Patient recruitment

Patient recruitment to this study was initiated by a genetic counsellor at Genome Services Western Australia, followed by written informed consent. The study adhered with the Declaration of Helsinki and the NHMRC National Statement on Ethical Conduct Human Ethics Research, and was approved by the Child and Adolescent Health Services, Human Research Ethics Committee, RGS0000000166.

Patient genome sequencing

Target enrichment from Illumina TruSight One Expanded Kit, with massively parallel sequencing (Illumina NextSeq550), followed by secondary analysis with Illumina BWA Enrichment (v2.1.1), and tertiary analysis with Alissa Interpret (v5.1). The variant was classified as a Variant of unknown pathogenicity according to the ACMG guidelines [14].

Cell culture

HEK cell culture. HEK293T cell lines were maintained in Complete Media (RPMI1640 (Gibco, Australia) with 10% Heat-Inactivated Foetal Bovine Serum (FCS) (CellSera, Australia), 1% sodium pyruvate (Gibco, Australia), 1% Penicillin/ Streptomycin Antibiotic (Gibco Australia), and 1% Glutamine (Gibco, Australia). Cells were dissociated with TryplE Express (Gibco, Australia). Cell cryopreservation was in Complete Media/10% DMSO.

Stem cell culture. KOLF2-C1 (KOLF2) cells were grown in 24-well plates coated with CellAdhere Vitronectin XF (STEMcell technologies), maintained in TeSR-E8 media (STEMCELL technologies), and media changed daily. Cells were split with Gentle Cell Dissociation Reagent with 10 µM ROCK Inhibitor (Y-27632, STEMCELL Technologies) for 24 h [13]. Cell cryopreservation was in CryoStor CS10 (STEMCELL Technologies).

All cultures were grown in a 37 °C humidified CO2 (5%) incubator, unless otherwise stated, and routinely checked for mycoplasma.

CRISPR/Cas9 GATA4_HDR transfection and cloning

HEK293T cell CRISPR_HDR transfection. HEK293T cells were grown to 30–50% confluence, and CRISPR_HDR transfected with click chemistry [15]. Constructs were GATA4_Bg5_crRNA (5′ Dibenzocyclooctyne N-Hydroxysuccinimide, DBCON, AGACCACCACCACCACGCTG 3′, PAM: TGG; GATA4 NM_002052.4) and ssDNA GATA4_HDR_Bg5(+) (5′AzideNCTCCTGTGCCAACTGCCAGACCACCACCACTACACTCTGGTGCCGCAATGCGGAGGGCGAGCCTGTGTGCAATGCCTGCGG 3′). The GATA4_VUS HEK293T variant sequence mutalyzer description is NG_008177.2 (GATA4):g.78246_78264delinsTACACTCTGGTGTAGAAAC with affected protein NG_008177.2 (GATA4_i001):p. (Arg283Cys).

iPSC CRISPR HDR transfection. KOLF2 cells grown to 30–50% confluence, were dissociated with Gentle Cell Dissociation Reagent and 1 × 105 cells in 400 µl mTeSR1 with 10 µM Y-27632 (STEMCELL technologies) aliquoted in 24 well plates. The 10 µM GATA4_AF_HDR, ssDNA (5′Alt-R-HDR1GTGGGCCTCTCCTGTGCCAACTGCCAGACCACCACCACCACGCTCTGGTGCCGCAATGCGGAGGGCGAGCCTGTGTGCAAAlt-R-HDR2 3′), and 10 µM GATA4_AF crRNA (5′AGACCACCACCACCACGCTG 3′, PAM: TGG; GATA4 NM_002052.4) were purchased from Integrated DNA Technology (IDT). The gRNA was generated in Duplex Buffer (IDT, USA) with 1 µM crRNA (IDT), 1 µM tracrRNA-ATTO550 (IDT), heated to 95 °C for 5 min and cooled to RT. CRISPR ribonucleoprotein (RNP complexes) were formed with 63 nM gRNA and 63 nM high-fidelity Cas9 protein (IDT) in OPTIMEM, followed by addition of 63 nM HDR. RNPs were transfected into KOLF2 cells with STEM Lipofectamine (Life Technologies), according to the manufacturer’s instruction. Cultures were rendered 30 µM ALT-R HDR enhancer, incubated 48 h at 32 °C with a media change to remove Alt-R HDR Enhancer at 24 h. The introduced GATA4_HDR iPSC variant sequence mutalyzer description is NM_002052.4 (GATA4):c.843_847delinsCTGGT with affected protein NM_002052.4 (GATA4_i001):p. (Arg283Cys).

Following transfection cells were cultured at least 7 days prior to cell freeze down and genomic DNA extraction (PureLink™ Genomic DNA Mini Kit, Life Technologies). Percentage frequency of HDR in gDNA was determined by amplicon sequencing [13, 15, 16]. Transfected cells were single cell cloned by limiting dilution in 96 well plates, replica plated, and DNA lysate prepared [13, 15]. A second round of amplicon sequencing on DNA lysate determined clonal cell lines.

The top six off-target CRISPR crRNA gene cut sites were confirmed as WT by forward and reverse Sanger sequencing for HEK293 and KOLF2 cell clones (AGRF, WA; Additional file 2: Fig. S2A).

Fluorescent immunohistochemistry

GATA4_WT and GATA4_HDR KOLF2 iPSC clones were cultured on Matrigel-coated chamber slides (ibidi) and differentiated into cardiomyocytes. Cardiomyocytes were fixed with 3.7% formaldehyde (Sigma-Aldrich) 20 min at room temperature and stored at 4 °C in DPBS. Subsequently, cells were permeabilised for 15 min with 0.1% Triton-X-100 (Sigma-Aldrich) and blocked with Intercept® Blocking Buffer (LI-COR) for 1 h at room temperature, incubated overnight at 4 °C with rabbit polyclonal anti-GATA4 antibody (1:400; D3A3M, Cell Signalling Technology), washed with 0.05% Tween20 (Sigma-Aldrich) and treated with Alexa-Fluor 488-conjugated anti-rabbit antibody (1:1000; Invitrogen) for 1 h at room temperature. Residual secondary antibody was removed by washing and the cell nuclei stained using NucBlue stain (Invitrogen). Antibody staining was visualised using a Nikon C2+ inverted confocal microscope with a Nikon 40× objective. Monochrome immunofluorescence images were captured and processed using NIS-Elements software (v.5.21.00) and Adobe Photoshop (v.22.3.1) to add colour and merge channels.

Amplicon sequencing

Next-generation amplicon sequencing was carried out on the MiniSeq Sequencing System (Illumina©). In brief, a 250 bp GATA4_HDR site PCR product was amplified, from gDNA or DNA lysates, with GATA4 pAMPF1 (5′ACACTCTTTCCCTACACGACGCTCTTCCGATCTcaccttttacttggacatgaagc3′) and GATA4 pAMPR1 (5′GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTgtacaaaggaagaagacaaggga 3′) [16] for 150 bp, paired-end, > 10,000 reads (MiniSeq, Illumina, Australia) and reads aligned to the HDR or WT amplicon with CRISPResso2 software [17]. In addition, Sanger sequencing was performed across a GATA4 516 bp PCR product with GATA4_F1 primer 5′cggtcagttctcctctcagg3′ and GATA4_R1 primer 5′gagagatgggcatcagaagg3′ (AGRF, Perth, Western Australia).

GATA4 Western Blot and Immunoprecipitation

Proteins were extracted from cells lysed with Pierce’s co-IP protein lysis buffer (ThermoFisher Scientific), quantified (Direct-Detect® Infrared Spectrometer, Merck Millipore), electrophoresed on NuPAGE Bis–Tris 4–12% protein gels (Life Technologies) and blotted onto Polyvinylidene fluoride (PVDF) membranes (Life Technologies). Membranes were blocked with Intercept® (TBS) Blocking Buffer (LI-COR® Biosciences) at 4 °C overnight, stained with rabbit anti-human GATA4 (1:1000, clone D3A3M, Cell Signalling Technology, Australia), and/or b-Actin (1:5000; MA5-15729; Life Technologies, Australia), with secondary stain with goat anti-rabbit IRDye® 800CW (1:10,000; LI-COR® Biosciences) or goat anti-mouse IRDye® 680RD (1:10,000; LI-COR® Biosciences) and imaged on the Odyssey® Infrared Imaging System (LI-COR® Biosciences).

IP protocol. Protein G Dynabeads (Invitrogen, Australia) were washed, and resuspended in 5 µg NKX2.5 antibody (clone F-2, Sana Cruz) or isotype IgG control (Sigma, Australia) in 200 µL PBS/0.02% Tween20, incubated 10 min at room temperature with rotation, placed on the DynaMag-2 magnet, and the supernatant removed. The bead-antibody complex was washed once in 200 µL PBS (0.02% Tween20), twice in 200 µL of Conjugation Buffer (20 mM Sodium Phosphate, 0.15 M NaCl pH 7–9), 250 µL of Pierce BS3 crosslinker (prepared in DMSO [Sigma-Aldrich, Australia]) added, and incubated 20 min at room temperature with rotation. The cross-linking was quenched with Quenching Buffer (1 M Tris HCl ph7.5) for 15 min at room temperature with rotation, washed three times in 200 µL of Li-Cor IP Lysis Buffer (Invitrogen, Australia), then 1 mL of sample lysate was added prior to incubation overnight at 4 °C with gentle rotation. Samples were then placed on the DynaMag-2 magnet (supernatant retained), and beads washed 3 times with PBS. Sample was eluted with 20 µL of Elution Buffer (50 mM Glycine pH2.8), then 7.5 µL of NuPAGE LDS Sample Buffer (Invitrogen, Australia), and 3 µL NuPAGE Reducing agent (Invitrogen, Australia) were added, and sample incubated 70 °C for 10 min prior to gel electrophoresis.

Cardiomyocyte differentiation

KOLF2 iPSCs, genetic variant and normal, were stimulated to differentiate with STEMdiff Cardiomyocyte Differentiation and Maintenance Kit (STEMCELL, Vic, Australia). At day 0 and 20 of cardiomyocyte differentiation 5 × 105 cells were collected, fix/permeabilised according to Transcription factor staining buffer set (eBioscience, US) and stained for expression markers for stem cells: OCT3/4 (OCT3/4-AF488, 1:20, clone 40/Oct-3 (RUO), BD Pharmingen) and NANOG (NANOG-BV421, 1:20, clone 16H3A48, BioLegend); and/or cardiomyocytes: GATA4 (GATA4-PE, 1:20, clone L97-56, BD Pharmingen) and cTNNT (cTnT-AF647, 1:20, clone 13-11, RUO, BD Pharmingen). Specifically, LIVEDEAD ef780 stain for 10 min (eBioscience, US), fix/permeabilised according to Transcription factor staining buffer set (eBioscience, US) and stained for stem cell expression markers (OCT3/4, NANOG), or stained with cardiac cell markers cTNNT and GATA4 (1:20, clone L97-56, BD Pharmingen). Samples were collected using an LSRII X20 flow cytometer (BD, Biosciences) with BD FACSDiva™ software (BD Biosciences), and analysed with FlowJo software (TreeStar Inc, Ashlan, OR, USA). Statistical analysis of data was with one-way ANOVA and Bonferroni’s correction for multiple testing performed using GraphPad Prism Version 8 software (GraphPad Software Inc., La Jolla, CA, USA).

In addition, cells were visualised by light microscopy, and representative video captured of beating cardiomyocytes. Beating cardiomyocytes were scored (1 – 5) according to percentage frequency beating cells: 1, ≥ 1%; 2, ≥ 20%; 3, ≥ 40%; 4, ≥ 60%; and 5, ≥ 80%.

Assessment of intracellular calcium

Cells cultured in StemDiff Cardiomyocyte Maintenance Basal Media (STEMCELL) were exposed to Fluo-4 (5 µM, Life Technologies) for 20 min, followed by 20 min de-esterification step in Hepes-Buffered Solution (HBS) containing (in mM): 5.3 KCl, 0.4 MgSO4.7H2O, 139 NaCl, 1.8 CaCl2, 5.6 Na2HPO4.2H2O, 5 glucose, 20 Hepes and 2 glutamine (pH = 7.4 at 37 °C), not supplemented with Fluo-4. Fluo-4 fluorescence was then recorded in HBS at 37 °C in stream acquisition mode, recording 400 images at 10 ms intervals, using a Zyla 5.5 sCMOS camera attached to an inverted Nikon TE2000-U microscope (ex 480 nm, em 535 nm). MetaMorph® 7.10.3 was used to quantify the signal by manually tracing myocytes. An equivalent region not containing cells was used as background and subtracted. The signal for each cell (F) was normalised to basal signal (F0), yielding F/F0, therefore basal signal approximated 1.0. Frequency of spontaneous calcium transients were calculated as the number of transients over the total length of recording (4 s), FWHM (full width of half maximum) was calculated by fitting a Lorentzian function to the first calcium transient on each recording using GraphPad Prism9.4.1 software [18,19,20].

RNA Sequencing

At day 0 and day 21 of cardiac differentiation cells were collected for RNA extraction (RNeasy Minikit, with DNase on column treatment; Qiagen). RNA integrity was determined on the bioanalyser (Australian Genomics Research Facility (AGRF), Perth, Western Australia). RNA sequencing was performed according to SureSelect Strand-Specific XT HS2 poly A RNA Library preparation for Illumina Multiplexed Sequencing, paired-end, 100 bp, 30 M reads on the NOVAseq 6000 platform (Illumina, USA) at Genomics WA (Perth, Australia).

Data processing and alignment

We used UMItools [21] to de-duplicate our reads prior to alignment. All reads were aligned to the human genome (build GRCh38), using a modified version of the ENCODE ‘rna-seq-pipeline’ (https://github.com/ENCODE-DCC/rna-seq-pipeline), via the Cromwell wrapper software ‘caper’ (https://github.com/ENCODE-DCC/caper). Within the pipeline, reads were aligned to GRCh38 using the STAR aligner (v2.5.1b) [22] and known transcripts were quantified using Kallisto (v0.44.0) [23]. We performed basic QC analysis of mapped reads using SAMStat [24].

Differential gene expression analysis

We imported estimated gene abundances into R 4.1.2 (https://www.R-project.org/) using the tximport function [25]. We applied the standard RNA-seq analysis using edgeR, limma and voom as previously described [26].

Biological interpretation of sample differences

We used clusterProfiler [27], DOSE [28], and Camera [29] via the EGSEA wrapper [30] for gene set enrichment analysis. The full analysis script is reported in Additional file 1.