Mutation-Specific Cardiomyocyte Lines from Patients with Fabry Disease: A Sustainable In Vitro Model to Investigate Structure, Function, and Disease Mechanisms ^†

Abstract

Background: Fabry disease (FD) results from pathogenic GLA variants, causing lysosomal α-galactosidase A (α-GalA) deficiency and sphingolipid ceramide trihexoside (Gb3 or THC) accumulation. Disease phenotype varies widely but cardiomyopathy is commonly life-limiting. As a multisystemic disorder, FD initiates at the cellular level; however, the mechanism/s underlying Gb3-induced cell dysfunction remains largely unknown. This study established an in vitro mutation-specific model of Fabry cardiomyopathy using human-induced pluripotent stem cell (iPSC)-derived cardiomyocytes to explore underlying cell pathology. Methods: Skin biopsies from consenting Fabry patients and normal control subjects were reprogrammed to iPSCs then differentiated into cardiomyocytes. The GLA mutations in Fabry cell lines were corrected using CRISP-Cas9. Phenotypic characteristics, α-Gal A activity, Gb3 accumulation, functional status, and lipid analysis were assessed. Cardiomyocytes derived from two patients with severe clinical phenotype and genotypes, GLA^c.851T>C, GLA^{c.1193_1196del}, and their respective corrected lines, GLA^{corr c.851T>C}, GLA^{corr c.1193_1196del}, were selected for further studies. Results: Cardiomyocytes derived from individuals with FD iPSCs exhibited stable expression of cardiomyocyte markers and spontaneous contraction, morphological features of FD, reduced α-Gal A activity, and accumulation of Gb3. Lipidomic profiling revealed differences in the Gb3 isoform profile between the control and FD patient iPSC-derived cardiomyocytes. Contraction strength was unchanged but relaxation after contraction was delayed, mimicking the diastolic dysfunction typical of Fabry cardiomyopathy. Conclusions: iPSC-derived cardiomyocytes provide a useful model to explore aspects of Fabry cardiomyopathy, including disruptions in sphingolipid pathways, proteomics, and multigene expression that together link genotype to phenotype. The platform potentially offers broad applicability across many genetic diseases and offers the prospect of testing and implementation of individualised therapies.

1. Introduction

In Fabry disease (FD), the most common lysosomal storage disorder worldwide, pathogenic variants in the GLA gene (MIM: 300644) lead to absolute or functional deficiency in lysosomal α-galactosidase A (α-Gal A) activity. Despite being frequently under-diagnosed, FD manifests across all racial and ethnic backgrounds. In Australia, its recently revised estimate of prevalence is 1 in 7700 live births [1], 60% higher than previously documented in 1996 [2], attributed to a 10-fold rise in de novo FD diagnoses [1]. Low or absent α-Gal A activity causes progressive intracellular build-up of its sphingolipid substrate, globotriaosylceramide (Gb3), also known as THC (ceramide trihexoside). Gb3 accumulation in the lysosomes of target tissues ultimately induces a varied phenotype that may include painful acroparaesthesia, nephropathy, cardiomyopathy, cerebrovascular disease, gastrointestinal dysfunction, and angiokeratoma, collectively contributing to diminished quality of life and heightened mortality risk. Cardiovascular complications remain the most common cause of both significant clinical events and premature death in FD.

No current treatments can reverse or cure established FD. Enzyme replacement therapy (ERT) can replace functional α-Gal A deficiency but does not fully correct the underlying pathology [3], especially if commenced beyond early adulthood. Organ dysfunction often progresses despite ERT. Additionally, ~40% of treated patients develop antibodies to recombinant α-Gal A that can neutralise its effects, induce infusion-related responses, or trigger anaphylactoid reactions [4]. The subgroup of patients who have amenable mutations may be treated with the oral pharmacological chaperone therapy migalastat, which selectively and reversibly binds to the active site of endogenous α-Gal A and supports the intracellular trafficking of misfolded protein to the lysosomes [5]. Substrate reduction therapies and gene therapies are in clinical trials, but their place in overall management of Fabry disease is yet to be determined. Induced pluripotent somatic stem cells (iPSCs) can be generated from a single skin biopsy, maintained in culture indefinitely, and expanded without losing pluripotency [6,7]. iPSCs can be differentiated into organ-specific cells including cardiomyocytes and used to investigate pathophysiologic mechanisms and mutation-specific response to therapies [7,8,9,10,11]. In all relevant target cell types, FD phenotype including substrate accumulation can be confirmed, and substrate reduction by potential therapies are demonstrated in vitro. Using well-established techniques, cardiomyocytes derived from iPSCs allow for the study of patient-specific genetic variations and phenotypic expression and provide a powerful tool to probe the pathophysiology of Fabry cardiomyopathy.

In a rare and heterogeneous disease of 1000 genotypes and wide interpatient variability, there is enhanced value in studying individual patients. We aimed to study the effects of specific GLA mutations associated with severe FD on cardiomyocyte structure and function in 2D and 3D models.

2. Materials and Methods

2.1. Study Participants

This research was carried out in accordance with the Declaration of Helsinki (2008) of the World Medical Association. Ethical approval was obtained from the Royal Melbourne Hospital Human Research Ethics Committee approval 66294/MH-2020, Monash University Human Research Ethics Committee (Project number CF16/1247–2016000663) and the Murdoch Children’s Research Institute (RCH HREC project number 33001A) approved 9 January 2013. Informed consent was obtained from the participants, including permission to publish anonymised results. All participants were male, and all had a molecular diagnosis of FD. Control participants were also males, of similar ages.

2.2. Derivation and Characterisation of iPSCs

Primary dermal fibroblast cultures were established from 4 mm skin punch biopsies performed (KN) under local anaesthetic. Fibroblasts were cultured in DMEM medium supplemented with 10% foetal bovine serum (FBS), 1% L-glutamine (200 mM), 1% penicillin-streptomycin (10,000 U/mL), and 0.1% fungizone.

As previously described [6,7], reprogramming of fibroblasts to pluripotency was achieved using messenger RNA-based overexpression of the Yamanaka reprogramming factors. Transfection was performed using the Reprocell StemRNA-NM Reprogramming Kit (Reprocell, Beltsville, MD, USA), consisting of reprogramming transcription factors (OSKMNL NM-RNA), immune evasion factors (EKB NM-RNA), and NM-microRNAs.

At days 10–14 following transfection, the newly formed iPSC colonies were manually dissected into pieces, using a needle, then replated onto organ culture dishes containing MEFs with ES medium or Geltrex matrix with Essential 8 (E8) medium and incubated at 37 °C in 5% CO₂. Once established, all lines were subsequently cultured on Geltrex matrix in E8 medium warmed to 37 °C., with the medium changed every 1–2 days.

2.3. Confirmation of Pluripotency and Normal Ploidy Status

Pluripotency of both Fabry-specific iPSC and control lines was confirmed using a Leukocyte Alkaline Phosphatase staining kit (Sigma-Aldrich Co., Truganina, Australia), by PCR for pluripotency genes OCT3/4, SOX2, KLF4, and NANOG, and further confirmed by their abilities to form three germ layers after passage 4, to form embryoid bodies expressing markers of all these three germ layers by immunostaining and qPCR, and to induce teratoma formation when injected into immunodeficient mice, as previously described [10,11]. Karyotype analysis of both control and Fabry iPSCs was performed to ensure normal ploidy status. Established iPSC lines were cultured on GelTrex (Thermo Scientific, Scoresby, Australia, #A1413302) coated plates in mTeSR Plus (STEMCELL Technologies, Tullamarine, Australia, #100-0276) at low density. The medium was changed every 1–2 days and the cells were passaged every 4–5 days using TrypLe (Thermo Fisher Scientific, Waltham, MA, USA, #12604039). Upon seeding, iPSCs were grown in the presence of Rock-Inhibitor (Selleck Chemicals, Houston, TX, USA, #S1049) for the first 24 h.

2.4. Generation and Phenotypic Characterisation of Cardiomyocytes

Cardiomyocytes were derived from IPSCs in monolayer cultures, as previously described [10,11,12].

In brief, at 70–80% confluency, cardiac differentiation was induced (day 0) using basal medium (RPMI 1640 (Thermo Fisher Scientific, #21-870-092) plus 1× Glutamax (Thermo Fisher Scientific, #35050061), B27-Vitamin A (Thermo Scientific, #12587010), and 50 U/mL Pen/Strep (Thermo Fisher Scientific, #15-140.122), containing 100 ng/mL Activin-A (R&D systems, Minneapolis, MN, USA, 338-AC-050/CF), 8 μM CHIR 99021 (Tocris, Victoria, Australia, #RDS4423/10), and 50 mg/mL ascorbic acid (Sigma-Aldrich, A8960). After 24 h, the induction medium 299 was replaced with basal medium supplemented with 5 μM IWR-1 (Sigma-Aldrich, #I0161) and 50 mg/mL ascorbic acid for 6 days, with medium changes every 2 days. From day 7 onwards, the cells were kept in basal media, with media changes every 2 days. At day 14, cardiomyocytes were harvested using Trypsin EDTA (Thermo Fisher Scientific, #25-200-056) and replated and kept in maintenance medium (aMEM + GlutaMAX (Thermo Scientific, #32561037), 10% FCS, 50 mg/mL ascorbic acid, Pen/Strep) for at least five days. Next, the cardiomyocytes were kept in maturation medium (DMEM no glucose (Thermo Fisher Scientific, #A14430), 1× Glutamax (Thermo Fisher Scientific, #35050061), 50 mg/mL ascorbic acid (Sigma-Aldrich, A8960), 1 mM glucose, B27-Insulin (Thermo Scientific, A1895601), 10 μM palmitic acid (Sigma Aldrich, #P0500)).

The cells were positively identified as cardiomyocytes by their characteristic sarcomeric structures positive for troponin I (TNNI) and alpha-actinin, and by observing their spontaneous contraction.

2.5. Immunofluorescence Microscopy for Troponin and Gb3

For immunofluorescence, cardiomyocytes were grown on Matrigel-coated 96-well glass bottom plates, washed once with PBS, and fixed with 2% PFA for 30 min at room temperature. The cells were washed once with PBS before permeabilization with PBS-T (PBS + 0.1% Tween and 0.1% Triton) for 15 min at room temperature. After blocking (10% normal goat serum in PBS T) for 1 h at room temperature, the samples were incubated with primary antibodies (anti cardiac Troponin-T, Abcam, Waltham, MA, USA, #ab8295; anti-Gb3, Fisher Scientific, #BDB551352; anti-a Galactosidase A, Abcam, #ab168341) in blocking buffer over night at 4dC. The next day, the cells were washed three times with PBS-T and incubated with secondary antibodies (Goat-α-mouse IgG Alexa Fluor 555, Invitrogen, Carlsbad, CA, USA, #A21422; Goat-α-mouse IgM Alexa Fluor 647, Thermo Fisher Scientifc, #A21238; Goat-α-rabbit IgG Alexa Fluor 647, Invitrogen, #A21244) in blocking buffer supplemented with Hoechst solution 3570 (Life Technologies, Carlsbad, CA, USA, #H3570). The cells were then washed three times and imaged using the Zeiss LSM-780 confocal microscope (Jena, Germany). The expression of cTnT was used to locate cardiomyocytes, then the expression of Gb3 was acquired. Identical settings were used for Fabry and control cardiomyocytes.

2.6. Transmission Electron Microscopy (TEM)

TEM was performed to assess morphological differences in the cell ultrastructure of iPSC-derived cardiomyocytes from FD patients compared to the control or GLA^corr cardiomyocytes. Differentiated cardiomyocytes were isolated and fixed in Karnovsky’s fixative and washed in 1 mL 0.1 M cacodylate buffer before post-fixation in 1% osmium tetroxide for one hour at room temperature. Serial dehydration was performed before embedding in resin. Pellets were thin sectioned with an ultra-microtome and placed in copper orthogonal grids before staining with 4% uranyl acetate. All sections were imaged using a Hitachi H7500 TEM (Hitachi, Tokyo, Japan) or Tecnai T12 TEM (FEI, Hillsboro, OR, USA) with Gatan Microscopy Suite Software GMS3.5 (Gatan Incorporated, Pleasanton, CA, USA).

2.7. Generation of Isogenic Control Cells

As per the current cell study standard, corrected isogenic cells differ only in the GLA variants. Each isogenic control cell line, GLA^corr, was generated via CRISPR-CAS9-mediated homology-directed repair, as previously described [12].

2.8. α-Gal A Activity Assay and Gb3 Accumulation

α-Gal A activity was assessed in fibroblasts and iPSCs using the Alpha Galactosidase Activity Assay Kit (Abcam, Cambridge, UK, ab239716), following the manufacturer’s instructions with variations. Specifically, the cells were lysed using six rounds of freezing and thawing in an ethanol, dry ice slurry, and the assay reaction mix included 100 mM N-acetylgalactosamine (Merck, Darmstadt, Germany) to inhibit any α-Gal B activity [13]. Cell protein concentration was measured using a micro bicinchoninic acid (BCA) assay kit (Thermo Scientific, Scoresby, Australia, 23235) and activity was calculated in pmol/mg protein/hour. Assays of α-galactosidase levels and Gb3 in both fibroblasts and iPSCs were undertaken in triplicate and means for each participant were analysed.

2.9. Sphingolipid Analysis

For lipid analysis, confluent fibroblasts, iPSCs, and iPSC-derived cardiomyocytes were harvested as cell pellets in triplicate and homogenised by sonication. Lipids were extracted from 50 μg of protein by the addition of 0.2 mL chloroform/methanol (2/1) and 10 pmol THC (d18:1/17:0) [N-heptadecanoyl ceramide trihexoside or Gb3] as internal standard. The samples were mixed for 10 min, sonicated for 30 min, and allowed to stand at room temperature for 20 min. The samples were then centrifuged (13,000× g; 10 min at room temperature) and the supernatant was removed and dried down under nitrogen at 40 °C. Lipid extracts were reconstituted in 10 mM ammonium formate in methanol, and individual species of Gb3 were quantified by multiple reaction monitoring (MRM), as described previously [14]. Total Gb3/THC was calculated by the sum of all isoforms measured.

2.10. Functional Studies in Cardiomyocytes

Contraction and relaxation times in selected mutant and corrected Fabry cardiomyocytes were studied using 3D cardiac organoids. These were formed by populating elastic scaffolds with differentiated cardiomyocytes. Muscular and contractile force and the rate of relaxation were measured using the heart-dyno system [12] in at least 10 organoids in each experiment.

2.11. Statistical Analysis

Statistical analyses were performed using Prism v10.2 (Graph-Pad Software V10 2LLC). Data are presented as the mean ± SEM. Data were identified as outliers if the z-score was more than 3 or less than −3 and subsequently excluded. Experiments were performed a minimum of 3 times in triplicate and significance was calculated with Prism v10.2 (Graph-Pad Software, LLC) using one-way ANOVA and Tukey’s multiple comparison test. A p value of ≤0.05 was considered statistically significant.

3. Results

3.1. Cardiomyocyte Generation and Confirmation of Fabry Phenotype

Fibroblast cultures were obtained from skin biopsies from six Fabry patients and age-matched normal males, and iPSCs were generated. Low α-galactosidase levels and Gb3 accumulation were confirmed in both fibroblasts and iPSCs prior to differentiation of iPSCs into cardiomyocytes. Fibroblast α-galactosidase activity in Fabry vs. control samples (n = 6) was 0.65 ± 1.0 vs. 33.2 ± 13.0 vs. nmol/mg protein/h, t = 6.132, p = 0.0001. Fibroblast Gb3 levels were 7613 ± 2649 pmol/mg in Fabry cells and 2007 ± 18.4 pmol/mg protein in the controls (t = 4.73, p = 0.0015).

In iPSCs, α-galactosidase activity in Fabry vs. control samples was 0.05 ± 0.06 vs. 7.31 ± 1.38 nmol/mg protein/h, t = 14.04, p = 0.0001. Gb3 levels in Fabry iPSCs were significantly higher than the controls at 850 ± 129.1 vs. 360 ± 0.49.7 pmol/mg protein (t = 7.085, p = 0.0004).

After differentiation, cardiomyocyte morphology and phenotype were also confirmed by observing spontaneous contraction in 2D and 3D cultures, and by the characteristic expression of both intracellular (TNNT2, NKX2-5, and α-actinin) and cell surface proteins (SIRPA and VCAM1). Cardiomyocytes (CMs) differentiated from GLA-mutant iPSCs showed preserved cellular CM phenotype, as assessed by the expression of troponin, peri-nuclear accumulation of Gb3-positive vesicles on confocal microscopy (Figure 1), and the presence of lamellar bodies on transmission electron microscopy. Decreased α-Gal A and Gb3 accumulation in mutated cells was confirmed by mass spectroscopy as well as the findings on confocal microscopy and transmission electron microscopy.

3.2. Selection of Cell Lines for In-Depth Study with CRSPR-Cas 9-Mediated Homology Repair

CRISPR-Cas9-mediated homology-directed repair was undertaken in Fabry cell lines (n = 7). The resulting iPSC lines were checked for normal karyotype and retained expression of pluripotency markers, as well as their ability to maintain stability throughout long-term culture. These corrected iPSC lines showed increased enzyme activity and a reduction in Gb3 content compared with their mutant lines. α-galactosidase activity was 7.3 ± 1.4 nmol/mg protein/h in the control lines, 0.51 ± 0.06 nmol/mg protein/h in the Fabry lines, and 5.6 ± 1.5 nmol/mg protein/h in the corrected Fabry lines (ANOVA F = 75.8, p < 0.00001. Corresponding Gb3 levels were 523.6 ± 199 pmol/mg protein control, 739 ± 176 Fabry, 270 ± 72.6 Fabry corrected (ANOVA F = 9.09, p = 0.0047).

Two cell lines and their corrected lines were then selected for in-depth studies based on their robust correction, karyotypic normality, stable pluripotent markers after genetic modification, and comparable cardiac differentiation capacity. These were a GLA^{c.1193_1196del} iPSC line, its corrected line GLA^{corr c.1193_1196del}, and a GLA^c.851T>C cell line and its corrected line GLA^{corr c.851T>C}. Patient-specific iPSC cell lines and their isogenic control cell lines were then differentiated into cardiomyocytes as specific matched experimental controls [12].

Both FD patients from whom these cell lines were derived had developed advanced disease, with renal and cardiovascular disease indicated by proteinuria, renal impairment, left ventricular hypertrophy with diastolic dysfunction, and a history of stroke (summarised in

Table 1. Clinical characteristics, patients with GLA^c.851T>C or GLA^{c.1193_1196del} variants, with reference values. Reproduced with permission from Wise AF et al., Kidney International Reports. 2024. Elsevier [15].

Characteristic (Normal Values)	Fabry GLAc.851T>C	Fabry GLAc.1193_1196del
Sex	Male	Male
Age at diagnosis	4	49
GLA mutation c	c.851T>C	c.1193_1196del
GLA mutation p	p.M284T	p.Gln398fs
Leukocyte A-galactosidase (ref range ≥1.2 nmol/mL/h)	<0.1	0.2 (truncated)
LysoGb3 (ref range <0.81 ng/mL)	27 ng/mL	24 ng/mL
Proteinuria (ref range <0.15 g/24 h)	>2 g/24 h	>2 g/24 h
LVH on ECG	Severe	Severe
Angiokeratoma	Yes	Yes
Anhidrosis	Yes	Yes
Chronic neuropathic pain	Yes	No
Brain involvement	Multiple strokes	Stroke and ischemia

). As is consistent with the diagnosis of FD, both patients exhibited characteristically reduced plasma α-GalA enzyme activity. Additionally, elevated concentrations of lysoGb3, the deacetylated derivative of the α-GalA substrate, were observed in both patients (Table 1).

3.3. Cardiomyocyte Phenotype Post CRSPR-Cas 9-Mediated Homology Repair

Gb3-positive vesicles were rarely observed in isogenic-corrected control lines on confocal microscopy. Abnormal myelin figures and dilated endoplasmic reticulum were typical of FD lines but were markedly less evident in corrected lines on transmission electron microscopy (Figure 1).

On direct measurement, Gb3 was increased in FD cardiomyocytes but reduced after mutation correction: mean of four measurements in triplicate was 1550 ± 900 in Fabry GLA^{c.1193_1196del}, correcting to 240 ± 60 in isogenic-corrected line, and in GLA^c.851T>C mutant cells was 739 ± 176, reducing to 270 ± 72.6 in isogenic controls.

3.4. Sphingolipid Analysis in Patient and Corrected Cardiomyocyte Lines

In cardiac organoids, total Gb3 (pmol/mg protein) on mass spectroscopy (four separate organoids populated for each of the four cell lines with each measurement a mean of triplicates) was 1600 ± 952 in Fabry GLA^{c.1193_1196del}, correcting to 221 ± 37 in isogenic-corrected line (p = 0.023), and in GLA^c.851T>C mutant cells was 702 ± 548, reducing to 73 ± 53 in isogenic controls (paired t = 24.1, p = 0.0002). For total Gb3, differing patterns within isoforms were noted. The pattern of Gb3 isoform distribution in mutant and corrected cardiomyocyte lines was similar in both patients studied, but compared with corrected cell lines, each mutant line contained more of the Gb3 (THC) isoform species THC d18:1/18:0 and THC18.1/16.0 and less THC d18.1/14.0 (Figure 2,

Table 2. Gb3 (THC) 18:1 isoform levels (pmol/mg) vary in patient and corrected cell lines. Measurements are averages of triplicates.

GLA Cell Line	THC d18:1/14:0	THC d18:1/16:0	THC d18:1/18:0	THC d18:1/20:0	THC d18:1/22:0	THC d18:1/24:0	THC d18:1/24:1
GLAc.851T>C	1	16	7	6	4	2	19
GLAcorr c.851T>C	9	267	129	76	38	8	189
GLAc.1193_1196del	1	13	7	6	7	2	24
GLAcorr c.1193_1196del	4	94	35	25	20	4	66

).

3.5. Functional Studies in 3D Organoids

Functional studies were undertaken using GLA^{c.1193_1196del} cells and their corrected controls. Despite clear accumulation of Gb3, studies using the heart-dyno system showed no significant differences in contractile force or rate of contraction between Fabry iPSC-derived cardiac organoids and their isogenic controls. However, acceleration and relaxation times were significantly reduced in Fabry organoids (Figure 3).

4. Discussion

Our results endorse the validity and utility of iPSC-derived cardiomyocytes as a platform to investigate Fabry cardiomyopathy. Identification of the histologic, biochemical, and functional cardiomyocyte phenotype indicate reproducible recapitulation of many features of clinical FD. Within this model, a novel potential readout for future interventional experiments studies is lipidomic analysis, although the significance of our preliminary findings presented here remains unclear, requiring further work. Altered lipidomics may provide useful pathophysiological insights as to downstream effects on cell signalling. While quantitation of cellular Gb3 accumulation remains the cornerstone for evaluation of disease intervention, our studies also suggest that a functional abnormality in cardiomyocyte relaxation could be usefully modelled and warrants further assessment. It is not surprising that we did not demonstrate uniform effects of FD across different mutations, or that not all the parameters that were studied were uniformly abnormal. For instance, we hypothesised that contractility of cardiomyocytes may be impacted in cells derived from patients with cardiomyopathy. However, despite clear accumulation of Gb3, we found no change in contractile force or speed of contraction between Fabry iPSC-derived cardiac organoids and their isogenic controls. However, the delay in organoid relaxation time mimics Fabry cardiomyopathy, where systolic function is preserved until very late, and diastolic dysfunction due to cardiac remodelling impairs diastolic relaxation and filling of the ventricle [16,17].

The iPSC-derived cardiomyocyte platform could also facilitate exploration of the roles of other genes, including those mapping to cardiac extracellular matrix pathways. It has long been recognised that Gb3 itself does not quantitatively explain the increased cardiac mass in Fabry cardiomyopathy, but that extracellular matrix accumulation plays an important role in the development of left ventricular hypertrophy [16,17].

We have recently extended these studies to include label-free proteomic analysis in this CM system, reporting significant differences in proteomic expression between Fabry and normal CMs and between Fabry and corrected cell lines [12]. Pathway analysis for over-expressed proteins revealed significant enrichment for lysosomal proteins including beta-glucocerebrosidase (encoded by GBA), the upregulation of which was previously reported in cardiomyocytes from patients with Fabry disease [18].

Overall, our studies endorse the value of iPSC-derived CMs in both recapitulating the Fabry disease phenotype and facilitating depth studies of individual GLA mutations. Other groups of researchers have also confirmed multiple aspects of the FD phenotype to be reproducible in iPSC-derived cardiomyocytes, including reduced GLA activity and Gb3 accumulation [18,19,20,21]. Cellular hypertrophy, impaired contractility, and altered energy metabolism have also been reported [19].

Variations between individual patients with FD are very common. In the classic mutations studied here, the missense GLA^c.851T>C and the truncated GLA^{c.1193_1196del}, frame readouts of the individual peptide components of α-galactosidase varies [20], further highlighting the complexities involved in cellular pathways downstream of the GLA mutation itself.

To date, we have only studied male patients with FD due to the X-linked nature of the disease and its gender-specific but variable clinical presentation in female patients. However, at a single cell level, the consequences of GLA mutation would be expected to be similar to male cells with the same mutation, even when the clinical phenotype varies. Understanding how specific GLA variants disrupt cellular function in Fabry disease can potentially offer new logical personalised therapies and humanuman iPSC disease modelling can offer fresh insights into organ-specific effects. We and many others also use this model to study Fabry podocytes, the key cell inducing the proteinuric Fabry nephropathy phenotype [15,20,21].

In summary, these studies confirm that the cellular phenotype of Fabry disease is recapitulated in iPSC-derived cardiomyocytes. The platform offers utility across many diseases to further investigate genotype–phenotype links, and to use identified pathogenetic pathways to personalise therapy.