing monophasic and HF-PFA protocols (Figure 1) we found that, per single pulse, HF-PFA was less efficacious generating smaller functional lesions and requiring higher field-intensity thresholds for lesion formation and sustainability.This may be related to decreased total time at which the trans-membrane potential is above irreversibility threshold, as well as with the specific HF-PFA pulse settings. 37ncreasing HF-PFA pulse number, thereby increasing the cumulative cardiomyocyte exposure to electric energy, had reduced the level of reversible electroporation and enabled achieving lesions comparable to monophasic PFA in size and field intensity threshold, at a pulse ratio of ≈1:6.These findings are consistent with previous in vivo studies, 34,41 and imply that HF-PFA protocols may achieve comparable results to monophasic-PFA, maintaining the significant advantage of reduced muscle stimulation.Interestingly, monophasic-PFA generated the largest initial lesions (as also reflected by the significantly lower threshold for acute lesion formation) while also characterized by the most prominent rapid recovery component, with significant reversible electroporation occurring within 5-minutes post-PFA, similar to previous reports. 39In contrast, HF-PFA protocols were associated with smaller initial lesions, but also a significantly slower initial recovery.Hence, cardiomyocyte recovery (and especially early phase reversible electroporation) is different between monophasic and HF-PFA.
Several clinically available PFA systems allow for biphasic-and HF-PFA.Nevertheless, such optimized biphasic protocols although used clinically, remain not fully disclosed by the industry, and each manufacturer offers different pulse setups and protocols which are modifiable by the operating clinician only to a limited extent.Hence, HF-PFA protocol advantages, disadvantages and temporal kinetics should be further studied, characterized, and standardized.Importantly, full disclosure of clinically used PFA protocols is needed, as even minor protocol changes may lead to significant differences in PFA outcomes.
Assessing longer term lesion sustainability, persistent conduction blocks were documented at 24-hour post-PFA, reflecting irreversible electroporation with an irreversibility threshold of ≈0.9 kV/cm.This is consistent with previous in vivo and in vitro reports, including recent studies utilizing hiPSC-CMs. 13,14,33,34,37,41In regions of the culture subjected to lower field intensities, a delayed recovery of electric activity could be detected.These findings further emphasize that initial electric silencing following PFA does not necessarily translate into durable lesions, and that recovery of electric activity may occur over the course of hours/days.Considering recent reports of 1-year clinical PFA outcomes, 16,17 our findings emphasize the importance of further research into identifying robust predictors of long-term lesion persistence during acute PFA application.
We demonstrated that pure necrotic cell death did not play a significant role nor immediately, neither late following PFA.Regulated cell-death markers, on the contrary, were significantly increased (>2-fold) at 24-hour post-PFA.These findings reflect the ongoing nature of the cumulative electroporation-induced cellular damage, leading to time dependent, gradual, and regulated cell death.Although cell-viability assays often rely on membrane integrity which may potentially limit their relevance in the context of electroporation, we did not observe significant differences in Annexin-V or PI staining comparing untreated control hiPSC-CCSs with hiPSC-CCSs collected immediately post-PFA.This excludes a potential bias attributed to membrane electroporation per se.
Our results highlight the different mechanisms that may underlie PFA electrophysiological effects.Although the late PFA effect probably stems from irreversible electroporation and regulated cell death, 7 the acute effects may stem from a combination of cellular electrophysiological changes and cardiomyocyte injury induced by membrane damage.Voltage-and calcium-based optical imaging showed that the acute PFA lesions were characterized by increased resting membrane potential and intracellular Ca 2+ levels, which may lead to electric silencing through Na + -channel inactivation.
We also demonstrated that by increasing extracellular Ca 2+ levels, we could augment PFA lesions to an extent comparable with doubling the pulse number under standard Ca 2+ concentrations.Interestingly, PFA augmentation with Ca 2+ occurred mainly during the later stages postablation and not immediately, which suggests a role for Ca 2+ in the gradual and regulated cellular damage induced by PFA.This PFA sensitization approach may allow for further clinical protocol optimization, for instance by irrigating the catheter tip with a Ca 2+ rich solution during ablation, thereby increasing its efficacy.
Using the hiPSC-CCS model, we demonstrated the effective targeting and elimination of single or multiple arrhythmic rotors with PFA, through the formation of sustainable bidirectional linear conduction blocks.Such continuous lesions were also able to isolate arrhythmogenic substrates within the cultures.Nevertheless, it is noteworthy that the role of rotors in clinical atrial fibrillation remains controversial and the interpretation of these in vitro findings to clinical arrhythmias should be done with caution.
Finally, the generation of targeted lesions sized ≤0.7 cm 2 , and the ability to affect only the midline portion of ≤3 cm 2 cultures, underscore the immense potential of PFA in targeting cardiomyocytes at a desired location with high precision.This is also supported by the exponential relationship between the field intensity and the culture area exposed to that intensity as revealed by numerical modeling.
Our study is not without limitations.We were unable to exclude significant thermal effects following PFA, which could be present especially in the immediate electrode vicinity, yet the protocols used were previously shown to cause only minor temperature changes. 34Although collecting the supernatant medium in flow-cytometry experiments to avoid loss of detached cells, we could not exclude some loss of fragmented cell debris during centrifugation, washes, and acquired data gating, which could potentially underestimate necrotic cell death.Additionally, characterizing the exact programmed/regulated cell-death pathway (apoptosis, necroptosis, etc), and the role of the immune system in cell death, was beyond the scope of this work.We evaluated lesion sustainability and regulated cell death within 24-hour post-PFA.As additional cellular recovery or ongoing cellular damage may potentially occur over the course of days, further studies are required assessing PFA lesions dynamics and durability over the course of days to weeks.Additionally, the irreversibility thresholds as derived from numerical models are influenced by multiple parameters affecting electroporation (such as voltage, pulse number, pulse shape, pulse duration, pulse polarity, and the electrode setup).Subsequently, such thresholds are protocol specific and may vary significantly between different protocols.Further research is warranted for the development of statistical models, accurately predicting the probability of irreversible electroporation as influenced by PFA protocols or tissue composition.
Further research is also warranted into the role of specific currents, including Na + , K + , and Ca 2+ in PFAmediated electric silencing, cellular recovery, and cell death.Interestingly, we noted occasional and transient increase in automaticity at the border zone of the PFA lesions, during cardiomyocyte recovery.This implies that PFA lesions may be transiently proarrhythmic, which also requires further investigation.
In summary, our study provides novel insights into PFA temporal and electrophysiological characteristics and facilitates electroporation protocol optimization, screening for potential PFA-sensitizers, and studying the mechanistic nature of its antiarrhythmic properties.It provides further support that hiPSCbased models may be used to study multiple aspects of PFA and should facilitate future bedside to bench and back studies in this rapidly evolving field of cardiac electrophysiology.

Figure 1 .
Figure 1.Summary of the pulsed-field ablation (PFA) protocols used in the current study.Standard (monophasic) and high-frequency (HF) biphasic PFA (HF-PFA) protocols were used and compared.Monophasic pulses were delivered as square pulses of 550 V, and 100 µsec duration per single pulse.Trains of 20 pulses were delivered sequentially at 1 Hz frequency (1 pulse delivered each second), via 2 needle-shaped electrodes with 0.5 cm inter-electrode spacing.HF biphasic pulses were delivered as biphasic bursts of ±550 V, and 100 µsec duration per single pulse, with 150 kHz biphasic frequency, and no isoelectric delay between phases (hence each single pulse consisted of 15 biphasic bursts, with 3.33 µsec positive and negative phases occurring interchangeably).Depending upon the experiment, trains of 20, 50, 60, or 120 pulses were delivered sequentially at 1 Hz frequency (1 pulse delivered each second) via 2 needle-shaped electrodes with 0.5 cm inter-electrode spacing.

Figure 2 .
Figure 2. Optical mapping of human-induced pluripotent stem cell (hiPSC)-derived cardiac circular cell sheets (hiPSC-CCSs).A, Sequential fluorescent images (from left-to-right panels) taken from dynamic optical mapping display of FlouVolt-loaded hiPSC-CCS during point electric stimulation.Note the uniform depolarization wave (white) signal propagation from the pacing site.B, The same hiPSC-CCS following monophasic PFA (20 pulses).Note that the PFA lesion (asterisks) is relatively depolarized (white), electrically inactive, and leads to conduction block.Subsequently, the paced impulse now propagates around the lesion (white arrows).C, Optical signal traces from 2 different regions of interest (ROI) within the same hiPSC-CCS showing typical action-potentials from the unaffected tissue (yellow ROI) and no electric activity within the PFA lesion (red ROI).D, Activation maps of the same hiPSC-CCS at baseline (top) and post-PFA (bottom).Note uniform conduction at baseline and the formation of a conduction block following PFA, with signal propagation around the lesion (arrows), including an early conduction block in the counterclockwise direction (oval arrow-cap).Isochrones = 20 msec.E, Numerical modeling of the field intensity distribution within the hiPSC-CCSs while applying pulsed electric fields via 2 needle-shaped electrodes.Note that the highest field strength is formed around each of the electrodes and tappers down with distance.Also note the elliptical shape of the isointensity regions (isochrones=0.2 kV/cm), which correlates with the shape of the PFA lesions as recorded by optical mapping.F, The correlation between electric field intensities and the culture area exposed to each intensity could be fitted to an exponential function ([1295e (−4.03 * Ef ) + 179], where Ef represents the electric field; R 2 =0.99,P<0.001).

Figure 4 .
Figure 4. Pulsed-field ablation (PFA) lesion sustainability and PFA-mediated cell-death mechanisms.A, Mapping of functional lesions immediately (T 0 ) and 24 hours following monophasic PFA (20 pulses).Electric activity recovery could be documented from T 0 to 24-hour post-PFA, occurring mainly at the lesion periphery.The regions within the cultures that were in greatest proximity to the PFA-electrodes (white arrows) remained electrically silent at 24-hour post-PFA, reflecting irreversible electroporation.The middle-most portion of the lesion remained electrically silent in 70% of the cultures (top-panel, orange arrow), yet reversible electroporation and breach of the conduction block could be noted in this region in the remaining 30% (bottom-panel, black arrow).B, Flow-cytometry results comparing propidium iodine (PI) and Annexin-V staining in untreated control human-induced pluripotent stem cell (hiPSC)-derived cardiac circular cell sheets (hiPSC-CCSs; n=4) as compared with hiPSC-CCSs immediately (T 0 , n=4) and 24-hour (n=5) post-PFA.Accidental necrotic cell death (PI+/Annexin-V−, left) did not play a significant role, with no significant changes noted from T 0 to 24 hour.Late regulated cell-death markers (PI+/Annexin-V+, right) significantly increased at 24-hour post-PFA, rising >2-fold (*P<0.05,**P<0.01).Importantly, there were no significant differences in Annexin-V and PI positive staining immediately following PFA (T 0 ) as compared with untreated control hiPSC-CCSs.

Figure 6 .
Figure 6.Arrhythmia targeting with pulsed-field ablation (PFA).Optical mapping of arrhythmogenic activity in human-induced pluripotent stem cell (hiPSC)-derived cardiac circular cell sheets (hiPSC-CCSs).A, A single rotor (rotor-core encircled) and its wave front (arrow), rotating clockwise.B through E, The same hiPSC-CCS following application of ablation lesions at the mid-horizontal plane (red lines) using a point-by-point HF-PFA approach (50 pulses repeatedly deployed in 5 consecutive locations).Ablation eliminated the rotor-core while generating a midline-block isolating the 2 culture portions.Asynchronous spontaneous electric activity is evident in the upper (B) and lower (C) culture portions.D through E, Persistent midline conduction block, 3 hours post-PFA.Pacing the culture from the upper (D) or lower (E) portions confirms a bidirectional conduction block.F through G, Optical mapping demonstrating multiple spontaneous rotors in the hiPSC-CCS.H through I, Optical mapping of the same hiPSC-CCS immediately (H) and 3 hours (I) following creation of a linear midline lesion (red-lines) using a point-by-point HF-PFA approach (50 pulses repeatedly deployed in 5 consecutive locations).Ablation eliminated arrhythmic activity within the treated area.The remaining arrhythmic substrate was isolated and confined by the linear lesion.J, Phase-mapping analysis of the same culture (rotor cores encircled).Following HF-PFA (bottom-panel), the lower right rotor (blue arrows in H and I) is mapped, rotating clockwise.Its propagating wave fronts (asterisks) are now blocked at midline.