Abstract
Background
Fabry disease (FD) is a rare X-linked lysosomal storage disorder caused by pathogenic variants in the GLA gene and commonly involves the heart. This study aimed to investigate the pathogenicity of a novel missense variant and its amenability to migalastat.
Methods
A retrospective analysis was conducted on 28 patients with FD collected over the past 10 years. Pathogenicity was assessed within a family carrying a de novo GLA missense variant through myocardial histopathology, bioinformatic predictions, protein structural modelling and in vitro cellular assays.
Results
In a cohort of 28 patients (15 males, 13 females), cardiac involvement was prevalent. We further reported a family with GLA c.227T>A (p.Met76Lys) variant. Myocardial biopsy showed cardiomyocyte hypertrophy with diffuse vacuolisation, interstitial fibrosis and microvascular dysfunction. Immunohistochemistry demonstrated markedly decreased expression of α-galactosidase A (α-Gal A) in vacuolated cardiomyocytes. Biochemical testing revealed low-to-normal α-Gal A enzymatic activity together with elevated plasma globotriaosylsphingosine (lyso-Gb3). Bioinformatic analyses suggested this variant was probably pathogenic by altering the three-dimensional structure of GLA protein. AC16 cardiomyocytes overexpressing wild-type and mutant GLA were constructed via lentiviral vectors. Immunofluorescence staining revealed that the intracellular localisation of the mutant protein remained unaltered, whereas α-Gal A enzyme activity was significantly reduced. Treatment with migalastat, a pharmacological chaperone for α-Gal A, partially restored the mutant α-Gal A enzymatic activity.
Conclusions
This study identified GLA c.227T>A (p.Met76Lys) as a novel pathogenic variant that primarily causes cardiac dysfunction. Migalastat may represent a therapeutic option for this variant.
Keywords: Genetic Association Studies; Genetic Diseases, Inborn; Magnetic Resonance Imaging; Echocardiography
WHAT IS ALREADY KNOWN ON THIS TOPIC.
WHAT THIS STUDY ADDS
This study identifies a novel GLA variant, c.227T>A (p.Met76Lys), and supports it is pathogenic based on family analysis, myocardial biopsy findings and bioinformatic modelling.
Experiments in AC16 cardiomyocytes further indicate that migalastat treatment partially restores the enzyme activity reduced by this variant.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Reclassifying GLA p.Met76Lys from VUS to pathogenic variation enhances diagnostic precision.
Functional assays in AC16 cells suggest that migalastat may serve as a potential therapeutic option.
Migalastat is an oral chaperone therapy that stabilises specific mutant enzymes. However, it is effective only for patients with amenable (drug-responsive) genetic variants.
Introduction
Fabry disease (FD; OMIM #301500) is an X-linked lysosomal storage disorder caused by GLA variants, leading to α-galactosidase A (α-Gal A) deficiency and systemic accumulation of globotriaosylceramide (Gb3) and lyso-Gb3.1 2 While hundreds of pathogenic variants exist, genotype–phenotype correlations remain complex, particularly in females due to X-chromosome inactivation.1,4 Cardiac involvement drives morbidity in FD, manifesting as left ventricular hypertrophy (LVH), arrhythmias and fibrosis. Advanced imaging, including two-dimensional speckle-tracking echocardiography (2D-STE) and cardiovascular magnetic resonance (CMR) with characteristic low native T1 and late gadolinium enhancement (LGE), is crucial for diagnosis. However, phenotypic overlap with other cardiomyopathies necessitates integrating imaging with genetic and biochemical assessments.5,7
Current treatments include enzyme-replacement therapy (ERT) and the oral pharmacological chaperone migalastat. Migalastat stabilises specific ‘amenable’ mutant enzymes to facilitate lysosomal trafficking, offering an alternative to ERT, though genotype-response relationships require precise characterisation.1 8 9
In this study, we identified a novel GLA variant, c.227T>A (p.Met76Lys), initially classified as a variant of uncertain significance (VUS), in a proband with severe cardiomyopathy from a cohort of 28 patients with FD. Through pedigree analysis, myocardial biopsy, structural modelling and functional validation in AC16 cardiomyocytes, we aimed to confirm its pathogenicity and evaluate its responsiveness to migalastat, providing evidence for a precise diagnosis and treatment selection.10 11
Materials and methods
Patient selection
In this retrospective study, 28 unrelated patients with FD treated at the Affiliated Drum Tower Hospital of Nanjing University Medical School (2015–2025) were enrolled. Diagnoses were established based on the 2021 Chinese Expert Consensus.12
Clinical data collection
Clinical data were retrospectively extracted from electronic medical records. Categories included cardiac assessments (symptoms, ECG, echocardiography with global longitudinal strain (GLS), CMR, biopsy), renal findings (proteinuria, chronic kidney diseas (CKD) stage and neurological involvement (stroke/transient ischemic attack (TIA), neuropathic pain, hypohidrosis, hearing loss). Data availability was 23/28 (82.1%) for ECG, 24/28 (85.7%) for echocardiography, 14/28 (50.0%) for 2D-STE and 9/28 (32.1%) for CMR. One proband underwent biopsy. Overall, 25 of 28 patients (89.3%) underwent at least one cardiac evaluation. All datasets were anonymised prior to analysis.
Cardiac involvement evaluation
Standard 12-lead ECGs were reviewed by two cardiologists for arrhythmias, conduction blocks and morphological abnormalities. Echocardiography was performed using a Philips EPIQ 7 system following the American Society of Echocardiography (ASE) guidelines. 2D-STE for GLS was analysed offline using EchoPAC (GE Healthcare). CMR was performed on a 3.0-T Siemens Skyra scanner. The protocol included cine imaging, native T1 mapping (MOLLI) from the interventricular septal myocardium on the short-axis, and LGE imaging (0.1 mmol/kg gadoterate meglumine), with analysis performed using cvi42 (Circle Cardiovascular Imaging). Septal myocardium obtained from the proband during myectomy was paraffin-embedded for H&E, histochemistry and immunohistochemistry staining.
Biochemical assays
GLA activity and Lyso-Gb3 levels were measured in dried blood spots (DBSs) using liquid chromatography–tandem mass spectrometry (LC–MS/MS) (Waters UPLC/Xevo TQD). GLA activity was quantified using the NeoLSD MS/MS kit (PerkinElmer), while Lyso-Gb3 concentrations were determined via multiple reaction monitoring against standard calibration curves.
Genetic testing
Genomic DNA isolated from DBS underwent next-generation sequencing on an Illumina NovaSeq platform following long-range PCR enrichment of the GLA locus. Variants were confirmed by Sanger sequencing and reported according to Human Genome Variation Society nomenclature.
Bioinformatic predictions
Variants were annotated (NM_000169.3) and queried in public databases (ClinVar, gnomAD, HGMD) to assess population frequency. Evolutionary conservation was evaluated via multiple-sequence alignment. Pathogenicity was predicted using SIFT, PolyPhen-2, MutationTaster and CADD. Structural impact was analysed using a SWISS-MODEL homology model13 visualised in PyMOL to inspect local side-chain interactions and hydrogen-bonding networks.
Cell experiments
Cell culture and transfection
AC16 human cardiomyocytes (Procell Life Science, China) were cultured in Dulbecco’s Modified Eagle’s Medium/F-12 (A4192001, Thermo Fisher) supplemented with 12.5% fetal bovine serum (10099–141, Thermo Fisher) and 1% penicillin–streptomycin (15140122, Thermo Fisher). Lentiviral vectors (Figure S1) encoding NC, wild-type or mutant GLA (c.227T>A) (Nanjing Zebrafish Biotech, China) were transfected at a multiplicity of infection (MOI) of 50. Stable cell lines were established following selection with puromycin (S7417, Selleck, China) for 14 days.
Migalastat Drug Intervention
To assess pharmacological chaperone responsiveness, AC16 cells were incubated with 10 µM migalastat (GR181413A, MedChemExpress) for 5 days.10 14 Based on established GLP-HEK standards, the variant was classified as amenable if α-Gal A activity showed a ≥1.2-fold increase over baseline and a ≥3.0% absolute increase relative to wild-type activity.8 9 14
Quantitative real-time PCR
Total RNA was extracted using TRIzol (Invitrogen) and reverse-transcribed using a complementary DNA synthesis kit (Vazyme). Quantitative PCR (qPCR) was performed on an ABI Prism 7000 instrument (Thermo Fisher) using Vazyme SYBR qPCR Master Mix. Relative GLA messenger RNA (mRNA) expression was calculated using the 2−ΔΔCT method with GAPDH as the internal control.
Western blot analysis
Cells were lysed in RIPA buffer supplemented with 1 mM phenylmethanesulfonyl fluoride (PMSF) (Vazyme). Equal amounts of protein (20 µg) were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) (PG212, Epizyme), transferred to polyvinylidene fluoride (PVDF) membranes (Bio-Rad) and blocked with 5% non-fat milk. Membranes were incubated with rabbit anti-α-Gal A (1:1000, ab168341, Abcam) and mouse anti-GAPDH (1:5000, 60004–1 Ig, Proteintech). Bands were visualised by chemiluminescence and quantified using Image J.
α-Gal A activity assay
Cells were sonicated in extraction buffer, and supernatants were assayed for α-Gal A activity using a commercial kit (Boxbio, AKSU041C) according to the manufacturer’s instructions. Absorbance was measured at 400 nm. Enzyme activity was calculated via a p-nitrophenol standard curve and expressed as p-nitrophenol production per 10 000 cells per hour.
Statistical analysis
Data were analysed using GraphPad Prism V.9.3.0. Results from three independent biological replicates are presented as mean±SD. Differences were assessed using unpaired Student’s t-tests for two groups and one-way analysis of variance followed by Tukey’s post hoc test for multiple groups. Statistical significance was defined as p<0.05.
Results
Clinical characteristics and cardiac imaging
This retrospective cohort included 28 patients with FD, with an onset age ranging from 10 to 55 years . Clinical features are summarised in table 1 according to the expert consensus on FD in China (2021).12 Cardiac involvement was most common (84.0%), including conduction abnormalities (30.4%) and arrhythmias (17.4%). Echocardiography showed LVH in 45.8%, while 2D-STE revealed impaired GLS (85.7%).
Table 1. Clinical manifestations of multiorgan system involvement in patients with Fabry disease (n=28).
| System | Manifestation | Criteria | N/N(%) |
|---|---|---|---|
| Cardiac | Any cardiac involvement | 21/25 (84.0) | |
| Arrhythmias (ECG) | AF, AFL, SVT or ventricular arrhythmia | 4/23 (17.4) | |
| Conduction abnormalities (ECG) | PR interval <120 ms, AV block or bundle branch block | 7/23 (30.4) | |
| LVH (Echo) | IVSd ≥13 mm or LVMi >115/95 g/m² (male/female) | 11/24 (45.8) | |
| Impaired GLS (2D-STE) | GLS > −18% | 12/14 (85.7) | |
| LGE (CMR) | Any myocardial late gadolinium enhancement | 6/9 (66.7) | |
| Reduced native T1 (CMR) | Native T1 <1100 ms | 4/9 (44.4) | |
| Renal | Any renal involvement | 14/28 (50.0) | |
| Proteinuria | 24 hours urine protein >150 mg | 10/28 (35.7) | |
| CKD stage ≥2 | eGFR <90 mL/min/1.73 m² | 7/28 (25.0) | |
| Neurological | Any neurological involvement | 5/28 (17.9) | |
| Stroke/TIA | Confirmed by clinical evaluation or imaging | 0/28 (0.0) | |
| Neuropathic pain | Confirmed by clinical evaluation | 2/28 (7.1) | |
| Hypohidrosis | Confirmed by clinical evaluation | 1/28 (3.6) | |
| Hearing loss | Confirmed by clinical evaluation or audiometry | 2/28 (7.1) |
Data are presented as n/N (%). Denominators differ by modality (ECG n=23, echocardiography n=24, 2D-STE n=14, CMR n=9). ‘Cardiac involvement’ was defined as the presence of at least one specified abnormality on ECG, TTE or CMR.
AF, atrial fibrillation; AFL, atrial flutter; AV, atrioventricular; CKD, chronic kidney disease; CMR, cardiovascular magnetic resonance; 2D-STE, two-dimensional speckle-tracking echocardiography; eGFR, estimated glomerular filtration rate; GLS, global longitudinal strain; IVSd, interventricular septal thickness in diastole; LGE, late gadolinium enhancement; LVH, left ventricular hypertrophy; LVMi, left ventricular mass index; SVT, supraventricular tachycardia; TIA, transient ischemic attack; TTE, transthoracic echocardiography.
To illustrate cardiac involvement in this cohort, figure 1 presents multimodal cardiac imaging from a representative 51-year-old male. Echocardiography demonstrated diffuse myocardial thickening (figure 1A–B), and colour Doppler showed moderate mitral regurgitation (figure 1C). Two-dimensional speckle-tracking revealed reduced GLS (−6.6%) (figure 1D). CMR confirmed LVH on cine imaging (figure 1E–F), along with LGE and reduced native T1 values in the interventricular septum (figure 1G–H), consistent with Fabry-related cardiac involvement.
Figure 1. Representative ECHO and CMR findings in patients with FD. (a–b) Apical four-chamber and parasternal long-axis views show diffuse myocardial hypertrophy affecting the left and right ventricles as well as the interventricular septum, with marked thickening observed in the left ventricle and the septum. (c) Colour Doppler image demonstrates moderate mitral regurgitation. (d) Two-dimensional speckle-tracking bull’s-eye plot shows significantly reduced global longitudinal strain (GLS = −6.6%, reference range −18% to −22%). (e–f) Four-chamber and short-axis cine (SSFP): concentric left ventricular myocardial hypertrophy. (g) Short-axis LGE (PSIR): Mid-wall enhancement of the left ventricular inferior and lateral myocardium. (h) Short-axis T1 mapping: significantly reduced myocardial native T1 values (897 ms, reference <1100 ms). CMR, cardiovascular magnetic resonance; FD, Fabry disease; GLS, global longitudinal strain; LGE, late gadolinium enhancement, PSIR, phase-sensitive inversion recovery; SSFP, steady-state free precession.
A novel GLA variant (c.227T>A, p.Met76Lys)
A 44-year-old female proband presented with a 1-month history of back pain and dyspnoea. B-type natriuretic peptide was markedly elevated (768.0 pg/mL). ECG revealed sinus rhythm, LVH with ST-T abnormalities and ventricular pre-excitation, while echocardiography suggested obstructive hypertrophic cardiomyopathy. Renal function tests showed no significant abnormalities. She subsequently underwent surgical septal myectomy (SSM) with a concurrent septal biopsy. DBS testing showed elevated lyso-Gb3 in both the proband and her 19-year-old son. Sequencing revealed a novel GLA p.Met76Lys. Together with supportive pathological findings and markedly elevated lyso-Gb3 levels, these data strengthen the evidence for pathogenicity of this p.Met76Lys variant.
Detailed endomyocardial biopsy
Interventricular septum biopsy demonstrated features consistent with lysosomal storage disease (LSD)-associated cardiomyopathy and/or hypertrophic cardiomyopathy. Cardiomyocytes showed marked hypertrophy with enlarged, hyperchromatic, irregular nuclei and occasional multinucleation (figure 2A). Cardiomyocyte diameter exceeded 30 µm, approximately twice normal. Cardiac myocytes were focally arranged in non-parallel intersecting fascicles with storiform-type disarray centred on collagen (figure 2B). Widespread sarcoplasmic vacuolisation was prominent (figure 2C) and periodic acid-Schiff staining showed diffuse vacuolar positivity indicative of intralysosomal glycolipid storage (figure 2D). Pericellular interstitial fibrosis, characterised by collagen enveloping individual myocytes, was observed (figure 2E). Masson trichrome staining demonstrated replacement fibrosis, with collagen replacing portions of the myocardium (figure 2F). Marked fibrotic endocardial thickening was evident (figure 2G), with small foci of chronic lymphocytic infiltration (figure 2H). Vascular dilatation was observed at the endocardial–myocardial junction (figure 2I), and in the myocardium with dilated vascular channels beneath the endocardium (figure 2J). Small intramural veins showed thickening of the smooth-muscle walls with mild luminal narrowing (figure 2K). Elastic fibre staining demonstrated elastic fibres within the vascular smooth muscle (figure 2L). Immunohistochemistry (IHC) revealed α-Gal A expression was markedly reduced within the cytoplasmic vacuoles of cardiomyocytes (figure 2M–N), whereas control myocardium exhibited granular cytoplasmic positivity (figure 2O–P). To exclude cardiomyopathies with overlapping features, additional IHC markers were assessed (online supplemental figure S2). There was no decrease in expression of desmin, dystrophin, lysosome-associated membrane protein 2 and antimitochondrial antibody (online supplemental figure S2), ruling out other myocardial diseases similar to FD, such as muscular dystrophies, Danon disease and mitochondrial cardiomyopathy.
Figure 2. Endomyocardial biopsy from the proband carrying GLA p.M76K variant. (a) H&E, ×400: cardiomyocyte hypertrophy with enlarged nuclei. (b) H&E, ×200: cardiac myocytes in non-parallel, storiform-disarray. (c) H&E, ×400: diffuse sarcoplasmic vacuolisation. (d) PAS, ×400: indicative of glycosphingolipid storage. (e) H&E, ×200: pericellular-type interstitial fibrosis. (f) Masson trichrome, ×200: myocytes was focal replaced by fibrosis. (g) H&E, ×100: marked endocardial fibrotic thickening. (h) H&E, ×100: small foci of chronic lymphocytic infiltration. (i) H&E, ×100: vascular dilatation at the endocardial–myocardial junction. (j) H&E, ×200: dilated vascular channels beneath endocardium. (k) H&E, ×400: small intramural vessels with mild luminal narrowing. (l) Elastic fibre staining, ×400: smooth-muscle wall thickening. (m–n) H&E and α-Gal A IHC, ×200: markedly reduced α-Gal A expression within the cytoplasmic vacuoles of cardiomyocytes. (o–p) H&E and α-Gal A IHC, ×200: normal control myocardium exhibited granular cytoplasmic positivity. α-Gal A, α-galactosidase A; IHC, immunohistochemistry; PAS, periodic acid-Schiff.
Genetic and biochemical marker testing
The biochemical results were concordant with the clinical and pathological findings (table 2). The proband exhibited α-Gal A activity in a low-normal range (3.56; reference range 2.20–17.65 µmol/L/hour), together with an elevated plasma lyso-Gb3 (11.84; reference <1.11 µg/L). The affected son showed markedly reduced α-Gal A activity, below 30% of the reference value (0.48; reference range 2.20–17.65 µmol/L/hour) and a significantly elevated plasma lyso-Gb3 level (72.58; reference <1.11 µg/L). The pedigree demonstrated X-linked inheritance and showed a novel GLA variant (c.227T>A, p.Met76Lys) present in the heterozygous mother (figure 3A) as well as her hemizygous son (figure 3B). The variant is absent from clinical databases and is currently classified as a VUS under American College of Medical Genetics and Genomics/Association for Molecular Pathology criteria.11
Table 2. Genetic, enzymatic and biomarker data for the proband and the affected son carrying GLA c.227T>A (p.Met76Lys) variant.
| Patient | Genomic position | Variant | Exon/intron | gnomAD max AF | Classification | α-Gal A activity (μmol/L/hour) | Lyso-Gb3 (μg/L) |
|---|---|---|---|---|---|---|---|
| Proband | chrX:100 658 941 | c.227T>A | exon2 | — | VUS | 3.56 | 11.84 |
| Son | chrX:100 658 941 | c.227T>A | exon 2 | — | VUS | 0.48 | 72.58 |
Note: α-Gal A activity assessment of dried blood spots (nmol/mL/hour; reference range 2.20–17.65) and plasma lyso-Gb3 (ng/mL; reference <1.11) are shown; Sanger sequencing confirmed heterozygosity in the proband and hemizygosity in her affected son.
AF, allele frequency; gnomAD, genome Aggregation Database; VUS, variant of uncertain significance; α-Gal A, α-galactosidase A.
Figure 3. Bioinformatic and structural evidence supporting the pathogenicity of the novel GLA missense variant. (a–b) Sanger sequence confirming GLA c.227T>A (p.M76K) in the proband (heterozygous) and her affected son (hemizygous). (c) Analysis of GLA gene conservation at position 76 across different species. The aligned amino-acid position is highly conserved (7/7) among Homo sapiens, Mus musculus, Gallus gallus, Danio rerio, Rattus norvegicus, Pan troglodytes and Canis lupus familiaris. (d) PolyPhen-2 in predictions for the GLA c.227T>A (p.M76K) substitution. HumVar models predicted the variant as ‘probably damaging’ with HumVar score 0.971 (sensitivity 0.60; specificity 0.93). This novel c.227T>A variant is predicted to be probably damaging. (e) Three-dimensional structural modelling of p.M76K in α-Gal A. Left: overall ribbon diagram with the mutated region boxed. Colour code: cyan represents α-helix; purple represents β-sheet; red dashed lines represents hydrogen bonds. Middle and right: close-ups comparing WT and MU residues at position 76, highlighting local geometry and hydrogen-bonding. MU, mutant; WT, wild-type.
Bioinformatic evidence for p.Met76Lys variant
A sequence comparison of the GLA protein among seven vertebrate species (human, chimpanzee, dog, mouse, rat, chicken and zebrafish) shows that methionine (Met, M) at residue 76 is highly conserved (figure 3C). This position received a multiple sequence alignment (MSA) conservation score of 5 (table 3). The conservation suggests that the residue is probably crucial for protein function. In contrast, the patient’s allele encodes lysine at this locus (p.M76K). Four bioinformatic prediction algorithms classify p.M76K variation as high probability of pathogenicity. PolyPhen-2 predicted this p.M76K substitution to be ‘probably damaging’ according to HumVar1 models, with HumVar giving a score of 0.971 (sensitivity 0.60; specificity 0.93) (figure 3D). Other tools yielded concordant results, including SIFT (damaging), MutationTaster (deleterious) and CADD (probably damaging) (Table 3)
Table 3. Summary of the pathogenicity prediction scores of the GLA c.227T>A variant using bioinformatics online software tools.
| Nucleotide change | Amino acid change | PolyPhen-2 | Mutation taster | CADD | SIFT | Grantham Matrix Score | MSA score | Total score |
|---|---|---|---|---|---|---|---|---|
| C.227T>A | p.M76K | Probably damaging | Deleterious | Probably damaging | Damaging | 95 (4) | 7/7 (5) | 9 |
Grantham Matrix Scores15 16 (shown in parentheses) are converted to points as follows: <60.0 = 1 point; 60.0–78.3 = 2 points; 78.4–93.4 = 3 points; >93.4 = 4 points; and any substitution to cysteine = 5 points. MSA scores indicate the number of conserved species among the seven tested species: 7/7 = 5 points; 6/7 = 4 points; 4–5/7 = 3 points; 1–3/7 = 1 point. Total score ≥8 indicates a high probability of pathogenicity; 6–7 indicates a moderate probability; ≤5 indicates likely non-pathogenic. Our mutation scores 9, high pathogenicity.
MSA, multiple sequence alignment.
The Grantham distance quantifies physicochemical differences between amino acids,15 16 integrating composition, polarity and molecular volume into a single score to evaluate substitution impact. Higher scores indicate greater physicochemical changes. This metric supports variant functional impact assessment, alongside evolutionary conservation, structural context and functional data. In our study the Grantham score for the Met to Lys substitution (p.M76K) is 95, indicating a moderately large physicochemical change due to the longer, positively charged lysine replacing the neutral, non-polar methionine. To further evaluate the potential pathogenicity of p.M76K, we used PyMOL to model the three-dimensional structures of the wild type and mutant proteins. The Met to Lys substitution is predicted to introduce a new hydrogen bond with the backbone near Tyr86. This small local shift suggests partial destabilisation of the local structure (figure 3E). Moreover, this mutation site may lead to the disruption of the disulfide bond between C94 and C52, resulting in the loosening of the overall protein structure and compromising its capacity to bind the lyso-Gb3 substrate effectively. Taken together with the conservation and functional data above, these results support the notion that the substitution could impair protein function.
Functional validation of p.Met76Lys
To further assess the impact of the missense variant on α-Gal A expression and activity in vitro, we established AC16 cardiomyocyte lines stably overexpressing wild-type GLA (GLA-WT), mutant GLA carrying p.Met76Lys (GLA-MU) or an empty-vector control (NC) through lentiviral vectors carrying GFP markers.
Quantitative real-time PCR (RT-qPCR) (figure 4A) demonstrated effective transfection in both GLA-MU and GLA-WT mRNA levels. Western blotting analysis was performed using anti-GLA antibody (figure 4B). Although GLA-MU demonstrated a slight reduction, this was not statistically significant (figure 4C). Representative immunofluorescence images at 100×magnification (figure 4D) show comparable transfection efficiency among AC16 cells carrying GLA-WT, GLA-MU or NC vectors, as indicated by similar proportions and fluorescence intensity of GFP-positive cells across fields. At 400× magnification (figure 4E). Immunofluorescence in AC16 cells showed no obvious difference in cellular localisation between GLA-MU and GLA-WT.
Figure 4. Functional validation of the GLA p.M76K variant in AC16 cardiomyocytes. (a) Relative GLA mRNA expression in AC16 cells transfected with NC, GLA-WT or GLA-MU constructs. (b–c) Representative Western blots and corresponding quantification of α-Gal A protein expression. (d) Immunofluorescence (100×) shows comparable transfection efficiency in AC16 cardiomyocytes of GLA-WT, GLA-MU or NC by lentiviral vectors. (e) Immunofluorescence (400×) of AC16 cardiomyocytes transfected with GLA-WT or GLA-MU lentivirus showed that both wild-type and mutant GLA localised to the cytoplasm. (f) α-Gal A enzymatic activity in AC16 cardiomyocytes transfected with GLA-WT, GLA-MU or NC. (G) α-Gal A enzymatic activity before and after treatment with 10 µM DGJ. *p<0.05, **p<0.01. α-Gal A, α-galactosidase A; DAPI, 4',6-diamidino-2-phenylindole; mRNA, messenger RNA; ns, not significant; MU, mutant; WT, wild-type.
To further validate whether GLA-MU protein impairs α-Gal A function, biological activity of α-Gal A was measured in cell lysates. Importantly, α-Gal A enzymic activity in GLA-MU cells was markedly reduced compared with GLA-WT (figure 4F), and this reduction is independent of protein expression levels. These findings suggested the p.Met76Lys variant potentially impairs enzyme function rather than protein quantity, likely through conformational alterations that diminish catalytic activity. This mutation causes the amino acid at position 76 of the GLA protein to be substituted from methionine to lysine, which leads to the formation of a new hydrogen bond and disruption of the original formed disulfide bond. This alteration may result in changes of three-dimensional structure of the GLA protein, impairing its ability to bind tightly with the substrate, leading to the accumulation of the substrate GB3 and subsequent disease progression.
Migalastat enhances α-Gal A activity
It is reported that the oral molecular chaperone migalastat (DGJ) binds to the active site of unstable mutant GLA, enabling its transport to the lysosome and restoring biological activity.17 To assess DGJ responsiveness, we measured α-Gal A activity before and after drug exposure in AC16 human cardiomyocytes. Treatment of cells with 10 µM DGJ for 5 days increased α-Gal A activity in both GLA-MU and GLA-WT cells (figure 4G), but to different degrees. This increase was more pronounced in GLA-MU and reached statistical significance (p<0.05). Before drug treatment the ratio of GLA enzymatic activity in MU to WT was 37.36%. After DGJ treatment this ratio increased to 45.7%, representing an absolute increase of 8.36 percentage points and a 1.36-fold change relative to baseline. Both the absolute increase and the fold change surpass the GLP-HEK criteria for amenability, which require an absolute increase greater than 3% and a fold change greater than 1.2 (table 4). These data indicate that the p.Met76Lys missense variant responds favourably to DGJ in AC16 cells.
Table 4. Comparison of α-Gal A activity in WT and MU AC16 cardiomyocytes before and after DGJ intervention.
| Nucleotide alteration | Amino acid change | −10 µM DGJ | +10 µM DGJ | Absolute increase at 10 µM (%WT) | α-Gal A activity at 10 µM—fold over baseline | Meet amenable GLA mutation criteria? (Yes/No) | ||
|---|---|---|---|---|---|---|---|---|
| α-Gal A activity | MU/WT% | α-Gal A activity | MU/WT% | |||||
| GLA-WT | 0.773±0.058 | / | 0.859±0.044 | / | / | / | / | |
| GLA-MU c.227T>A |
M76K | 0.289±0.030 | 37.387 | 0.393±0.04* | 45.750 | 8.363 | 1.360 | Yes |
Note: amenable mutation criteria for GLA gene mutations applicable to migalastat are proposed by the GLP HEK assay.
Asterisk indicates that the enzyme activity of MU after (+DGJ) treatment was statistically significant compared with that before treatment (−DGJ), p<0.05. The 1.36-fold increase was calculated as α-Gal A activity in MT with 10 µM (+DGJ)/α-Gal A activity in MT without (−DGJ). The absolute increase in α-Gal A activity (8.339) is calculated by the percentage of MU/WT with (+DGJ, 45.7%) minus the baseline (−DGJ, 37.36%). Both the absolute increase and the fold change surpass the criteria for drug amenability, which require an absolute increase greater than 3% percentage points and a fold change greater than 1.2 times.
MU, mutant; WT, wild-type; α-Gal A, α-galactosidase A.
Discussion
In our cohort, cardiac involvement was substantial, with 21/25 (84.0%) assessed patients showing abnormalities on multimodality imaging. Consistent with prior reports, males presented earlier with greater structural burden,1 while female phenotypes ranged from asymptomatic to severe. Notably, advanced imaging detected early myocardial involvement even in the absence of LVH: 12/14 (85.7%) had impaired GLS and 4/9 (44.4%) showed low native T1.5 6 18 These findings support a proactive approach for early detection and therapeutic intervention. The higher prevalence of cardiac involvement in our study compared with global registries (Fabry Outcome Survey, Fabry Registry)19 20 likely reflects the increased sensitivity of 2D-STE and CMR.
We specifically highlight a 44-year-old female proband initially misdiagnosed with hypertrophic obstructive cardiomyopathy (HOCM). While HOCM is typically caused by sarcomeric mutations (prevalence ∼1:500),21 FD is an important differential diagnosis, accounting for 0.5–2.0% of suspected HOCM cases.21 Screening for FD is crucial, particularly in patients lacking a definitive family history or responding poorly to conventional therapy. In this case, the diagnosis was confirmed via endomyocardial biopsy performed during SSM to relieve outflow tract obstruction.
Our study presents comprehensive myocardial pathological alterations in FD for the first time. The characteristic histopathological feature was extensive vacuolisation of cardiomyocytes. More than half of cardiomyocytes exhibited cytoplasmic vacuoles of varying sizes, and these vacuoles extended beyond the subendocardial layer, indicating a potential association with LSD. Moreover, myocardial fibres lost their normal parallel arrangement and displayed myocyte disarray, occasionally forming vortex-like or herringbone patterns. Although these structural patterns lack specific diagnostic value, they reflect underlying myocardial structural remodelling. Myocardial interstitial fibrosis and endocardial thickening manifested focal pericellular and replacement fibrosis with scattered chronic inflammatory cell infiltrates. Additionally, vascular dilatation and microvascular abnormalities were noted in the interventricular septum, including eccentric thickening of the smooth muscle walls of intramural microvessels and mild luminal stenosis. Such vascular lesions were more commonly seen in areas of myocyte disarray and fibrosis, suggesting that they may be secondary to myocardial structural abnormalities during disease progression rather than originating from primary vascular disease. Long-term vascular injury may contribute to myocardial hypertrophy and subsequent fibrosis, thereby forming a vicious cycle.
This study systematically applied four widely used prediction algorithms, including PolyPhen-2, MutationTaster, CADD and SIFT, to analyse the novel variant c.227T>A. These tools integrated analysis strategies such as evolutionary conservation and structural modelling. The results consistently indicated that the p.Met76Lys substitution occurs at a highly conserved residue essential for GLA protein folding. This crucial structural alteration may induce local conformational instability, thereby disrupting the spatial arrangement and dimerisation ability of the α-Gal A protein.
α-Gal A functions as a homodimer with two domains per monomer, with the active site located in the first domain.22 23 Although the c.227T>A variant lies outside this catalytic site, structural modelling suggests that changes in polarity and hydrogen bonding disrupt local conformation and stability. This instability likely leads to the retention and degradation of misfolded α-Gal A in the endoplasmic reticulum, limiting lysosomal delivery and resulting in enzyme deficiency.24 25
Functional validation is essential because bioinformatics predictions remain theoretical. We transfected AC16 cardiomyocytes with GLA-MU, GLA-WT or NC lentiviral vectors and confirmed efficiency via RT-qPCR, Western blotting and immunofluorescence. GLA-MU and GLA-WT cells displayed similar protein levels and localisation. However, α-Gal A enzymatic activity was significantly reduced in the GLA-MU group. This indicates that the p.Met76Lys variant impairs catalytic function rather than protein expression, confirming its pathogenicity. Importantly, in vitro treatment with DGJ restored α-Gal A enzymatic activity. This suggests the mutant enzyme retains sufficient conformational plasticity to be stabilised by DGJ. Since DGJ facilitates protein folding and endoplasmic reticulum exit,26,28 these findings highlight the variant’s potential amenability to chaperone therapy.
This study has certain limitations. CMR was not performed in the proband due to financial constraints, limiting further evaluation of familial cardiac involvement. Electron microscopy and Gb3 staining were unavailable because of formalin-fixed paraffin-embedded (FFPE) specimens and lack of suitable commercial primary antibody. Sarcomeric gene testing was also not conducted, as only FFPE tissue remained, which is less optimal than blood-based analysis. Additionally, although in vitro findings suggested potential responsiveness of the GLA variant to migalastat, the drug is not yet approved in China, and its clinical efficacy and safety require confirmation through longitudinal follow-up.
In conclusion, this study validated the pathogenicity of a novel GLA missense variant c.227T>A (p.Met76Lys) through myocardial biopsy pathology, enzymatic and genetic testing, bioinformatics analysis, as well as subsequent functional validation in AC16 cardiomyocytes and DGJ rescue experiments in vitro.
Supplementary material
Acknowledgements
The pathological specimens of FD used in this study were provided by the department of biobank, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School. This study was supported by Nanjing Health Science and Technology Development Major Project (ZDX22001).
Footnotes
Funding: This study was supported by fundings for Clinical Trials from the Affiliated Drum Tower Hospital, Medical School of Nanjing University (2025-LCYJ-PY-09, 2025- LCYJ-MS-09), the National Natural Science Foundation of China (81671113), the Natural Science Foundation of Jiangsu Province (BK20231120) and the Nanjing Medical Science and Technology Development Foundation (ZKX22025).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by Medical Ethics Committee of Nanjing Drum Tower Hospital, Nanjing University Medical School, 2025-1136-01. Participants gave informed consent to participate in the study before taking part.
Data availability free text: Data supporting the study findings are available from the corresponding author upon reasonable request.
Data availability statement
Data are available upon reasonable request.
References
- 1.Germain DP, Altarescu G, Barriales-Villa R, et al. An expert consensus on practical clinical recommendations and guidance for patients with classic Fabry disease. Mol Genet Metab. 2022;137:49–61. doi: 10.1016/j.ymgme.2022.07.010. [DOI] [PubMed] [Google Scholar]
- 2.Aerts JM, Groener JE, Kuiper S, et al. Elevated globotriaosylsphingosine is a hallmark of Fabry disease. Proc Natl Acad Sci U S A. 2008;105:2812–7. doi: 10.1073/pnas.0712309105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.National Center for Biotechnology I ClinVar: GLA (galactosidase alpha) gene. [24-Sep-2025]. https://www.ncbi.nlm.nih.gov/clinvar?term=GLA%5Bgene%5D Available. Accessed.
- 4.Viggiano E, Politano L. X Chromosome Inactivation in Carriers of Fabry Disease: Review and Meta-Analysis. Int J Mol Sci. 2021;22:7663. doi: 10.3390/ijms22147663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lanzillo C, Fedele E, Martino A, et al. Cardiac magnetic resonance in Fabry disease. Eur Heart J Suppl. 2023;25:C200–4. doi: 10.1093/eurheartjsupp/suad045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Nordin S, Kozor R, Medina-Menacho K, et al. Proposed Stages of Myocardial Phenotype Development in Fabry Disease. JACC Cardiovasc Imaging. 2019;12:1673–83. doi: 10.1016/j.jcmg.2018.03.020. [DOI] [PubMed] [Google Scholar]
- 7.Cotella J, Randazzo M, Maurer MS, et al. Limitations of apical sparing pattern in cardiac amyloidosis: a multicentre echocardiographic study. Eur Heart J Cardiovasc Imaging. 2024;25:754–61. doi: 10.1093/ehjci/jeae021. [DOI] [PubMed] [Google Scholar]
- 8.Benjamin ER, Della Valle MC, Wu X, et al. The validation of pharmacogenetics for the identification of Fabry patients to be treated with migalastat. Genet Med. 2017;19:430–8. doi: 10.1038/gim.2016.122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lukas J, Cimmaruta C, Liguori L, et al. Assessment of Gene Variant Amenability for Pharmacological Chaperone Therapy with 1-Deoxygalactonojirimycin in Fabry Disease. Int J Mol Sci. 2020;21:956. doi: 10.3390/ijms21030956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405–24. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Brnich SE, Abou Tayoun AN, Couch FJ, et al. Recommendations for application of the functional evidence PS3/BS3 criterion using the ACMG/AMP sequence variant interpretation framework. Genome Med. 2019;12:3. doi: 10.1186/s13073-019-0690-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Expert consensus for diagnosis and treatment of Fabry disease in China. Zhonghua Nei Ke Za Zhi. 2021;60:321–30. doi: 10.3760/cma.j.cn112138-20201218-01028. [DOI] [PubMed] [Google Scholar]
- 13.Waterhouse AM, Studer G, Robin X, et al. The structure assessment web server: for proteins, complexes and more. Nucleic Acids Res. 2024;52:W318–23. doi: 10.1093/nar/gkae270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Johnson FK, Mudd PN, Jr, Bragat A, et al. Pharmacokinetics and Safety of Migalastat HCl and Effects on Agalsidase Activity in Healthy Volunteers. Clin Pharmacol Drug Dev. 2013;2:120–32. doi: 10.1002/cpdd.1. [DOI] [PubMed] [Google Scholar]
- 15.Grantham R. Amino acid difference formula to help explain protein evolution. Science. 1974;185:862–4. doi: 10.1126/science.185.4154.862. [DOI] [PubMed] [Google Scholar]
- 16.Guo W, Ji P, Li Y, et al. Pathogenicity of novel GLA gene missense mutations in Fabry disease and the therapeutic impact of migalastat. J Adv Res. 2026;79:549–61. doi: 10.1016/j.jare.2025.03.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Asano N, Ishii S, Kizu H, et al. In vitro inhibition and intracellular enhancement of lysosomal alpha-galactosidase A activity in Fabry lymphoblasts by 1-deoxygalactonojirimycin and its derivatives. Eur J Biochem. 2000;267:4179–86. doi: 10.1046/j.1432-1327.2000.01457.x. [DOI] [PubMed] [Google Scholar]
- 18.Vijapurapu R, Nordin S, Baig S, et al. Global longitudinal strain, myocardial storage and hypertrophy in Fabry disease. Heart. 2019;105:470–6. doi: 10.1136/heartjnl-2018-313699. [DOI] [PubMed] [Google Scholar]
- 19.Linhart A, Kampmann C, Zamorano JL, et al. Cardiac manifestations of Anderson-Fabry disease: results from the international Fabry outcome survey. Eur Heart J. 2007;28:1228–35. doi: 10.1093/eurheartj/ehm153. [DOI] [PubMed] [Google Scholar]
- 20.Beck M, Ramaswami U, Hernberg-Ståhl E, et al. Twenty years of the Fabry Outcome Survey (FOS): insights, achievements, and lessons learned from a global patient registry. Orphanet J Rare Dis. 2022;17:238. doi: 10.1186/s13023-022-02392-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Elliott P, Baker R, Pasquale F, et al. Prevalence of Anderson-Fabry disease in patients with hypertrophic cardiomyopathy: the European Anderson-Fabry Disease survey. Heart. 2011;97:1957–60. doi: 10.1136/heartjnl-2011-300364. [DOI] [PubMed] [Google Scholar]
- 22.Garman SC, Garboczi DN. The molecular defect leading to Fabry disease: structure of human alpha-galactosidase. J Mol Biol. 2004;337:319–35. doi: 10.1016/j.jmb.2004.01.035. [DOI] [PubMed] [Google Scholar]
- 23.Saito S, Ohno K, Sakuraba H. Comparative Study of Structural Changes Caused by Different Substitutions at the Same Residue on α-Galactosidase A. PLoS ONE. 2013;8:e84267. doi: 10.1371/journal.pone.0084267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Varela P, Mastroianni Kirsztajn G, Ferrer H, et al. Functional Characterization and Pharmacological Evaluation of a Novel GLA Missense Mutation Found in a Severely Affected Fabry Disease Family. Nephron. 2020;144:147–55. doi: 10.1159/000503998. [DOI] [PubMed] [Google Scholar]
- 25.Hurtley SM, Helenius A. Protein oligomerization in the endoplasmic reticulum. Annu Rev Cell Biol. 1989;5:277–307. doi: 10.1146/annurev.cb.05.110189.001425. [DOI] [PubMed] [Google Scholar]
- 26.Yam GH-F, Zuber C, Roth J. A synthetic chaperone corrects the trafficking defect and disease phenotype in a protein misfolding disorder. FASEB J. 2005;19:12–8. doi: 10.1096/fj.04-2375com. [DOI] [PubMed] [Google Scholar]
- 27.Khanna R, Soska R, Lun Y, et al. The pharmacological chaperone 1-deoxygalactonojirimycin reduces tissue globotriaosylceramide levels in a mouse model of Fabry disease. Mol Ther. 2010;18:23–33. doi: 10.1038/mt.2009.220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Germain DP, Hughes DA, Nicholls K, et al. Treatment of Fabry’s Disease with the Pharmacologic Chaperone Migalastat. N Engl J Med. 2016;375:545–55. doi: 10.1056/NEJMoa1510198. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.