Congenital disorder of glycosylation with defective fucosylation 2 (FCSK gene defect): The third report in the literature with a mild phenotype

Abstract

Background

Congenital disorders of glycosylation (CDG) are a group of heterogeneous disorders caused by abnormal lipid or protein glycosylation. Variants in the FCSK gene have been reported to cause CDG. Defective FCSK-induced CDG (FCSK–CDG) has only been reported previously in three unrelated children.

Methods

In this study, we genetically and clinically examined a 3-year-old proband with resolved infantile spasms and normal development. Standard whole-exome sequencing (WES) and Sanger sequencing were performed to identify the functional impact of the variant.

Results

WES revealed a rare biallelic missense variant (c.3013G>C; p.Val1005Leu) in FCSK. RT-qPCR showed a significant depletion in FCSK gene expression in the affected individual. Western blotting revealed reduced FCSK expression at the protein level compared to that in the control. Furthermore, 3D protein modeling suggested changes in the secondary structure, which might affect the overall FCSK protein function.

Conclusion

This study broadens the mutation and phenotypic spectrum of FCSK-associated developmental disorders.

INTRODUCTION

Glycosylation—a type of post-translational modification in proteins—is an essential cellular process that adds glycans to proteins via an amide or hydroxyl group (Bertozzi & Rabuka, 2009). Approximately 2% of all genes in the human genome encode proteins that undergo different types of glycosylation (Apweiler et al., 1999). CDG are a subset of rare metabolic disorders resulting from abnormal protein or lipid glycosylation (Ng, Rosenfeld, et al., 2018; Ng, Xu, et al., 2018). More than 130 different CDG have been identified across several glycosylation pathways, and the majority are accounted for by genetic defects (Jaeken & Péanne, 2017; Sosicka et al., 2020).

To date, only three variants of the FCSK gene have been associated with FCSK-CDG (OMIM 618324) (Ng, Rosenfeld, et al., 2018; Özgün & Şahin, 2022). Ng, Rosenfeld, et al. (2018) reported two patients who presented severe developmental delays, encephalopathy, intractable seizures, and hypotonia. The first patient had a compound heterozygous missense variants (c.667T>C; p.Ser223Pro and c.2047C>T; p.Arg683Cys) and the second a homozygous missense variant (c.2980A>C; p.Lys994Gln) (Ng, Rosenfeld, et al., 2018). Later, Özgün and Şahin (2022), reported a patient with a homozygous frameshift variant (c.993_1011del; p.Glu335Hisfs*55) in the FCSK gene with overlapping phenotypes (Özgün & Şahin, 2022).

Fucosylation is a common form of glycosylation where the monosaccharide l-fucose is integrated into the terminal fragment of N, O, or lipid-linked glycans (Schneider et al., 2017). Fucosyltransferase enzymes utilize the nucleotide diphosphate sugar guanosine diphosphate L-fucose (GDP-fucose) as a donor substrate to alter the targeted proteins or lipids (Ng, Rosenfeld, et al., 2018; Ng, Xu, et al., 2018). Glycans are essential for cellular development, and pathogenic variants in glycan pathways are lethal (Sosicka et al., 2020).

Here, we describe the case of an index patient with resolved infantile spasms and normal development. Whole-exome sequencing (WES) revealed a homozygous missense variant in the FCSK gene and its pathogenicity was confirmed through functional studies.

MATERIALS AND METHODS Study approval

This study was approved by the IRB of KAIMRC, Riyadh, Saudi Arabia. The parents provided written informed consent for conducting the research study and publication of clinical data.

WES

Genomic DNA was extracted and quantified using standard methods. WES was performed at CENTOGENE laboratory (Germany). Briefly, the generated DNA library was sequenced using the HiSeqX platform (Illumina) to obtain at least 20x coverage/depth for >98% of the targeted bases by following standard protocols (Alhamoudi et al., 2021; Barhoumi et al., 2019; Umair, Khan, et al., 2021). The bioinformatics pipeline was used, and data were aligned with the GRCh37/hg19 genome assembly. Basespace (Illumina) software was used to analyze the VCF files. Disease-associated variants in HGMD®, CentoMD®, OMIM®, ClinVar, and variants with MAF <1% in ExAC, gnomAD, and 1000genomes databases were considered. In addition, ACMG variant classification was performed. The WES variant filtration steps are summarized in Supplementary Figure S1.

Sanger sequencing

Sanger sequencing was performed to check for the segregation of the identified variant. The Primer3 online tool was used to design primers; the details of the primer sequences would be shared upon request.

RNA extraction

RNA was extracted from the peripheral blood mononuclear cells (PBMCs) and fibroblast samples of the patient, parents, and control samples using the RNA isolation protocol (TRIzol). Primers used for qPCR were F1: GCT GTG GAT GCT GGA CCA and R1: AAG GGC CGC ATC CAC TCA. To quantify gene expression, quantitative real-time PCR (RT-qPCR) was performed on a QuantStudio 6 Flex Real-Time PCR System using SYBR® Green PCR Master Mix (Applied Biosystems). The cycle conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and annealing at 60°C for 30 s. A no-template control (NTC) was used as a negative control for each experiment. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control, and the 2^−ΔΔCt method was used for gene quantification (Asiri et al., 2020).

Cell culture

Skin biopsies were obtained from the index patient, and fibroblast cultures were established in Dulbecco's modified Eagle's medium complemented with 10% fetal bovine serum (FBS), 10 U/ml penicillin/streptomycin, and 2 mM l-glutamine. The cultures were maintained at 37°C in a 5% CO₂ atmosphere.

Western blotting analysis

Proteins were extracted from the fibroblast cell lines of the control and index patient using the Tris–HCl method. For the detection of the targeted protein, FCSK mouse monoclonal antibody [1/1000; Thermo Fisher (MA5-15847)] and mouse monoclonal antibody against α-tubulin antibody [1/2000] were used (Mutairi et al., 2020). α -tubulin was used as a loading control.

3D modeling

The FCSK amino acid sequence was retrieved from the UniProt database [Q8NOW3-1]. The FCSK structure sequence was submitted to the I-TASSER server, and the model was selected based on the evaluation score. The selected structure was optimized using UCSF Chimera version1 via the AMBERff14 SB force field3D. The overall quality of the model was evaluated using an ERRAT plot (Asiri et al., 2021).

Statistical analysis

All statistical analyses were performed using GraphPad Prism (version 8.1). The qPCR results were tested for normal distribution, and analysis of variance (one-way ANOVA) was applied (significant if p < .05). WB results were analyzed using an unpaired t-test (significant if p < .05).

RESULTS Clinical evaluation

The patient was a 3-year-old male second child born to healthy Saudi parents; he was born at full term with a birth weight of 2.6 kg (5th–10th percentile), length of 49 cm (25th–50th%), and head circumference (HC) of 34 cm (10th–25th%). The parents became concerned for the first time at the age of 6 months when the mother noticed abnormal eye movements that progressed over time to involve the entire body; the condition is described as infantile spasm (Figure 1b). An electroencephalogram (EEG) revealed modified hypsarrhythmias. The patient was administered vigabatrin (50 mg/kg/day) and cosyntropin injections (0.25 mg every 12 h). After 14 days, the cosyntropin injection was discontinued, but vigabatrin was continued, and the patient improved dramatically. After 3 months, the EEGs were repeated and were normal, vigabatrin was discontinued, and the patient has remained seizure-free (Supplementary Figure S2). The patient has reached the developmental milestones. The parents were first cousins and had no history of illness in the family. They had another older son who was 4.5 years old and healthy.

FIGURE 1. (a) Family pedigree showing an autosomal recessive inheritance pattern. (b) Facial photographs of the patient. (c) Schematic representation of the FCSK domains reflecting position of identified mutation p.Val1005Leu. (d) Sanger sequencing electropherograms of normal, affected, and carrier samples. (e) qPCR: Patient sample showing reduced FCSK-expression in fibroblast. (f) qPCR: FCSK gene expression in patient's and parents' blood sample. (g) Western blotting (WB): Reduction in the FCSK protein expression in the patient's sample compared with that of the control. (h) Conservation of Val1005 across several species

The proband could run, climb stairs independently, ride a tricycle, stand briefly on one foot, scribble, say his full name, and form three-word sentences. His current growth parameters are as follows: weight: 15 kg (50th–75th percentile), length: 100 cm (75th–90th percentile), and HC: 50 cm (50th–75th percentile). The results of his neurological and other system examinations are unremarkable.

Molecular analysis

Molecular analyses were performed using WES following Sanger sequencing using standard methods. WES revealed a rare homozygous missense variant [c.3013G>C, p.(Val1005Leu)] in exon 23 of the FCSK gene [16q22.1] (NM_145059.3) (Supplementary Table S1). Sanger sequencing showed perfect segregation of the identified variant (Figure 1d) (Umair, Kharfy, et al., 2021). Pathogenicity and deleterious effects were confirmed using online software (Supplementary Table S2). Furthermore, in gnomAD (exomes) “c.3013G>C” was identified 8 times in the heterozygous state (frequency: 0.0000321) and 16 times in gnomAD (genomes) (frequency: 0.000105). However, this has not been reported in the homozygous state. The Val1005 amino acid sequence is conserved across several species (Figure 1h).

qPCR and WB

qPCR data revealed that the affected proband (c.3013G>C) showed a significant reduction in FCSK expression than in the parents and control (Figure 1e,f). WB analysis was performed using anti-FCSK and anti-α-tubulin antibodies. A confirmed reduction in FCSK expression was observed in the index patient (Figure 1g).

In silico analysis

Homology modeling for the mutant FCSK^Val1005Leu and FCSK^wild-type proteins were performed (Figure 2a–c). Approximately 90% and 93% of residues in the FCSK^Val1005Leu and FCSK^wild-type structures lie in the allowed torsion angles according to the Ramachandran plot.

FIGURE 2. (a) FCSKWildtype model. (b) FCSKVal1005Leu model. (c, d) exhibited: Different interatomic interactions between the FCSKWildtype amino acid Val and FCSKVal1005Leu leu at position 1005 with the surroundings residues

The residue [Valine1005] is located in the C-terminal domain of GHMP kinase, and the variant at this position might affect the secondary FCSK structure. Val1005 interacts with Trp990, Glu1001, Pro1002, Thr1004, Met1008, and Asp1010. Both leucine and valine have a neutral side chain (non-polar); however, substitution of smaller valine with a larger leucine might change the surrounding interactions that might disrupt FCSK function (Figure 2d,e). According to the mCSM, ENCoM, and DUET, FCSK^Val10054Leu, respectively, cause a − 0.566, 0.244, and − 0.964 kcal/mole change in ΔΔG, indicating destabilization of the FCSK structure and function.

DISCUSSION

In all mammalian cells, fucosylation is the process by which fucosyltransferase transports fucose from GDP-fucose to its substrates, which include specific proteins and N- and O-linked glycans in glycoproteins or glycolipids. FCSK encodes a fucokinase enzyme that phosphorylates β-l-fucose to produce β-l-fucose-1-phosphate (β-l-Fuc1p), which helps synthesize GDP-fucose and mediates fucosylation of global cellular proteins (Ng, Rosenfeld, et al., 2018).

In the present study, we evaluated a patient with infantile spasms and a homozygous missense variant [c.3013G>C p.(Val1005Leu)] in the FCSK gene. qPCR and WB revealed reduced FCSK gene and protein expression in the patient compared with that in the controls (Figure 1e–g). Similarly, 3D modeling suggested changes in the protein secondary structure; thus, the variant may affect the overall function and downstream interactions (Figure 2). Ng, Rosenfeld, et al. (2018) reported the FCSK variant in two unrelated children with encephalopathy, severe GDD, hypotonia, and intractable seizures (Ng, Rosenfeld, et al., 2018). Similarly, Özgün and Şahin (2022), reported a case with a homozygous frameshift variant in the FCSK gene and hypotonia, developmental delay, abnormal brain MRI, and blindness. The patient reported here had mild phenotypes, including normal brain MRI and normal development, in contrast to the brain anomalies, such as dysplastic corpus callosum, delayed myelination, and cerebellar atrophy, reported in the previous cases (Table 1). Variants in the FCSK gene showed loss-of-function in the salvage pathway and thus were responsible for the severe multisystemic disease (Ng, Rosenfeld, et al., 2018). Similarly, the patient reported here showed a 40% reduction in the FCSK gene expression at the mRNA level and changes in the protein secondary structure, which might explain the underlying disease mechanism.

TABLE 1 Summary of phenotypic characteristics of patients with congenital disorders of glycosylation (CDG) caused by pathogenic variants in the FCSK gene compared with that in the current patient

	Ng, Rosenfeld, et al. (2018) (n = 1)	Ng, Rosenfeld, et al. (2018) (n = 1)	Özgün and Şahin (2022) (n = 1)	Current patient (n = 1)
Ancestry	Hispanic	Middle Eastern (Qatari)	Somalian	Middle Eastern (Saudi)
Gender/age mean	6 years	7 years	11 years	2 years
cDNA change	c.667T>C; c.2047C>T	c.2980A>C	c.993_1011del	c.3013G>C
Protein change	(p.Ser223Pro); (p.Arg683Cys)	(p.Lys994Gln)	(p.Glu335Hisfs*55)	p.(Val1005Leu)
Mutation type	Compound heterozygous	Missense homozygous	Homozygous deletion	Missense homozygous
Consanguinity	No	Yes	No	Yes
IUGR	No	No	No	No
Dysmorphic features	No	No	No	No
Intellectual disability	Yes	Yes	Yes	No
Developmental delay	Yes	Yes	Yes	No
Brain anomalies	Dysplastic corpus callosum. Delayed myelination	Cerebellar atrophy Agenesis of corpus callosum White matter abnormalities	Hyperintensities in T2 and fluid attenuated Inversion recovery series of bilateral frontal, parietal, and occipital periventricular white matter, Cortical sulcal prominence and lateral ventricular enlargement	Normal
Eye disorder	Symmetric maculopathy Severe visual impairment	Strabismus Optic nerve atrophy Cortical blindness	Blindness	Normal
Seizures/epilepsy	Generalized tonic–clonic type seizures in the setting of epileptic encephalopathy	Seizures, infantile spasms	N/A	Infantile spasms, family history of seizure
Immunology	Recurrent infections	No	No	No
Hypotonia	Yes	Yes	Yes	No
Cardiac	No	No	No	No
Type of Mutation	Compound heterozygous, missense	Homozygous, missense	Homozygous, deletion	Homozygous, missense

The two main pathways in vertebrates where GDP-fucose is synthesized are: the de novo pathway and fucose salvage pathway (Figure 3). In the de novo pathway, two enzymes—GDP-mannose 4,6-dehydratase (GMDS) and GDP-keto-6-deoxymannose 3,5 epimerase (TSTA3) (aka FX protein)—convert GDP mannose to GDP-fucose. Approximately 90% of cellular GDP-fucose are synthesized via this pathway (Ng, Rosenfeld, et al., 2018; Ng, Xu, et al., 2018). In the salvage pathway, a two-step process converts free fucose to GDP-fucose. Later, GDP-fucose is transported into the lumen of the Golgi apparatus and modified for the catalytic domains of fucosyltransferase (Becker & Lowe, 2003).

FIGURE 3. Synthetic fucose metabolic pathways. Two different pathways produce GDP-fucose in cells: The de novo synthesis pathway and the fucose salvage pathway

In conclusion, we have reported the incidence of CDG with a rare homozygous variant in the FCSK gene in a child from a Saudi family. Proper genetic counseling for the affected family is essential in the case of rare genetic diseases. Furthermore, parenteral genetic screening/diagnosis is the best strategy for managing this disease, which currently has no therapy (Alfadhel et al., 2019; Alyafee, Al Tuwaijri, et al., 2021; Alyafee, Alam, et al., 2021). Reporting additional cases associated with this gene would help identify genotype–phenotype correlations and lead to clinical trials in the future (Alfadhel et al., 2021). This study further proves with essential evidence that variants of FCSK cause mild-to-severe CDG in humans.

AUTHOR CONTRIBUTIONS

Abeer Al Tuwaijri and Yusra Alyafee: wrote the manuscript; Qamre Alam, Muhammad Umair, and Majid Alfadhel: reviewed and edited the manuscript; Majid Alfadhel: supervision; Arwa Alsubait, Mariam Ballow, Hamad AlEidi, Mohammed Aldrees, Muhammad Talal Alrifai, and Mashael Alharbi: performed the experiments.

ACKNOWLEDGMENTS

This research study was funded by KAIMRC.

FUNDING INFORMATION

KAIMRC, Riyadh, KSA.

CONFLICT OF INTEREST

None to declare.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT

KAIMRC, Riyadh, KSA.

PATIENT CONSENT STATEMENT

Yes, obtained from parents.