Clinical and molecular investigations of Japanese cases of glutaric acidemia type 2

Clinical and molecular investigations of Japanese cases of glutaric acidemia type 2

Available online at www.sciencedirect.com Molecular Genetics and Metabolism 94 (2008) 61–67 www.elsevier.com/locate/ymgme Clinical and molecular inv...

167KB Sizes 2 Downloads 117 Views

Available online at www.sciencedirect.com

Molecular Genetics and Metabolism 94 (2008) 61–67 www.elsevier.com/locate/ymgme

Clinical and molecular investigations of Japanese cases of glutaric acidemia type 2 Yuka Yotsumoto a,*, Yuki Hasegawa a, Seiji Fukuda a, Hironori Kobayashi a, Mitsuru Endo a, Toshiyuki Fukao b, Seiji Yamaguchi a a

Department of Pediatrics, Shimane University School of Medicine, 89-1 En-ya-cho, Izumo, Shimane 693-8501, Japan b Department of Pediatrics, Graduate School of Medicine, Gifu University, Gifu 501-1194, Japan Received 4 November 2007; received in revised form 6 January 2008; accepted 6 January 2008 Available online 4 March 2008

Abstract Glutaric acidemia type 2 (GA2) is an autosomal recessive disorder resulting from a deficiency of electron transfer flavoprotein (ETF) or ETF dehydrogenase (ETFDH) that manifests from most severe neonatal to late-onset forms. However, the genetic defect responsible for the disease and clinical severity is not well-characterized. In order to understand the relationship between the phenotype and genetic defect, we investigated the clinical and molecular features of 15 Japanese patients, including 4 previously reported cases. Three patients had the neonatal form and 8 patients had the late-onset form, 1 of whom presented an extremely mild phenotype. Immunoblot analysis showed that either ETFa, ETFb, or ETFDH was significantly reduced or absent in all patients. However, no specific enzyme deficiency predominated, and there were no associations with the clinical severity. Genetic analyses identified 15 mutations including non-sense, missense, splice site mutations, and small deletions, in ETFA, ETFB and ETFDH genes. Although almost all mutations were unique to Japanese patients and no common mutations were found, some of them appeared to be associated with a specific phenotype. Our results suggest that clinical and mutational spectrums of Japanese GA2 patients are heterogeneous and that genetic diagnoses may help to predict a prognosis and provide more accurate diagnostic information for patients and families with GA2. Ó 2008 Elsevier Inc. All rights reserved. Keywords: Glutaric acidemia type 2 (GA2); Electron transfer flavoprotein (ETF); Electron transfer flavoprotein dehydrogenase (ETFDH); Organic acidemia; Fatty acid metabolism disorder

Glutaric acidemia type 2 (GA2) is an inherited autosomal recessive disorder of fatty and organic acid metabolism caused by a defect of electron transfer flavoprotein (ETF) or ETF dehydrogenase (ETFDH) [1,2]. GA2 is roughly divided into 2 clinical forms: a neonatal onset form (severe form) and a late onset form (milder form). In the severe form, hypotonia, hypoglycemia, metabolic acidosis and/ or hyperammonemia are present with/without anomalies such as dysplastic kidney and/or congenital heart disease. Children with the severe form show a fatal course in the neonatal period regardless of intensive treatments. In the mild form, the age at onset is often after infancy, and

*

Corresponding author. Fax: +81 853 20 2215. E-mail address: [email protected] (Y. Yotsumoto).

1096-7192/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2008.01.002

the symptoms are variable with intermittent episodes of hypotonia, tachypnea, hypoglycemia and/or hyperammonemia, which are often trigged by metabolic stress [3,4]. In Japanese cases of fatty acid disorders, GA2 is relatively common, followed by VLCAD deficiency and CPT2 deficiency in terms of frequency [5]. In the acute phase of the disease, increased excretions of suberylglycine, isovalerylglycine, hexanoylglycine, ethylmalonic acid, and hypoketotic dicarboxylic aciduria are noted on urine organic acid analysis using GC/MS. However, they may not be detected in the stable phase of many cases, making a precise diagnosis difficult. With ESI-MS/MS, an increase of some acylcarnitines with specific carboxylases ranging from medium to long chains (C4, C5, C8:1, C8, C12, C14, C16, and C5DC) is detected, but in the stable asymptomatic phase, no abnormalities may be seen.

62

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

A defective protein in GA2, ETF, is a heterodimetric mitochondrial matrix enzyme consisting of alpha and beta subunits (ETFa and ETFb, respectively). ETFDH is a monomer integrated in the inner mitochondrial membrane. At the molecular level, the defective enzyme in GA2 is either ETFa, ETFb or ETFDH. These enzymes are required for electron transfer from at least 9 mitochondrial flavin-containing dehydrogenases to the main respiratory chain [6,7]. The genes for ETFa, ETFb and ETFDH proteins (ETFA, ETFB and ETFDH, respectively) were cloned and mapped to 15p23–25, 19q13.1 and 4q33, respectively [8–10]. Up to now, 55 mutations have been reported in the literature, including 4 Japanese mutations [11–14], although GA2 is one of the most common defects in fatty acid oxidation and organic acids. Since one of the three genes is affected in GA2, it is essential to accumulate information on genetic mutations to determine any genotype/ phenotype correlation and to identify defective enzymes for an accurate diagnosis/prenatal diagnosis of GA2. In this study, we investigated the relationship between clinical and molecular aspects of Japanese patients with GA2, in which typical profile of urinary organic acids were observed at least in the acute stage, and found 15 mutations in 11 Japanese patients, referring 4 previously reported cases [11,12]. Materials and methods Patients GA2 was diagnosed in 11 Japanese children, based on the characteristic metabolic profiles of urinary organic acids analyzed using gas chromatography/mass spectrometry (GC/MS). Informed consent was obtained from all patients’ families. Our study protocol was approved by the Ethics Committee of the Shimane University Faculty of Medicine. Previously reported Japanese patients (cases 12–15) were compared.

Cell culture Skin fibroblasts obtained from the patients were cultured in Eagle’s essential medium (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) containing 10% fetal bovine serum (10% FCS/MEM) and antibiotics (penicillin-streptomycin: 100 lg/mL).

Immunoblotting The pellets of cultured fibroblasts were dissolved in 0.1 M potassium phosphate buffer, pH 7.2, 1% Triton X-100 and 0.2 M NaCl, and then sonicated. After centrifugation, the supernatant was subjected to 12.5% SDS/ PAGE. Immunoblot analysis was carried out as previously described [15]. The protein concentration was determined by the method of Lowry et al. [16], according to the Bio-Rad protein assay protocol (Bio-Rad Laboratories, Hercules, CA, USA). Fifty micrograms of protein of fibroblasts were subjected to each lane.

DNA sequencing The genomic DNAs were isolated from fibroblasts using a Qiamp DNA Microkit (QIAGEN GmbH, Hilden, Germany). Control genomic DNA from 50 unaffected Japanese individuals was obtained from peripheral blood lymphocytes using Blood and Cell Culture DNA Midi Kits (QIAGEN GmbH, Hilden, Germany). Each exon of ETFA, ETFB, and

ETFDH including intron/exon boundaries was PCR-amplified for 30 cycles using the conditions shown in Table 1. Primers of ETFDH were prepared as previously reported [17]. The PCR products were purified by a QIAquick PCR Purification Kit (QIAGEN GmbH, Hilden, Germany) and sequenced using ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems, Foster City, CA, USA) or CEQ 8000 Genetic Analysis System (Beckman Coulter Inc., Fullerton, CA, USA). The structures of human ETFA, ETFB, and ETFDH genes were obtained from the GenBank database (ETFA: AF436646–AF436657, ETFB: AF436658–AF436663, and ETFDH: AF449432–AF449444).

Results Clinical aspects Clinical phenotypes of 11 Japanese patients (cases 1–11) and 4 reported cases (cases 12–15) with GA2 are summarized in Table 2. Three patients showed neonatal onset with a severe phenotype (cases 1–3) and 8 showed the late onset form (cases 4–11). Case 12 had neonatal onset, whereas cases 13, 14, and 15 had late onset [11,12]. None of the patients had a familial history of consanguineous marriage. All neonatal patients died upon the initial development of symptoms, demonstrating the lethality of neonatal GA2. Polycystic kidney and congenital heart anomaly were seen in 2 cases with the neonatal form (cases 1 and 3). In contrast, 7 out of 8 cases with the late onset form remain alive even after the onset of disease. At least 2 patients with the late onset form (cases 4 and 5) responded to vitamin B2 (riboflavin-responsive). Case 7 and his father were affected with osteogenesis imperfecta. While polycystic kidney and congenital heart anomaly have been previously documented, GA2 with osteogenesis imperfecta has never been reported. It is not clear whether these combinations are merely coincidental or are closely associated. Among the 8 late onset patients, cases 4–8 showed the onset of symptoms before 2 years of age, whereas cases 9, 10, and 11 showed a later onset compared to the other patients. Of note, case 11 demonstrated an extremely late onset with the mildest phenotype. He stayed asymptomatic until his early 40’s, when muscle pain, hypotonia, and CPK elevation were noticed. He had 8 siblings, but his younger brother died suddenly at the age of 30 due to an unidentified disease. All other patients with late onset form developed symptoms upon infection or metabolic stress. Protein defect in ETF or ETFDH In order to confirm the diagnosis of GA2 of these patients, immunoblot analyses were performed using antibodies against ETF and ETFDH in the present cases along with 4 previously reported patients (Fig. 1). ETF or ETFDH protein was absent or significantly reduced in all patients, validating the GA2 diagnosis. ETF protein was absent in cases 1–3 and 6 whereas ETFDH was observed in these patients. In contrast, ETFDH was significantly

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

63

Table 1 Primers and conditions to amplify genes of ETFA, ETFB, and ETFDH Gene

Exon

Sense primer

Antisense primer

Annealing temperature (°C)

Product length (bp)

ETFA

1 2 3 4 5 6 7 8 9 10 11 12

ttatctcgttcgcgcgtct ctgtgacaacattctttacc tcttaactggcccttttgct aaggcctacttccaggtga agtcttccatcactatgtatgtg gagttagaagacaactggtc gtctttccattatcagagag ggtgacatgcttaatatggc cagtgaatatcctaactagc acagcaaacactttggaaag agaactatcagtctgtgtga ctgtgaccatatcacagtga

cagtgatctttgcaagacc tcctgtctcttgatggagat caagcgtaatttagacactac ccagattggctacataaacag gggccaccagcaattctagactaa tgtaggcagaggtcatctct cttcccaacttcttacatttg ccatttacttggaggcttaag cactgagatcactgttcaac acaaggtaagccaaatccta cctaaccatttctgagagtt ctctagaggctaggcatatt

53 53 55 55 55 53 51 53 51 55 51 53

227 251 268 303 312 325 380 359 230 199 215 389

ETFB

1 2 3 4 5 6

attgtatcaccggcggaagcggagacc ttgcctgtttcctcctgacc tgaggatagcagccaagtca tgaatggtccatgagctcca tcggacctgaattagcctca gagttccacagccctgtgaa

gttaaagcggcagaggtcg aatccaggctatcagcccat tcagcctggagtggtgaatg ttccaggaggagagaacag gctggcaatgtgcttgcca tgggtctctaggaaataag

51 57 55 55 55 57

300 347 318 222 266 282

ctgcagcagagttcttgctt tttcagtctactgaggaaaac gggttatattaatcccag cacttgcaaatataaact gtgaccatcaatgtagcact gaaacctaaggctgtttactgttttt tctgaccagatgtgaatgtatttt ttatgcattgtggtgacataaa cctacatgttttctgata gggtattctgttgttctt cagtttcgcacttaacat gggctagtcatatttctttggtg aagttaggcacttcaata

gcctgagaaagctgatgaga caaagtatccagaaaagtctc gggaacaattactgaaat ccttccagctgtggaattc catgaggaattcaagtactc tttcactttgatgcccacac ccccttgaaaaatatcgcaaa aatataactctagcagcagaat acatacttttcctattcc aaatacacataacccagc atcatgtcactcatactc cacattcctaaaatgtttaaagcaaa aaactggctagctgcagt

60 50 50 50 57 55 55 55 48 53 53 55 50

271 253 330 178 401 235 374 280 351 288 251 390 221

ETFDH

1 2 3 4 5 6 7 8 9 10 11 12 13

reduced or barely detectable in cases 4, 5, 7, 8, 9, 10, and 11, all of whom showed the late onset form. ETF was observed at comparable levels in these patients (cases 4, 5, 7, 8, 9, 10, and 11) with controls. These results suggest that cases 1, 2, 3, and 6 are deficient in ETF, while patients 4, 5, 7, 8, 9, 10, and 11 exhibit ETFDH deficiency. Gene mutations in ETFA, ETFB or ETFDH ETFA and ETFB Genetic defects in these patients were determined by DNA sequencing (Table 3). Direct sequencing of all the intron/exon boundaries of ETFA and ETFB genes in patients 1, 2, 3, and 6 with ETF deficiency revealed 3 different mutations in the ETFA gene and 4 mutations in the ETFB gene. Patients 1 and 2 with the neonatal phenotype were compound heterozygotes of 77delG/R174stop and K19stop/80delC in the ETFB gene, respectively. A novel homozygous IVS6-1G>C mutation at the splice acceptor site of intron 6 in the ETFA gene was identified in case 3 with the neonatal phenotype. The mRNA of the patient lacked the N-terminus in the first 7 nucleotides of exon 7, resulting in a truncated peptide as a consequence of a frame shift. In addition, a shorter transcript that lacks exons 6 and 7 was identified by direct sequencing of the whole

PCR product, although this shorter product was barely detectable in the RT-PCR gel. A normal transcript was not detected. Familial analysis revealed that her parents were heterozygous carries for IVS6-1G>C. A novel homozygous mutation of L95V was found in the ETFA gene in patient 6 with the late onset phenotype. A homology search of the ETFA protein in different species demonstrated that leucine 95 was highly conserved from zebrafish, Xenopus, rats, and mice, to humans. ETFDH All patients with ETFDH deficiency had the late onset form. Sequencing of their ETFDH gene demonstrated 9 mutations. A homozygous C to T transition in exon 11 that substitutes proline at 456 with leucine (P456L) was found in patient 11 with an extremely mild phenotype. Patient 10, who manifested liver dysfunction during early adolescence, harbored a G to A transition at nucleotide 524 and a T to C transition at 1774. They introduced a substitution of arginine 175 in the FAD binding domain with histidine (R175H) and cysteine 592 with arginine (C592R), respectively. Patient 7 with osteogenesis imperfecta who experienced a hyperammonemia attack at 1.4 years old was compound heterozygous for G362R and P534L. Patient 4 who showed hypotonia and hypoglycemia at 5

64

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

Table 2 Clinical manifestations of Japanese patients with GA2 Patient Neonatal 1 2 3

Age at onset onset form F 0d F 0d F 0d

Late onset form 4 M 5 M 6 M 7 M 8 F 9 F 10 F 11 M

Family history of sudden death

Congenital anomaly

Clinical findings at onset

Outcome

+ + +

PCK, CHD ? PCK

Dyspnea, hypoglycemia, and liver dysfunction Sudden death Liver dysfunction and cardiomyopathy

Dead (3d) Dead (5d) Dead (21d)

Hypotonia and hypoglycemia, and riboflavin-responsive Hypotonia and riboflavin-responsive Poor feeding, CPK elevation, and cardiomyopathy Hypotonia, hypoglycemia, and hyperammonemia Vomiting, hypoglycemia, and liver dysfunction Convulsion, hypoglycemia, and liver dysfunction Vomiting, hypotonia, and liver dysfunction Hypotonia, muscle pain, and CPK elevation

Alive n.d. (1y) Alive n.d. (15y) Dead (1y) Alive n.d. (5y) Alive n.d. (17y) Alive n.d. (7y) Alive (14y) Alive n.d. (60y)

5m 6m 8m 1y4m 1y10m 5y 13 y 58 y

+

— — — OI — — — —

Previously reported cases (Neonatal onset form) 12* M 0d

+

PCK

Dyspnea, hypoglycemia, and hyperammonemia

Dead (3d)

+ +

— — —

No remarkable symptoms but died suddenly Poor feeding, hypoglycemia and liver dysfunction Reye-like illness and riboflavin-responsive

Dead (3y) Dead (2y) Alive n.d. (2y)

(Late onset form) M 4m 13** 14** M 5m 15* M 1y0m

+ +

PCK, polycystic kidney; CHD, congenital heart disease; OI, osteogenesis imperfecta; n.d., normal development. * Purevjav E. et al. [12]. ** Colombo I. et al. [11]; sibling case.

ETFα ETFβ ETFDH C

1

2

3

4

5

6

7

8

9

10

11

ETFα ETFβ ETFDH 12

13

14

15

Fig. 1. Immunoblotting of electron transfer flavoprotein (ETF) and ETF dehydrogenase (ETFDH) in the fibroblasts. Fifty micrograms of protein were loaded per lane. ETFa and ETFb represents a- and b-subunit of ETF, respectively; ETFDH represents ETF dehydrogenase; C, a normal control; Patient numbers are shown at the bottom of the gel. Arrowheads indicate the protein detected by immunoblot analyses.

months old was heterozygous for the F308V mutation in exon 9, but no other mutation was identified. Patient 5 was homozygous for the A403V mutation in exon 10 and developed hypotonia at 6 months old. A heterozygous codon termination mutation (R559stop) and L366F mutation were identified in exons 12 and 9, respectively, in patient 8. L366F was also identified in patient 9, but no other mutation was detected. All mutations identified in this study were novel except for P456L in ETFDH, that was reported in the late onset form [17]. A homology search of the EST data base for human ETFDH identified several potential ETFDH transcripts in the beetle, pea aphid, and zebrafish and demonstrated that the mutated amino acids are highly conserved among insects, zebrafish, mice, rats, and humans. Screening of normal Japanese indi-

viduals revealed the presence of none of these mutations in 100 alleles of Japanese controls. Discussion This study identified 3 patients with the neonatal onset form and 8 cases with the late onset form in the Japanese population. Biochemical and genetic analyses including 4 previously reported cases revealed that there were 4 cases with ETFa deficiency, 4 with ETFb deficiency, and 7 with ETFDH deficiency. There were 15 genetic alterations: 14 novel and 1 reported mutations were identified. Eight missense mutations (R175H, F308V, G362R, L366F, A403V, P534L, P456L, and C592R) and a nonsense mutation (R559stop) were identified in the ETFDH gene, respec-

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

65

Table 3 Genetic aspects of Japanese patients with GA2 Patient

Mutant gene

Neonatal onset form 1

ETFB

2

ETFB

3

ETFA

Late onset form 4

ETFDH

cDNA position

Effect

Exon 2 Exon 5 Exon 1 Exon 2 Intron 6 Intron 6

77delG 490C>T 55A>T 80delC IVS6-1G>C IVS6-1G>C

Truncated Truncated Truncated Truncated Truncated Truncated

Exon ? Exon Exon Exon Exon Exon Exon Exon Exon Exon ? Exon Exon Exon Exon

922T>G ? 1208C>T 1208C>T 283T>G 283T>G 1084G>A 1601C>T 1096C>T 1675C>T 1096C>T ? 524G>A 1774T>C 1367C>T 1367C>T

F308V

9

5

ETFDH

6

ETFA

7

ETFDH

8

ETFDH

9

ETFDH

10

ETFDH

11

ETFDH

Previously reported cases (Neonatal onset form) 12*

ETFA

Exon 1 Exon 9

7C>T 799G>A

Truncated (R3stop) G267R

(Late onset form) 13**

ETFB

14**

ETFB

15*

ETFA

Exon 5 Intron 5 Exon 5 Intron 5 Exon 6 Exon 9

491G>A IVS5+1G>C 491G>A IVS5+1G>C 478delG 764G>T

R174Q G148-M200del R174Q G148-M200del Truncated (frame shift) G255V

* **

10 10 4 4 9 12 9 12 9

(frame shift) (R174stop) (K19stop) (frame shift) (frame shift) (frame shift)

5 13 11 11

A403V A403V L95V L95V G362R P534L L366F Truncated (R559stop) L366F R175H C592R P456L P456L

Purevjav E. et al. [12]. Colombo I. et al. [11]; sibling case.

tively. A splice site mutation (IVS6-1G>C) and a missense mutation (L95V) were found in 2 alleles of the ETFA gene, respectively. Two single nucleotide deletions (77delG and 80delC) and 2 nonsense mutations (K19stop and R174stop) were identified in the ETFB gene. Although the number of subjects limited, the frequency of ETFa, ETFb, and ETFDH deficiency in Japanese patients is 27, 27, and 47%, respectively, in our studies. The previous study showed a similar trend, although patients from various ethnic groups were analyzed (11, 33, and 56%, respectively) [13]. All neonatal forms were either ETFa or ETFb deficiency, but cases 6 and 15 with ETFa deficiency showed the late onset form. No cases of ETFDH deficiency showed the neonatal onset form in our study. Previous reports, however, identified patients with the neonatal onset form with ETFDH deficiency [13]. These findings suggest that the severity of the disease is unlikely associated with particular enzyme deficiency. Even if the late onset form appears to be predominant in GA2, the neonatal form may be

underestimated, since children with the neonatal onset form die early in the neonatal period before the diagnosis of GA2 is made. In addition to the current study, our previous studies revealed 4 mutations in the ETFA gene and 2 mutations in the ETFB gene in 4 Japanese patients (Table 3) [11,12]. In the 15 cases studied, only 2 cases with ETFa deficiency and 2 with ETFDH deficiency had homozygous mutations, while the rest of them were compound heterozygotes. There were no common mutations except for L366F in the ETFDH gene that was shared by 2 unrelated patients. Other mutations were different in each case, except for the sibling cases (cases 13 and 14). In addition, mutations identified in our study were different from those reported in other countries except for P456L. These findings suggest that Japanese patients with GA2 show allelic heterogeneity without frequent mutations. Similarly, multiple mutations were identified in ETFA, ETFB, and ETFDH genes in non-Japanese populations (19, 7, and 29 mutations, respectively) [13,14], though T266M in the ETFA

66

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

gene was reported as the most frequent mutation associated with a severe phenotype [13]. All mutations described in this report are novel, except for P456L. P456L in the ETFDH gene of patient 11 who remained asymptomatic until his 40’s was previously reported in a late onset patient [17]. The patient was a compound heterozygote with a nonsense mutation (W182stop) and P456L and remained asymptomatic until 27 years old [17]. These findings suggest that P456L is associated with a mild phenotype. Since our case was homozygous for this mutation, P456L most likely prevented the development of symptoms until 40 years of age, whereas its presence in combination with the W182stop mutation that can destroy the structure of the protein due to codon termination presumably accelerated the onset of the disease at 27 years [17]. However, immunoblot analysis failed to detect residual protein in our patient. The mutation may only affect the stability of the ETFDH protein with a short half-life without impairing its activity. Although a younger brother of case 11 is alleged to have died of an unknown disease at 30 years old, it was not clarified whether he was affected with the same mutation. A homozygous IVS6-1G>C mutation in patient 3 produced 2 aberrant ETFA transcripts. One of the transcripts lacked the first 7 nucleotides of exon 7, which appears to be generated by alternative splicing using the AG at the 6th and 7th nucleotides of exon 7 as a splice acceptor site as a consequence of the loss of the normal splice acceptor site in intron 6 (IVS6-1G>C). This introduces a frame shift and alters the protein structure, explaining the severe phenotype. The second aberrant transcript containing the skipping of exon 6 and 7 was barely detectable by RT-PCR, suggesting that it represents an extremely minor transcript. Although the mechanism responsible for exon 6 and 7 skipping in the minor transcript is not clear, the sequence analysis of other intron/exon boundaries detected no additional mutations. While exon 6 and 7 skipping does not introduce a frame shift, it is likely that the protein structure and function will be lost as a result of the loss of 71 amino acids. Two patients with the neonatal onset form (cases 1 and 2) were heterozygous for frame shift and codon termination mutations (77delG/R174stop and K19stop/80delC in the ETFB gene, respectively), both of which destroy the protein structure and/or function as a consequence of the frame shift or truncation of the ETFB mRNA and explain the severe neonatal onset phenotype. All patients with mutations on one of the ETF protein showed loss of both ETFa and ETFb protein, presumably because degradation of one subunit may also be induced by the other subunit deficiency. Although cause of the lability of ETF protein is not clear, an ETF complex may not be formed in the mitochondrial matrix due to abnormality in the peptides of either ETFa or ETFb, and the defect results in the lability of both subunits [4]. Although the polarity of the mutated amino acids was not affected by almost all missense mutations except for

G362R and C592R in ETFDH that alter the non-polar side chain to a basic side chain, all mutated amino acids are highly conserved among species from insect to humans. Furthermore, none of the mutations were found in 100 alleles from normal Japanese controls. These findings strongly suggest that they are specifically associated with the disease and are the most likely disease-producing mutations. Cases 5 and 6 with the late onset form were homozygous for A403V of ETFDH and L95V of ETFA, respectively, suggesting that these mutations may be associated with the late onset phenotype. Similarly, L366F in ETFDH that was shared by 2 unrelated patients with the late onset form appears to be associated with a mild phenotype since 1 of the patients (case 8) was a compound heterozygote with L366F and an R559stop nonsense mutation that likely destroyed the protein structure. These results suggest that some mutations may be associated with a specific phenotype, although it is necessary to accumulate additional cases with the same genotype to identify a definitive relationship. It is not feasible to make a definitive diagnosis for GA2 by GC/MS or ESI-MS/MS, although the technique is very useful to confirm the disease. All patients reported here demonstrated reduced or barely detectable ETF or ETFDH protein, making the deficient enzyme predictable. While biochemical diagnosis using immunoblot analysis is useful to identify cases with unstable or reduced protein levels, it does not rule out the functional deficiency of enzymes if patients show comparable levels of protein with controls. Since the activity of ETF and ETFDH may not be determined on a routine basis due to technical difficulties, the easiest and most reliable means to determine specific enzyme deficiency is a genetic diagnosis. Furthermore, this it will facilitate a prenatal diagnosis for families with affected children. In summary, our results demonstrate that the mutational spectrum is heterogeneous in Japanese patients with GA2, and their phenotypes were not associated with a specific enzyme deficiency. Although no common mutation was found, some mutations appeared to be associated with a specific phenotype. A genetic diagnosis may help to predict the potential outcome of patients and provide more accurate diagnostic information for patients and families with GA2. Acknowledgments We thank all of the attending physicians for providing clinical information on the patients, Dr. T. Hashimoto for providing anti-ETF antibody, and Ms. M. Furui for her technical assistance. This study was supported in part by a Grant from the Ministry of Education, Science, Technology, Sports and Culture of Japan, the Ministry of Health, Labor and Welfare of Japan, and the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (NIBIO).

Y. Yotsumoto et al. / Molecular Genetics and Metabolism 94 (2008) 61–67

References [1] F.E. Frerman, S.L. Goodman, Deficiency of electron transfer flavoprotein or electron transfer flavoprotein:ubiquinone oxidoreductase in glutaric acidemia type II fibroblasts, Proc. Natl. Acad. Sci. USA 82 (1985) 4517–4520. [2] Y. Ikeda, S.M. Keese, K. Tanaka, Biosynthesis of electron transfer flavoprotein in a cell-free system and in cultured human fibroblasts. Defect in the alpha subunit synthesis is a primary lesion in glutaric aciduria type II, J. Clin. Invest. 78 (1986) 997– 1002. [3] J.P. Loehr, S.I. Goodman, F.E. Frerman, Glutaric acidemia type II: heterogeneity of clinical and biochemical phenotypes, Pediatr. Res. 27 (1990) 311–315. [4] S. Yamaguchi, T. Orii, Y. Suzuki, K. Maeda, M. Oshima, T. Hashimoto, Newly identified forms of electron transfer flavoprotein deficiency in two patients with glutaric aciduria type II, Pediatr. Res. 29 (1991) 60–63. [5] Y. Tamaoki, M. Kimura, Y. Hasegawa, M. Iga, M. Inoue, S. Yamaguchi, A survey of Japanese patients with mitochondrial fatty acid beta-oxidation and related disorders as detected from 1985 to 2000, Brain Dev. 24 (2002) 675–680. [6] F.J. Ruzicka, H. Beinert, A new iron–sulfur flavoprotein of the respiratory chain. A component of the fatty acid beta oxidation pathway, J. Biol. Chem. 252 (1977) 8440–8445. [7] D.L. Roberts, F.E. Frerman, J.J. Kim, Three-dimensional structure of human electron transfer flavoprotein to 2.1-A resolution, Proc. Natl. Acad. Sci. USA 93 (1996) 14355–14360. [8] G. Finocchiaro, M. Ito, Y. Ikeda, K. Tanaka, Molecular cloning and nucleotide sequence of cDNAs encoding the alpha-subunit of human electron transfer flavoprotein, J. Biol. Chem. 263 (1988) 15773–15780. [9] G. Finocchiaro, I. Colombo, B. Garavaglia, C. Gellera, G. Valdameri, N. Garbuglio, S. Didonato, cDNA cloning and mitochondrial

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

67

import of the beta-subunit of the human electron-transfer flavoprotein, Eur. J. Biochem. 213 (1993) 1003–1008. S.I. Goodman, K.M. Axtell, L.A. Bindoff, S.E. Beard, R.E. Gill, F.E. Frerman, Molecular cloning and expression of a cDNA encoding human electron transfer flavoprotein-ubiquinone oxidoreductase, Eur. J. Biochem. 219 (1994) 277–286. I. Colombo, G. Finocchiaro, B. Garavaglia, N. Garbuglio, S. Yamaguchi, F.E. Frerman, B. Berra, S. DiDonato, Mutations and polymorphisms of the gene encoding the beta-subunit of the electron transfer flavoprotein in three patients with glutaric acidemia type II, Hum. Mol. Genet. 3 (1994) 429–435. E. Purevjav, M. Kimura, Y. Takusa, T. Ohura, M. Tsuchiya, N. Hara, T. Fukao, S. Yamaguchi, Molecular study of electron transfer flavoprotein alpha-subunit deficiency in two Japanese children with different phenotypes of glutaric acidemia type II, Eur. J. Clin. Invest. 32 (2002) 707–712. R.K. Olsen, B.S. Andresen, E. Christensen, P. Bross, F. Skovby, N. Gregersen, Clear relationship between ETF/ETFDH genotype and phenotype in patients with multiple acyl-CoA dehydrogenation deficiency, Hum. Mutat. 22 (2003) 12–23. M. Schiff, R. Froissart, R.K. Olsen, C. Acquaviva, C. Vianey-Saban, Electron transfer flavoprotein deficiency: functional and molecular aspects, Mol. Genet. Metab. 88 (2006) 153–158. S. Yamaguchi, T. Orii, N. Sakura, S. Miyazawa, T. Hashimoto, Defect in biosynthesis of mitochondrial acetoacetyl-coenzyme A thiolase in cultured fibroblasts from a boy with 3-ketothiolase deficiency, J. Clin. Invest. 81 (1988) 813–817. O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. S.I. Goodman, R.J. Binard, M.R. Woontner, F.E. Frerman, Glutaric acidemia type II: gene structure and mutations of the electron transfer flavoprotein:ubiquinone oxidoreductase (ETF:QO) gene, Mol. Genet. Metab. 77 (2002) 86–90.