The role of iPSC research for insight in inherited arrhythmia conditions

Introduction

The invention of human induced pluripotent stem cells (iPSCs) 17 years ago revealed a new, uniquely human model system to study disease, test therapeutics and introduced the potential for regeneration of diseased tissue.¹ This technology allows for cell samples from either skin or blood to be efficiently and reproducibly reprogrammed to a state of pluripotency and subsequently derived into any somatic cell type in the body.^2–4 This exciting development revolutionized precision medicine in the investigation of cardiac arrhythmic disease with human-specific and patient-specific in vitro models.⁵ Patient specific or gene-edited iPSCs have significant advantages over alternate in vitro or animal models including the maintenance of human-specific genetic and epigenetic regulatory factors which may play a role in disease pathophysiology and the ability to integrate with other previously developed computational analyses (Figure 1).^6–9 Coupled with huge advances in culture techniques and multi-cellular systems that have unlocked a more physiologic cellular microenvironment, many cardiac disease phenotypes have been recapitulated in iPSCs.^10–13 As with any new model, there has been much effort in both defending as well as criticizing the use of iPSCs to study cardiac disease.^12,14 Here, we aim instead to focus on the extent to which iPSC research has changed our fundamental understanding of human cardiac arrhythmic disease biology. We discuss ways that cardiac research using iPSCs has progressed beyond disease phenotypic recapitulation, how our modeling capacity for arrhythmic heart disease has been advanced through iPSCs, and we consider the impact and potential benefits of using iPSCs for therapeutic testing.

Figure 1: The role of iPSCs in studying inherited arrhythmia syndromes.

Patient-specific iPSCs are generated from individuals with inherited arrythmia disorders. These iPSCs may be differentiated into a range of iPSC-derived cardiovascular cells to be used for disease modeling or therapeutic development.

Revealing arrhythmic pathophysiology with iPSCs

One of the significant advantages of using iPSCs is their capacity to uncover previously unknown biology and provide mechanistic insights into inherited arrhythmias. Inherited arrhythmia syndromes encompass a range of genetic disorders that predispose individuals to life-threatening arrhythmias. Several studies have attempted to recapitulate the human phenotype for arrhythmic disorders in iPSC models and have been reviewed previously (Supp. Table 1).^11–13 Below, we aim to highlight novel biological and therapeutic insights achieved through iPSC research.

Recent iPSC work has uncovered new biology for long QT syndrome (LQTS). LQT2 is caused by mutations in KCNH2, which encodes the human ether-à-go-go (hERG) potassium channel and is important for normal cardiac repolarization.¹⁵ The hERG channel is present in the heart as two splice variants of the KCNH2 transcript, referred to as hERG1a (full length transcript) and hERG1b (truncation of the majority of the N-terminus including the PAS domain).¹⁶ The two splice variants demonstrate different kinetics that effect repolarization but the impact of hERG1a/1b ratio in LQT2 remained unclear.¹⁷ A follow-up study by Ukachukwu et al. further implicated the PAS domain in electrophysiological dysfunction and demonstrated the therapeutic potential of hERG1 PAS disruption using a Jervell and Lange-Nielsen (JLN) syndrome patient-derived iPSC line with complex homozygous loss of KCNQ1.¹⁸ In JLN iPSC-CMs, PAS targeting increased I_Kr, shortened APDs, and decreased the propensity of arrhythmogenic events. Recent work by Feng et al. compared iPSCs from a LQT2 patient harboring KCNH2 (H70R) and an isogenic patient line with a CRISPR/Cas9-edited “correction” of the H70R mutation.¹⁹ Notably, this variant is within the hERG N-terminal PAS domain, a well reported hot-spot for hERG mutations that is absent in hERG1b isoforms. In their model, LQT2 was recapitulated in the cellular phenotype. In addition, they found that both hERG1b mRNA levels and the hERG1b/hERG1a ratio were significantly elevated in the patient line. This altered ratio leads to faster deactivation of I_Kr and longer repolarization time. Further, they demonstrated impaired channel trafficking of hERG1a in H70R iPSC-CMs, revealing a dual effect on loss of normal I_Kr and novel insight into the pathogenesis of LQT2 (Figure 2A).

Figure 2: Working models of arrhythmia pathophysiology developed using iPSCs.

(A) Model of human ether-à-go-go related gene 1 (hERG1)-H70R pathogenesis: The H70R variant, located in the PAS domain of the hERG1a subunit, results in misfolding of hERG1a and ER-associated degradation (ERAD) along with cotranslated subunits. This effect disproportionately affects 1a-1a complexes in comparison to 1a-1b complexes, resulting in an increased to the relative ratio of 1a and 1b subunits in the tetrameric channel. (Adapted from Feng et al. ¹⁹) (B) Proposed model of LMNA-A388fs haploinsufficiency in cardiomyopathy: Mutant lamin A/C protein in iPSC-CMs accelerates the degradation of lamin A-binding sirtuin 1 (SIRT1) to introduce mitochondrial dysfunction and elevated reactive oxygen species (ROS). ROS activate Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) to cause SR Ca²⁺ leakage through RYR2, increasing intracellular diastolic Ca²⁺ concentrations and arrhythmic events. Meanwhile, ROS production leads to the accumulation of nuclear protein SUN1, further destabilizing the nuclear envelope. Both the arrhythmogenic and nuclear defects can be ameliorated through the application of SIRT1 activators or ROS scavengers. (Adapted from Qiu et al.²¹)

The LMNA gene encodes the nuclear envelope proteins lamin A and lamin C, which are essential for maintaining nuclear architecture and cellular integrity. Mutations in LMNA are implicated in a spectrum of diseases collectively referred to as laminopathies. This includes the condition of arrhythmogenic cardiomyopathy (ACM), underscoring the critical role of nuclear structural proteins in the pathophysiology of cardiac disease.²⁰ Recently, patient-specific iPSCs were utilized by Qiu et al. to generate iPSC-derived cardiomyocytes (iPSC-CMs) exhibiting the characteristic arrhythmic phenotype of LMNA-related cardiomyopathy (heterozygous A388fs).²¹ With this model system, the investigators mechanistically linked mutant lamin A-induced accelerated degradation of SIRT1, which in turn led to increased oxidative stress, elevated reactive oxygen species (ROS), and mitochondrial dysfunction. ROS-activated Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) was shown to mediate sarcoplasmic reticulum (SR) Ca²⁺ leak promoting arrhythmogenesis and uncovering a novel mechanism underlying laminopathies. Notably, elevated ROS feeds forward to increase to the accumulation of SUN1, which disrupts nuclear envelope structure. The authors advanced their proposed pathogenic model by manipulating SIRT1 in iPSC-CMs to interrupt the cycle of ROS, CaMKII, and RYR2-induced arrhythmogenesis, as well as nuclear envelope deformation (Figure 2B). This new biological insight helps to point the way for therapeutic methodologies.²²

Timothy Syndrome is an aggressive arrhythmia syndrome characterized by QT prolongation, syndactyly, and multiorgan disfuction.²³ Yazawa et al. utilized iPSCs derived from a patient with Timothy Syndrome (CACNA1C-G1216A) to replicate the cardiac phenotype in vitro and demonstrated that roscovitine, which enhances the voltage-dependent inactivation of CaV1.2, restored the electrical and Ca²⁺ signaling properties of patient-specific iPSC-CMs.^24,25 However, the mechanism by which roscovitine influences L-type Ca²⁺ channels (LTCC) to rescue the phenotype remained unclear. In a subsequent study, the authors employed roscovitine analogs to elucidate the molecular mechanism, proposing that roscovitine partially inhibits CDK5 to regulate CaV1.2.²⁶ This elegant series of studies introduced the regulation of LTCCs by CDK5 as a pathogenic mechanism and highlighted a potential therapeutic target for Timothy Syndrome.

iPSCs to interrogate Variants of Unknown Significance (VUS)

The etiologies of arrhythmic conditions are often rooted in mutations in genes encoding ion channels or Ca²⁺ regulatory proteins, which disrupt the delicate balance of cardiac electrophysiology.^27,28 As a result, genetic testing is an important component of diagnosis, treatment and prognosis for many inherited arrhythmia syndromes.²⁹ Particularly when the phenotype is not strong, unclassified variants or variants of uncertain significance (VUS) are often identified and present a complex challenge due to the ambiguous relationship between the genetic underpinnings and mechanistic causes. ^30–32 iPSCs provide a valuable tool for addressing this challenge, as they allow for the functional characterization of genetic variants in a specifically human cellular context. Patient-derived or healthy WT iPSCs may be utilized to investigate the causative role of VUS, either through VUS “correction” in the patient-line or by introducing the variant into WT iPSCs using CRISPR/Cas9 (Figure 3).

Figure 3: Generation of isogenic controls for VUS interrogation.

The iPSC lines employed for investigating VUS may be derived either directly from patients or from a healthy WT individual. In the case of patient-derived iPSCs, CRISPR/Cas9 is used to correct the VUS, thereby generating an isogenic control line. For WT iPSCs, the variant is introduced to examine the causal relationship between the VUS and the resulting in vivo phenotype.

Terrenoire et al. investigated familial LQTS using patient-specific iPSCs harboring two VUS from both SCN5A (F1473C) and KCNH2 (K897T).³³ Patch-clamp recordings of patient-specific iPSC-CMs demonstrated enhanced late Na⁺ current (I_NaL) driving a primary LQT3 phenotype, while the KCNH2-K897T variant showed normal current density. This work revealed the pathogenic SCN5A polymorphism, which can strongly support ClinVar classification per ACMG Guidelines.³⁴ (Figure 4A). Benzoni et al. reported on three siblings who developed refractory atrial fibrillation (AFib) with an onset <55 years of age and harbored multiple VUS identified through whole-exome sequencing.³⁵ The authors created iPSCs from two of these siblings and their iPSC-CMs studies revealed functional enhancements to the I_f and I_CaL, suggesting a cellular mechanism for the underlying the disorder. iPSCs have also been utilized to screen VUS in LQT1^36,37, LQT2^38,39, LQT3⁴⁰, Timothy Syndrome²⁵, Brugada syndrome (BrS)^41,42, and others. These findings underscore the utility of iPSC-based systems in unraveling the complex genetics of arrhythmic syndromes in which the identified variant is unclassified.

Figure 4: Interrogation of clinical dual channel VUS, transcription factor VUS, or idiopathic VF phenotype using patient-specific iPSCs.

(A) iPSCs may be used to unravel the complex genetics of familial LQTS and other channelopathies. Sequencing revealed both a de novo SCN5A-F1473C and a KCNH2-K897T variant. However, in vitro iPSC-CM screening demonstrated that Na_V1.5 steady state availability was increased while normalized hERG channel current was unchanged, uncovering the pathogenic SCN5A polymorphism and supporting ClinVar classification.³⁴ (Adapted from Terrenoire et al.³³) (B) iPSCs also serve as useful tools for screening variants in non-channel loci. Sequencing of a BrS patient demonstrated a missense mutation in the cardiac transcription factor TBX5. While the reduction in TBX5 expression was not significant, this alteration results in a significant downregulation of SCN5A as the pathologic BrS mechanism. (Adapted from Bersell et al.⁴³) (C) iPSCs are an important model platform for idiopathic arrhythmias, including IVF. In vitro functional assays may be used in tandem with computational modeling to understand the IVF mechanism, and therefore potential therapeutic targets and mechanistic causes. (Adapted from Reilly et al.⁵⁵)

iPSCs can also uniquely provide insight into the potential mechanistic role for VUS identified outside of known ion channel or cardiac-related genes, including non-protein coding loci. Bersell et al. reported a BrS patient-specific G145R variant in T-box transcription factor 5 (TBX5). TBX5 haploinsufficiency has previously been associated with other developmental disorders, but the relationship to BrS had not been defined. The authors found that TBX5 G145R reduced transcriptional activity, downregulated SCN5A and platelet-derived growth factor receptor α (PDGFR), blunted peak I_Na and enhanced I_NaL (Figure 4B).⁴³ Atypically for both BrS and for the patients carrying this TBX5 variant, the action potential duration was notably prolonged compared to controls. The implications of this observation remain unclear. CRISPR/Cas9 correction of TBX5 ameliorated the transcriptional, ionic current and arrhythmogenic defects, demonstrating that heritable mutations in cardiac-associated transcription factors (TFs) may result in arrhythmogenesis in the absence of channel variants. The TFs tumor necrosis factor alpha (TNF-α) and atrial natriuretic peptide (ANP) have also been implicated in arrhythmogenesis, and genetic regulatory elements have been associated strongly with AFib.^44–46 Moreover, Salvarani et al. identified the LMNA K219T mutation as a genetic modifier, showing that the mutant lamin A/C proteins have enhanced binding to the SCN5A promoter in patient-derived iPSC-CMs.⁴⁷ Consequently, K219T mutant CMs exhibit histone repression at SCN5A and a concomitant reduction in Na_V1.5 expression, effects that are restored following CRISPR/Cas9-mediated correction of the variant. Since much of the human transcriptional and epigenetic machinery is not conserved across species, these collective findings emphasize the necessity of iPSC-based human cellular systems to model human disease.^48–51

Patients with ventricular fibrillation (VF) who fall outside diagnostic criteria for specified syndromes are often grouped into the subcategory of Idiopathic Ventricular Fibrillation (IVF). This heterogeneous cohort represents ~20–50% of patients in SCA databases and lack obvious genetic, gross structural, or biomarker links to SCA.^52–54 IVF patient-derived iPSCs offer a source to model the arrhythmic phenotype and identify the origins of the disorder. Our group has recently demonstrated the utility of using iPSC-CMs from a patient with IVF in tandem with a biophysically-detailed computational model demonstrate ionic alterations compared to healthy controls.⁵⁵ The “reverse cellular phenotyping” technique begins with the final phenotype and systematically works backward, utilizing patient-specific iPSC-CM electrophysiology to identify the underlying abnormal features responsible for the arrhythmic mechanisms (Figure 4C). This innovative approach emphasizes the need for inventive frameworks to study VUS and idiopathic arrhythmic syndromes, where the only readily available distinguishing features that can be incorporated into a precise model are patient-specific iPSCs.

Chamber specificity and cellular crosstalk

The human atria and ventricles represent distinct tissues in their structural, transcriptional and electrophysiological properties.^56,57 Yet, much of the in vitro arrhythmia modeling conducted thus far has been performed using protocols which predominantly generate mixed or ventricular-like iPSC-CMs (vCMs) features.⁵⁸ Multiple groups have reported reproducible protocols to derive atrial-like iPSC-CMs (aCMs) which primarily focus on retinoic acid signaling.^59–61 These aCMs have facilitated more accurate recapitulation of the human atria in vitro, as demonstrated by Kayser et al, who modeled sinus node dysfunction using patient-derived iPSCs harboring a KCNJ5 variant (W101C), which encodes a subunit of cardiac G protein–gated inward rectifying K⁺ (GIRK) channels (Figure 5A).⁶² The authors observed that aCMs, but not vCMs, responded to small molecule GIRK antagonists and agonists. iPSC-aCMs also offer the ability to interrogate atrial fibrillation mechanisms, which can originate from an underlying genetic component.^63–66 Brown et al. used iPSC-aCMs and a micropatterned culture platform to demonstrate that primary atrial, but not ventricular, cardiac fibroblasts (CFs) induce greater aCM functional maturity.⁶⁷ In addition, patterned aCF-aCM coculture had a greater sensitivity in detecting drug efficacy compared to all other conditions, highlighting the power of iPSCs to investigate both chamber-specific therapeutics and mechanisms of disease.⁶⁸ Others have also demonstrated that CM-CF coculture enhances myocyte function and exacerbates the arrhythmogenic phenotype in vitro.^69,70 Considering the role of structure in arrhythmogenesis, future cardiac tissue engineering should focus in part on developing multicellular and anatomically-informed models to study atrioventricular conduction and complex arrhythmias.^71–73 At present, observations of reentrant or spiral wave phenomena must be approached with caution, as existing models fail to accurately represent the three-dimensional architecture of native tissue.^74,75

Figure 5: iPSC-CMs to study chamber specific and multicellular effects.

(A) The KCNJ5 variant p.W101C has been previously linked to sinus node dysfunction (SND). Atrial-like iPSC-CMs (aCMs) from a patient harboring W101C exhibit the arrhythmogenic phenotype and demonstrate responsiveness to both pharmacological and CRISPR-based interventions. In contrast, W101C ventricular-like CMs (vCMs) do not display the phenotype nor show any response to treatment. (Adapted from Kayser et al.⁶²) (B) iPSC- from a patient carrying the ARVC-associated S358L mutation in TMEM43, when cultured in isolation, do not exhibit electrophysiological abnormalities relative to control CMs. However, the introduction of S358L mesenchymal stem cells (MSCs) or adipogenic factors induces the development of an arrhythmogenic substrate, characterized by alterations in conduction and action potentials. Treatment with IGF-1 effectively mitigates the impact of MSC inclusion in this model system. (Adapted from Vasireddi et al.⁸¹) (C) Similarly, healthy aCMs cultured with M0 (unactivated) monocytes do display irregular activity. However, the inclusion of M1 (activated) macrophages with aCMs is sufficient to downregulate channel-encoding genes such as SCN5A and KCNA5 as well as induce electrophysiological irregularities. The authors demonstrated that treatment with glucocorticoids may be sufficient to suppress these M1-mediated arrhythmogenic effects. (Adapted from Hutschalik et al.⁸²).

Cellular crosstalk is not limited to a chamber-specific context or solely between CFs and CMs. In addition to vCMs, aCMs, and CFs, the cellular composition of the adult heart includes a variety of other cell phenotypes, many of which play significant roles in the heart’s pathophysiological processes.⁷⁶ Protocols have been developed to derive sinoatrial (SA) node cardiomyocytes, Purkinje cells (PCs), endothelial cells (ECs), and adipogenic cell phenotypes from iPSCs.^77–80 In iPSCs from a patient with the transmembrane protein 43 (TMEM43) mutation S358L associated with ACM, Vasireddi et al. observed no significant functional differences between patient-specific and control iPSC-CMs under basal conditions (Figure 5B).⁸¹ However, following the introduction of either patient-specific adipogenic progenitors or adipogenic factors, TMEM43-S358L CMs exhibited an arrhythmogenic phenotype, including reduced conduction velocity (CV) and action potential duration (APD), along with unidirectional block. In another study, Hutschalik et al. differentiated aCMs and iPSC-derived M1 macrophages to generate an isogenic model of AFib (Figure 5C).⁸² The inclusion of M1 macrophages and M1 conditioned medium resulted in irregular beating and decreased CV, suggesting that macrophage-mediated inflammatory factors may induce an arrhythmogenic phenotype in aCMs. In addition, aCMs had reduced expression of SCN5A, KCNA5, ATP1A1, and GJA5, correlating highly with a meta-analysis of clinical AFib transcriptomes.⁸³ In addition, in a model of dilated cardiomyopathy (DCM; SCN5A-R222Q), Wauchop et al. observed no DCM phenotype in embryoid bodies or cell sheets due to poor SCN5A expression.⁸⁴ Yet, when CFs and ECs are included and cultured with CMs in a biowire system, SCN5A is sufficiently expressed to reveal the arrhythmogenic phenotype in R222Q preparations. Taken together, these iPSCs studies enhance our understanding of the role of cellular crosstalk in arrhythmogenesis and may provide insights for future therapeutic strategies.

iPSCs as a pharmacologic testbed

In addition to their role in disease modeling, iPSCs have transformed the landscape of drug screening and therapeutic development. The capacity to generate cardiomyocytes from patients with specific arrhythmogenic mutations allows for high-throughput target identification and drug screening in a human context, addressing the limitations of conventional cell lines and animal models. Lee et al. employed DCM patient-derived iPSCs harboring a premature stop codon in LMNA (R225X, Q354X, or T518fs) to explore the iPSC-CM treatment response to PTC124 (Figure 6A).⁸⁵ PTC124 (Ataluren) is a small molecule known to promote ribosomal readthrough of premature stop codons, a mechanism initially exploited in the treatment of cystic fibrosis.⁸⁶ Notably, PTC124 promoted full-length LMNA protein which mitigated the downstream measured effects only in the R225X mutant. This highlights the use of iPSCs as a tool to detect and treat mutation-specific disease mechanisms. Ma et al. utilized LQT1 patient-derived KCNQ1 mutant iPSCs (heterozygous exon 7 deletion) alongside a healthy familial control to demonstrate the efficacy of the small molecule ML277 in restoring I_Ks and partially rescuing APD.⁸⁷ These findings have recently led to the identification of a novel, singular binding site for ML277 on the Kv7.1 channel, which may hold therapeutic potential for the treatment of LQT1.⁸⁸ Activation of serum/glucocorticoid-regulated kinase 1 (SGK1) has been shown to enhance the activity of Na⁺ channels and CaMKII, extending APD.⁸⁹ Giannetti et al. proposed the inhibition of SGK1 (SGK1-Inh) as a potential therapeutic strategy for LQT1 and LQT2.⁹⁰ Following SGK1-Inh, the authors observed a reduction in corrected field potential durations (cFPDs) in KCNQ1-R594Q, KCNH2-A561V and KCNH2-IVS9–28A/G iPSC-CMs (Figure 6B). Interestingly, SGK-1-Inh did not alter cFPD for KCNQ1-A341V iPSC-CMs nor the LQT1 rabbit model KCNQ1-Y315S. Similarly, Kim et al. reported comparable findings in iPSC-CMs derived from patients with LQT1 (KCNQ1-V254M), LQT2 (KCNH2-G604S), and LQT3 (SCN5A-R1623Q), demonstrating the therapeutic potential of SGK1-Inh. In addition to its therapeutic potential for LQTS, Hutschalik et al. recently suggested that glucocorticoid treatment may also have the potential to reverse the in vitro AFib phenotype in aCMs.^82,91

Figure 6: iPSC-CMs as a mutation-specific platform for therapeutic screening.

(A) In 3 LMNA variants (2 premature stop codon, 1 frameshift), PTC124 (Ataluren) promotes ribosomal readthrough and full-length Lamin A/C protein in R225X only. R225X also demonstrate reduced nuclear blebbing and apoptosis, in addition to improvements in excitation-contraction coupling. (Adapted from Lee et al.⁸⁵) (B) Inhibition of serum/glucocorticoid-regulation kinase 1 (SGK1) has LQT-specific and variant specific efficacy. In LQT1, KCNQ1-R594Q patient-specific iPSC-CMs demonstrate corrected field potential durations (cFPD) shortening while A341V variants show no response to SGK1 modulation. Interestingly, SGK1 inhibition shortens cFPD in both LQT2 KCNH2 mutants (IVS9–28A/G and A561V) (Adapted from Giannetti et al.⁹⁰) (C) Lumacaftor, known to improve channel trafficking, also shows variant-dependent efficacy in the treatment of LQTS. Of 5 patient-derived iPSC lines harboring KCNH2 mutations, Lumacaftor rescues electrophysiological alterations in 3 (Adapted from Mehta et al.⁹³)

In a more recent study, Mehta et al. proposed that Lumacaftor, a pharmacological agent known for its role as a channel trafficking chaperone and with an established safety profile, effectively corrects hERG trafficking defects in patient-specific iPSC-CMs from individuals with LQT2 who are not protected by ß-blockers. The improved trafficking of hERG thereby addresses the associated arrhythmogenic phenotype in 3 of 5 variants (Figure 6C).^92,93 Interestingly, this effect may be mutation-specific. O’Hare et al. investigated Lumacaftor treatment on three trafficking-deficient LQT2 patient-derived iPSC lines and reported contradictory results with amelioration of two variants (KCNH2-N633S and KCNH2-R685P), and worsening of a third (KCNH2-G604S).⁹⁴ These inconsistent findings could be related to the degree of protein misfolding that prevents normal channel trafficking and thus the capacity for channel rescue.^95,96

Short QT syndrome (SQTS) is linked to gain-of-function mutations in KCNH2, KCNQ1 and KCNJ2.⁹⁷ iPSC models have been employed for pharmacological screenings, consistently demonstrating that quinidine can normalize the pathologically shortened APD and suppress arrhythmias in SQTS iPSC-CMs, whereas sotalol and metoprolol do not alleviate the phenotype.^74,75,98,99 Interestingly, after screening antiarrhythmic drugs, Shiti et al. found that venakalant, an I_Kur blocker, showed strong reversal effects in only iPSC-aCMs, once again emphasizing the importance of chamber-specific screening.⁷⁴ In addition, disopyramide and the short peptide scorpion toxin BmKKx2 (which targets KCNH2) have also shown potential for in vitro correction of the SQTS phenotype.^75,99

These findings underscore the utility of iPSC-based cardiac models as reliable, mutation-specific platforms for both evaluating novel therapeutics and recontextualizing established drugs within alternative therapeutic frameworks. Furthermore, these models offer advantages including cost-effectiveness, increased throughput, and greater relevance to human physiology compared to traditional animal models, while also serving as critical precursors to clinical trials.

Exploring gene therapies through iPSCs

Given that the pathogenesis of most inherited arrhythmia syndromes is primarily, if not exclusively, associated with a single gene, the prospect of using gene therapy to address these disorders is a compelling concept. Gene therapy techniques for the treatment of inherited arrhythmia syndromes thus far have harnessed the concepts of suppression (of the mutant transcript), replacement (with healthy transcript), or both using dual suppression-replacement. CALM1, CALM2, and CALM3 encode identical CaM proteins, with each harboring known pathogenic variants linked to LQTS.¹⁰⁰ Bortolin et al. recently leveraged this genetic redundancy and demonstrated that antisense oligonucleotides targeting only the mutant gene are sufficient to restore AP and CaT features in heterozygous CALM1^F142L/+ iPSC-CMs without altering total CaM protein levels (Figure 7A).¹⁰¹ This result suggests that CALM2 and CALM3 are capable of compensatory effects to rescue the phenotype. Similarly, in the case of total protein loss, replacement-only strategies may be capable of attenuating arrhythmogenic phenotypes. Lodola et al. differentiated iPSC-CMs from a catecholaminergic polymorphic ventricular tachycardia type 2 (CPVT2) patient carrying a homozygous CASQ2 mutation (G112+5X), which generates a premature stop codon and a total loss of CASQ2 expression leading to arrhythmogenic Ca²⁺ handling abnormalities (Figure 7B).¹⁰² After administering the human WT CASQ2 gene through an adeno-associated viral vector serotype 9 (AAV9), the authors observed near physiologic expression of CASQ2 protein and a decrease in delayed afterdepolarizations (DADs). These results may be unique to specific homozygous or complex heterozygous loss of function mutations which disrupt WT transcripts and no proteins are made. In other circumstances, Kyriakopoulou et al demonstrated that AAV-mediated delivery of WT PKP2 for patient-specific PKP2^{c.2013delC/WT} iPSC-CMs was sufficient to restore both PKP2 expression as well as other desmosome proteins that were decreased in response.¹⁰³ These effects improved junctional sodium conduction and enhanced contractile function in iPSC-CMs.

Figure 7: Suppression, Replacement, and dual Suppression-Replacement gene therapies in patient-derived iPSC-CMs.

(A) Suppression of pathogenic LQTS-related CALM1-F142L/+ variant. Patient-derived F142L/+ iPSC-CMs display abnormal CaTs and APs. Hybridization and inhibition of mutant CALM1 using antisense oligonucleotides (ASO) reduces expression of CALM1 mRNAs in a dose-dependent manner but does not affect CaM protein levels. Treatment with ASO is sufficient to partially rescue both Ca²⁺ handling and AP alterations in F142L/+ iPSC-CMs. (Adapted from Bortolin et al.¹⁰¹) (B) Replacement of CASQ2 in CPVT2 patient-derived iPSC-CMs. Viral gene transfer of WT hCASQ2 using AAV9 is capable of restoring CASQ2 protein expression in homozygous CASQ2-G112+5X iPSC-CMs. The increase in CASQ2 protein levels restores AP features and reduces arrhythmogenic events in vitro. (Adapted from Lodola et al.¹⁰²) (C) Suppression and replacements in LQT2 patient-derived KCNH2-G604S iPSC-CMs. A dual lentiviral vector, containing KCNH2-repressing short hairpin RNAs (shRNAs) and shRNA-immune KCNH2 cDNA (KCNH2-shIMM) can effectively replace WT/variant KCNH2 expression with shIMM. As a result, restoration of the pathologically extended AP is observed in G604S+SupRep treatment group. (Adapted from Bains et al.¹⁰⁵)

However, in the absence of total protein loss or genetic redundancy, a tandem suppression and replacement technique may be necessary.¹⁰⁴ As an example, a dual-component vector was used to silence LQT1 mutations in KCNQ1 (Y171X, V254M, I567S, and A344A/spl) with short hairpin RNAs (shRNAs), while simultaneously replacing the deficient gene function with shRNA-immune KCNQ1 cDNA in an effort to restore normal repolarization. Bains et al. proposed the therapeutic potential of this dual strategy to address arrhythmias caused by mutations in KCNH2, demonstrating significant improvements to iPSC-CM APDs following treatment (Figure 7C).¹⁰⁵ Furthermore, Hamrick et al. explored this unified approach to simultaneously suppress and replace function in CALM1, CALM2, and CALM3-related arrhythmias.¹⁰⁶ These studies collectively act as proof-of-principal for suppression-replacement therapies and underscore the versatility of these gene therapy techniques across multiple genetic contexts.

Collectively, these studies illuminate the utility of patient-specific iPSCs to model the effectiveness of suppression, replacement, and dual suppression-replacement strategies to correct underlying genetic defects. This provides a cost-effective and promising avenue for the development and testing of gene therapies for management of inherited arrhythmias.

Addressing limitations of hiPSC-based systems

Despite the many advantages of hiPSC technology, there are still some limitations that researchers must navigate. It is well reported that iPSC-CMs are often less mature than native adult CMs, affecting their functional properties and electrophysiological behavior.¹⁰⁷ For example, in the absence of maturing cues, KCNJ2 expression in iPSC-CMs only reaches near-adult levels by day 80 post-differentiation.¹⁰⁸ This immaturity may hinder the ability to effectively model pathologies which develop later, including arrhythmogenic cardiomyopathies.¹⁰⁹ However, the ‘younger’ iPSC phenotype offers a unique look at early disease characteristics and biomarkers. Such findings can be harnessed for early diagnosis and/or targeted prior to advanced disease expression.¹¹⁰

The combination of iPSC models with organ-on-a-chip and coculture techniques provides a more physiologically-relevant microenvironment and aids in inducing electrical maturity in iPSC-CMs.^67,69,70,84 In addition, we and others have demonstrated that the inclusion of I_K1 through either Kir2.1 enhancement or current injection serves to improve AP dynamics and enhance electrical maturity of iPSC-CMs.^111,112 The implementation of these advanced model systems, including micropatterned platforms, biowires, EHTs, and microfluidic devices, facilitates CM maturation and promotes more adult-like phenotypes in vitro.^113,114 This, in turn, contributes to the increased reliability of experimental findings. Less sophisticated culture systems, such as traditional glass or tissue culture plastic, may even introduce proarrhythmic phenotypes in vitro, including alternans and spiral wave activity.¹¹⁵

The heterogeneity of iPSC lines can introduce variability in experimental outcomes, making standardization, and the inclusion of multiple iPSC clones, a crucial experimental consideration.⁷² Moreover, the majority of cell cultures, engineered tissue or complex multicellular constructs do not incorporate day/night cycle with cell synchronization and/or light/dark variability despitat that these are features known to affect biological responses. Still, iPSC-CMs remain more robust in their ability to identify cardiac-specific pathways and ionic abnormalities that are undetectable in non-human or simple human systems, including tsaA-201 and HEK cells.¹¹⁶ iPSC-based platforms also hold significant advantages over animal models, such as the retention of human-specific genetic and epigenetic regulators, conservation of ion channel structure and phenotype, and the ability to generate patient-specific cells.^6,94,117,118 While no gold-standard model iPSC-based model has emerged, engineered systems offers a more biomimetic human microenvironment and promote the electrical maturation of iPSC-CMs to enhance the physiological relevance of our findings. Combined approaches, including complex models, coculture, and computational analytics, offer significant potential to advance our understanding of the cellular and molecular mechanisms that underpin cardiac arrhythmias.

Conclusion

The advent of iPSC technology provided human-specific, patient-derived systems that enhance our understanding of disease mechanisms, enable high-throughput drug screening, and offer insights into the effects of genetic mutations on cardiac function completely through the human cardiac cell lens. Research including new derivation methods and advanced culturing techniques continues to evolve and facilitate the development of more robust human heart-in-a-dish models. As we refine these models and integrate them with other advanced technologies, the promise of iPSCs for personalized cardiovascular medicine becomes increasingly attainable. However, the critical next step must be to effectively translate these findings to the clinic to represent tangible gains in our ability to treat arrythmia syndromes and pave the way for next-generation therapeutics and improved patient outcomes.