Cardiac ventricular and atrial tissue specimens were obtained with informed consent from a 15-day-old patient with d-TGA (dextro-Transposition of the Great Arteries) severe pulmonary outflow obstruction who underwent open chest surgery; then, cardiac fibroblasts were isolated as described previously (Park et al., 2008b). Briefly, tissue specimens were cut into small pieces, placed in a 6-well plate containing hFib media (DMEM containing 10% FBS, 2 mM L-GIn, 50 U ml^-1 penicillin and 50 mg mL^-1 streptomycin), and covered with a cover slip. The culture medium was changed every three days for 2 weeks, and fibroblast outgrowth was observed after 10 days of culture. Fibroblast identification was confirmed via immunofluorescence staining for vimentin and TE-7 expression (Gao et al., 2018).

Immunofluorescence staining was performed as previously described (Wang and Zhang, 2022). Briefly, cells were fixed in 4% paraformaldehyde (PFA) for 15 min at room temperature, permeabilized with 90% acetone for 3 min, and then blocked with 10% donkey serum for 20 min. The fixed cells were incubated with primary antibodies (Supplementary Table S1) overnight at 4°C and with corresponding fluorescently conjugated secondary antibodies at room temperature for 2 h; then, the cells were mounted with mounting medium containing 4,6-diamidino-2-phenyl- indole (DAPI) (Vector Laboratories; H-1200) and imaged with a confocal microscope.

iPSC reprogramming was performed with a CytoTune™-iPS 2.0 Sendai Reprogramming Kit as directed by the manufacturer’s instructions. Briefly, cells (passage 2) were transduced with CytoTune™-iPS 2.0 Sendai reprogramming vectors and then transferred to feeder cells in fibroblast medium (DMEM supplemented with 10% FBS, 1% MEM Non-Essential Amino Acids Solution, and 0.1% 2-mercaptoethanol). Putative hiPSC colonies were identified via Tra1-60 live staining, transferred into Matrigel-coated wells containing mTeSR medium, and expanded without feeder cells.

OCT4, Nanog, SSEA4, and SOX2 expression in reprogrammed iPSCs was evaluated via immunofluorescence staining. The absence of residual Sendai virus was confirmed after 10 passages by qRT-PCR with a primer for SeV amplification as previously described; (Grossmann et al., 2021); RNA from passage 0 iPSCs and from somatic fibroblasts served as the positive and negative controls, respectively. The absence of mycoplasma was confirmed at passage 12 with a LookOut Mycoplasma PCR Detection Kit (Sigma-Aldrich—Merck) as directed by the manufacturer’s instructions. The PCR products were loaded on a 1% agarose gel containing SYBR Safe DNA Gel Stain (Invitrogen, S33102) and visualized with a ChemiDoc MP Imaging System (Bio-Rad).

The teratoma formation assay was performed as described previously (Nelakanti et al., 2015). Briefly, ∼1 × 10⁶ iPSCs were collected with Accutase, suspended in 50 μL Matrigel, and slowly injected into the gastrocnemius muscle of NOD/SCID Gamma mice. Teratomas were monitored and surgically removed 6–8 weeks after injection, fixed with 4% formaldehyde, and embedded in paraffin. The preserved samples were sectioned and stained with hematoxylin and eosin at the Pathology Core Research Laboratory in the Department of Pathology, University of Alabama, Birmingham.

Karyotype analysis was conducted in the Cytogenetics Lab at the WiCell Research Institute (Madison, WI). Metaphases were analyzed for each sample by using a brightfield microscope after G-banding. Chromosome identification and karyotype descriptions were made according to the International System for Human Cytogenetic Nomenclature (McGowan-Jordan, 2016). Short tandem repeat (STR) analysis was performed in the Histocompatibility/Molecular Diagnostics laboratory at the University of Wisconsin Hospital and Clinics (Madison, WI).

^AiPSC and ^ViPSC were differentiated into cardiomyocytes (CMs) as previously described (Lian et al., 2013). Briefly, ^AiPSC and ^ViPSC were maintained with mTeSR™ Plus (Stem Cell Technologies) cell media in GelTrex-coated (Thermo Fisher Scientific) plate until meeting 80% confluency. Cardiomyocyte differentiation was induced by culturing iPSCs with CHIR99021 for 24 h in RPMI 1640 medium and B27 without insulin (B27-) media. Cells were then recovered for 48 h in RPMI 1640 medium and B27- media, and cultured with IWR-1 for 48 h in RPMI 1640 medium and B27- media. Beating cardiomyocytes typically appeared nine days after differentiation initiation.

Total RNA was extracted using RNeasy mini kits (Qiagen, United States) as directed by the manufacturer’s instructions and quantified via Nanodrop. cDNA was synthesized with SuperScriptTM II Reverse Transcriptase (Thermo Scientific, United States), and qRT-PCR was performed on a QuantStudio three real-time PCR system (Eppendorf, United States) with appropriate primers (Supplementary Table S2) and the Power Up SYBR Green PCR Mix (Thermo Fisher Scientific). Measurements were determined via the 2^−ΔΔCT method and normalized to the abundance of glyceraldehyde phosphate dehydrogenase (GAPDH) RNA.

Purity of differentiated hiPSC-CMs was determined via flow cytometry analysis as previously described (Gao et al., 2020). Briefly, iPSC-CMs were trypsinized into single cells, fixed with fixation and permeabilization solution (51-2090KZ) for 30 min at 4°C, and blocked in Human BD Fc Block (564219) at room temperature for 10 min. Cells were then incubated with primary antibodies and isotype control antibodies at room temperature for 40 min, incubated with corresponding fluorescently conjugated secondary antibodies for 30 min, and resuspended in wash buffer (554723). Flow cytometry was performed with an LSR Fortessa instrument (BD Biosciences, United States).

hiPSC-CM action and field potentials were assessed with a MaestroEdge multi-electrode array system as previously described (Wickramasinghe et al., 2022). Briefly, cells were plated on 24-well CytoView MRA plates (Axion BioSystems) coated with GelTrex at a density of 4 × 10⁴ cells per well. One week later, the plates were equilibrated for 10 min, data were recorded using the Axion Integrated Studio (AxIS) software and finally analyzed using the Cardiac Analysis Tool (Axion Biosystems, Atlanta, GA, United States).

Ca²⁺ transients were measured as previously described (Gao et al., 2020). Briefly, hiPSC-CMs were plated as individual cells on cover glasses (25 × 25 mm) coated with Geltrex and incubated with Fura-2 AM (0.5 μM, Invitrogen, United States) for 10 min in Tyrode’s solution. Cells were then stimulated at 1 Hz and 2 Hz, and the ratio of fluorescence emitted at 340 and 380 nm was determined with a Ca²⁺ recording system and analyzed using IonWizard (IonOptix, United States) software.

^AiPSC, ^ViPSC, ^AiPSC-CM, and ^ViPSC-CM RNA sequencing was performed by Novogene—Advancing Genomics using Illumina NovaSeq 6,000 platforms. RNA quality control was achieved by removing reads containing adapters, having more than 10% of undetermined bases, and low-quality reads. Less than 2% of these reads were removed in all samples. Reads were mapped to human GRCh38 reference genome (Nurk et al., 2022) using STAR pipeline (Dobin et al., 2013); the unique mapping rate was above 94% in all samples (Supplementary Table S3). Then, for each gene, the count of transcripts for each sample was recorded.

The phrase “hiPSC derived cardiomyocyte” was used to search for iPSC-CM sequencing data in Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/); this query yields 69 data sources. Publications associated with these data sources were manually read to select data sets that 1) had the same cardiomyocyte differentiation (from iPSC) protocol as ours, without further chemical or genetic modification treatment; 2) iPSC-CMs were harvested 30 days or later after differentiation; 3) reported iPSC-CM electrophysiology; 4) have the iPSC-CM bulk-RNA sequencing data generated by Illumina platforms; and 5) the transcript raw count format (required for DeSeq2 gene expression normalization) was available. Two datasets (Zhao et al., 2017; Garay et al., 2022) were selected to be analyzed with ^AiPSC, ^ViPSC, ^AiPSC-CM, and ^ViPSC-CM RNA sequencing data, which were summarized in Supplementary Table S4. Dataset (Garay et al., 2022) contains iPSC-CM derived from skin tissue after 90 days of differentiation from skin (^S−D90iPSC-CM) and kidney tissue (^K−D90iPSC-CM). Dataset (Zhao et al., 2017) contains iPSC derived from the dermal fibroblast (skin) tissue (^SiPSC) and iPSC-CM after 30 days of differentiation (^S−D30iPSC-CM).

Both atrial and ventricular heart tissues were obtained from the same male newborn infant patient during open-chest surgery for d-TGA and cut into small pieces to induce fibroblast outgrowth (Figure 1A); then, atrial and ventricular fibroblasts were collected, expanded for two weeks (Figure 1B), and stained for the expression of vimentin and TE-7 to confirm fibroblast identity (Figure 1C). Fibroblasts were reprogrammed into ^AiPSC and ^ViPSC via transfection with non-integrating Sendai virus vectors coding for expression of the human variants of Oct3/4, Sox2, Klf4, and c-Myc. Two weeks later, colonies with an ESC-like morphology (Figure 1D) expressing the pluripotent/stem-cell marker Tra1-60 (Figure 1E) were mechanically isolated and expanded. The morphologies of the ^AiPSC and ^ViPSC lineages were indistinguishable (Figure 1F). Both ^AiPSC and ^ViPSC expressed the pluripotency genes OCT4, Nanog, SSEA4, and SOX2 (Figure 2A) and generated cells from all three germ layers when evaluated via teratoma formation assay (Figure 2B). Similar results for ^AiPSC see Supplementary Figure S1. Sendai virus (Figure 2C) and mycoplasma (Figure 2D) were undetectable after 10 and 12 passages, respectively, and the cells’ karyotypes were normal (Figure 2E). (Grossmann et al., 2021)

mRNA assessments of the expression of pluripotency genes (OCT4, NANOG, and SOX2; Figure 3A), the early mesodermal marker Brachyury (Figure 3B), the cardiac mesodermal markers MESP1 (Figure 3C) and Gata4 (Figure 3D), and the cardiovascular progenitor-cell transcription factor NKX2.5 (Figure 3E) were conducted before the ^AiPSC and ^ViPSC were differentiated into hiPSC-CMs (Day 0), three and six days after the differentiation protocol (Lian et al., 2012) was initiated (D-Day 3 and D-Day 6), and 3 and 12 days after the cells began beating (B-Day 3 and B-Day 12). Measurements in both hiPSC lineages indicated that pluripotency gene expression declined from Day 0 to D-Day 6. At the same time, the abundance of Brachyury and Gata4 mRNA peaked on D-Day 3 and B-Day 3, respectively, and NKX2.5 expression progressively increased from D-Day 6 through B-Day 12. However, whereas measures of MESP1 abundance gradually increased through B-Day 3 in ^ViPSC, MESP1 expression in ^AiPSC peaked, and was significantly greater than in ^ViPSC, on D-Day 3. Nevertheless, when the expression of cardiac troponin T (cTnT) (Figure 3F) was evaluated via immunofluorescence, the proportions of positively stained ^AiPSC-CM and ^ViPSC-CM populations were similar (Figure 3G) on B-Day 12, indicating that the differentiation protocol was equally efficient for both hiPSC lines.

Electrophysiological properties of ^AiPSC-CM and ^ViPSC-CM were characterized via action potential (Figure 4A) and field potential (Figure 4E) recordings acquired 30 days after the differentiation protocol was initiated. Action potential durations to 30% (APD₃₀, Figure 4B), 60% (APD₆₀, Figure 4C), and 90% (APD₉₀, Figure 4D) recovery, as well as measurements of beat period (Figure 4F), spike amplitude (Figure 4G), and conduction velocity (Figure 4H) were nearly identical in ^AiPSC-CM and ^ViPSC-CM. Field-potential durations (Figure 4I) were detected to be significantly longer in ^ViPSC-CM. Calcium transients tended to have higher peak amplitudes in ^AiPSC-CM than in ^ViPSC-CM when the cells were paced at 1 or 2 Hz (Figures 4J,K), but the differences between groups did not reach statistical significance.

On the other hand, Table 1 shows that our ^AiPSC-CM and ^ViPSC-CM APD₉₀ (measured at 1 Hz) were within the range of the iPSC-CM derived from the reported skin, kidney, and peripheral blood tissues. Meanwhile, ^AiPSC-CM/^ViPSC-CM had significantly higher conduction velocity (Table 2) than iPSC-CM derived from other tissues.

RNA-seq experiments were performed to compare the transcriptional profiles of iPSC and derived cardiomyocytes from this study (^AiPSC-CM, ^ViPSC-CM) and two works where iPSC were generated and differentiated to cardiomyocytes using skin (^SD30iPSC-CM, ^S−D90iPSC-CM) and kidney fibroblast (^K−D90iPSC-CM) (Zhao et al., 2017; Garay et al., 2022). Uniform Manifold Approximation and Projection (UMAP) analysis of the data showed clear separation of cardiomyocytes and their original induced pluripotent stem cells, suggesting that the differentiation process was successful in all cases (Figure 5A). Transcriptional analysis comparing derived cardiomyocytes and iPSC identified 4,970 differentiated expressed genes between both groups (Figure 5B). When focused on this difference, a significant upregulation of genes belonging to the calcium signaling, Actin, Myosin, and Troponin in cardiomyocytes was detected in iPSC-CM. iPSC samples also showed upregulation of genes related to signaling pathways regulating stem cell pluripotency and cell differentiation (Figure 5C). Comparison between ^AiPSC-CM/^ViPSC-CM (derived from cardiac tissue), ^S−D30iPSC-CM/^S−D90iPSC-CM (derived from skin tissue) and ^K−D90iPSC-CM (derived from kidney) revealed minor transcriptional differences between cardiomyocytes derived from heart tissue (124 genes differentially expressed) and moderated change when compared to those differentiated from skin or kidney (1,081 differentially expressed genes) (Figure 5B). A clustergram to study the pairwise similarity among the iPSC and iPSC-CM samples demonstrated that while the detected transcriptional changes are mild, these are enough to establish significant differences between heart and non-heart tissue differentiated cardiomyocytes. iPSC and derived cardiomyocytes form two very different clusters. When focusing only on derived cardiomyocytes, clear subclusters can be observed separating cardiac-tissue-derived cardiomyocytes (^AiPSC-CM and ^ViPSC-CM) from ^S−D30iPSC-CM, ^S−D90iPSC-CM and ^K−D90iPSC-CM (Figure 5D). Gene ontology analysis of the 1081 DE genes (fold change >2 see method section) identified 44 terms, some of them related to heart function (Supplementary Table S5). Further analysis identified 12 genes (AKAP9, BIN1, CACNA1G, CACNB2, CAV1, DSP, GJA5, IRX3, NKX2-5, RANGRF, RYR2 and ISL1) involved in the electrical conduction system of the heart (Zhou et al., 2011) (Supplementary Figure S3), suggesting that the differential expression of these genes play a key role in explaining the distinct electrophysiological properties detected between cardiac and non-cardiac differentiated cardiomyocytes (Tables 1,2). Further analysis between ^AiPSC-CM and ^ViPSC-CM samples revealed no significant difference in gene expression for atrial (HEY2 and MYL2), ventricular (MYL7 and NPPA), and pacemaker (HCN4, TBX3, and GJC1) cardiomyocyte markers (Supplementary Figure S4).