Biochemical and Biophysical Research Communications 361 (2007) 445–450

www.elsevier.com/locate/ybbrc

Null mutations and lethal congenital form of glycogen

storage disease type IV
Stefania Assereto a, Otto P. van Diggelen b, Luisa Diogo c, Eva Morava d,
Denise Cassandrini a, Isabel Carreira c, Willem-Pieter de Boode d, Jildau Dilling d,
Paula Garcia c, Margarida Henriques c, Olinda Rebelo c, Henk ter Laak d,
Carlo Minetti a, Claudio Bruno a,*
a
Muscular and Neurodegenerative Disease Unit, Department of Pediatrics, Istituto Giannina Gaslini,
University of Genova, Largo G. Gaslini 5, I-16147 Genova, Italy
b
Department of Clinical Genetics, Erasmus Medical Center, Rotterdam, The Netherlands
c
Unidade de Doencas Metabolicas, Hospital Pediatrico de Coimbra, Coimbra, Portugal
d
Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Received 6 July 2007

Available online 24 July 2007

Abstract

Glycogen branching enzyme deficiency (glycogen storage disease type IV, GSD-IV) is a rare autosomal recessive disorder of the gly-
cogen synthesis with high mortality. Two female newborns showed severe hypotonia at birth and both died of cardiorespiratory failure,
at 4 and 12 weeks, respectively. In both patients, muscle biopsies showed deposits of PAS-positive diastase-resistant material and bio-
chemical analysis in cultured fibroblasts showed markedly reduced glycogen branching enzyme activity. Direct sequencing of GBE1 gene
revealed that patient 1 was homozygous for a novel c.691 + 5 g > c in intron 5 (IVS5 + 5 g > c). RT-PCR analysis of GBE1 transcripts
from fibroblasts cDNA showed that this mutation produce aberrant splicing. Patient 2 was homozygous for a novel c.1643G > A muta-
tion leading to a stop at codon 548 in exon 13 (p.W548X). These data underscore that in GSD-IV a severe phenotype correlates with null
mutations, and indicate that RNA analysis is necessary to characterize functional consequences of intronic mutations.
 2007 Published by Elsevier Inc.

Keywords: Glycogen storage disease type IV; Glycogen branching enzyme deficiency; GBE1 gene; Polyglucosan body

Glycogen storage disease type IV (GSD-IV; MIM#
232500) is a rare autosomal recessive disorder due to defi-
ciency of the enzyme which catalyzes the last step in glyco-
gen biosynthesis, the glycogen branching enzyme (GBE;
EC 2.4.1.18) [1,2]. Its deficiency produces accumulation
of abnormal glycogen consisting of long chains of glucose
units infrequently branched, resembling starch, and known
as polyglucosan bodies (PBs). PBs, which stain strongly
with periodic acid-Schiff, and are partially resistant to dia-
stase digestion due to their dense packing, accumulate in
almost all tissues, but to a different degree [1].
*
Corresponding author. Fax: +39 010 3538265.
E-mail address: claudiobruno@ospedale-gaslini.ge.it (C. Bruno).
The clinical presentation is variable and involves either
the liver, leading to cirrhosis and death in early childhood,
or the neuromuscular system, in its three variants, congen-
ital, juvenile and adult [1]. The diagnosis is confirmed by
the determination of the activity of GBE in affected tissues
[3]. The human GBE1 gene, which has been cloned,
sequenced and localized on chromosome 3p14, codify for
a 702 amino acids protein [4].
Different mutations in the GBE1 gene have been identi-
fied in all different GSD-IV phenotypes [5–14], and in par-
ticular, 13 infants with the fatal congenital form have been
characterized genetically in the last few years [5,8–12].
Here, we report two additional infants with the lethal
congenital neuromuscular form of glycogen branching

0006-291X/$ - see front matter  2007 Published by Elsevier Inc.

doi:10.1016/j.bbrc.2007.07.074
446 S. Assereto et al. / Biochemical and Biophysical Research Communications 361 (2007) 445–450

enzyme deficiency, who had novel mutations in the GBE1
gene.
Patient reports
Patient 1
This infant girl was the second child of healthy non-con-
sanguineous parents. She was born at 34 weeks of gestation
after a pregnancy complicated by polyhydramnios and by
reduced fetal movements. At birth, she presented with
severe hypotonia, hyporeflexia, and had to be intubated
and mechanically ventilated. In addition, she had equino-
varus feet with flexion contractures. There was no organo-
megaly. Electrocardiogram and 2D echocardiogram did
not show any cardiac dysfunctions, and cerebral ultra-
sound was normal. Extensive laboratory investigations
did not reveal any abnormalities. Karyotype, screening
for metabolic diseases, quantitative serum amino acids
and urine organic acids were all normal. Muscle biopsy
showed basophilic intracytoplasmic inclusions in several
fibers deposits (Fig. 1A), and PAS-positive diastase-resis-
tant material, compatible with glycogen storage disease.
The baby died at 4 weeks of age due to cardiorespiratory
failure. At autopsy, PAS-positive, diastase-resistant depos-
its were present in almost all tissues (Fig. 1C–F).
Patient 2
This infant girl was the third child of healthy non-con-
sanguineous parents. The first two pregnancies resulted in
healthy siblings. The pregnancy was complicated with
decreased fetal movements and polyhydramnios. She was
born at 34 weeks of gestation with a birth weight of
1840 g by caesarean section due to fetal bradycardia.
Apgar scores were 1, 3, and 7 at 1, 5, and 10 min on full
cardiorespiratory resuscitation, including adrenalin treat-
ment for severe bradycardia. The patient was immediately

Fig. 1. Hematoxylin and eosin (H&E) staining in muscle biopsies of patient 1 (A) and patient 2 (B). Basophilic intracytoplasmic inclusions in several
fibers. PAS-diastase in liver (C), muscle (D), myocardium (E), and brain (F) of patient 1. PAS-positive and diastase-resistant pale eosinophilic inclusions
(arrows). Original magnification, (A) and (B), 40·; (C) and (E), 400·; (D) and (F), 1000·.
 S. Assereto et al. / Biochemical and Biophysical Research Communications 361 (2007) 445–450 447

intubated due to the absence of spontaneous respiratory chemiluminescence detection system (Amersham Pharmacia Biotech). For
activity. The patient had no liver function anomalies, and control protein was used a mouse anti-b-actin monoclonal antibody at a
1:5000 dilution.
no cardiomyopathy was present. The EMG showed no
characteristic alterations or abnormalities comparable to
Werdnig-Hoffman syndrome. Muscle biopsy showed baso- Results
philic intracytoplasmic inclusions in several fibers deposits
(Fig. 1B), and accumulation of periodic acid Schiff-posi- Branching enzyme activity
tive, diastase-resistant storage material. The patient died
at the age of 12 weeks due to cardiorespiratory insuffi- In cultured skin fibroblasts of both patients, GBE activ-
ciency. Autopsy was not performed. ity was less than 5% of normal activity.
Blood, fibroblasts, muscle, and autoptic samples were
collected after fulfilments of our institutional guidelines Molecular analysis
for genetic analysis.
In patient 1, direct sequencing of the promoter region,
Methods the entire coding region sequence, and exon–intron bound-
aries of the GBE1 gene identified an intronic homozygous
Cell culture. Primary cultures of fibroblasts from skin biopsy of g > c transversion 5-bp downstream of exon 5
patients were grown in RPMI 1640 medium containing 20% FCS, 100 U/ (c.691 + 5 g > c, IVS5 + 5 g > c) (Fig. 2A). Both parents
ml penicillin, and 0.1 mg/ml streptomycin. Cells were maintained in an resulted to be heterozygotes for the same mutation.
humidified atmosphere of 5% CO2 at 37 C.
By RT-PCR of exon 3–7 (fragment C), we obtained
Branching enzyme activity. In both patients, GBE activity was assayed
in cultured fibroblasts, as previously described [9]. PCR products with size of 382 and 291-bp, being the
Molecular analysis. Genomic DNA and total RNA were extracted
from the patients’ fibroblasts according to standard protocols or using a
ToTALLY RNA Total RNA Isolation Kit (Ambion, Inc., Austin,
USA), respectively.
For mutation analysis at the genomic DNA level, direct sequencing of
the GBE1 gene was performed after amplification of all exons and flanking
intronic sequences using primer sets and conditions as previously descri-
bed [9].
In addition, mRNA was amplified by RT-PCR using the Superscript
First strand System (Invitrogen Life Technologies, Carlsbad, CA), and
subjected to agarose gel electrophoresis as previously described. The
GBE1 cDNA was amplified in six overlapping fragments (A–F) using the
primers listed in Supplementary Table 1.
The relevant fragments were subcloned using the TOPO TA Cloning
Kit (Invitrogen Life Technologies, Carlsbad, CA) and directly sequenced
with the ABI PRISM 3100 Genetic analyzer (Applied Biosystems, Foster
City, CA).
Sequences of analyzed fragments were compared with the GBE1
cDNA sequence where the A of the ATG translation initiation codon
represents nucleotide +1. The initiation codon is numbered as codon +1.
Nomenclature of mutations follows the recommendations of the
Nomenclature Working Group [15].
To confirm identified mutations and assess parental origin, specific
primer extension techniques were employed using minisequencing analysis
(ABI Prism SNaPshot multiplex kit) [16].
Western blot analysis. Fibroblasts cells of patients and controls were
lysed in 100 ll of RIPA lysis buffer (10 mM Tris–HCl pH 8, 140 mM
NaCl, 1% Triton X-100, 1% Na-desoxycholate, 0.1% SDS, 1 mg/ml
PMSF, 5 ll/ml Protease Inhibitor Cocktail, 1 mM Na3VO4, 1 mM NaF,
and 1 mM EDTA), by sonication and centrifuged at 10,000 rpm for
15 min at 4 C. Supernatants were collected and proteins concentration
was determined by using Bradford protein assay method. Equal amount of
proteins (40 lg) were loaded into each lane and resolved in a 10% SDS–
polyacrylamide gel.
The proteins were blotted onto nitrocellulose membrane (Immobilon
PVDF Millipore).
The membrane was blocked in 5% Non-Fat Dry Milk powder in PBS-
tween 0.1% (PBSt). A Goat anti-rabbit liver GBE polyclonal antibody
(generous gift from Dr. JR Mickelson, University of Minnesota, St. Paul, Fig. 2. (A) Electropherogram of the GBE1 genomic DNA of a normal
MN), diluted 1/400, was used as primary antibody. The secondary anti- control (NC) and of patient 1 (P1), encompassing the exon/intron 5
body was an anti-goat Ig biotinylated from donkey at a 1:750 dilution, and boundary, showing an homozygous mutation (arrow). (B) Electrophero-
an incubation with streptavidin–horseradish peroxidase conjugate (1:6000 gram of the GBE1 genomic DNA of a normal control (NC) and of patient
dilution) was performed [17]. The signal was developed by using the ECL 2 (P2), encompassing exon 13, showing an homozygous mutation (arrow).
448 S. Assereto et al. / Biochemical and Biophysical Research Communications 361 (2007) 445–450

expected size of the wild-type fragment of 518-bp. These Discussion

PCR products have been subcloned and direct sequenced.
Sequencing comparison revealed alternatively spliced prod- GSD-IV is a rare recessive disorder involving the liver or
ucts: a transcript that lacks exon 5, and a transcript lacking the neuromuscular system with different degree of severity
exon 5 plus 6. Of the two mRNA species, the one with skip- depending on the age of onset [1,2].
ping of exon 5 was more prevalent (21 clones against 10 A congenital phenotype can be characterized by severe
clones of the other transcript) (Fig. 3). All other fragments perinatal disorder presenting as fetal akinesia deformation
showed no differences in sequence from that of a normal sequence (FADS) with multiple congenital contractures,
control.
Direct sequencing of patient 2 DNA revealed homozy-
gosity for a G-to-A transition in exon 13 (c.1643G > A), NC1 NC2 P1 P2
leading to a premature stop (p.W548X) (Fig. 2B). By
RT-PCR, mRNA was normally amplified and the muta-
tion was confirmed at cDNA level. Both parents of patient
2 were shown to be carriers of the mutation.
80 kD- GBE
Western blot analysis

To explore the effect on the protein of the identified

mutations in the patients, we performed Western blot
analysis for glycogen branching enzyme (GBE) from
human fibroblasts of patients and normal controls.
Immunoblot analysis showed that the normal
80.4 kDa protein was absent in both patients, whereas
it was present in the wild-type fibroblasts lysates
(Fig. 4).
β-actin
Fig. 4. Immunoblot analysis for glycogen branching enzyme (GBE) on
total lysates of human fibroblasts. NC1 and NC2, normal controls; P1,
patient 1; P2, patient 2. In patients 1 and 2, GBE protein levels were
completely absent. Equal loading control was assessed with an antibody
against b-actin (42 kDa).

Fig. 3. Exon skipping caused by the c.691 + 5 g > c (IVS5 + 5 g > c). Fragment C of cDNA from patient 1 (P1) and normal control (NC) fibroblasts were
amplified and analyzed by agarose gel electrophoresis. The PCR products were then subcloned and direct sequenced. The normal sequence for fragment C
of 518-bp is illustrated in (A). Patient 1 showed different bands: (i) a 382-bp band indicative of exon 5 skipping leading to premature termination 7 codon
downstream (B) and (ii) a 291-bp band indicative of exon 5 plus 6 skipping leading to premature termination 6 codon downstream (C).
 S. Assereto et al. / Biochemical and Biophysical Research Communications 361 (2007) 445–450
hydrops fetalis, and perinatal death [17–20], or by severe
myopathy, often simulating Werding-Hoffman disease,
and inconsistently associated with cardiopathy [21,22].
Liver involvement usually includes failure to thrive, hepa-
tosplenomegaly, and progressive liver cirrhosis leading to
death in early childhood [23,24]. The juvenile phenotype
is characterized by myopathy or by cardiopathy, while in
adulthood the defect is associated with central and periph-
eral nervous system dysfunction (adult polyglucosan body
disease, APBD) [25–29].
Mutations in the GBE1 gene have been identified in all
the different clinical variants, and the same mutation have
been identified in unrelated patients with different clinical
presentations. These data suggest to consider GSD-IV as
a clinical continuum, with different degrees of involvement
of each organ system, rather than splitting the disease in
separate clinical variants, and provide evidence that factors
other than the single variants in the GBE1 gene might influ-
ence clinical presentation in the patients.
To date, in humans, 24 mutations have been described
in the GBE1 gene, distributed throughout the gene [5–14];
Human Gene Mutation Database (http://www.hgmd.cf.a-
c.uk) accessed June 2007] (see Supplementary Table 2).
All cases characterized by perinatal death or by fatal infan-
tile hypotonia have been associated with almost complete
absence of GBE activity and with severe mutations in the
GBE1 gene [5,8–12]. Reduced enzyme activity and mild
or heterozygous GBE1 mutations result in mild congenital
hypotonia and in APBD [6,7,13,14].
We report two unrelated patients with the lethal congen-
ital form of GBE deficiency caused by two homozygous
null mutations, c.691 + 5 g > c, and c.1643G > A, respec-
tively. Both mutations have not been previously described
and were not identified in 200 chromosomes from normal
controls.
The intronic base alteration, c.691 + 5 g > c, affect a
conserved nucleotide at the donor splice site consensus
sequence (gt[a/g]agt), with a maximal predicted splicing
score of 0.99 [http://www.fruitfly.org/seq_tools/splice.
html].
To determine the potential pathogenicity of the
c.691 + 5 g > c mutation, we analyzed GBE transcripts
from the patient’s fibroblasts, which revealed skipping of
exon 5 or skipping of both exons 5 and 6. Skipping of exon
5 (136-bp) leads to a frameshift and premature termination
7 codons downstream. Skipping of both exons 5 and 6
(227-bp) leads to a frameshift and premature termination
6 codons downstream.
The nonsense mutation, c.1643G > A, cause premature
termination of translation, generating truncated 548-amino
acid peptide.
Both of the reported mutations are therefore predicted
to produce truncated protein. The lack of the proteins
showed by Western blot analysis suggest that these proteins
are likely to be degraded.
In conclusion, the identified mutations enlarge the list of
GBE1 mutations, and confirm that in GSD-IV a severe
449
phenotype correlates with null mutations. We underline
the importance to evaluate intronic variants as putative
mutations, especially in those cases where the clinical and
biochemical data are strongly diagnostic. Thus, the cDNA
analysis can help to identify unexpected splicing aberra-
tions due to intronic mutation, allowing the genetic
diagnosis.
Finally, because of the high lethality associated with this
disorder, molecular characterization of these patients is of
particular value for genetic counselling of patients and
their families.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bbrc.
2007.07.074.
References
[1] S. DiMauro, A.P. Hays, S. Tsujino, Nonlysosomal glycogenoses, in:
A.G. Engel, C. Franzini-Armstrong (Eds.), Myology, McGraw-Hill,
New York, 2004, pp. 1535–1558.
[2] S.W. Moses, R. Parvari, The variable presentations of glycogen
storage disease type IV: a review of clinical, enzymatic and molecular
studies, Curr. Mol. Med. 2 (2002) 177–188.
[3] B.I. Brown, D.H. Brown, Branching enzyme activity of cultured
amniocytes and chorionic villi: prenatal testing for type IV glycogen
storage disease, Am. J. Hum. Genet. 44 (1989) 378–381.
[4] V.J. Thon, M. Khalil, J.F. Cannon, Isolation of human glycogen
branching enzyme cDNAs by screening complementation in yeast, J.
Biol. Chem. 268 (1993) 7509–7513.
[5] Y. Bao, P. Kishnani, J.Y. Wu, Y.T. Chen, Hepatic and neuromus-
cular forms of glycogen storage disease type IV caused by mutations
in the same glycogen-branching enzyme gene, J. Clin. Invest. 97
(1996) 941–948.
[6] A. Lossos, Z. Meiner, V. Barash, D. Soffer, I. Schlesinger, O.
Abramsky, Z. Argov, S. Shpitzen, V. Meiner, Adult polyglucosan
body disease in Ashkenazi Jewish patients carrying the Tyr329Ser
mutation in the glycogen-branching enzyme gene, Ann. Neurol. 44
(1998) 867–872.
[7] F. Ziemssen, E. Sindern, J.M. Schroder, Y.S. Shin, J. Zange, M.W.
Kilimann, J.P. Malin, M. Vorgerd, Novel missense mutations in the
glycogen-branching enzyme gene in adult polyglucosan body disease,
Ann. Neurol. 47 (2000) 536–540.
[8] M. Nambu, K. Kawabe, T. Fukuda, T.B. Okuno, S. Ohta, I. Nonaka,
H. Sugie, I. Nishino, A neonatal form of glycogen storage disease
type IV, Neurology 61 (2003) 392–394.
[9] C. Bruno, O.P. van Diggelen, D. Cassandrini, M. Gimpelev, B.
Giuffre, M.A. Donati, P. Introvini, A. Alegria, S. Assereto, L.
Morandi, M. Mora, E. Tonoli, S. Mascelli, M. Traverso, E. Pasquini,
M. Bado, L. Vilarinho, G. van Noort, F. Mosca, S. DiMauro, F.
Zara, C. Minetti, Clinical and genetic heterogeneity of branching
enzyme deficiency (glycogenosis type IV), Neurology 63 (2004)
1053–1058.
[10] S.K. Tay, H.O. Akman, W.K. Chung, M.G. Pike, F. Muntoni,
A.P. Hays, S. Shanske, S.J. Valberg, J.R. Mickelson, K. Tanji, S.
DiMauro, Fatal infantile neuromuscular presentation of glycogen
storage disease type IV, Neuromuscul. Disord. 14 (2004)
253–260.
[11] A.R. Janecke, S. Dertinger, U.P. Ketelsen, L. Bereuter, B. Simma, T.
Muller, W. Vogel, F.A. Offner, Neonatal type IV glycogen storage
disease associated with ‘‘null’’ mutations in glycogen branching
enzyme 1, J. Pediatr. 145 (2004) 705–709.
450 S. Assereto et al. / Biochemical and Biophysical Research Communications 361 (2007) 445–450
[12] A. L’hermine-Coulomb, F. Beuzen, R. Bouvier, M.O. Rolland, R.
Froissart, F. Menez, F. Audibert, P. Labrune, Fetal type IV glycogen
storage disease: clinical, enzymatic, and genetic data of a pure
muscular form with variable and early antenatal manifestations in the
same family, Am. J. Med. Genet. A 139 (2005) 118–122.
[13] E.E. Ubogu, S.T. Hong, H.O. Akman, S. DiMauro, B. Katirji, D.C.
Preston, B.E. Shapiro, Adult polyglucosan body disease: a case report
of a manifesting heterozygote, Muscle Nerve 32 (2005) 675–681.
[14] T.A. Burrow, R.J. Hopkin, K.E. Bove, L. Miles, B.L. Wong, A.
Choudhary, D. Bali, S.C. Li, Y.T. Chen, Non-lethal congenital
hypotonia due to glycogen storage disease type IV, Am. J. Med.
Genet. A 140 (2006) 878–882.
[15] J.T. den Dunnen, M.H. Paalman, Standardizing mutation nomen-
clature: why bother? Hum. Mutat. 22 (2003) 181–182.
[16] D. Cassandrini, M.G. Calevo, A. Tessa, G. Manfredi, F. Fattori,
M.C. Meschini, R. Carrozzo, E. Tonoli, M. Pedemonte, C. Minetti,
F. Zara, F.M. Santorelli, C. Bruno, A new method for analysis of
mitochondrial DNA point mutations and assess levels of hetero-
plasmy, Biochem. Biophys. Res. Commun. 342 (2006) 387–393.
[17] G. van Noort, W. Straks, O.P. van Diggelen, R.C. Hennekam, A
congenital variant of glycogenosis type IV, Pediatr. Pathol. 13 (1993)
685–698.
[18] E. Uro-Coste, M.C. Lelong-Tissier, I. Maire, C. Ceuterick, F.
Chausseray, M.B. Delisle, Congenital variant of type IV glycogenosis.
Anatomoclinical report of a case, Ann. Pathol. 16 (1996) 449–452.
[19] A. Alegria, E. Martins, M. Dias, A. Cunha, M.L. Cardoso, I. Maire,
Glycogen storage disease type IV presenting as hydrops fetalis, J.
Inherit. Metab. Dis. 22 (1999) 330–332.
[20] P.M. Cox, L.A. Brueton, K.W. Murphy, V.C. Worthington, P.
Bjelogrlic, E.J. Lazda, N.J. Sabire, C.A. Sewry, Early-onset fetal
hydrops and muscle degeneration in siblings due to a novel variant of
type IV glycogenosis, Am. J. Med. Genet. 86 (1999) 187–193.
[21] H. Zellweger, S. Mueller, V. Ionasescu, S.S. Schochet, W.F. McCor-
mick, Glycogenosis IV. A new cause of infantile hypotonia, J. Pediatr.
80 (1972) 842–844.
[22] T.T. Tang, A.D. Segura, Y.T. Chen, L.M. Ricci, R.A. Franciosi,
M.L. Splaingard, M.S. Lubinsky, Neonatal hypotonia and cardio-
myopathy secondary to type IV glycogenosis, Acta Neuropathol.
(Berl.) 87 (1994) 531–536.
[23] D.H. Andersen, Familial cirrhosis of the liver with storage of
abnormal glycogen, Lab. Invest. 5 (1956) 11–20.
[24] H.L. Greene, B.I. Brown, D.T. McClenahan, R.M. Agostini Jr., S.R.
Taylor, A new variant of type IV glycogenosis: deficiency of
branching enzyme activity without apparent progressive liver disease,
Hepatology 8 (1988) 302–306.
[25] S. Servidei, R.E. Riepe, C. Langston, L.Y. Tani, J.T. Bricker, N.
Crisp-Lindgren, H. Travers, D. Armstrong, S. DiMauro, Severe
cardiopathy in branching enzyme deficiency, J. Pediatr. 111 (1987)
51–56.
[26] E. Reusche, F. Aksu, H.H. Goebel, Y.S. Shin, T. Yokota, H.
Reichmann, A mild juvenile variant of type IV glycogenosis, Brain
Dev. 14 (1992) 36–43.
[27] H.H. Goebel, Y.S. Shin, F. Gullotta, T. Yokota, J. Alroy, T. Voit, P.
Haller, A. Schulz, Adult polyglucosan body myopathy, J. Neuropa-
thol. Exp. Neurol. 51 (1992) 24–35.
[28] A. Lossos, V. Barash, D. Soffer, Z. Argov, M. Gomori, Z. Ben-
Nariah, O. Abramsky, I. Steiner, Hereditary branching enzyme
dysfunction in adult polyglucosan body disease: a possible
metabolic cause in two patients, Ann. Neurol. 30 (1991)
655–662.
[29] C. Bruno, S. Servidei, S. Shanske, G. Karpati, S. Carpenter, D.
McKee, R.J. Barhon, M. Hirano, Z. Rifai, S. DiMauro, Glycogen
branching enzyme deficiency in adult polyglucosan body disease,
Ann. Neurol. 33 (1993) 88–93.