Identification of six novel mutations in the acid alpha-glucosidase gene in three Spanish patients with infantile onset glycogen storage disease type II (Pompe disease)

Neuromuscular Disorders 12 (2002) 159–166 www.elsevier.com/locate/nmd

Identification of six novel mutations in the acid alpha-glucosidase gene in three Spanish patients with infantile onset glycogen storage disease type II (Pompe disease) Roberto Fernandez-Hojas a,b, Maryann L. Huie a, Carmen Navarro b, Carmen Dominguez c, Manuel Roig d, Diana Lopez-Coronas e, Susana Teijeira b, Kwame Anyane-Yeboa f, Rochelle Hirschhorn a,* a

Division of Medical Genetics, Department of Medicine, New York University School of Medicine, 550 First Avenue, New York, NY 10016, USA b Department of Pathology and Neuropathology, Hospital Meixoeiro, Vigo, Spain c Biochemistry and Molecular Biology Research Unit, Hospital Vall d’Hebron, Barcelona, Spain d Section of Neuropediatrics, Hospital Vall d’ Hebron, Barcelona, Spain e Department of Pediatrics, Centro Materno-Infantil Ntra. Sra. de Belen, S.A., La Corun˜a, Spain f Columbia University Physicians & Surgeons, Dept. of Pediatrics, New York, NY 10032, USA Received 20 November 2000; received in revised form 8 May 2001; accepted 12 June 2001

Abstract Glycogen storage disease type II is an autosomal recessive muscle disorder due to deficiency of lysosomal acid a-glucosidase and the resulting intralysosomal accumulation of glycogen. We found six novel mutations in three Spanish classic infantile onset glycogen storage disease type II patients with involvement of both cardiac and skeletal muscle; three missense mutations (G219R, E262K, M408V), a nonsense mutation (Y191X), a donor splice site mutation (IVS18 12gt . ga) and an in frame deletion of an asparagine residue (nt1408– 1410). The missense mutations were not found in 100 normal chromosomes and therefore are not normal polymorphic variants. The splice site mutation was subsequently detected in an additional ‘Spanish’ infantile onset glycogen storage disease type II patient from El Salvador. Further studies will be required to determine if the IVS18 12gt . ga splice site mutation might in fact be a relatively common Spanish mutation. Mutations among Spanish glycogen storage disease type II patients appear to be genetically heterogeneous and differ from common mutations in neighboring countries. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Alpha-glucosidase; Pompe disease

1. Introduction Acid maltase deficiency (AMD) or glycogen storage disease type II (GSD II) is an autosomal recessive disease of glycogen metabolism resulting from deficiency of the lysosomal enzyme acid a-glucosidase in all tissues of affected individuals. Three different forms, infantile, juvenile and adult variants are recognized, differing as to the age of onset, rate of progression and extent of tissue involvement. The infantile form (Pompe disease), with onset in the first few months of life, is characterized by severe hypotonia, progressive weakness and massive cardiomegaly with variable hepatomegaly and macroglossia. Classic infantile onset GSDII with cardiomegaly is typically fatal before two

* Corresponding author. Fax: 11-212-263-7151. E-mail address: [email protected] (R. Hirschhorn).

years of age due to cardiac failure. In both the childhood/ juvenile and adult onset variants, symptoms are generally limited to skeletal muscle, with a slowly progressive proximal myopathy and marked clinical involvement of respiratory muscles (reviewed in [1]). In all forms of GSDII the enzyme deficiency results in intra-lysosomal accumulation of glycogen of normal structure in skeletal muscle [1]. In the classic infantile onset form, massive glycogen storage is also present in cardiac muscle, as well as in other tissues, including central nervous system (especially anterior horn cells and brainstem motor nuclei). Extensive clinical, biochemical and genetic heterogeneity have been demonstrated [2–5]. The structural gene encoding acid a-glucosidase (GAA) is localized to human chromosome 17q25.2-q25.3 and contains 20 exons, the first of which is non-coding [1,6]. The cDNA is over 3.6 kb in length, with 2856 nucleotides of coding sequence, predict-

0960-8966/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 0960-896 6(01)00247-4

160

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

ing a protein of 952 amino acids with a calculated molecular mass of 105 kDa for the non-glycosylated protein [7,8]. The enzyme precursor is extensively modified post-translationally by glycosylation on asparagine residues, phosphorylation on mannose residues for targeting to lysosomes and proteolytic processing [9,10]. The major forms seen by SDS electrophoresis are a precursor of 110 kDa and two mature forms of 76 and 70kDa. Genetic heterogeneity is extensive both in patients and in normals, since more than 40 different deleterious mutations and more than 15 normal polymorphic variants have been described to date [1]. Many mutations identified represent private variants although a few are common in specific ethnic groups, including Dutch, Chinese and African-American patients [11–17]. A splice site mutation in intron 1 (IVS1–13t . g) is present in over 2/3 of adult-onset Caucasian patients [18,19]. We performed a molecular genetic study of three Spanish patients with classical infantile onset GSDII. Although no single ‘common’ Spanish mutation was identified in the three infantile onset patients initially studied, six novel mutations can now be added to the spectrum of deleterious mutations in the acid a-glucosidase gene. These newly identified mutations include three missense mutations (G219R, E262K and M408V), one nonsense mutation (Y191X), a donor splice site mutation (IVS18 12gt . ga) and an inframe deletion of an asparagine at an N-linked glycosylation site (DN470). The IVS 18 12gt . ga splice site mutation was subsequently identified in an additional infantile onset GSDII patient of presumably Spanish descent from El Salvador.

[20] with modifications [21]. Protein was measured by the method of Lowry et al. [22]. Genomic DNA was extracted by standard methods from leukocytes or cultured fibroblasts. All the coding exons (either singly or in groups of 2-3 exons) as well as the flanking intron/exon junctions of the acid a-glucosidase gene were amplified by Polymerase Chain Reaction (PCR) from genomic DNA using intron primers. The PCR reaction was performed in a final volume of 50 ml containing 5 ml of 10 £ PCR buffer (10 mM Tris–HCl pH 9.0, 50 mM KCl and 0.1 % Tritonw X-100), 3 ml of 25 mM MgCl2, 1.5 units of Taq polymerase, (Promega, WI, USA), 5 ml of 2 mM of each dNTP, (Amersham-Pharmacia, NJ, USA), 1 ml each of 0.2 mM forward and reverse primers (Gibco-BioRad Technology, NY, USA) and 300–500 ng of genomic DNA. PCR was performed using a Perkin-Elmer MasterCycler 2400 with the following cycling program: 5 min at 948 followed by 35 cycles (948C, 30 seconds; 608C, 30 s; 728C, 30 s) and final extension at 728C for 7 min. For products larger than 450 nt, the extension time was increased from 30 s to 1 min. The PCR products were directly purified or gel purified using the QiAquick PCR purification kit (Qiagen, Chatsworth, CA, USA) according to the manufacturer’s directions. We sequenced the PCR products directly, using ABI Prism Big-Dye terminator sequencing kits and an ABI Prism 310 or 377 Genetic Analyzer, according to the protocol supplied by the manufacturer (Perkin Elmer, Foster City, CA, USA). All mutations were confirmed by sequencing of a second independent PCR. Restriction enzyme analyses with BsaAI, BsiHKAI, DdeI (New England Biolabs, MA, USA) and Eco57 I (Fermentas, Lithuania) were performed according to the manufacturer’s directions.

2. Patients and methods We studied three Spanish Pompe patients (two female, one male) with clinical presentation consistent with classic infantile onset GSDII. The three patients were not related to one another and were the offspring of non-consanguineous parents. All three patients exhibited marked generalized hypotonia and cardiomegaly within the first three months of life. Cardiac dysfunction resulting from severe cardiac hypertrophy and respiratory infection was the main cause of death prior to the age of two. Serum enzyme values (CK, LDH, GOT, GPT) were markedly elevated in all three patients. (Table 1). Dermal fibroblast cultures were established by explant technique and maintained in Eagle’s minimum essential medium (MEM), supplemented with 10% fetal calf serum, nonessential amino acids and antibiotics (100 U/ml penicillin and 100 mg/ml streptomycin) at 378C with 5% CO2. For each enzymatic analysis the cultures used (patient or controls) were of the same passage number. Alpha glucosidase enzyme activity at acid and neutral pH (4.0 and 6.5) was determined in aqueous extracts of muscle and/or fibroblasts using the artificial substrate, 4-methylumbelliferyl-a-d-glucoside (4MUG), as previously described

3. Results 3.1. Histology and acid alpha glucosidase enzyme activity Examination of muscle biopsies by light microscopy for all three patients and by electron microscopy for one patient revealed vacuolar myopathy with storage of glycogen consistent with classic GSDII. Extremely low a-glucosidase activity in fibroblasts (0.42–2.5% of normal) from all three patients was consistent with classic infantile onset GSDII. Enzyme activity in muscle was determined in two of the three patients; patient three had enzyme activity in muscle consistent with infantile onset GSDII (5.6% of normal) while patient two had unexpectedly relatively high activity (16% of normal) more typical of milder juvenile or adult onset GSDII (enzyme activity in muscle from patient one was not determined). (See Table 1). 3.2. Identification of six novel mutations in three unrelated infantile onset GSDII patients Detection of a nonsense mutation (Y191X) and a splice site mutation (IVS18 12 t . a) in Patient 1: Direct sequen-

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

161

Table 1 Summary of clinical findings, alpha glucosidase activity and mutations a

Sex Age of onset Consanguinity Clinical picture Hypotonia/weakness Cardiomegaly Hepatomegaly Macroglossia Serum enzymes levels (in UI/l) CK LDH GOT GPT Electrocardiographic findings (ECG) Electromyographic findings (EMG) Muscle biopsy Age and cause of death Enzyme activity (% of normal) Fibroblasts b Muscle c Mutations Mutation #1 Site Enzyme change Mutation #2 Site Enzyme change a b c

Case 1

Case 2

Case 3

Female Neonatal (2.5–3 m) No

Male Neonatal (2 m) No

Female Neonatal (2.5–4 m) No

Severe Yes Yes (2–3 cm) Yes

Mild Yes Yes (8 cm) No

Severe Yes Yes (2 cm) No

296–1056 2284 260 165 Left ventricular hypertrophy Short PR interval Pseudomyotonic discharges. Myopathic pattern Vacuolar myopathy PAS and acid phosphatase positive 4.5 m. Heart failure and respiratory infection.

505 1416 212 156 Biventricular hypertrophy

1200 2302 350 209 Biventricular hypertrophy

Pseudomyotonic discharges. Diffuse myopathic pattern Vacuolar myopathy PAS and acid phosphatase positive , 2 years

Pseudomyotonic discharges. Myopathic pattern Vacuolar myopathy PAS and acid phosphatase positive 10 m. Heart failure and respiratory infection.

2.1% NA

0.42% 5.6%

2.5% 16.2%

Y191X; 573C . A exon 3 (1) DdeI IVS18 12gt . ga Intron 18 None

M408V; 1222A . G exon 8 (1) BsaAI Asn (nt1408-1410) Exon 9 None

G219R; 655G . A exon 3 (1) Eco57I 784G . A; E262K Exon 4 (2) BsiHKAI

m, months; cm, centimetre; UI/l, international units per liter. Control fibroblasts (Mean ^ SEM): 1.65 ^ 0.15 (n ¼ 18). Control muscle (Mean ^ SEM): 0.51 ^ 0.1 (n ¼ 4).

cing of all 19 coding exons and splice junctions revealed a heterozygous 573 C . A transversion in exon 3 (TAC . TAA) (Fig. 1a), predicting substitution of tyrosine by a stop codon (Y191X) and the gain of a site for DdeI. (Fig. 2a). The second allele in this patient showed a t . a transversion at the second nucleotide of intron 18, part of the invariant gt of the consensus splice site (IVS18 12gt . ga) (Fig. 1b). This mutation was not found by sequence analysis of 53 additional chromosomes. Most significantly, the splice site mutation was subsequently found in an additional Classic infantile onset GSDII patient from El Salvador of presumed Spanish descent. (In work in progress, this patient appears to have a large deletion not involving exon 18 on the other chromosome.) 3.3. Detection of a missense mutation (M408V) and deletion of a codon (D N470) in Patient 2 A heterozygous 1222 A . G transition in exon 8 (ATG . GTG), predicting substitution of methionine by valine at codon 408 (M408V) and a new site for BsaI was found on one allele of this patient (Fig. 1c). The M408V missense

mutation was maternally derived and absent in both the father and sibling as determined by restriction enzyme analysis (Fig. 2b). The M408V missense mutation was not found by similar restriction enzyme analysis of 100 normal Spanish chromosomes indicating that the M408V missense mutation is not a common normal polymorphic variant. The second mutation in this patient was a three nt ‘in frame’ deletion (Dnt1408–1410) in exon 9 of an asparagine residue at codon 470 (D N470), a site for N-linked glycosylation on the protein (Fig 1d) [10]. Sequence analysis of exon 9 from the family confirmed that the deletion (D nt1408-1410) was paternally derived (data not shown). 3.4. Detection of a missense mutations G219R and E262K in Patient 3 The third patient carried two missense mutations: a 655 G . A transition in exon 3 (GGG . AGG), predicting substitution of glycine by arginine (G219R) (Fig. 1e) and a 784 G . A transition in exon 4 (GAG . AAG), predicting substitution of glutamic acid by lysine (E262K) (Fig. 1f). Both these missense mutations are non-conservative

162

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

Fig. 1. Direct sequence analysis of PCR amplified genomic DNA demonstrating six novel mutations in the acid a-glucosidase gene. (a) Sequence analysis of exon 3 in case 1, demonstrating heterozygosity for a C . A transversion at nt 573, predicting substitution of tyrosine191 by a premature stop codon (Y191X; TAC . TAA). (b) Detection of heterozygosity in case 1 for an IVS18 12 t . a splice site mutation at the donor splice site in intron 18. Sequence of the antisense strand is shown, (3 0 –5 0 ) but the complementary, sense nucleotides are indicated and the mutation at the second nucleotide of intron 18 (t . a) is identified by an asterisk. (c) Sequence analysis of exon 8 in case 2, demonstrating heterozygosity for an A . G transition at nt1222, predicting the substitution of methionine 408 by valine (M408V; ATG . GTG). Sequence of the antisense strand is shown (3 0 –5 0 ) but the complementary, sense nucleotides are indicated and the 1222A . G transition identified by an asterisk. (d) Sequence analysis of exon 9 in case 2, demonstrating heterozygosity for the deletion in-frame of the nucleotides 1408-1410 coding for asparagine (N470) indicated by the asterisk. The overlapping peaks on the sequence correspond to the deleted paternal and normal maternal alleles downstream of the deletion. Sequencing in the reverse direction (not shown) confirmed the site of the deletion. (e) Sequence analysis of exon 3 in case 3, demonstrating heterozygosity for a G . A transition at nt 655, predicting substitution of glycine 219 by arginine (G219R; GGG . AGG). (f) Sequence analysis of exon 4 in case 3, demonstrating heterozygosity for a G . A transition at nt 784, predicting the substitution of glutamic acid 262 by lysine (E262K; GAG . AAG).

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

163

Fig. 2. Detection of the nonsense mutation (Y191X) and two missense mutations (M408V and G219R) by restriction enzyme digestion. (a) Identification of the Y191X; 573C . A nonsense mutation by gain of a site for DdeI: We amplified a 316 nt fragment containing exon 3 (forward primer 5 0 -tgcccatggtcccacatccatgtgtg-3 0 , and reverse primer 5 0 -agaatggggtcgccctccccatcatgct-3 0 ) and digested the fragment with DdeI. Presence of the Y191X nonsense mutation generates a new site for DdeI (CTACG . CTAAG) and results in two fragments of 171 and 145 nt. However exon 3 of the GAA gene contains three polymorphic sites (nt 596, 642 and 668) and presence of T at nt 642, a normal silent RFLP, also creates a restriction site for DdeI, resulting in fragments of 216 and 100 nt. This normal polymorphism can easily be distinguished from the Y191X nonsense mutation. The DdeI digests were resolved in 4.5% agarose gel and visualized by ethidium bromide staining. Lane 1 is a 100 nt ladder, lane 2 is undigested product, lane 3 is from case 1, heterozygous for the Y191X nonsense mutation and lane 4 is from a normal control heterozygous at the 642 polymorphic C/T. (b) Identification of the M408V; 1222 A . G missense mutation in case 2 by gain of a new site for BsaAI: A 703 nt fragment containing exons 6, 7 and 8 was amplified from genomic DNA of the patient, both parents and an unaffected sibling (forward primer 5 0 -tcaactctccgcctgtgattggcccat-3 0 , and reverse primer 5 0 -tgcacagagaaggagccactgggca-3 0 ) and digested the fragment with BsaAI. There are no sites for BsaAI in the normal fragment. Presence of the M408V; 1222 A . G missense mutation creates a new site for BsaAI (TACATG . TACGTG) and results in two fragments of 527 and 176 nt. The BsaAI digest was resolved in 3 % agarose gel and visualized by ethidium bromide staining. Lane 1 is a 100 nt ladder, lane 2 is the proband, heterozygous for the maternally inherited M408V missense mutation, lane 3 is from the sibling, normal for both mutations found in the patient and lane 4 is the father (known to carry the DAsn470 in exon 8, see Fig. 1d), lane 5 is the mother, showing presence of the 1222 A . G/M408V missense mutation and lane 6 is a normal control. (c) Identification of the G219R; 655G . A missense mutation by gain of a site for Eco57I. We amplified exon 3 as described for Fig. 2a (see above) and digested the fragment with Eco57 I. There is no site for Eco57I in the normal DNA fragment. The mutation creates a new site for Eco57I (CTTCGG . CTTCAG) and results in two fragments of 211 and 105 nt. Electrophoresis was carried out in 3% agarose gel and the DNA fragments were visualized by ethidium bromide staining. Lane 1 is a 100 nt ladder, lane 2 is undigested, lane 3 is from case 3, heterozygous for the G219R missense mutation and lane 4 is from a normal control.

changes and result in radical changes in residue charge. Neither the G219R missense mutation which predicts gain of a site for the restriction enzyme Eco57I nor the E262K missense mutation which predicts loss of a site for the restriction enzyme BsiHKAI was detected by restriction enzyme analysis of 100 normal chromosomes, indicating that both these missense mutations are not normal polymorphic variants. (Table 1, Fig. 2c,d).

4. Discussion We conducted molecular genetic studies (including sequence analysis of the complete coding region and surrounding splice sites) in three Spanish patients with the classic infantile onset form of Glycogen Storage Disease Type II (GSD II or Pompe disease). Six novel mutations were identified: three amino acid substitutions (missense

Table 2 Phylogenetic conservation of acid a-glucosidase at the site of missense mutations

Human a Murine b Quail c Quaile d Bovine e Conserved a b c d e

G219R

E262K

M408V

SE E P FGVI VHR SE E P FGVI V RR SQDP FGVLL RR LQDPFG I VV FR SE E PFGVVVRR * * * PFG * * * * R

I I L I I *

NDLDYMD S RRD NDLDYMDARRD ND I DYMDGYRD NDLDYMAKRRD NDLDYMDARRD ND* D YM * * * R D

Homo sapiens: GENBANK# CAA68764. Mus musculus: GENBANK# NP032090. Coturnix coturnix japonica: GENBANK# BAA25884. Coturnix coturnix japonica: GENBANK# BAA25890. Bos taurus: GENBANK # AF171665.

TGLA EHLS PL TGLG EHLS PL YGLGEHRSTL CGLGE RLAPL TGLAE HLGSL *GL *E * * * *L

164

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

mutations G219R, E262K, M408V), a nonsense mutation coding for a premature termination signal of translation (Y191X), a donor splice site mutation at the conserved splice site consensus sequence (IVS18 12gt . ga) and an in-frame deletion of an asparagine residue Dnt1408-1410) (Table 1). Based upon several considerations, all six mutations can be considered to be sufficiently deleterious to explain the observed severe phenotype. Firstly, for all six mutations, these were the only changes observed by sequence analysis of the entire coding region and exon/ intron boundaries. We considered the missense mutations G219R, E262K and M408V to be deleterious since (1) the amino acids glycine219, glutamic acid262 and methionine408 are conserved in acid a-glucosidase of species phylogenetically far removed from man (Table 2) as well as in other related enzymes such as intestinal sucraseisomaltase [2] and (2) these changes were not found in DNA from 100 normal chromosomes, indicating that they are not common polymorphic variants. Of the three additional mutations, the nonsense mutation (Y191X) codes for a premature termination of translation upstream of the catalytic site, and therefore is by definition deleterious. The IVS18 12gt . ga changes the 5 0 donor consensus splice site at the virtually invariant t (based on previous tabulation [23] and predicts disruption of normal splicing. More significantly, the same splice site mutation (IVS 18 12gt . ga) was subsequently found in an additional classic infantile onset GSDII patient of apparent Spanish descent from El Salvador. Furthermore, we have found evidence supporting the hypothesis that the IVS 18 12gt . ga splice site mutation arose from a common founder. When mutations arise from a common founder, the normal single nucleotide polymorphisms (SNPs) are identical at other contiguous sites within the gene. Family studies demonstrate that the splice site mutation is carried together with a G at the polymorphic site (SNP nt 2553) in exon 18. Additionally, the presence of a large deletion in the El Salvadorian patient carrying the splice site mutation allows us to determine the haplotype of 13 SNPs in the area of the deletion ([1] and unpublished). This haplotype is identical with that of patient 1 who is homozygous for these 13 polymorphisms. Lastly, the three nucleotide deletion in exon 9 (nt1408-1410), while inframe, deletes an asparagine at codon 470, which is one of seven N-linked glycosylation sites in the acid a-glucosidase protein. Site directed mutagenesis of the asparagine to a glutamine residue at codon 470 failed to show disruption of targeting of the enzyme to lysosomes [10]. However, the asparagine to glutamine substitution is a conservative change whereas the non-conservative loss of the site for N-linked glycosylation by deletion could give rise to an unstable protein. Additionally, in support of this hypothesis, the in frame loss of an asparagine residue in exon 14 (DAsn675), albeit not a glycosylation site, has been previously reported as a deleterious mutation in an infantile onset GSDII patient [24]. Earlier studies of GSDII patients appeared to indicate that

mutations were clustered in three critical regions of the gene; exon 2 which contains the start codon, exon 10–11 containing the enzyme catalytic site and exon 14 which corresponds to a highly conserved area of the protein [1,25]. However, the mutations reported herein are distributed throughout the gene, consistent with more recently published and unpublished studies. Of note, the Y191X nonsense and G219R missense mutations are the first mutations reported for exon 3, an exon that is less highly conserved and contains two normal polymorphic amino acid substitutions. All three Spanish patients and the fourth patient from El Salvador, were compound heterozygotes. Although no single common mutation was found among all the patients studied, the IVS18 1 2gt . ga was found in two of the eight chromosomes studied. Further studies will be required to determine if the IVS18 1 2gt . ga splice site mutation might in fact be a relatively common Spanish mutation. None of the patients carry Y292C, a missense mutation previously reported in an offspring from a Spanish father (Galician) carrying the Y292C [26] and a Dominican mother who carried the R854X nonsense mutation common in the African-American population [28]. We and others have not found any of the mutations described in this study in neighboring countries such as Portugal, France [27–29] or Italy (unpublished) or in other countries, including The Netherlands where investigations in GSD II have been extensively conducted [29]. This suggests different mutational origins in the Spanish population as compared to geographically proximate countries, as well as genetic heterogeneity, although more patients need to be studied. In most populations, genetic heterogeneity is one of the hallmarks in GSD II, thus identification of the molecular defect of the disease is arduous and time consuming. Most mutations identified have been ‘private’ variants distributed along the 19 coding exons. However, in certain ethnic groups some mutations appear to be common. Two deletions account for over 50% of the mutant alleles among the Dutch population (525T and exon 18); two missense mutations account for 50–80% of mutant alleles in Chinese from Taiwan (D645E, G615R) and a nonsense mutation is very frequent among African–Americans (R854X) [11– 17]. As previously mentioned, a specific splice site mutation (IVS1-13t . g) is found in approximately 70% of Caucasian adult-onset GSDII patients [18,19]. When constructing strategies for genetic screening, the ethnic origin should be one factor under consideration in order to use the appropriate technology. One common strategy for screening is PCR amplification of genomic DNA followed by automatic sequencing of the entire coding region of the gene and exon/intron boundaries. An alternative approach utilizes mutational screening of mRNA by RT-PCR, recognizing the fact that specific changes in intronic sequences and some types of mutations leading to absence of mRNA such as nonsense mutations and

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

frame altering small insertions/deletions cannot be detected by such methods [30]. Large intragenomic deletions, such as the recently described 9kb Alu–Alu mediated deletion [31] are also not easily detected by any PCR based methods and should be considered. A more precise genotypephenotype correlation, which currently continues to be difficult to establish, would be important for a better understanding of the disease and its therapy.

Acknowledgements This study was supported by a grant from the March of Dimes (R.H.), the NIH GCRC at NYU School of Medicine (# NCRRM01RR00096), the Spanish Fondo de Investigacion Sanitaria (FIS 96/0229) and Xunta de Galicia 99/ PX190501B. The authors are grateful to Professor Salvatore DiMauro (Department of Neurology, Columbia University Physicians & Surgeons, NY) for critical reading of the manuscript and arranging the opportunity to study these patients and C. O’Hara for help in the initial English version of this paper.

References [1] Hirschhorn R, Reuser AJJ. Glycogen storage disease type II: acid aglucosidase (Acid maltase) deficiency. In: Scriver CR, Beaudet AL, Valle D, Sly WS, editors. The metabolic and molecular bases of inherited disease. 8th ed. New York: McGraw–Hill, 2001. pp. 3389–3420. [2] Reuser AJ, Kroos MA, Willemsen R, Swallow D, Tager JM, Galjaard H. Clinical diversity in glycogenosis type II. J Clin Invest 1987;79:1689–1699. [3] Martiniuk F, Mehler M, Tzall S, Meredith G, Hirschhorn R. Extensive heterogeneity in patients with Acid Alpha Glucosidase deficiency as detected by abnormalities of DNA and mRNA. Am J Hum Genet 1990;47:73–78. [4] Reuser AJ, Kroos MA, Hermans MM, et al. Glycogenosis type II (acid maltase deficiency). Muscle Nerve 1995;Suppl 3:S61–S69. [5] Raben N, Nichols RC, Boerkoel C, Plotz P. Genetic defects in patients with glycogenosis type II (acid maltase deficiency). Muscle Nerve 1995;Suppl 3:S70–S74. [6] Kuo W-L, Hirschhorn R, Huie ML, Hirschhorn K. Localization and ordering of acid alpha-glucosidase (GAA) and thymidine kinase (TK1) by fluorescence in situ hybridization. Hum Genet 1996;97:404–406. [7] Martiniuk F, Bodkin M, Tzall S, Hirschhorn R. Isolation and partial characterization of the structural gene for human acid alpha glucosidase. DNA Cell Biol 1991;10(4):283–292. [8] Martiniuk F, Mehler M, Tzall S, Meredith G, Hirschhorn R. Sequence of the cDNA and 5’-flanking region for human acid a-glucosidase, detection of an intron in the 5 00 untranslated leader sequence, definition of 18-nt polymorphisms, and differences with previous cDNA and amino acid sequences. DNA Cell Biol 1990;9:85–94. [9] Wisselaar HA, Kroos MA, Hermans MM, van Beeumen J, Reuser AJ. Structural and functional changes of lysosomal acid alpha-glucosidase during intracellular transport and maturation. J Biol Chem 1993;268:2223–2231. [10] Hermans MM, Wisselaar HA, Kroos MA, Oostra BA, Reuser AJ. Human lysosomal a-glucosidase: functional characterization of the glycosylation sites. Biochem J 1993;289:681–686.

165

[11] Hirschhorn R, Huie ML. Frequency of mutations for glycogen storage disease type II in different populations: the DT525 and Dexon 18 mutations are not generally ‘common’ in white populations. J Med Genet 1999;36:85–86. [12] Van der Kraan M, Kroos MA, Joosse M, et al. Deletion of exon 18 is a frequent mutation in glycogen storage disease type II. Biochem Byophys Res Commun 1994;203:1535–1541. [13] Hermans MM, de Graaff E, Kroos MA, et al. The effect of a single base pair deletion (DT525) and a C1634T missense mutation (pro545leu) on the expression of lysosomal a-glucosidase in patients with glycogen storage disease type II. Hum Mol Genet 1994;3:2213– 2218. [14] Shieh JJ, Lin CY. Frequent mutation in Chinese patients with infantile type of GSD II in Taiwan: evidence for a founder effect. Hum Mutat 1998;11:306–312. [15] Hermans MM, de Graaff E, Kroos MA, et al. The conservative substitution Asp-645 ! Glu in lysosomal alpha-glucosidase affects transport and phosphorylation of the enzyme in an adult patient with glycogen storage disease type II. Biochem J 1993;289:687– 693. [16] Adams EM, Becker JA, Griffith L, Segal A, Plotz PH, Raben N. Glycogenosis type II: a juvenile specific mutation with an unusual splicing pattern and a shared mutation in African Americans. Hum Mutat 1997;10:128–134. [17] Becker JA, Vlach J, Raben N, et al. The African origin of the common mutation in African American patients with glycogen storage disease type II. Am J Hum Genet 1998;62:991–994. [18] Huie ML, Chen AS, Tsujino S, et al. Aberrant splicing in adult onset glycogen storage disease type II (GSD II): molecular identification of an IVS1 (213T ! G) mutation in a majority of patients and a novel IVS10 (121GT ! CT) mutation. Hum Mol Genet 1994;3:2231– 2236. [19] Boerkoel CF, Exelbert R, Nicastri C, et al. Leaky splicing mutation in the acid maltase gene is associated with delayed onset of glycogenosis type II. Am J Hum Genet 1995;56:887–897. [20] Dreyfus JC, Poenaru L. White blood cells and the diagnosis of aGlucosidase deficiency. Pediatr Res 1980;14:342–344. [21] Shanske S, DiMauro S. Late-onset acid maltase deficiency. Biochemical studies in leukocytes. J Neurol Sci 1981;50:57–62. [22] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265– 275. [23] Cooper DN, Krawczak M. Human gene mutation. Oxford, UK: Bios Scientific Publishers, 1993. [24] Ko TM, Hwu WL, Lin YW, et al. Molecular genetic study of Pompe disease in Chinese patients in Taiwan. Hum Mut 1999;13:380– 384. [25] Huie ML, Tsujino S, Sklower Brooks S, et al. Glycogen storage disease type II: identification of four novel missense mutations (D645N, G648S, R672W and R672Q) and two insertion/deletions in the acid a-glucosidase locus of patients of differing phenotype. Biochem Biophys Res Commun 1998;244:921–927. [26] Castro-Gago M, Eirı´s J, Rodrı´guez A, Pintos E, Benlloch T, Barros F. Forma grave de glucogenosis tipo II juvenil en un nin˜ o heterocigoto compuesto (Tyr-292 ! Cys/Arg-854 ! Stop). Rev Neurol 1999; 29:46–49. [27] Huie ML, Chen AS, Sklower Brooks S, Grix A, Hirschhorn R. A de novo 13 nt deletion, a newly identified C647W missense mutation and a deletion of exon 18 in infantile onset glycogen storage disease type II (GSD II). Hum Mol Genet 1994;3:1081–1087. [28] Nicolino M, Puech JP, Letourneur F, Fardeau M, Kahn A, Poenaru L. Glycogen storage disease type II (acid maltase deficiency): identification of a novel small deletion (delCC482 1 2483) in French patients. Biochem Byophys Res Commun 1997;235:138–141. [29] Kroos MA, van der Krans M, van Diggelen OP, Kleijer WJ, Reuser AJ. Glycogen storage disease type II: frequency of three common

166

R. Fernandez-Hojas et al. / Neuromuscular Disorders 12 (2002) 159–166

mutation alleles and their associated clinical phenotypes studied in 121 patients. J Med Genet 1995;32:836–840. [30] Hermans MM, van Leenen D, Kroos MA, Reuser AJ. Mutation detection in glycogen storage disease type II by RT-PCR and automated sequencing. Biochem Biophys Res Commun 1997;241:414–418.

[31] Huie ML, Shanske AL, Kasper JS, Marion RW, Hirschhorn R. A large Alu-mediated deletion, identified by PCR, as the molecular basis for glycogen storage disease type II (GSDII). Hum Genet 1999;104:94– 98.