- Email: [email protected]
α-Dystroglycan does not play a major pathogenic role in autosomal recessive hereditary inclusion-body myopathy
Neuromuscular Disorders 15 (2005) 177–184 www.elsevier.com/locate/nmd
a-Dystroglycan does not play a major pathogenic role in autosomal recessive hereditary inclusion-body myopathy Aldobrando Broccolinia, Carla Gliubizzia, Ernesto Pavonic, Teresa Gidaroa, Roberta Morosettia, Francesca Sciandrac, Bruno Giardinac, Pietro Tonalia, Enzo Riccia,b, Andrea Brancaccioc,1, Massimiliano Mirabellaa,* a
Department of Neuroscience, Catholic University, L.go A. Gemelli 8, 00168 Rome, Italy b U.I.L.D.M.-Rome Section, Rome, Italy c CNR—Istituto di Chimica del Riconoscimento Molecolare c/o Istituto di Biochimica e Biochimica Clinica, Catholic University, L.go F. Vito 1, 00168 Rome, Italy Received 19 August 2004; received in revised form 27 September 2004; accepted 4 October 2004
Abstract Mutations of the GNE gene are responsible for autosomal recessive hereditary inclusion-body myopathy (HIBM). In this study we searched for the presence of any significant abnormality of a-dystroglycan (a-DG), a highly glycosylated component of the dystrophinglycoprotein complex, in 5 HIBM patients which were previously clinically and genetically characterized. Immunocytochemical and immunoblot analysis showed that a-DG extracted from muscle biopsies was normally expressed and displayed its typical molecular mass. Immunoblot analysis on the wheat germ lectin-enriched glycoprotein fraction of muscles and primary myotubes showed a reduced amount of a-DG in 4 out of 5 HIBM patients, compared to normal and other diseased muscles. However, such altered lectin-binding behaviour, possibly reflecting a partial hyposialylation of a-DG, did not affect the laminin binding properties of a-DG. Therefore, the subtle changes within the a-DG glycosylation pattern, detected in HIBM muscles, likely do not play a key pathogenic role in this disorder. q 2004 Elsevier B.V. All rights reserved. Keywords: Inclusion body myopathy; HIBM; GNE; Dystrophin-glycoprotein complex; Dystroglycan
1. Introduction Autosomal recessive (AR) hereditary inclusion-body myopathy (HIBM; MIM# 600737), originally described in Persian–Jewish families, is a neuromuscular disorder characterized by onset in the early adult life with weakness and atrophy of distal lower limb muscles and relative sparing of the quadriceps [1]. HIBM is associated with mutations in the UDP-N-acetylglucosamine 2-epimerase/ N-acetylmannosamine kinase gene (GNE; MIM# 603824) on chromosome 9p12-13 [2,3]. GNE mutations have also been found in Japanese patients previously diagnosed as
* Corresponding author. Tel.: C39 6 3015 4303; fax: C39 6 3550 1909. E-mail addresses: [email protected] (A. Brancaccio), [email protected] (M. Mirabella). 1 Tel.: C39 6 305 7612; fax: C39 6 305 3598. 0960-8966/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.nmd.2004.10.001
having distal myopathy with rimmed vacuoles (DMRV; MIM# 605820), thus confirming that HIBM and DMRV represent indeed the same disease [4–7]. UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (UDP-GlcNAc epimerase/ManNAc kinase) is a bifunctional enzyme, composed of two distinct epimerase and kinase domains, that is expressed in different tissues and has a critical role in sialic acid biosynthesis [8,9]. Sialic acid is normally present on the distal ends of N- and O-glycans and is involved in many biological functions including cellular adhesion, stabilization of glycoproteins structure and signal transduction [10]. Although previous studies have shown that HIBM-associated GNE mutations result in a reduced enzymatic activity [6,11,12], the pathogenic mechanism, triggered by a possible disturbance of sialic acid metabolism and leading to muscle fiber degeneration, remains to be elucidated.
178
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
In recent years, several lines of evidence have pointed out a key role of abnormal carbohydrate metabolism and protein glycosylation in the pathogenesis of different neuromuscular disorders [13–15]. For example, Fukuyama congenital muscular dystrophy (FCMD) and muscle-eye-brain disease (MEB) are caused by mutations in genes encoding for glycosyltransferases that result in a reduced glycosylation of a-dystroglycan (a-DG), a central component of the dystrophin-glycoprotein complex (DGC) [16]. The DGC is a multimeric transmembrane complex, that provides a tight connection between the extracellular matrix and the cytoskeleton and has a crucial role in the stability of muscle cell membrane during cycles of contraction and relaxation. DG is generated from a single gene (DAG1) and is subsequently cleaved into two subunits: transmembrane b-DG and peripheral a-DG [17]. a-DG undergoes extensive post-translational N-linked and O-linked glycosylation that is thought to be critical for its interaction with proteins of the extracellular matrix such as laminin, agrin, perlecan and neurexin [18]. Abnormal glycosylation of a-DG, as observed in FCMD and MEB, results in reduced ligand binding and this is likely responsible for the muscle and brain pathology observed in these disorders [16,19]. Whether a similar mechanism plays a role also in HIBM pathology has not been to date unequivocally elucidated [11,20,21]. In the present study we investigated a-DG expression in the muscle of 5 patients with HIBM associated with mutations of the GNE gene.
2. Materials and methods 2.1. Patients Five patients from 4 unrelated families were diagnosed as having HIBM based on clinical findings, muscle pathology and genetic study [24,25]. Clinical criteria
included (i) onset in the second-third decade of life, (ii) initial weakness and atrophy of distal lower limb muscles with distal-proximal progression and sparing of the quadriceps muscles, (iii) later involvement of upper limb muscles and (iv) normal or slightly increased creatin-kinase blood levels [1]. All muscle biopsies showed morphological abnormalities indicative of HIBM [22,23]. For patient 1 different specimens were obtained from the deltoid muscle, that was clinically affected (strength grade 3/5, Medical Research Council scale), and the quadriceps, that had instead normal strength. Mutational analysis of the GNE gene in our patients has been described in details in previous studies [24,25] and is summarized in Table 1. Disease control muscles were: sporadic inclusion-body myositis (sIBM, nZ2), facio-scapulo-humeral muscular dystrophy (FSHD, nZ2), late-onset acid maltase deficiency (nZ2). Muscle biopsies from 6 patients proven to be free of any neuromuscular disorder, after all diagnostic tests were performed, were used as normal controls. All muscle samples used in this study were obtained for diagnostic purposes with informed consent. 2.2. Primary muscle cultures Human aneural primary muscle cultures were obtained from the biopsies of patients 2–4 using the explantation reexplantation method, as previously described [26]. The fusion of confluent mononucleated myoblasts into multinucleated myotubes was obtained using a culturing medium containing 5% fetal bovine serum (FBS, Cambrex Bioscience, Baltimore, MD). Fusion index was expressed as number of myonuclei/number of total nuclei, visualized by Hoechst 33258 staining (Molecular Probes Inc., Eugene, OR), and varied between 0.8 and 0.9 for all the culture sets used in this study. Before any experimental procedure, myotubes from HIBM and control patients were maintained
Table 1 Summary of clinical and genetic data of HIBM patients Patient/ gender
Age at onset (years)
Age at examination (years)
Clinical involvement
GNE mutation Nucleotide substitution (c.)a
Consequence (p.)b
Protein domain
p.N519S
Kinase
p.N519S
Kinase
p.P27S p.R246Q p.Q355_C357del/ p.I377fsX16c p.M171V p.M712T
Epimerase Epimerase Kinase
1/F
37
48
Moderate
Exon 9
2/M
25
55
Severe
Exon 9
3/F 4/F
35 21
42 49
Moderate Moderate
Exon 2 Exon 4 Intron 6
Homozygous c.1556AOG Homozygous c.1556AOG Homozygous c.79COT c.737GOAC c.1070C2dupT
5/F
23
38
Moderate
Exon 3 Exon 12
c.511AOGC c.2135TOC
a b c
Epimerase Kinase
Nucleotide position in GNE cDNA, Genbank accession# NM_005476 version 2; nucleotide numbering starts with the A of the start codon. Protein id. NP_005467.1. Predicted protein products of the two alternatively spliced GNE mRNAs.
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
for 48 h in culturing medium containing only serum replacement factors (Ultraculture, Cambrex Bioscience) to avoid the uptake of sialic acid and glycoproteins from the FBS. 2.3. Immunohistochemistry on muscle biopsies This was performed on 8 mm-thick unfixed cryostat sections. The following primary antibodies were used: (i) monoclonal anti-a-DG (clone VIA4-1, Upstate Biotechnology, Waltham, MA) recognizing only glycosylated a-DG, at the concentration of 3 mg/ml; (ii) monoclonal anti-b-DG (clone 43DAG1/8D5, YLEM, Newcastle upon Tyne, UK), diluted 1/25. The reaction with all primary antibodies was performed for 16–18 h at 4 8C in a humid chamber, followed by incubation with the appropriate horseradish peroxidase (HRP)-conjugated secondary antiserum (Dakocytomation, Glostrup, Denmark). 2.4. Immunocytochemistry on primary muscle cultures This was performed according to our previously established protocol [27] with minor modifications. In brief, myotubes were grown on glass coverslips placed on the bottom of Petri dishes. Coverslips were then collected and myotubes were washed in PBS and fixed with 2% paraformaldehyde in phosphate buffer for 15 min. at room temperature. For b-DG immunostaining only, myotubes were post-fixed with PBS containing 1% saponin and 10% normal goat serum. Incubation with either the anti-a-DG (clone VIA4-1, at the concentration of 3 mg/ml) or the antib-DG (clone 43DAG1/8D5, diluted 1/10) antibodies was performed for 16–18 h at 4 8C in a humid chamber. Detection of immunocomplexes was performed using a Texas Redw-conjugated anti-mouse antibody (Jackson Immunoresearch Laboratories, West Grove, PA). Cultures were photographed using the same exposure time to allow a correct comparison of staining intensities. 2.5. Western blot analysis on muscle biopsies and primary muscle cultures Diagnostic muscle biopsies and primary muscle cultures from HIBM and control patients were homogenized in Trisbuffered saline (TBS) containing 1% Triton X-100 and protease inhibitors; supernatants were collected upon centrifugation as previously described [28]. Glycoprotein enrichment was performed on both muscle biopsies and muscle culture extracts [16]. Briefly, supernatants were incubated with Wheat Germ lectin (WGL) Sepharosee 6MB (Amersham Bioscience, Piscataway, NJ) for 16 h at 4 8C and then centrifuged at 3000 rpm for 15 min. After supernatant removal, pellets formed from the beads were washed three times and proteins were then eluted in TBS containing 250 mM N-acetylglucosamine. Equal amounts of proteins of either total extracts or glycoprotein-enriched
179
samples were separated by SDS-PAGE and blotted onto a nitrocellulose membrane (Schleicher and Schuell, Relliehausen, Germany). Blots were incubated overnight using one of the following antibodies: (i) monoclonal anti-a-DG (clone VIA4-1, Upstate Biotechnology) at the concentration of 1.5 mg/ml, (ii) monoclonal anti-b-DG (clone 43DAG1/ 8D5, YLEM) diluted 1/50. After incubation with the appropriate HRP-conjugated secondary antibody (Amersham Bioscience), blots were developed using the enhanced chemiluminescence technique (Amersham Bioscience). Densitometry on autoradiographic films was carried out using the TotalLab 2.01 software (Nonlinear Dynamics Ltd, Newcastle upon Tyne, UK). The mean densitometric value of a-DG signal in control samples was arbitrarily set as 1 and all the measured values of patients’ samples were then expressed as a percentage of it. 2.6. Laminin overlay assay Experiments were performed according to previously described protocols [16,29]. In brief, 10 mg of proteins of total extracts were separated by SDS-PAGE and blotted onto a nitrocellulose membrane. Membranes were then blocked in 5% non-fat dry milk in laminin binding buffer (10 mM triethanolamine, 140 mM MgCl2, 1 mM CaCl2, pH 7.6) and then incubated with mouse Engelbreth–Holm– Swarm laminin (Invitrogen, Carlsbad, CA) for 16–18 h at 4 8C. Membranes were then washed, incubated with a rabbit polyclonal anti-laminin antibody (SIGMA, St Louis, MO) diluted 1/1000, followed by the appropriate HRP-conjugated secondary antibody (Chemicon, Temecula, CA) and then developed by enhanced chemiluminescence.
3. Results 3.1. Muscle biopsies In all HIBM biopsies a-DG was normally detected along the sarcolemma of muscle fibers. No difference in staining intensity was observed between HIBM and control biopsies. A normal a-DG immunosignal was also detected in all vacuole-bearing HIBM muscle fibers. Characteristically, within these fibers, the a-DG immunosignal was also located along the rim of the cytoplasmic vacuoles (Fig. 1A). Immunocytochemistry of b-DG also showed a normal signal along the sarcolemma of fibers in all HIBMs compared to control biopsies (Fig. 1A). In agreement with these results, western blot analysis on total protein extracts showed a-DG as a broad band of 150–200 kDa that was present in similar amounts in all HIBM muscle biopsies compared to normal controls (Fig. 1B). The electrophoretic mobility indicates that a-DG is not hypoglycosylated in all HIBM muscles. In addition, in patient 1 no difference of a-DG expression was observed between the deltoid and the quadriceps muscles (Fig. 1B).
180
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
Fig. 1. (A) Peroxidase immunohistochemistry for a- and b-DG on muscle biopsies. A normal immunosignal for a- and b-DG was detected along the sarcolemma of muscle fibers in all HIBM patients. Arrow indicates a vaculated muscle fiber showing that the a-DG immunosignal was also located along the rim of the cytoplasmic vacuoles. Original magnification 100!. (B) Western blot analysis of a-DG on total protein extracts from muscle biopsies showing that a-DG was present in similar amount in all HIBM muscle compared to normal controls. In patient 1 no difference of a-DG immunosignal was observed between the deltoid and the quadriceps muscles. Data shown are representative of the 5 HIBM patients studied.
Muscle glycoproteins were then purified by WGLSepharosee 6MB from total extracts of HIBM and control biopsies and used for western blot analysis of a- and b-DG. Differently from what observed in total tissue extracts, in
glycoprotein samples obtained from HIBM patients 1, 2, 3 and 5 we detected lower amounts of a-DG, compared to normal and disease control muscles (Fig. 2). Densitometric analysis showed 21% in the deltoid and 20% in
Fig. 2. Western blot analysis of a- and b-DG on muscle glycoproteins showing lower amounts of a-DG in experimental samples of HIBM patients 1, 2, 3 and 5 compared to what observed in normal and disease control muscles. No difference of a-DG immunosignal was observed between samples obtained from the deltoid and the quadriceps muscles, respectively in patient 1. On the contrary, analysis of a-DG in glycoprotein extracts from HIBM patient 4 did not show any abnormality. b-DG was quantitatively reduced in samples from HIBM patients 1 and 3 whereas it was present in normal amounts in those from the other HIBM patients.
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
181
Fig. 3. Graph representation of a-DG quantity in glycoprotein extracts from muscle biopsies of normal controls, disease controls and HIBM patients, measured by densitometry.
the quadriceps from patient 1, 43% in patient 2, 44% in patient 3 and 42% in patient 5, respectively, compared to the a-DG mean value detected in normal muscles (Fig. 3). a-DG displayed a normal molecular mass and, similarly to what observed in total tissue extracts, no difference of a-DG immunosignal was found between samples from the deltoid and the quadriceps muscles of patient 1. It should be noted that in glycoprotein extract from HIBM patient 4 a-DG was detected in amount similar to control samples whereas b-DG was reduced only in samples from HIBM patients 1 and 3 (Fig. 2). Moreover, we performed a ligand overlay assay using laminin-1 on blots of HIBM total protein extracts. In muscle from HIBM patients the ligand-bindind activity of laminin was retained similarly to what observed in normal control muscles (Fig. 4).
Fig. 4. Laminin overlay assay showing a similar a-DG-laminin interaction in HIBM muscles compared to normal controls. Data shown are representative of the 5 HIBM patients studied.
previously suggested in other cell systems [30]. In agreement with the results of western blot studies on muscle biopsies, in glycoprotein samples from HIBM patients 2 and 3 a-DG appeared as a broad band of normal molecular weight but quantitatively reduced compared to control samples (49% in patient 2 and 20% in patient 3 compared to the mean normal value), whereas it was present in normal amount in HIBM patient 4 (Fig. 5B). Western blot analysis for b-DG showed a reduced amount of protein in HIBM patient 3 whereas it resulted normal in all the others (Fig. 5B).
4. Discussion 3.2. Primary muscle cultures Cultured myotubes obtained from the muscle biopsies of patients 2–4 and normal controls were used in this study. a-DG immunocytochemistry produced a signal that was mainly localized along the plasma membrane of cultured myotubes whereas immunolabeling of permeabilized myotubes with the anti-b-DG antibody resulted in a more diffuse staining. In agreement with the results of immunocytochemical studies performed on muscle biopsies, cultured myotubes from HIBM patients showed a normal immunoreactivity for both a- and b-DG (Fig. 5A). Western blot for a-DG in glycoproteins extracted from aneurally cultivated normal primary myotubes showed a broad band with an approximate molecular mass of 120–150 kDa, lower than what observed in mature muscle. This possibly reflects the fact that the post-translational glycosylation of a-DG is developmentally regulated, as
Despite the identification of the genetic defect responsible for AR HIBM, the pathogenic mechanisms, triggered by a possible impairment of sialic acid metabolism and leading to progressive muscle degeneration, are still elusive. In recent years, a growing body of evidence has pointed out the role of a-DG, a key component of the DGC, in the pathogenic scenario of various neuromuscular disorders [13–16]. In fact, although mutations in the DAG1 gene have never been identified to date, numerous reports have shown that an abnormal glycosylation of its protein product a-DG represents the common final pathogenic pathway of muscle disorders associated with mutations of known or putative glycosyltransferases [19]. A dysfunctional a-DG could result in disruption of the link between the extracellular matrix and the cytoskeleton, thus promoting muscle fiber degeneration [13–16]. Sialic acid is normally present on the distal ends of O-glycans that decorate a-DG although its precise role in a-DG physiology remains to be elucidated
182
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
Fig. 5. (A) Fluorescence immunocytochemistry for a- and b-DG on primary muscle cultures. A normal staining for a- and b-DG was observed in all HIBM cultured myotubes compared to normal controls. Original magnification 200!. (B) Western blot analysis on glycoprotein extracts showing reduced amounts of a-DG in muscle cultures of HIBM patients 2 and 3 and a normal amount in HIBM patient 4.
[19]. Nevertheless, these considerations make a-DG a potential candidate in the attempt to elucidate the molecular mechanisms underlying HIBM pathology. By immunohistochemistry, we found that, in all HIBM biopsies, a-DG was normally expressed along the sarcolemma of muscle fibers, including vacuolated
ones. Identical results were obtained using an antibody directed against b-DG. This is in agreement with the western blot study on the corresponding total protein extracts showing normal amounts of a-DG with a molecular mass similar to that of control muscle biopsies. This latter evidence suggests that a-DG is not hypoglycosylated in HIBM muscle, in contrast with a previous report indicating hypoglycosylation of a-DG as a possible downstream causative mechanism in HIBM pathophysiology [21]. Interestingly, both dystroglycans were detected along the rim of cytoplasmic vacuoles. The functional significance of this evidence is not known. However, it should be noted that the presence of several cytoskeletal proteins, including dystrophin and laminin, has been previously demonstrated on vacuolar boundaries in different vacuolar myopathies and a possible role in strengthening the vacuolar membranes has been proposed [31]. To date, it has not been elucidated whether mutations in the GNE gene produce hyposialylation of specific muscle glycoproteins and all the studies that have attempted to tackle this issue have provided differing results [11,12,20]. Abnormal sialylation of a-DG may occur in the muscle of 4 of our HIBM patients, as suggested by the reduced binding of a-DG to the Wheat Germ lectin (Figs. 2 and 3). In fact, such lectin is known to avidly bind glycopeptides displaying clusters of sialyl oligosaccarides, since removal of sialic acid terminal residues abolishes the interaction with glycoproteins [32]. In addition, it has been shown that the digestion of the skeletal muscle dystrophin-glycoprotein complex with V. Cholerae sialidase determines a loss of staining of a-DG by Wheat Germ lectin on western blot [33]. It is likely that, in our HIBM patients, a-DG O-linked glyconjugates are only partially devoid of sialic acid residues. Indeed, it has been previously demonstrated in L6 myotubes that the total enzymatic removal of sialic acid from glycoprotein extracts should result in a w10 kDa decrease of a-DG molecular mass [34] whereas in our patients we found that a-DG always had a normal electrophoretic mobility. This is likely to depend on some residual enzymatic activity of mutated UDP-GlcNAc epimerase/ManNAc kinase in HIBM patients, as previously demonstrated [11,12]. In glycoprotein extracts from muscle biopsies of patients 1 and 3 we also observed a reduction of b-DG. Since sialic acid is also present on the distal ends of N-glycans that decorate b-DG, it is therefore possible that, in these biopsies, a reduced presence of sialic acid moieties also results in a reduced binding of b-DG to the Wheat Germ lectin, similarly to what postulated for a-DG. Our data are in agreement with other studies showing a partial reduction in the content of sialic acid in HIBM muscles and a weaker binding of Wheat Germ lectin on cultured HIBM myotubes [11]. It should be pointed out that, in our patients, the observed a-DG abnormality is not connected with the impairment of a specific domain of the enzyme, as it was detected in patients harboring a homozygous mutation in either the epimerase domain (patient 3) or the kinase domain
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
(patients 1 and 2) and in a patient who was a compound heterozygous for GNE mutations affecting different domains (patient 5). The specific role of sialic acid caps on the distal end of aDG O-linked glycoconjugates has not been fully elucidated. In peripheral nerve it has been demonstrated that sialic acid competitively inhibits laminin binding to a-DG [35], whereas in skeletal muscle the enzymatic removal of sialic acid residues does not affect this interaction [33]. In our study, a-DG extracted from HIBM muscle biopsies displayed a strong laminin binding, thus ruling out a possible impairment of the functional link between the cytoskeleton and the extracellular matrix due to a partial hyposialylation of a-DG. Furthermore, a normal signal for a-DG was detected also in glycoprotein enriched sample from HIBM patient 4 and in HIBM patient 1 we found no difference of a-DG abundance in samples obtained from the deltoid and quadriceps muscles respectively, that were instead differently affected. Taken together these data would strongly argue against a pivotal role of a-DG in the molecular pathophysiology of HIBM muscle. To the best of the current knowledge, it appears that a partial hyposialylation of a-DG, as observed in 4 of our HIBM patients, would represent only a minor byproduct of a metabolic impairment that may instead crucially affect other subcellular compartments and may not exclusively be associated with the synthesis of sialic acid. In fact, it has been previously postulated that UDP-GlcNAc epimerase/ ManNAc kinase could be also involved in other, yet unknown, metabolic pathways possibly more important in HIBM pathophysiology [12]. Whether an hyposialylated a-DG may also play a role in the activation of abnormal cellular mechanisms remains to be determined and additional studies will be required to tackle this issue.
Acknowledgements Study supported in part by grants from The Italian Ministry of Health 2002 and 2003 to Massimiliano Mirabella. The grant Telethon GGP030332 to Andrea Brancaccio is gratefully acknowledged.
References [1] Argov Z, Yarom Y. Rimmed vacuole myopathy sparing the quadriceps. A unique disorder in Iranian Jewish. J Neurol Sci 1984; 64:33–43. [2] Eisenberg I, Avidan N, Potikha T, et al. The UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase gene is mutated in recessive hereditary inclusion body myopathy. Nat Genet 2001;29:83–7. [3] Eisenberg I, Grabov-Nardini G, Hochner M, et al. Mutation spectrum of GNE in hereditary inclusion body myopathy sparing the quadriceps. Hum Mutat 2003;21:99–105.
183
[4] Kayashima T, Matsuo H, Satoh A, et al. Nonaka myopathy is caused by mutations in the UDP-N-acetylglucosamine-2-epimerase/N-acetylmannosamine kinase gene (GNE). J Hum Genet 2002;47:77–9. [5] Tomimitsu H, Ishikawa K, Shimizu J, Ohkoshi N, Kanazawa I, Mizusawa H. Distal myopathy with rimmed vacuoles: novel mutations in the GNE gene. Neurology 2002;59:451–4. [6] Nishino I, Noguchi S, Muruyama K, et al. Distal myopathy with rimmed vacuoles is allelic to hereditary inclusion body myopathy. Neurology 2002;59:1689–93. [7] Arai A, Tanaka K, Ikeuchi T, et al. A novel mutation in the GNE gene and a linkage disequilibrium in Japanese pedigrees. Ann Neurol 2002; 52:516–9. [8] Hinderlich S, Sta¨sche R, Zeitler R, Reutter W. A bifunctional enzyme catalyses the first two steps in N-acetylneuraminic acid biosynthesis of rat liver. Purification and characterization of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase. J Biol Chem 1997; 272:24313–8. [9] Efferts K, Hinderlich S, Reutter W. Selective loss of either the epimerase or kinase activity of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase due to site-directed mutagenesis based on sequence alignments. J Biol Chem 1999;274:28771–8. [10] Keppler OT, Hinderlich S, Langner J, Schwartz-Albiez R, Reutter W, Pawlita M. UDP-GlcNAc 2-epimerase: a regulator of cell surface sialylation. Science 1999;284:1372–6. [11] Noguchi S, Keira Y, Murayama K, et al. Reduction of UDP-Nacetylglucosamine 2-epimerase/N-acetylmannosamine kinase activity and sialylation in distal myopathy with rimmed vacuoles. J Biol Chem 2004;279:11402–7. [12] Hinderlich S, Salama I, Eisenberg I, et al. The homozygous M712T mutation of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase results in reduced enzyme activities but not in altered overall cellular sialylation in hereditary inclusion body myopathy. Fed Eur Biochem Soc Lett 2004;566:105–9. [13] Grewal PK, Hewitt JE. Glycosylation defects: a new mechanism for muscular dystrophy? Hum Mol Genet 2003;12:R259–R64. [14] Martin PT, Freeze HH. Glycobiology of neuromuscular disorders. Glycobiology 2003;13:67R–775. [15] Muntoni F, Brockington M, Torelli S, Brown SC. Defective glycosylation in congenital muscular dystrophies. Curr Opin Neurol 2004;17:205–9. [16] Michele DE, Barresi R, Kanagawa M, et al. Post-translational disruption of dystroglycan-ligand interactions in congenital muscular dystrophies. Nature 2002;418:417–22. [17] Ibraghimov-Beskrovnaya O, Ervasti JM, Leveille CJ, Slaughter CA, Sernett SW, Campbell KP. Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature 1992;355:696–702. [18] Winder SJ. The complexities of dystroglycan. Trends Biochem Sci 2001;26:118–24. [19] Michele DE, Campbell KP. Dystrophin-glycoprotein complex: posttranslational processing and dystroglycan function. J Biol Chem 2003; 278:15457–60. [20] Saito F, Tomimitsu H, Arai K, et al. A Japanese patient with distal myopathy with rimmed vacuoles: missense mutations in the epimerase domain of the UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (GNE) gene accompanied by hyposialylation of skeletal muscle glycoproteins. Neuromuscul Disord 2004;14:158–61. [21] Huizing M, Rakocevic G, Sparks SE, et al. Hypoglycosylation of alpha-dystroglycan in patients with hereditary IBM due to GNE mutations. Mol Genet Metab 2004;81:196–202. [22] Askanas V, Engel WK. Sporadic inclusion-body myositis and hereditary inclusion-body myopathies: current concepts of diagnosis and pathogenesis. Curr Opin Rheumatol 1998;10:530–42. [23] Mirabella M, Alvarez RB, Bilak M, Engel WK, Askanas V. Difference in expression of phosphorylated tau epitopes between sporadic inclusion-body myositis and hereditary inclusion-body myopathies. J Neuropathol Exp Neurol 1996;55:774–86.
184
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
[24] Broccolini A, Pescatori M, D’Amico A, et al. An Italian family with autosomal recessive inclusion-body myopathy and mutations in the GNE gene. Neurology 2002;59:1808–9. [25] Broccolini A, Ricci E, Cassandrini D, et al. Novel GNE mutations in Italian families with autosomal recessive hereditary inclusion-body myopathy. Hum Mutat 2004;23:632–7. [26] Askanas V, Engel WK. Culturing normal and genetically abnormal human muscle. In: Rowland LP, DiMauro S, editors. Handbook of Clinical Neurology, vol. 18. Amsterdam: Elsevier; 1992. p. 85–116. [27] Broccolini A, Ricci E, Pescatori M, et al. Insulin-like growth factor I in inclusion-body myositis and human muscle cultures. J Neuropathol Exp Neurol 2004;63:650–9. [28] Losasso C, Di Tommaso F, Sgambato A, et al. Anomalous dystroglycan in carcinoma cell lines. Fed Eur Biochem Soc Lett 2000;484:194–8. [29] Longman C, Brockington M, Torelli S, et al. Mutations in the human LARGE gene causes MDC1D, a novel form of congenital muscular dystrophy with severe mental retardation and abnormal glycosylation of a-dystroglycan. Hum Mol Genet 2003;12:2853–61.
[30] Leschziner A, Moukhles H, Lindembaum M, et al. Neural regulation of a-dystroglycan biosynthesis and glycosylation in skeletal muscle. J Neurochem 2000;74:70–80. [31] De Bleecker JL, Engel AG, Winkelmann JC. Localization of dystrophin and b-spectrin in vacuolar myopathies. Am J Pathol 1993;143:1200–8. [32] Bhavanandan VP, Katlic AW. The interaction of wheat germ agglutinin with sialoglycoproteins. The role of sialic acid. J Biol Chem 1979;254:4000–8. [33] Ervasti JM, Burwell AL, Geissler AL. Tissue-specific heterogeneity in b-dystroglycan sialoglycosylation. Skeletal muscle alpha-dystroglycan is a latent receptor for Vicia villosa agglutinin B4 masked by sialic acid modification. J Biol Chem 1997;272:22315–21. [34] McDearmon EL, Combs AC, Ervasti JM. Differential vicia villosa agglutinin reactivity identifies three distinct dystroglycan complexes in skeletal muscle. J Biol Chem 2001;276:35078–6086. [35] Yamada H, Chiba A, Endo T, et al. Characterization of dystroglycanlaminin interaction in peripheral nerve. J Neurochem 1996;66: 1518–24.
a-Dystroglycan does not play a major pathogenic role in autosomal recessive hereditary inclusion-body myopathy Aldobrando Broccolinia, Carla Gliubizzia, Ernesto Pavonic, Teresa Gidaroa, Roberta Morosettia, Francesca Sciandrac, Bruno Giardinac, Pietro Tonalia, Enzo Riccia,b, Andrea Brancaccioc,1, Massimiliano Mirabellaa,* a
Department of Neuroscience, Catholic University, L.go A. Gemelli 8, 00168 Rome, Italy b U.I.L.D.M.-Rome Section, Rome, Italy c CNR—Istituto di Chimica del Riconoscimento Molecolare c/o Istituto di Biochimica e Biochimica Clinica, Catholic University, L.go F. Vito 1, 00168 Rome, Italy Received 19 August 2004; received in revised form 27 September 2004; accepted 4 October 2004
Abstract Mutations of the GNE gene are responsible for autosomal recessive hereditary inclusion-body myopathy (HIBM). In this study we searched for the presence of any significant abnormality of a-dystroglycan (a-DG), a highly glycosylated component of the dystrophinglycoprotein complex, in 5 HIBM patients which were previously clinically and genetically characterized. Immunocytochemical and immunoblot analysis showed that a-DG extracted from muscle biopsies was normally expressed and displayed its typical molecular mass. Immunoblot analysis on the wheat germ lectin-enriched glycoprotein fraction of muscles and primary myotubes showed a reduced amount of a-DG in 4 out of 5 HIBM patients, compared to normal and other diseased muscles. However, such altered lectin-binding behaviour, possibly reflecting a partial hyposialylation of a-DG, did not affect the laminin binding properties of a-DG. Therefore, the subtle changes within the a-DG glycosylation pattern, detected in HIBM muscles, likely do not play a key pathogenic role in this disorder. q 2004 Elsevier B.V. All rights reserved. Keywords: Inclusion body myopathy; HIBM; GNE; Dystrophin-glycoprotein complex; Dystroglycan
1. Introduction Autosomal recessive (AR) hereditary inclusion-body myopathy (HIBM; MIM# 600737), originally described in Persian–Jewish families, is a neuromuscular disorder characterized by onset in the early adult life with weakness and atrophy of distal lower limb muscles and relative sparing of the quadriceps [1]. HIBM is associated with mutations in the UDP-N-acetylglucosamine 2-epimerase/ N-acetylmannosamine kinase gene (GNE; MIM# 603824) on chromosome 9p12-13 [2,3]. GNE mutations have also been found in Japanese patients previously diagnosed as
* Corresponding author. Tel.: C39 6 3015 4303; fax: C39 6 3550 1909. E-mail addresses: [email protected] (A. Brancaccio), [email protected] (M. Mirabella). 1 Tel.: C39 6 305 7612; fax: C39 6 305 3598. 0960-8966/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.nmd.2004.10.001
having distal myopathy with rimmed vacuoles (DMRV; MIM# 605820), thus confirming that HIBM and DMRV represent indeed the same disease [4–7]. UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (UDP-GlcNAc epimerase/ManNAc kinase) is a bifunctional enzyme, composed of two distinct epimerase and kinase domains, that is expressed in different tissues and has a critical role in sialic acid biosynthesis [8,9]. Sialic acid is normally present on the distal ends of N- and O-glycans and is involved in many biological functions including cellular adhesion, stabilization of glycoproteins structure and signal transduction [10]. Although previous studies have shown that HIBM-associated GNE mutations result in a reduced enzymatic activity [6,11,12], the pathogenic mechanism, triggered by a possible disturbance of sialic acid metabolism and leading to muscle fiber degeneration, remains to be elucidated.
178
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
In recent years, several lines of evidence have pointed out a key role of abnormal carbohydrate metabolism and protein glycosylation in the pathogenesis of different neuromuscular disorders [13–15]. For example, Fukuyama congenital muscular dystrophy (FCMD) and muscle-eye-brain disease (MEB) are caused by mutations in genes encoding for glycosyltransferases that result in a reduced glycosylation of a-dystroglycan (a-DG), a central component of the dystrophin-glycoprotein complex (DGC) [16]. The DGC is a multimeric transmembrane complex, that provides a tight connection between the extracellular matrix and the cytoskeleton and has a crucial role in the stability of muscle cell membrane during cycles of contraction and relaxation. DG is generated from a single gene (DAG1) and is subsequently cleaved into two subunits: transmembrane b-DG and peripheral a-DG [17]. a-DG undergoes extensive post-translational N-linked and O-linked glycosylation that is thought to be critical for its interaction with proteins of the extracellular matrix such as laminin, agrin, perlecan and neurexin [18]. Abnormal glycosylation of a-DG, as observed in FCMD and MEB, results in reduced ligand binding and this is likely responsible for the muscle and brain pathology observed in these disorders [16,19]. Whether a similar mechanism plays a role also in HIBM pathology has not been to date unequivocally elucidated [11,20,21]. In the present study we investigated a-DG expression in the muscle of 5 patients with HIBM associated with mutations of the GNE gene.
2. Materials and methods 2.1. Patients Five patients from 4 unrelated families were diagnosed as having HIBM based on clinical findings, muscle pathology and genetic study [24,25]. Clinical criteria
included (i) onset in the second-third decade of life, (ii) initial weakness and atrophy of distal lower limb muscles with distal-proximal progression and sparing of the quadriceps muscles, (iii) later involvement of upper limb muscles and (iv) normal or slightly increased creatin-kinase blood levels [1]. All muscle biopsies showed morphological abnormalities indicative of HIBM [22,23]. For patient 1 different specimens were obtained from the deltoid muscle, that was clinically affected (strength grade 3/5, Medical Research Council scale), and the quadriceps, that had instead normal strength. Mutational analysis of the GNE gene in our patients has been described in details in previous studies [24,25] and is summarized in Table 1. Disease control muscles were: sporadic inclusion-body myositis (sIBM, nZ2), facio-scapulo-humeral muscular dystrophy (FSHD, nZ2), late-onset acid maltase deficiency (nZ2). Muscle biopsies from 6 patients proven to be free of any neuromuscular disorder, after all diagnostic tests were performed, were used as normal controls. All muscle samples used in this study were obtained for diagnostic purposes with informed consent. 2.2. Primary muscle cultures Human aneural primary muscle cultures were obtained from the biopsies of patients 2–4 using the explantation reexplantation method, as previously described [26]. The fusion of confluent mononucleated myoblasts into multinucleated myotubes was obtained using a culturing medium containing 5% fetal bovine serum (FBS, Cambrex Bioscience, Baltimore, MD). Fusion index was expressed as number of myonuclei/number of total nuclei, visualized by Hoechst 33258 staining (Molecular Probes Inc., Eugene, OR), and varied between 0.8 and 0.9 for all the culture sets used in this study. Before any experimental procedure, myotubes from HIBM and control patients were maintained
Table 1 Summary of clinical and genetic data of HIBM patients Patient/ gender
Age at onset (years)
Age at examination (years)
Clinical involvement
GNE mutation Nucleotide substitution (c.)a
Consequence (p.)b
Protein domain
p.N519S
Kinase
p.N519S
Kinase
p.P27S p.R246Q p.Q355_C357del/ p.I377fsX16c p.M171V p.M712T
Epimerase Epimerase Kinase
1/F
37
48
Moderate
Exon 9
2/M
25
55
Severe
Exon 9
3/F 4/F
35 21
42 49
Moderate Moderate
Exon 2 Exon 4 Intron 6
Homozygous c.1556AOG Homozygous c.1556AOG Homozygous c.79COT c.737GOAC c.1070C2dupT
5/F
23
38
Moderate
Exon 3 Exon 12
c.511AOGC c.2135TOC
a b c
Epimerase Kinase
Nucleotide position in GNE cDNA, Genbank accession# NM_005476 version 2; nucleotide numbering starts with the A of the start codon. Protein id. NP_005467.1. Predicted protein products of the two alternatively spliced GNE mRNAs.
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
for 48 h in culturing medium containing only serum replacement factors (Ultraculture, Cambrex Bioscience) to avoid the uptake of sialic acid and glycoproteins from the FBS. 2.3. Immunohistochemistry on muscle biopsies This was performed on 8 mm-thick unfixed cryostat sections. The following primary antibodies were used: (i) monoclonal anti-a-DG (clone VIA4-1, Upstate Biotechnology, Waltham, MA) recognizing only glycosylated a-DG, at the concentration of 3 mg/ml; (ii) monoclonal anti-b-DG (clone 43DAG1/8D5, YLEM, Newcastle upon Tyne, UK), diluted 1/25. The reaction with all primary antibodies was performed for 16–18 h at 4 8C in a humid chamber, followed by incubation with the appropriate horseradish peroxidase (HRP)-conjugated secondary antiserum (Dakocytomation, Glostrup, Denmark). 2.4. Immunocytochemistry on primary muscle cultures This was performed according to our previously established protocol [27] with minor modifications. In brief, myotubes were grown on glass coverslips placed on the bottom of Petri dishes. Coverslips were then collected and myotubes were washed in PBS and fixed with 2% paraformaldehyde in phosphate buffer for 15 min. at room temperature. For b-DG immunostaining only, myotubes were post-fixed with PBS containing 1% saponin and 10% normal goat serum. Incubation with either the anti-a-DG (clone VIA4-1, at the concentration of 3 mg/ml) or the antib-DG (clone 43DAG1/8D5, diluted 1/10) antibodies was performed for 16–18 h at 4 8C in a humid chamber. Detection of immunocomplexes was performed using a Texas Redw-conjugated anti-mouse antibody (Jackson Immunoresearch Laboratories, West Grove, PA). Cultures were photographed using the same exposure time to allow a correct comparison of staining intensities. 2.5. Western blot analysis on muscle biopsies and primary muscle cultures Diagnostic muscle biopsies and primary muscle cultures from HIBM and control patients were homogenized in Trisbuffered saline (TBS) containing 1% Triton X-100 and protease inhibitors; supernatants were collected upon centrifugation as previously described [28]. Glycoprotein enrichment was performed on both muscle biopsies and muscle culture extracts [16]. Briefly, supernatants were incubated with Wheat Germ lectin (WGL) Sepharosee 6MB (Amersham Bioscience, Piscataway, NJ) for 16 h at 4 8C and then centrifuged at 3000 rpm for 15 min. After supernatant removal, pellets formed from the beads were washed three times and proteins were then eluted in TBS containing 250 mM N-acetylglucosamine. Equal amounts of proteins of either total extracts or glycoprotein-enriched
179
samples were separated by SDS-PAGE and blotted onto a nitrocellulose membrane (Schleicher and Schuell, Relliehausen, Germany). Blots were incubated overnight using one of the following antibodies: (i) monoclonal anti-a-DG (clone VIA4-1, Upstate Biotechnology) at the concentration of 1.5 mg/ml, (ii) monoclonal anti-b-DG (clone 43DAG1/ 8D5, YLEM) diluted 1/50. After incubation with the appropriate HRP-conjugated secondary antibody (Amersham Bioscience), blots were developed using the enhanced chemiluminescence technique (Amersham Bioscience). Densitometry on autoradiographic films was carried out using the TotalLab 2.01 software (Nonlinear Dynamics Ltd, Newcastle upon Tyne, UK). The mean densitometric value of a-DG signal in control samples was arbitrarily set as 1 and all the measured values of patients’ samples were then expressed as a percentage of it. 2.6. Laminin overlay assay Experiments were performed according to previously described protocols [16,29]. In brief, 10 mg of proteins of total extracts were separated by SDS-PAGE and blotted onto a nitrocellulose membrane. Membranes were then blocked in 5% non-fat dry milk in laminin binding buffer (10 mM triethanolamine, 140 mM MgCl2, 1 mM CaCl2, pH 7.6) and then incubated with mouse Engelbreth–Holm– Swarm laminin (Invitrogen, Carlsbad, CA) for 16–18 h at 4 8C. Membranes were then washed, incubated with a rabbit polyclonal anti-laminin antibody (SIGMA, St Louis, MO) diluted 1/1000, followed by the appropriate HRP-conjugated secondary antibody (Chemicon, Temecula, CA) and then developed by enhanced chemiluminescence.
3. Results 3.1. Muscle biopsies In all HIBM biopsies a-DG was normally detected along the sarcolemma of muscle fibers. No difference in staining intensity was observed between HIBM and control biopsies. A normal a-DG immunosignal was also detected in all vacuole-bearing HIBM muscle fibers. Characteristically, within these fibers, the a-DG immunosignal was also located along the rim of the cytoplasmic vacuoles (Fig. 1A). Immunocytochemistry of b-DG also showed a normal signal along the sarcolemma of fibers in all HIBMs compared to control biopsies (Fig. 1A). In agreement with these results, western blot analysis on total protein extracts showed a-DG as a broad band of 150–200 kDa that was present in similar amounts in all HIBM muscle biopsies compared to normal controls (Fig. 1B). The electrophoretic mobility indicates that a-DG is not hypoglycosylated in all HIBM muscles. In addition, in patient 1 no difference of a-DG expression was observed between the deltoid and the quadriceps muscles (Fig. 1B).
180
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
Fig. 1. (A) Peroxidase immunohistochemistry for a- and b-DG on muscle biopsies. A normal immunosignal for a- and b-DG was detected along the sarcolemma of muscle fibers in all HIBM patients. Arrow indicates a vaculated muscle fiber showing that the a-DG immunosignal was also located along the rim of the cytoplasmic vacuoles. Original magnification 100!. (B) Western blot analysis of a-DG on total protein extracts from muscle biopsies showing that a-DG was present in similar amount in all HIBM muscle compared to normal controls. In patient 1 no difference of a-DG immunosignal was observed between the deltoid and the quadriceps muscles. Data shown are representative of the 5 HIBM patients studied.
Muscle glycoproteins were then purified by WGLSepharosee 6MB from total extracts of HIBM and control biopsies and used for western blot analysis of a- and b-DG. Differently from what observed in total tissue extracts, in
glycoprotein samples obtained from HIBM patients 1, 2, 3 and 5 we detected lower amounts of a-DG, compared to normal and disease control muscles (Fig. 2). Densitometric analysis showed 21% in the deltoid and 20% in
Fig. 2. Western blot analysis of a- and b-DG on muscle glycoproteins showing lower amounts of a-DG in experimental samples of HIBM patients 1, 2, 3 and 5 compared to what observed in normal and disease control muscles. No difference of a-DG immunosignal was observed between samples obtained from the deltoid and the quadriceps muscles, respectively in patient 1. On the contrary, analysis of a-DG in glycoprotein extracts from HIBM patient 4 did not show any abnormality. b-DG was quantitatively reduced in samples from HIBM patients 1 and 3 whereas it was present in normal amounts in those from the other HIBM patients.
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
181
Fig. 3. Graph representation of a-DG quantity in glycoprotein extracts from muscle biopsies of normal controls, disease controls and HIBM patients, measured by densitometry.
the quadriceps from patient 1, 43% in patient 2, 44% in patient 3 and 42% in patient 5, respectively, compared to the a-DG mean value detected in normal muscles (Fig. 3). a-DG displayed a normal molecular mass and, similarly to what observed in total tissue extracts, no difference of a-DG immunosignal was found between samples from the deltoid and the quadriceps muscles of patient 1. It should be noted that in glycoprotein extract from HIBM patient 4 a-DG was detected in amount similar to control samples whereas b-DG was reduced only in samples from HIBM patients 1 and 3 (Fig. 2). Moreover, we performed a ligand overlay assay using laminin-1 on blots of HIBM total protein extracts. In muscle from HIBM patients the ligand-bindind activity of laminin was retained similarly to what observed in normal control muscles (Fig. 4).
Fig. 4. Laminin overlay assay showing a similar a-DG-laminin interaction in HIBM muscles compared to normal controls. Data shown are representative of the 5 HIBM patients studied.
previously suggested in other cell systems [30]. In agreement with the results of western blot studies on muscle biopsies, in glycoprotein samples from HIBM patients 2 and 3 a-DG appeared as a broad band of normal molecular weight but quantitatively reduced compared to control samples (49% in patient 2 and 20% in patient 3 compared to the mean normal value), whereas it was present in normal amount in HIBM patient 4 (Fig. 5B). Western blot analysis for b-DG showed a reduced amount of protein in HIBM patient 3 whereas it resulted normal in all the others (Fig. 5B).
4. Discussion 3.2. Primary muscle cultures Cultured myotubes obtained from the muscle biopsies of patients 2–4 and normal controls were used in this study. a-DG immunocytochemistry produced a signal that was mainly localized along the plasma membrane of cultured myotubes whereas immunolabeling of permeabilized myotubes with the anti-b-DG antibody resulted in a more diffuse staining. In agreement with the results of immunocytochemical studies performed on muscle biopsies, cultured myotubes from HIBM patients showed a normal immunoreactivity for both a- and b-DG (Fig. 5A). Western blot for a-DG in glycoproteins extracted from aneurally cultivated normal primary myotubes showed a broad band with an approximate molecular mass of 120–150 kDa, lower than what observed in mature muscle. This possibly reflects the fact that the post-translational glycosylation of a-DG is developmentally regulated, as
Despite the identification of the genetic defect responsible for AR HIBM, the pathogenic mechanisms, triggered by a possible impairment of sialic acid metabolism and leading to progressive muscle degeneration, are still elusive. In recent years, a growing body of evidence has pointed out the role of a-DG, a key component of the DGC, in the pathogenic scenario of various neuromuscular disorders [13–16]. In fact, although mutations in the DAG1 gene have never been identified to date, numerous reports have shown that an abnormal glycosylation of its protein product a-DG represents the common final pathogenic pathway of muscle disorders associated with mutations of known or putative glycosyltransferases [19]. A dysfunctional a-DG could result in disruption of the link between the extracellular matrix and the cytoskeleton, thus promoting muscle fiber degeneration [13–16]. Sialic acid is normally present on the distal ends of O-glycans that decorate a-DG although its precise role in a-DG physiology remains to be elucidated
182
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
Fig. 5. (A) Fluorescence immunocytochemistry for a- and b-DG on primary muscle cultures. A normal staining for a- and b-DG was observed in all HIBM cultured myotubes compared to normal controls. Original magnification 200!. (B) Western blot analysis on glycoprotein extracts showing reduced amounts of a-DG in muscle cultures of HIBM patients 2 and 3 and a normal amount in HIBM patient 4.
[19]. Nevertheless, these considerations make a-DG a potential candidate in the attempt to elucidate the molecular mechanisms underlying HIBM pathology. By immunohistochemistry, we found that, in all HIBM biopsies, a-DG was normally expressed along the sarcolemma of muscle fibers, including vacuolated
ones. Identical results were obtained using an antibody directed against b-DG. This is in agreement with the western blot study on the corresponding total protein extracts showing normal amounts of a-DG with a molecular mass similar to that of control muscle biopsies. This latter evidence suggests that a-DG is not hypoglycosylated in HIBM muscle, in contrast with a previous report indicating hypoglycosylation of a-DG as a possible downstream causative mechanism in HIBM pathophysiology [21]. Interestingly, both dystroglycans were detected along the rim of cytoplasmic vacuoles. The functional significance of this evidence is not known. However, it should be noted that the presence of several cytoskeletal proteins, including dystrophin and laminin, has been previously demonstrated on vacuolar boundaries in different vacuolar myopathies and a possible role in strengthening the vacuolar membranes has been proposed [31]. To date, it has not been elucidated whether mutations in the GNE gene produce hyposialylation of specific muscle glycoproteins and all the studies that have attempted to tackle this issue have provided differing results [11,12,20]. Abnormal sialylation of a-DG may occur in the muscle of 4 of our HIBM patients, as suggested by the reduced binding of a-DG to the Wheat Germ lectin (Figs. 2 and 3). In fact, such lectin is known to avidly bind glycopeptides displaying clusters of sialyl oligosaccarides, since removal of sialic acid terminal residues abolishes the interaction with glycoproteins [32]. In addition, it has been shown that the digestion of the skeletal muscle dystrophin-glycoprotein complex with V. Cholerae sialidase determines a loss of staining of a-DG by Wheat Germ lectin on western blot [33]. It is likely that, in our HIBM patients, a-DG O-linked glyconjugates are only partially devoid of sialic acid residues. Indeed, it has been previously demonstrated in L6 myotubes that the total enzymatic removal of sialic acid from glycoprotein extracts should result in a w10 kDa decrease of a-DG molecular mass [34] whereas in our patients we found that a-DG always had a normal electrophoretic mobility. This is likely to depend on some residual enzymatic activity of mutated UDP-GlcNAc epimerase/ManNAc kinase in HIBM patients, as previously demonstrated [11,12]. In glycoprotein extracts from muscle biopsies of patients 1 and 3 we also observed a reduction of b-DG. Since sialic acid is also present on the distal ends of N-glycans that decorate b-DG, it is therefore possible that, in these biopsies, a reduced presence of sialic acid moieties also results in a reduced binding of b-DG to the Wheat Germ lectin, similarly to what postulated for a-DG. Our data are in agreement with other studies showing a partial reduction in the content of sialic acid in HIBM muscles and a weaker binding of Wheat Germ lectin on cultured HIBM myotubes [11]. It should be pointed out that, in our patients, the observed a-DG abnormality is not connected with the impairment of a specific domain of the enzyme, as it was detected in patients harboring a homozygous mutation in either the epimerase domain (patient 3) or the kinase domain
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
(patients 1 and 2) and in a patient who was a compound heterozygous for GNE mutations affecting different domains (patient 5). The specific role of sialic acid caps on the distal end of aDG O-linked glycoconjugates has not been fully elucidated. In peripheral nerve it has been demonstrated that sialic acid competitively inhibits laminin binding to a-DG [35], whereas in skeletal muscle the enzymatic removal of sialic acid residues does not affect this interaction [33]. In our study, a-DG extracted from HIBM muscle biopsies displayed a strong laminin binding, thus ruling out a possible impairment of the functional link between the cytoskeleton and the extracellular matrix due to a partial hyposialylation of a-DG. Furthermore, a normal signal for a-DG was detected also in glycoprotein enriched sample from HIBM patient 4 and in HIBM patient 1 we found no difference of a-DG abundance in samples obtained from the deltoid and quadriceps muscles respectively, that were instead differently affected. Taken together these data would strongly argue against a pivotal role of a-DG in the molecular pathophysiology of HIBM muscle. To the best of the current knowledge, it appears that a partial hyposialylation of a-DG, as observed in 4 of our HIBM patients, would represent only a minor byproduct of a metabolic impairment that may instead crucially affect other subcellular compartments and may not exclusively be associated with the synthesis of sialic acid. In fact, it has been previously postulated that UDP-GlcNAc epimerase/ ManNAc kinase could be also involved in other, yet unknown, metabolic pathways possibly more important in HIBM pathophysiology [12]. Whether an hyposialylated a-DG may also play a role in the activation of abnormal cellular mechanisms remains to be determined and additional studies will be required to tackle this issue.
Acknowledgements Study supported in part by grants from The Italian Ministry of Health 2002 and 2003 to Massimiliano Mirabella. The grant Telethon GGP030332 to Andrea Brancaccio is gratefully acknowledged.
References [1] Argov Z, Yarom Y. Rimmed vacuole myopathy sparing the quadriceps. A unique disorder in Iranian Jewish. J Neurol Sci 1984; 64:33–43. [2] Eisenberg I, Avidan N, Potikha T, et al. The UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase gene is mutated in recessive hereditary inclusion body myopathy. Nat Genet 2001;29:83–7. [3] Eisenberg I, Grabov-Nardini G, Hochner M, et al. Mutation spectrum of GNE in hereditary inclusion body myopathy sparing the quadriceps. Hum Mutat 2003;21:99–105.
183
[4] Kayashima T, Matsuo H, Satoh A, et al. Nonaka myopathy is caused by mutations in the UDP-N-acetylglucosamine-2-epimerase/N-acetylmannosamine kinase gene (GNE). J Hum Genet 2002;47:77–9. [5] Tomimitsu H, Ishikawa K, Shimizu J, Ohkoshi N, Kanazawa I, Mizusawa H. Distal myopathy with rimmed vacuoles: novel mutations in the GNE gene. Neurology 2002;59:451–4. [6] Nishino I, Noguchi S, Muruyama K, et al. Distal myopathy with rimmed vacuoles is allelic to hereditary inclusion body myopathy. Neurology 2002;59:1689–93. [7] Arai A, Tanaka K, Ikeuchi T, et al. A novel mutation in the GNE gene and a linkage disequilibrium in Japanese pedigrees. Ann Neurol 2002; 52:516–9. [8] Hinderlich S, Sta¨sche R, Zeitler R, Reutter W. A bifunctional enzyme catalyses the first two steps in N-acetylneuraminic acid biosynthesis of rat liver. Purification and characterization of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase. J Biol Chem 1997; 272:24313–8. [9] Efferts K, Hinderlich S, Reutter W. Selective loss of either the epimerase or kinase activity of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase due to site-directed mutagenesis based on sequence alignments. J Biol Chem 1999;274:28771–8. [10] Keppler OT, Hinderlich S, Langner J, Schwartz-Albiez R, Reutter W, Pawlita M. UDP-GlcNAc 2-epimerase: a regulator of cell surface sialylation. Science 1999;284:1372–6. [11] Noguchi S, Keira Y, Murayama K, et al. Reduction of UDP-Nacetylglucosamine 2-epimerase/N-acetylmannosamine kinase activity and sialylation in distal myopathy with rimmed vacuoles. J Biol Chem 2004;279:11402–7. [12] Hinderlich S, Salama I, Eisenberg I, et al. The homozygous M712T mutation of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase results in reduced enzyme activities but not in altered overall cellular sialylation in hereditary inclusion body myopathy. Fed Eur Biochem Soc Lett 2004;566:105–9. [13] Grewal PK, Hewitt JE. Glycosylation defects: a new mechanism for muscular dystrophy? Hum Mol Genet 2003;12:R259–R64. [14] Martin PT, Freeze HH. Glycobiology of neuromuscular disorders. Glycobiology 2003;13:67R–775. [15] Muntoni F, Brockington M, Torelli S, Brown SC. Defective glycosylation in congenital muscular dystrophies. Curr Opin Neurol 2004;17:205–9. [16] Michele DE, Barresi R, Kanagawa M, et al. Post-translational disruption of dystroglycan-ligand interactions in congenital muscular dystrophies. Nature 2002;418:417–22. [17] Ibraghimov-Beskrovnaya O, Ervasti JM, Leveille CJ, Slaughter CA, Sernett SW, Campbell KP. Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature 1992;355:696–702. [18] Winder SJ. The complexities of dystroglycan. Trends Biochem Sci 2001;26:118–24. [19] Michele DE, Campbell KP. Dystrophin-glycoprotein complex: posttranslational processing and dystroglycan function. J Biol Chem 2003; 278:15457–60. [20] Saito F, Tomimitsu H, Arai K, et al. A Japanese patient with distal myopathy with rimmed vacuoles: missense mutations in the epimerase domain of the UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (GNE) gene accompanied by hyposialylation of skeletal muscle glycoproteins. Neuromuscul Disord 2004;14:158–61. [21] Huizing M, Rakocevic G, Sparks SE, et al. Hypoglycosylation of alpha-dystroglycan in patients with hereditary IBM due to GNE mutations. Mol Genet Metab 2004;81:196–202. [22] Askanas V, Engel WK. Sporadic inclusion-body myositis and hereditary inclusion-body myopathies: current concepts of diagnosis and pathogenesis. Curr Opin Rheumatol 1998;10:530–42. [23] Mirabella M, Alvarez RB, Bilak M, Engel WK, Askanas V. Difference in expression of phosphorylated tau epitopes between sporadic inclusion-body myositis and hereditary inclusion-body myopathies. J Neuropathol Exp Neurol 1996;55:774–86.
184
A. Broccolini et al. / Neuromuscular Disorders 15 (2005) 177–184
[24] Broccolini A, Pescatori M, D’Amico A, et al. An Italian family with autosomal recessive inclusion-body myopathy and mutations in the GNE gene. Neurology 2002;59:1808–9. [25] Broccolini A, Ricci E, Cassandrini D, et al. Novel GNE mutations in Italian families with autosomal recessive hereditary inclusion-body myopathy. Hum Mutat 2004;23:632–7. [26] Askanas V, Engel WK. Culturing normal and genetically abnormal human muscle. In: Rowland LP, DiMauro S, editors. Handbook of Clinical Neurology, vol. 18. Amsterdam: Elsevier; 1992. p. 85–116. [27] Broccolini A, Ricci E, Pescatori M, et al. Insulin-like growth factor I in inclusion-body myositis and human muscle cultures. J Neuropathol Exp Neurol 2004;63:650–9. [28] Losasso C, Di Tommaso F, Sgambato A, et al. Anomalous dystroglycan in carcinoma cell lines. Fed Eur Biochem Soc Lett 2000;484:194–8. [29] Longman C, Brockington M, Torelli S, et al. Mutations in the human LARGE gene causes MDC1D, a novel form of congenital muscular dystrophy with severe mental retardation and abnormal glycosylation of a-dystroglycan. Hum Mol Genet 2003;12:2853–61.
[30] Leschziner A, Moukhles H, Lindembaum M, et al. Neural regulation of a-dystroglycan biosynthesis and glycosylation in skeletal muscle. J Neurochem 2000;74:70–80. [31] De Bleecker JL, Engel AG, Winkelmann JC. Localization of dystrophin and b-spectrin in vacuolar myopathies. Am J Pathol 1993;143:1200–8. [32] Bhavanandan VP, Katlic AW. The interaction of wheat germ agglutinin with sialoglycoproteins. The role of sialic acid. J Biol Chem 1979;254:4000–8. [33] Ervasti JM, Burwell AL, Geissler AL. Tissue-specific heterogeneity in b-dystroglycan sialoglycosylation. Skeletal muscle alpha-dystroglycan is a latent receptor for Vicia villosa agglutinin B4 masked by sialic acid modification. J Biol Chem 1997;272:22315–21. [34] McDearmon EL, Combs AC, Ervasti JM. Differential vicia villosa agglutinin reactivity identifies three distinct dystroglycan complexes in skeletal muscle. J Biol Chem 2001;276:35078–6086. [35] Yamada H, Chiba A, Endo T, et al. Characterization of dystroglycanlaminin interaction in peripheral nerve. J Neurochem 1996;66: 1518–24.