A compound heterozygote case of glutaric aciduria type II in a patient carrying a novel candidate variant in ETFDH gene: A case report and literature review on compound heterozygote cases

Abstract

Background

Glutaric aciduria type II (GA2) is a rare genetic disorder inherited in an autosomal recessive manner. Double dosage mutations in GA2 corresponding genes, ETFDH, ETFA, and ETFB, lead to defects in the catabolism of fatty acids, and amino acids lead to broad‐spectrum phenotypes, including muscle weakness, developmental delay, and seizures. product of these three genes have crucial role in transferring electrons to the electron transport chain (ETC), but are not directly involve in ETC complexes.

Methods

Here, by using exome sequencing, the cause of periodic cryptic gastrointestinal complications in a 19‐year‐old girl was resolved after years of diagnostic odyssey. Protein modeling for the novel variant served as another line of validation for it.

Results

Exome Sequencing (ES) identified two variants in ETFDH: ETFDH:c.926T>G and ETFDH:c.1141G>C. These variants are likely contributing to the crisis in this case. To the best of our knowledge at the time of writing this manuscript, variant ETFDH:c.926T>G is reported here for the first time. Clinical manifestations of the case and pathological analysis are in consistent with molecular findings. Protein modeling provided another line of evidence proving the pathogenicity of the novel variant. ETFDH:c.926T>G is reported here for the first time in relation to the causation GA2.

Conclusion

Given the milder symptoms in this case, a review of GA2 cases caused by compound heterozygous mutations was conducted, highlighting the range of symptoms observed in these patients, from mild fatigue to more severe outcomes. The results underscore the importance of comprehensive genetic analysis in elucidating the spectrum of clinical presentations in GA2 and guiding personalized treatment strategies.

Glutaric aciduria type II (GA2) is a rare genetic disorder caused by mutations in specific genes. A 19‐year‐old girl with periodic gastrointestinal issues was diagnosed using whole‐exome sequencing, revealing one novel mutation (ETFDH:c.926T>G) in the ETFDH gene. These mutations were found to be the likely cause of her condition. The study also highlighted the diversity of symptoms in GA2 patients with compound heterozygous mutations, ranging from mild to severe outcomes.

1. INTRODUCTION

Glutaric acidemia type II (GA2; MIM 231680) is a rare autosomal recessive disorder primarily affecting fatty acid metabolism, leading to defects in fatty acid oxidation pathways (Przyrembel et al., 1976). GA2, also known as multiple acyl‐CoA dehydrogenase deficiency (MADD), is a heterogeneous disorder with variable expressions (Olsen et al., 2003). The primary manifestation of GA2 might be a severe metabolic crisis during infancy with hypoglycemia and acidosis. GA2 patients with more severe phenotypes may develop hepatomegaly, dilated cardiomyopathy, and even syncope. However, delayed and mild manifestations might occur in late‐onset forms. Weakness, muscle pain, vomiting, poor feeding, and weight loss are other common symptoms experienced by individuals with GA2 (Grünert, 2014).

This disease may be detected in newborn screening, or later, an infant or an adult may be referred to a clinic with relevant symptoms. In adult cases, the proband usually presents with episodes of nausea along with hypoglycemia, metabolic acidosis, and/or muscle weakness. Diagnosis can be confirmed through an acylcarnitines profile or genetic testing. NGS‐based methods are usually used for genetic diagnosis. These methods can scan the three relevant genes that cause GA2 (i.e., ETFDH (MIM 231675), ETFA, and ETFB), and if no suitable variants are found, an investigation of genetic‐based differential diagnoses for is warranted (Onkenhout et al., 2001). Loss‐of‐function mutations in both copies of the mentioned genes lead to deficiencies in the enzymes involved in the mitochondrial electron transfer flavoprotein (ETF) or ETF dehydrogenase (ETFDH), which are critical for the catabolism of fatty acids, some amino acids, and choline. The results of these deficiencies include the accumulation of acylcarnitines and organic acids, leading to the generation of oxidative stress, inflammation, and liver damage, all contributing to the characteristic phenotypes of GA II (Pfaller et al., 2021). Therefore, GA2 primarily is considered as catabolic defect rather than defect in ETC and energy production defect.

Cofactor plays a crucial function of enzymes. Cofactors of ETFDH include: FAD (Flavin Adenine Dinucleotide), which acts as an electron acceptor; SF4 (Sulfur–Iron Cluster), which facilitates electron transfer; and UQ5 (Ubiquinone), which transfers electrons to the electron transport chain (Henriques et al., 2021).

In cases of myopathic presentation, muscle biopsy may be indicated to observe lipid excess and myopathic atrophy (Li et al., 2021). Although there is no cure for GA2 so far, this disorder could be managed with restrictions on protein and fat, avoidance of fasting, and supplementation with L‐carnitine and high‐dose riboflavin. Riboflavin is more likely to be effective in late‐onset cases (Li et al., 2021; Prasun, 2020).

Defect of riboflavin metabolism (mutations in FLAD1, SLC52A1, and SLC52A2/3) are considered as differential diagnosis (DD) of GA2 because they mimic it clinical and biochemical features. There is a list of metabolic disorders and Lipid storage myopathy and Inflammatory myopathies which should be considered as DD for GA2 (Prasun, 2020).

Herein, we discuss a female patient with long‐lasting gastrointestinal complications since 6 years of age that was mismanaged for 13 years and diagnosed by ES as a GA2.

2. METHODS

2.1. Ethical compliance

Informed consent, in accordance with the principles outlined in the Declaration of Helsinki, from the patient was obtained prior to the collection of muscle biopsies and exome sequencing, and it was noted that the results may be published without disclosing any personal identifiable information.

2.2. Study participant

A 19‐year‐old female was referred to the genetic counseling clinic at the Academic Center for Education, Culture, and Research (ACECR) for long‐lasting episodes of nausea and vomiting as well as dull abdominal pain.

2.3. Clinical evaluation and laboratory tests

The patient was evaluated by subspecialty medical doctors and underwent an extensive array of clinical laboratory tests, including various diagnostic workups for gastrointestinal disorders. These tests included a duodenum biopsy. Additionally, a muscle biopsy was performed on the left quadriceps, and tandem mass spectrometry, to assess her health status and diagnose any underlying conditions.

2.4. Exome sequencing and data analysis

The exome enrichment was performed using the Agilent SureSelect V6 Target Enrichment Kit, followed by Next‐generation sequencing using the Illumina HiSeq4000 platform. All exons and flanking 10 bp were detected and analyzed. Based on the results of mass spectrometry and muscle biopsy, the ETFA, ETFB, and ETFDH genes were prioritized for analysis in the initial screening.

For segregation analysis, primers were designed using the Primer3 website (https://primer3.ut.ee) and validated for specificity through NCBI Primer‐Blast (https://www.ncbi.nlm.nih.gov/tools/primer‐blast). The thermodynamic stability (ΔG) of the primer secondary structures was assessed using the free version of Oligo Analyzer software 1.0.3. The designed primers for ETFDH:c.926T>G variant are GTGCCCAGCCTTCTTTGATA and GACTTAATGACACAGCAGCCA; and for ETFDH:c.1141G>C variant are TTCAGCCTTTCCCTACAG and TTGAAATACACATAACCCAGC.

2.5. Protein modeling

To investigate the effect of this novel variant in 3D protein structure, homology modeling was performed by applying the SWISS‐MODEL‐Expasy (https://swissmodel.expasy.org). Sus scrofa ETFDH protein (PDB ID: 2gmh.2) was served as the template for modeling wild‐type and novel missense variants (mutant) (Pfaller et al., 2021; Studer et al., 2020).

The quality of the 3D structures was verified using UCLA‐DOE LAB tools (https://www.doe‐mbi.ucla.edu/services) (Colovos, 1993). The Z‐scores of the models were analyzed using ProSA‐web (https://prosa.services.came.sbg.ac.at/prosa.php) (Wiederstein & Sippl., 2007).

TM‐align was utilized to superimpose two models for structural comparison, and the results were visualized at (https://www.rcsb.org/3d‐view) (Zhang & Skolnick., 2005).

To predict changes in protein stability, “STRUM: Structure‐based stability change prediction upon single‐point mutation” was used (Quan et al., 2016).

3. RESULTS

3.1. Clinical presentation and laboratory findings

The case was experienced long‐lasting episodes of nausea and vomiting as well as dull abdominal pain. The primary manifestations of the disease started at the age of six with episodes of recurrent vomiting occurring up to 3 times per day. The vomiting episodes declined by age seven and recurred every few months until the present. The vomiting episodes are usually accompanied by mild diffuse abdominal pain and lethargy. The pain severity during vomiting episodes varied each time and was often exacerbated by emotional stress. Duodenum biopsy showed no remarkable finding. During the disease episodes, the patient did not have decreased alertness, diarrhea, fever, or headache. The creatine kinase level increased to 1270 U/L (normal range 22–198 U/L) during the disease episodes. The serum aldolase, ammonia, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels increased during the disease episodes (Figure 1). Other serum biochemical parameters, including thyroid hormones, were within their normal range. The patient achieved standard growth and developmental milestone, and there was no notable complication during disease‐free intervals. During the last disease episode, the patient developed severe muscle weakness and had decreased deep tendon reflexes in her neurologic examination. The electromyography (EMG) study revealed myopathic changes, and the patient was recommended for a muscle biopsy. Figure 2 shows the results of muscle biopsy.

FIGURE 1.

Level of four metabolic agents (ALT (U/L), AST (U/L), lactate (mg/dL), and ammoniac (μmol/L)) in about one and half year every 3 months. SGOT and SGPT showed elevated level sync to disease climax.

FIGURE 2.

(a) Slight myopathic atrophy with presence of multiple small round vacuoles with variable sizes (hematoxyline and eosin × 400). (b) Lipid excess is present and cytoplasmic vacuoles are stained red (Oil‐Red‐O × 400).

The tandem mass spectrometry showed elevated levels of medium‐chain acylcarnitines (C4‐, C5‐, C5DC‐, C6‐, C8‐, C10:1‐, C12‐, C14‐, C14:1‐, C16‐, C16:1‐, C18‐, C18:1‐, C16‐OH‐, C16:1‐OH‐, C18‐OH‐, and C18:1‐OH‐acylcarnitines), and no evidence of peroxisomal disease (VLCFA and phytanic acid levels were in the normal range).

3.2. Exome sequencing results and variant confirmation

Two possible variants are candidates as possible causes of the patient's phenotype. The first previously known variant, ETFDH:c.1141G>C (p.GLy381Arg; rs1466787789), and the second, ETFDH:c.926T>G (p.Leu309Arg; rs1276785426), which are respectively pathogenic and likely pathogenic variant according to the American College of Medical Genetics (2015) guideline (Richards et al., 2015). Albeit the result seems satisfying, pursuing other possible causative variants continued. Among 19,682 variants, those with an allele frequency greater than 5% in gnomAD and Iranome databases were excluded. Among the remaining variants, those with a greater probability of pathogenicity, including variants with frameshift, stop/start gain/loss, and splice site in the primary gene list (other than ETFA, ETFB, and ETFDH) were candidates for further evaluation based on their relevance the patient's phenotypes. Finally, there were no other suggestive variants corresponding to her phenotype, and the two variants in the ETFDH gene considered the most probable candidate variants.

The segregation analysis by Sanger sequencing for these two variants showed that the c.1141G>C variant was inherited from the mother and the c.926T>G variant from the father (Figure 3).

FIGURE 3.

Data of segregation analysis status for compound heterozygote mutation performed by sanger sequencing. Top left, father; top right, mother; and in down the affected child variants.

The predictive pathogenically scoring and preservation scoring provided for the variants are presented in Tables 1 and 2, respectively.

TABLE 1.

Summary of the variants.

Allele	Paternal allele	Maternal allele
Gene name	ETFDH
NM number	NM_004453.4
Cytogenetic band	4q32.1
Chromosomal location (hg19)	Chr4‐159618805‐T‐G	Chr4‐159624599‐G‐C
Variant ID (dbSNP 155)	rs1276785426	rs1466787789
Nucleotide change	ETFDH:c.926T>G	ETFDH:c.1141G>C
Effect on protein	p.Leu309Arg	p.Gly381Arg
CDS	Exon 8 of 13	Exon 10 of 13
Alteration type	SNV
ACMG classification	Likely pathogenic	Pathogenic
ClinVar	‐	Pathogenic
Allele frequency (gnomAD)	ƒ = 0.00000398	This variant does not have a gnomAD exomes entry
Inheritance	Autosomal recessive
Literatures	–	PMID:31312603 PMID:32393189 PMID:31904027 PMID:28388738
Interacting proteins	ETFA, ETFB, PUS7, HSPD1, BCKDK, UBC, AHCY, C1orf151
Novelty	Novel	–
Mutation assessor	Pathogenic moderate	Pathogenic moderate
MutPred	Pathogenic moderate	Pathogenic moderate
SNP&GO	Disease causing probability: 0.927	Disease causing probability: 0.921
Phanter	Probably damaging	Probably damaging
SIFT	Pathogenic supporting	Pathogenic supporting
Mutation taster	Uncertain	Uncertain
Polyphen‐2	Probably damaging	Probably damaging
CAAD	Pathogenic moderate/high (28.3)	Pathogenic moderate/high (29.1)
Revel score	Pathogenic strong	Pathogenic moderate
Conservation (GERP)	Highly conserved (5.55)	Highly conserved (5.3)

Note: Protein sequence: GenBank NP_004444.2.

TABLE 2.

A brief literature review on studies reporting compound heterozygote ETFDH gene mutations.

Author, year	Gender/age of initial symptoms onset	Mutations				Symptoms	Laboratory findings
		First allele		Second allele
		Nucleotide	Protein	Nucleotide	Protein
Izumi et al., 2011	Male/46 years	c.1211T>C	p.D511N	c.1786G>A	p.G429Dfs*21	Fatigability and weakness	Increased CK, lactate, pyruvate
Sugai et al., 2012	Male/37 years	c.1531G>A	p. D130V	c.1809G>A	p.W343R	Exercise intolerance, muscle weakness	Increased CK, LDH, increased 2‐OH glutarate and 2‐OH adipate
Wen et al., 2015	Male/24 years	c.1211T>C	p.M404T	c.1450T>C	p.W484R	General fatigue and weakness, nausea and vomiting	Elevated CK
Zhuo et al., 2015	Female/9 years	c.389A>T	p.M404T	c.736G>A	p.E246K	Intermittent vomiting, muscle weakness, hepatosplenomegaly	Increased ALT, AST, and CK
Fu et al., 2016	Male/23 years	c.892C>T	p.P298S	c.453delA	p.E152Rfs*15	Progressive muscle weakness, intermittent nausea and vomiting, exercise intolerance, myalgia, and muscle atrophy	Increased AST, CK, CKMB, and LDH
	Male/15 years	c.453delA	p.E152Rfs*15	c.449_453delTAACA	p.L150X	Proximal muscle weakness, diarrhea, periodic abdominal pain, and muscle atrophy	Increased AST, CK, CKMB, and LDH
	Male/60 years	c.453delA	p.E152Rfs*15	c.449_453delTAACA	p.L150X	Diagnosed as irritable bowel syndrome, abdominal distension, nauseating and vomiting, and muscle atrophy	Increased CK, myoglobin, AST, and LDH
Béhin et al., 2016	Female/18 years	c.250G>A	p.A84T	c.524G>A	p.R175H	Muscle weakness and fatigue	Increase of C5 to C18, abnormal Acylglycines, dicarboxylic acids 2‐hydroxyglutaric and ethylmalonic acid
	Male/15 years	c.571G>A	p.G191S	c.1331T>C	p.V444A	Post‐exercise muscle pain and stiffness	Increase of C4 to C18
	Female/32 years	c.622G>C	p.D208H	c.1241_1246del	p.I414_P415del	Exercise intolerance, asthenia, and weight loss	Increase of C4 to C18
	Male/18 years	c.877C>G	p.H293D	c.1691‐3C>G	Splicing defect	Delirium, vomiting, asthenia, liver function abnormalities and myolysis	Increase of C4 to C18
	Male/39 years	c.79C>T	p.P27S	c.1471_1473delins8	p.S491Qfs*3	Exercise intolerance, and progressive proximal muscle weakness	Increase of C4 to C18
	Female/35 years	c.406‐2A>G	Splicing defect	c.728T>C	p.I243T	Exercise intolerance, and progressive proximal muscle weakness	Increase of C5 to C18; no C5DC, and no C6
Xue et al., 2017	Female/23 years	c.250G>A	p.A84T	c.920C>G	p.S307C	Muscle weakness and exercise intolerance, vomiting, and weight loss	Elevated CK, ALT, AST, CKMB, LDH, glutarate, adipate, 2‐hydroxyglutaric acid, 3‐hydroxyglutaric acid, 2‐hydroxy adipic acid, orotic acid, 4‐hydroxy‐phenyllactic acid, and ethylmalonic acid, normal Acylcarnitine analysis
Xue et al., 2017	Female/34 years	c.814G>A	p.Gly272Arg	c.389A>T	p.D130V	Muscle weakness, short breath, limb atrophy, dropped head syndrome, decreased deep tendon reflex, and myopathy in electrodiagnostic	Increased AST, CK, CKMB, and LDH
Hong et al., 2018	Male/22 years	c.265‐266delCA	p.Q89Vfs*5	c.1211T>C	p.M404T	Progressive muscle weakness after a viral infection, and distal sensory disturbance	Elevated CK, the Combined elevation of short‐, medium‐, and long‐chain acylcarnitines
	Male/55 years	c.34G>C	p.A12P	c.736G>A	p.E246K	Muscle weakness, and dysphagia	Elevation of 2‐hydroxyglutaric acid and 2‐hydroxyadipic acid
Goh et al., 2018	Male/17 years	c.770A>G	p.Y257C	c.250G>A	p.A250T	Muscle weakness, exercise intolerance, and dysphagia	Elevated CK, myoglobin, and elevated concentrations of several acylcarnitine species (C5–C18)
Van der Westhuizen et., 2018	Female/2 years	c.1067G>A	p.G356E	c.1448C>T	p.P483L	Progressive muscle weakness, hepatomegaly, migraine, exercise intolerance, and hypermobility	Elevated AST, ALT, muscle respiratory chain enzymes deficient (CI, CIII, CII + III, CIV); Muscle CoQ10 levels reduced, and increased C4, C5, C5‐DC
	Female/4 years	c.1067G>A	p.G356E	c.1448C>T	p.P483L	Progressive muscle weakness, hepatomegaly, and migraine	Elevated AST, ALT, Oxidation of C16 and C14 fatty acids reduced in fibroblasts, and increased C4, C5, C5‐DC, C8
Missaglia et al., 2020	Female/38 years	c.560C>T	p.A187V	c.1027T>C	p.W343R	Skeletal and respiratory muscle weakness	Increased AST, CK, CKMB, and LDH
	Female/2 years	c.560C>T	p.A187V	c.1285 + 1G>A	p.G429Dfs*21	Vomiting, appetite loss, drowsiness, asthenia and acetonemic breath, and muscle atrophy	–
Pan et al., 2020	Female/4 years	c.250G>A	p.A84T	c.872T>G	p.V291G	palpitation and exercise intolerance, dropped head syndrome, loss of consciousness, and cardiac syncope*	Increased ALT, AST, C2, C3, C4DC, C16
Zhu et al., 2023	Male/21 years	c.1034A>G	p.H345R	c.1448C>A	p.P483Q	Limb muscle tenderness, and neck and proximal muscle weakness, vomiting, and severe pain	Increased CK, and myoglobin

3.3. Protein modeling results

The SWISS‐MODEL‐Expasy result contained two suggested models for each protein, one without its ligands and the second, which was selected, included ligands (i.e., 1 × FAD, 1 × SF4, 1 × UQ5). QMEANDisCo Global score, implying high‐quality of modeling, was 0.92 ± 0.05 in both wild‐type and mutant models. In the following, comparing the two models revealed in the mutant model, the number of Ramachandran Outliers increased from 1.39 to 1.56 (one from 617 amino acids), and the number of bad angles increased from 56/6437 to 58/6434 (Figure 4a–c).

FIGURE 4.

Protein modeling output. (a) and (b) are the Ramachandaran's plot for wild and mutant protein, respectively. (c) Protein modeling in mutant; (d) superimposition of wild type and mutant protein.(e) the position of Arg 309 with respect to the cofactors (f, g,) Close up novel variant to mention observing novel hydrogenic and ionic bonds respectively.

Scoring in UCLA‐DOE LAB tools for both models are acceptable, with the following results in wild‐type and mutant models, respectively: ERRAT scores were 88.225 and 88.7522, VERIFY scores were 89.10% and 88.93%, and the G‐factor in PROCHECK was negative −0.29 and −0.28. he Z‐scores provided by ProSA‐web for both models were within the range of Z‐scores for models achieved by x‐ray of similar lengths (−9.39 in the wild‐type model and −9.29 in the mutant model). TM‐align result indicated a slightly different structure RMSD = 0.02, Seq_ID = n_identical/n_aligned = 0.998. Figure 4d shows the result of superimposing the two models, and Figure 4e shows the location of novel variant in protein structure 3D and its cofactors. Figure 4f,g, respectively, shows the probable and possible novel bond in protein with this novel missense variant.

STRUM predicted that ΔΔG was −0.93, which means an increase in ΔG and a decrease in stability due to this missense variant. Other calculation methods also indicate a decrease in stability (ΔΔG varies from −1.6 to −0.43).

4. DISCUSSION

The present report demonstrated a case of GA2 with novel compound heterozygote variants in the ETFDH gene. Our patient's symptoms developed when she was six, and her disease remained undiagnosed for 13 years. The variable phenotype of GA2 complicates diagnosis in many patients, leading to a wide range of differential diagnoses in mild and moderate cases that can result in significant delays in diagnosis.

GA2 is a rare disorder of amino acids, fatty acids, and choline metabolism with variable clinical manifestations (Przyrembel et al., 1976). Mutations in the three genes, that is, ETFA, ETFB, and ETFDH, lead to GA2. ETFA and ETFB genes, respectively, encode electron‐transferring‐flavoprotein subunits alpha and beta, and ETFDH encodes electron‐transferring‐flavoprotein dehydrogenase. These products are parts of the complex II mitochondrial respiratory chain. Any molecular defect in these products disrupts choline, amino acids, and fatty acid metabolisms, affecting the disease phenotype (Onkenhout et al., 2001). The correlation between genotype and phenotype in GA2 is complicated (Table 2). Patients with mutations in ETFA and ETFB genes are more likely to develop symptoms at an early age in contrast to those who have mutations in the ETFDH gene (Izumi et al., 2011; Olsen et al., 2003). Missense mutations in the ETFDH gene present with different degrees of GA2 phenotype, most likely depending on the level of preserved activity of the produced protein. Clinical phenotype, especially muscle atrophy, is related to decreases in ETFDH protein function (Missaglia et al., 2020). Based on the information available from ClinVar as of 5/8/2024, 229 mutations of ETFDH have been identified. Among these mutations, there are 181 pathogenic variants and 148 likely pathogenic variants. It is noted that the types and frequencies of these mutations vary among discrete populations (Fan et al., 2018); (Zhuo et al., 2015). Regardless of the classification of GA2 based on the genetic etiology, another classification is based on the age of onset (Yotsumoto et al., 2008). The early‐onset forms of GA2 generally occur during the neonatal and infantile periods and have a poor prognosis. The late‐onset forms usually present with progressive muscle weakness and have a more favorable prognosis (Yotsumoto et al., 2008). Based on the initial manifestation, GA2 might be misdiagnosed with a wide range of acute or chronic diseases according to its initial symptoms. Moreover, another possible explanation behind the delayed diagnosis of GA2 is the episodic nature of the disease. Metabolic stress induces the disease symptoms, and patients may develop various symptoms over the disease course (Yotsumoto et al., 2008). In the acute phase of the disease, mass spectrometry detects an exertion of many organic acids in urine and serum. However, these metabolites may not be detectable in the stable phases, and the diagnosis may become complicated (Yotsumoto et al., 2008).

It has been reported that GA2 may be presented as Guillain‐Barre syndrome, adult myopathy, chronic digestive disease, and even syncope (Fu et al., 2016; Hong et al., 2018). Gastrointestinal manifestations, including intermittent nausea, vomiting, and abdominal pain, were seen in one‐third of patients with compound heterozygote mutations in the ETFDH gene (Fu et al., 2016; Missaglia et al., 2020; Purevjav et al., 2002; Wen et al., 2015; Xue et al., 2017; Yotsumoto et al., 2008; Zhuo et al., 2015) (Table 2). Gastrointestinal symptoms uncommonly presented as the initial presentations of GA2. Although our patient's gastrointestinal symptoms started when she was 6, these symptoms in GA2 affected might commence as early as the first days of life to 60 years (Fu et al., 2016; Purevjav et al., 2002). Our patient underwent various gastrointestinal workups and remained an undiagnosed case of gastrointestinal disease for 13 years. Fu et al. reported three patients with compound heterozygote mutations in the Chinese population who developed gastrointestinal symptoms, muscle weakness, and muscle atrophy (Fu et al., 2016). They demonstrated a case of GA2 with first and dominant gastrointestinal symptoms (Fu et al., 2016). In contrast to our patient, their case was a late‐onset form of GA2 presenting in the sixth decade of life as irritable bowel syndrome, who received proton pump inhibitors for 6 years (Fu et al., 2016).

The primary concern in this case is the GI phenotype due to its significant impact on daily life compared to a 1–2 degree decrease in musculoskeletal power (MMT) and muscle weakness, which is more likely supposed to be a sequelae of poor feeding followed by repetitive vomiting, and probably is neglected in early examinations. Adult‐onset GA2 is presumed to be a spectrum of phenotypes where musculoskeletal symptoms are more frequent than metabolic complications. In this case, it should follow the same pattern. The wide spectrum of GA2 arises from two levels: first, the type of mutations, where residual enzyme activity correlates with the severity of the phenotype. For instance, in the case of null mutations compared to leaky mutations, more severe phenotypes are expected. Second, this spectrum arises from environmental conditions such as diet (calorie deficiency and/or high levels of fat and protein), dehydration, and other diseases like seasonal flu, which can diversify the phenotypes of a case and are difficult to estimate the effect of (Grünert, 2014; Olsen et al., 2003; Prasun, 2020).

Even though literatures do not mention specific mechanisms linking choline catabolism to GI phenotypes, defects in choline metabolism as a consequence of GA2 may be more prominently affect the GI system. These defects affect phospholipid synthesis, subsequently impacting the integrity and function of GI system cells, and increasing their sensitivity to oxidative stress. Impaired choline transport and choline‐dependent enzyme function can also contribute to GI symptoms in GA2 (Repetto et al., 2010). However, understanding whether choline metabolism is the sole or primary reason for the prevalence of the GI phenotype in this case remains challenging.

Multiple lines of evidence support the hypothesis that Leu309Arg is a causative variant. The first clue is provided by the clinical features (ms/ms profile and histopathology) compatible with GA2. The second line of evidence is provided by ES and segregation data. No other variant in GA2 genes was observed to be considered causative, but it should be noted that deep intronic variants would not be captured by ES. Leu309Arg is classified as likely pathogenic according to ACMG and SHERLOC guidelines for the classification of variants (Nykamp et al., 2017; Richards et al., 2015). This variant is conserved, and multiple prediction tools indicate that it could have a deleterious effect, but the functional analysis is the only conclusive approach. Protein modeling is the final method utilized in this analysis.

Ramachandaran's plot shows that arginine is placed in the non‐permissible areas of dihedral angles ϕ and ψ, probably due to its larger size than leucine, necessitating significant changes in the protein structure for stabilization. Superimposition of the wild‐type and mutant predicted structures revealed minor structural alterations in ETFDH. This alteration may involve in its interactions with the ubiquinone pool and other proteins participating in electron transfer to the Electron Transport Chain (ETC), potentially posing challenges for proper mitochondrial function and energy metabolism.

In addition to the larger size of arginine, it differs from alanine regarding charge and polarity (because of the guanidino group), and its replacement can have a drastic effect. The formation of hydrogen bonds (two hydrogen bonds are created with each of the amino acids aspartate 292 and histidine 311) and a potential new ionic bond (with aspartate 292; Figure 4f,g), along with changes in dG levels, provide evidence in support of this hypothesis. It must be noted that the new ionic bond between arginine 309 and aspartate 292 is created, assuming no change in the position and rotation of aspartate, and this bond is not observed in the mutant modeling. This bond was shown because it might be created under in vivo conditions.

Position 309 is close to the FAD and UQ5 binding site, and the change of the amino acid in this position could impact the beta‐sheet structure and consequently affect the binding of these ligands, warranting additional investigation. In addition, if this missense mutation affects the binding of cofactors, it can influence protein stability. In this case, the underlying mechanism could lead to binding disruption, altering protein conformation, dynamics, or thermodynamics, resulting in an unstable protein conformation. Simultaneously, the issue with cofactor binding due to this missense variant could decrease enzyme activity (Stefl et al., 2013).

The first report of ETFDH:p.Gly381Arg, a previously reported pathogenic variant present in this case, was described by Vieira et al. in 2016. It involved a homozygous neonatal boy with hepatomegaly, extreme brain deterioration during crisis as observed by MRI, and profound hypotonia. In contrast to other severe neonatal cases, this was riboflavin‐responsive. Extensive brain leukodystrophy is a rarely reported feature in MADD (Vieira et al., 2017).

The second report of ETFDH:p.Gly381Arg was by Fischer et al. in 2019. It involved a homozygous boy who experienced his first crisis at the age of 12 with massive hyperammonia and lactic acidosis. The child's development was mildly delayed, with leukodystrophy and a predisposition to seizures. He was wheelchair‐bound and has cognitive restrictions. The use of D/L‐β‐hydroxybutyric acid in severe MADD could be a beneficial addition to the use of classical ketone body salts (Fischer et al., 2019).

The third report was by van Rijt et al., who conducted an analysis of the effectiveness of D,L‐3‐hydroxybutyrate in a case series, including one case with ETFDH:p.Gly381Arg in a homozygous status. However, no details were provided by the authors (van Rijt et al., 2020).

5. CONCLUSION

Here, a case is presented where the individual exhibits relatively mild symptoms that periodically manifest under specific conditions. Disease crises can be due to an increase in energy demands during illness or other stresses, whereas the consumption of antibiotics can decrease the threshold for symptom manifestations (i.e., disease exacerbations may result from increased energy demands during illness or stress, whereas antibiotic use could exacerbate symptoms). Discovering accurate information about the pathogenesis of GA2 would provide logical connections between genotype and phenotype, the periodic nature of the disease, and forecasting the time of clinical crisis.

To the best of our knowledge, our patient is the first case of GA2 with the novel compound heterozygote mutation in the ETFDH gene presented with gastrointestinal symptoms as the initial manifestations of this relatively rare disease in Iran. The episodic nature of the disease complicated the diagnosis, and ES revealed the possible genetic cause of the disease after 13 years. The segregation analysis, prediction tools, and protein modeling suggested these variants in the ETFDH gene, ETFDH: L309R and ETFDH: G381R, as reasonable causative variants for GA2. The present case report demonstrated that intermittent gastrointestinal problems in childhood could be a manifestation of GA2, especially accompanied by muscle weakness.

AUTHOR CONTRIBUTIONS

Mohammad Reza Seyedtaghia: Contribution in data collecting, writing and editing the manuscript and data analysis. Reza Jafarzadeh‐Esfehani: Contribution in data collecting and writing the manuscript. Yalda Nilipour: Pathologist of the team and editing the manuscript. Seyedmojtaba Hosseini: protein modeling. Sepehr Kobravi: writing the manuscript. Mahdis Hakkaki: Corporating in revision.

ACKNOWLEDGMENTS

We would like to express our sincere gratitude to the patient who generously provided their diagnostic data for this publication, and to the Academic Center for Education, Culture and Research (ACECR) manager and staff for their support.

FUNDING INFORMATION

Not applicable.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

All tests performed were part of the patient's diagnostic process. This article collects data from these examinations and does not disclose any information that could lead to patient identification. During the diagnosis process, consent was obtained from the patient that this information may be used in research works without revealing the identity.

Seyedtaghia, M. R. , Jafarzadeh‐Esfehani, R. , Hosseini, S. , Kobravi, S. , Hakkaki, M. , & Nilipour, Y. (2024). A compound heterozygote case of glutaric aciduria type II in a patient carrying a novel candidate variant in ETFDH gene: A case report and literature review on compound heterozygote cases. Molecular Genetics & Genomic Medicine, 12, e2489. 10.1002/mgg3.2489

DATA AVAILABILITY STATEMENT

Raw data will be made available on reasonable request.