- Email: [email protected]
LGMD 2E in Tunisia is caused by a homozygous missense mutation in β-sarcoglycan exon 3
Neuromuscular Disorders 8 (1998) 193–197
LGMD 2E in Tunisia is caused by a homozygous missense mutation in b-sarcoglycan exon 3 C.G. Bo¨nnemann a , b, J. Wong a, Ch. Ben Hamida c, M. Ben Hamida c, F. Hentati c, L.M. Kunkel a ,* a Division of Genetics, Howard Hughes Medical Institute, Boston, MA, USA Department of Neurology, Children’s Hospital and Harvard Medical School, Boston, MA, USA c Institut National de Neurologie, Tunis, Tunisia
b
Received 15 December 1997; accepted 4 February 1998
Abstract Four of the currently recognized autosomal recessive limb-girdle muscular dystrophies (LGMD type 2C–F) are caused by mutations in the genes encoding components of the sarcoglycan complex. LGMD 2C, caused by mutations in g-sarcoglycan, is prevalent in northern Africa, especially in Tunisia, where this type of muscular dystrophy was originally described. Although the disease initially was assumed to be genetically homogeneous in this region, linkage to the a-sarcoglycan locus (LGMD 2D) has also been found. We have now identified the first Tunisian family with b-sarcoglycanopathy (LGMD 2E), further adding to the genetic heterogeneity of autosomal recessive LGMD in this population. Direct sequencing of the b-sarcoglycan gene revealed a homozygous mutation (G272 → T, Arg91Leu) in exon 3. This change affects the same arginine residue in the immediate extracellular domain of the protein that was mutated to a proline (G272 → C, Arg91Pro) in a Brazilian family with a severe form of the disease. Immunohistochemical analysis for the sarcoglycan complex demonstrates absence of the known components of the complex in both of these families. We postulate that the immediate extracellular domain of b-sarcoglycan may be important for the assembly and/or maintenance of this complex, potentially mediated by disulfide-bond formation to another sarcoglycan via the single cysteine residue in that domain. 1998 Elsevier Science B.V. Keywords: Limb-girdle muscular dystrophy; b-Sarcoglycan; Sarcoglycanopathy; LGMD 2E
1. Introduction The disorders of the sarcoglycan complex (sarcoglycanopathies) are a genetically heterogeneous subgroup within the autosomal recessive limb-girdle muscular dystrophies (LGMD) [1,2]. Although severe childhood presentations appear to be more common amongst the group of sarcoglycanopathies as a whole, milder disease manifestations are part of the spectrum [3–5] (C.G. Bo¨nnemann and L.M. Kunkel, unpublished data). With the identification of the four currently-known sarcoglycan genes encoding for a-, b-, g-, and d-sarcoglycan [6–10], mutation analysis, and * Corresponding author. Division of Genetics and Howard Hughes Medical Institute, Enders 570, Children’s Hospital, 300 Longwood Avenue, Boston, MA 02115, USA. Tel.: +1 617 3557576; fax: +1 617 3557588; e-mail: [email protected]
0960-8966/98/$19.00 1998 Elsevier Science B.V. All rights reserved PII S0960-8966 (98 )0 0014-5
thereby precise genetic diagnosis, has become possible. However, genotype/phenotype correlations have been difficult to achieve. These difficulties are partly due to the fact that only homozygous missense mutations are useful for structure/function predictions within the individual proteins. An additional difficulty is the possible variability in clinical severity, even with identical underlying mutations [3,11]. Autosomal recessive LGMD is prevalent in northern African countries [12], and in that population most of the families with this LGMD phenotype showed genetic linkage to chromosome 13q12 (LGMD 2C) [13–15] and are now known to be caused by a common single base-pair deletion in exon 6 of the g-sarcoglycan gene (del521T) [9]. However, even in this population, the disease appears to be genetically heterogeneous, with a low proportion of families linked to the a-sarcoglycan locus on chromosome 17q21 [16–18]. b-Sarcoglycan mutations (LGMD 2E) have not
194
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
yet been detected in northern Africa. Here we present the first family with LGMD 2E in the Tunisian population caused by a novel homozygous missense mutation. The phenotype and location of the mutation is compared with the previously-described b-sarcoglycan mutations.
extension at 72°C for 1 min for 36 cycles in a total reaction volume of 25 ml. A Taq polymerase allowing for antibodymediated hotstart of the reaction (Platinum Taq, Gibco, Gaithersburg, MD) was used. Products were separated by electrophoresis in 3% agarose and visualized with ethidium bromide for gel photography.
2. Methods 3. Results Linkage analysis was performed by standard techniques [13]; the polymorphic dinucleotide repeat markers for chromosome 4 used in this study were D4S405, D4S399, D4S395 [19]. The mutation was detected by direct sequencing of PCR products generated from genomic DNA with the primer sets and conditions described in [20]. Sequencing was performed by dRhodamine Terminator Cycle Sequencing on an ABI 377 automated sequencer (PE Applied Biosystems, Foster City, CA). The co-inheritance of the mutation with the disease was confirmed using the amplification resistant mutation system (ARMS) essentially as described [21,22]. A wild type-specific primer (5′ ACT CCA TAC TAT CAC AGC CAT TTG GTC CAA TAC 3′) or a mutation-specific primer (5′ ACT CCA TAC TAT CAC AGC CAT TTG GTC CAA TAA 3′) was used in conjunction with a common primer located in intron 2, (5′ TGG TGA TAA TAT TTT CTA CTT GTT TTC CAA TTA C 3′). Control primers were those given in [21,22] for a 360-bp fragment from exon III of the a1-antitrypsin gene. Amplification conditions included denaturation at 94°C for 1 min, annealing at 60°C for 1 min and
Of 20 Tunisian families with severe childhood onset autosomal recessive LGMD large enough for linkage analysis, 16 are linked to the g-sarcoglycan locus on chromosome 13q12, LGMD 2C, and carry the common northern African mutation del521T. In one family, linkage analysis indicated the a-sarcoglycan locus (LGMD 2D) on chromosome 17q21, and in another family there was linkage to the LGMD 2A locus on chromosome 15q15 (data not shown). One family was unlinked to any of the known loci for autosomal recessive LGMD. The remaining family (A) was a consanguineous marriage with three affected and four unaffected offspring. Segregation analysis with the markers D4S405, D4S399, D4S395 suggested linkage to proximal chromosome 4q, as shown by homozygosity at these loci in the affected individuals. Age of onset in patients 1 and 2 (22 and 24 years old) was dated to about 8 years of age in both, when difficulties in walking were first noted. These two patients were wheelchair-dependent at ages 17 (patient 1) and 12 (patient 2). Patient 3, now 14 years old, noted first symptoms at 9 years of age and is still ambulatory with
Fig. 1. Amplification resistant mutation system (ARMS) analysis of the G272 → T transversion in b-sarcoglycan exon 3 and pedigree structure of Tunisian family A. The affected offspring 1, 2 and 3 are depicted next to each other in this pedigree to facilitate the presentation of results, and does not correspond to the actual succession of offspring in this family. The band depicted with an asterisk (*) corresponds to the control amplification form from exon III of the a1antitrypsin gene. In patient 2 there was no amplifiable DNA template indicated by the failure of the control amplification. The bands indicated by the arrow correspond to the allele specific amplification products, with N specifying the wildtype and M the mutant reaction for each proband in the pedigree. As expected from the first cousin union in this pedigree, the two patients for whom there was amplifiable DNA (1 and 3) are homozygous for the mutant allele, whereas the parents and siblings are all heterozygous. A control specimen from a normal individual is homozygous for the wildtype allele.
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
significant Achilles tendon contractures and lordosis. CK values were significantly elevated in all affected siblings. A muscle biopsy from patient 2 revealed changes consistent with muscular dystrophy and clear progression of the histologic changes over an 8-year period. Immunohistochemistry on this biopsy specimen from patient 2 demonstrated normal dystrophin immunoreactivity, but absent immunoreactivity for a-, b-, and g-sarcoglycan (not shown and [23]). Evidence of linkage to the b-sarcoglycan locus on chromosome 4q12 suggested a primary b-sarcoglycan mutation in this family. b-Sarcoglycan exons were amplified from genomic DNA from patient 1 and sequenced directly. A homozygous G272 → T transversion was found in exon 3, which resulted in a predicted arginine to leucine missense change at position 91 in the immediate extracellular domain of the protein. Co-inheritance of the mutation with the disease in other family members was confirmed using an amplification-resistant mutation system (ARMS) based on the design of allele-specific (mutant versus wild type) primers. The parents and all unaffected siblings were found to be heterozygous for the mutation (Fig. 1). Both patients for whom the control amplification was positive were homozygous for the mutant allele (Fig. 1).
195
4. Discussion Here we present a not-previously-described b-sarcoglycan mutation (LGMD 2E) in northern Africa, further contributing to the genetic heterogeneity of autosomal recessive LGMD even within this fairly genetically-homogeneous appearing population [13–15]. Clinically, the Tunisian bsarcoglycanopathy patients were not recognizably different from the much more prevalent patients with LGMD 2C due to the del521T mutation in g-sarcoglycan. However, the absence of immunohistochemical detection of the sarcoglycan proteins in this family was different from the pattern usually observed in LGMD 2C in Tunisia, as well as in other populations with different g-sarcoglycan mutations, in which a partial preservation in a- and b-sarcoglycan immunoreactivity is often seen [11,17,19,24–26]. Truncating mutations in b-sarcoglycan (Fig. 2) are mostly associated with a severe phenotype and are as virtual null mutations less helpful for predictions about functional protein domains. Thus, the identification of homozygous missense mutations in b-sarcoglycan is particularly valuable for attempts at structure/function predictions. The novel mutation described here affects the same nucleotide as the one
Fig. 2. Schematic summary of b-sarcoglycan mutations published so far plus additional unpublished observations. The single cysteine residue is emphasized. Missense mutations are indicated by solid arrows above the schematic of the molecule, and truncating mutations by broken arrows below the molecule. The Tunisian Arg91Leu mutation is boxed. Missense mutations that appear in homozygous form are underlined. References to the mutations are given in brackets and follow the numbering in the reference section. Mutations designated by [*] refer to unpublished observations (CGB and LMK). Details are given in the main text.
196
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
described by us in a Brazilian pedigree with LGMD 2E, where G272 → C results in an Arg91Pro change [20]. The phenotype in the Brazilian family appears to be somewhat more severe with onset at or before 6 years of age and loss of ambulation by 10 years in all three affected patients. There was also total absence of the sarcoglycan complex by immunohistochemistry. The Brazilian Arg91Pro change not only abolishes a positive charge in the immediate extracellular domain, it also interrupts the betasheet structure in that part of the molecule, whereas the Tunisian Arg91Leu change abolishes the charge, but does not interrupt the beta sheet. This difference may contribute to the somewhat milder course of the disease in the Tunisian family reported here. However, such assumptions have to be made with caution, since there is known inherent variability of severity within the sarcoglycan disorders even on the basis of an identical mutation [3,11]. Indeed, there also is a suggestion of variability of severity in our family, as shown by the age of loss of ambulation in patient 1 (17 years) compared with patient 2 (12 years). Given the vicinity of these mutations to the single extracellular cysteine residue in b-sarcoglycan (Fig. 2) we speculate further that the observed disintegra-tion of the sarcoglycan complex may possibly be related to interference with disulfidebond mediated cross-linking between b-sarcoglycan and another component of the complex. a-, g- and d-sarcoglycan also have single cysteine residues in the extracellular domain that could function as cross-linking partners to bsarcoglycan, thus participating in the assembly and/or maintenance of the sarcoglycan complex. It is of interest to note that there appears to be a small cluster of missense mutations in this immediate extracellular domain encoded on exon 3 of b-sarcoglycan (Fig. 2). In all cases examined by us, we also found a total absence of the complex by immunohistochemistry [20,26]. These mutations appear to lead to a rather severe phenotype [20] and may interfere with the assembly and/or maintenance of the complex in a way similar to the mutations at residue 91. On the other hand, the recurrent mutation Ser114Phe located slightly further-out in the extracellular domain is much more variable in clinical expression, allowing ambulation into adult life in some patients [5] (C.G. Bo¨nnemann and L.M. Kunkel, unpublished data), while presenting with a severe childhood muscular dystrophy in another case [27]. Clinical variability is also seen in the Thr151Ala mutation found in the Indiana Amish population [8]. Tyr184Cys likewise presented with a somewhat milder phenotype [28], whereas Thr182Ala was more severe [27]. The intracellular Gln11Glu mutation was associated with a clearly severe phenotype [27], pointing to another potentially important domain of the molecule. Although genotype/phenotype correlations in the sarcoglycan genes have to be qualified by the intrinsic variability of the disease itself, they nonetheless may guide biochemical research into structure/function correlations in the corresponding proteins and the sarcoglycan complex as a whole.
Acknowledgements We would like to thank Richard Bennet, Julie Scarfo and Ivan Guerrero for expert help with sequencing as well as the members of the Kunkel lab for helpful discussions. The project described was supported by grant number 5 RO1 NS 23740-09 of the NINDS to L.M.K. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NINDS. L.M.K. is an investigator of the Howard Hughes Medical Institute.
References [1] Bushby KMD. Diagnostic criteria for the limb-girdle muscular dystrophies: report of the ENMC consortium on limb-girdle dystrophies. Neuromusc Disord 1995;5:71–74. [2] Bo¨nnemann CG, McNally EM, Kunkel LM. Beyond dystrophin: current progress in the muscular dystrophies. Curr Opin Pediatr 1996;8:569–582. [3] McNally EM, Passos-Bueno R, Bo¨nnemann CG, et al. Mild and severe muscular dystrophy caused by a single g-sarcoglycan mutation. Am J Hum Genet 1996;59:1040–1047. [4] Eymard B, Romero NB, Leturcq F, et al. Primary adhalinopathy (alpha-sarcoglycanopathy):clinical, pathologic, and genetic correlation in 20 patients with autosomal recessive muscular dystrophy. Neurology 1997;48:1227–1234. [5] Ginjaar HB, vd Kooi A, Ceelie H, et al. Sarcoglycanopathies in Dutch patients with autosomal recessive limb girdle muscular dystrophies. Neuromusc Disord 1997;7:440,GP1B.6. [6] Roberds SL, Leturcq F, Allamand V, et al. Missense mutations in the adhalin gene linked to autosomal recessive muscular dystrophy. Cell 1994;78:625–633. [7] Bo¨nnemann CG, Modi R, Noguchi S, et al. b-Sarcoglycan (A3b) mutations cause autosomal recessive muscular dystrophy with loss of the sarcoglycan complex. Nat Genet 1995;11:266–273. [8] Lim LE, Duclos F, Broux O, et al. b-Sarcoglycan (43 DAG): characterization and involvement in a recessive form of limb-girdle muscular dystrophy linked to chromosome 4q12. Nat Genet 1995;11:257–265. [9] Noguchi S, McNally EM, Ben Othmane K, et al. Mutations in the dystrophin-associated protein g-sarcoglycan in chromosome 13 muscular dystrophy. Science 1995;270:819–822. [10] Nigro V, de Sa Moriera E, Piluso G, et al. The 5q autosomal recessive limb-girdle muscular dystrophy (LGMD 2F) is caused by a mutation in the d-sarcoglycan gene. Nat Genet 1996;14:195–198. [11] Ben Hamida M, Ben Hamida C, Zouari M, Belal S, Hentati F. Limbgirdle muscular dystrophy 2C: clinical aspects. Neuromusc Disord 1996;6:493–494. [12] Ben Hamida M, Fardeau M, Attia N. Severe childhood muscular dystrophy affecting both sexes and frequent in Tunisia. Muscle Nerve 1983;6:469–480. [13] Ben Othmane K, Ben Hamida M, Pericak-Vance MA, et al. Linkage of Tunisian autosomal recessive Duchenne-like muscular dystrophy to the pericentromeric region of chromosome 13q. Nat Genet 1992;2:315–317. [14] Azibi K, Bachner L, Beckmann JS, et al. Severe childhood autosomal recessive muscular dystrophy with the deficiency of the 50 kDa dystrophin-associated glycoprotein maps to chromosome 13q12. Hum Mol Genet 1993;2:1423–1428. [15] el Kerch F, Sefiani A, Azibi K, et al. Linkage analysis of families with severe childhood autosomal recessive muscular dystrophy in Morocco indicates genetic homogeneity of the disease in North Africa. J Med Genet 1994;31:342–343.
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197 [16] Ben Hamida C, Belal S, Hchaichi O, Hentati F, Ben Hamida M. Linkage analysis of Tunisian Duchenne-like muscular dystrophies and other limb-girdle muscular dystrophies. 30/31 ENMC International workshop on LGMD/SCARMD. Naarden, The Netherlands, 1995. [17] Ben Hamida M, Ben Hamida C, MDA LGMD Workshop, Minneapolis, MA, 1995. [18] Azibi K, Chaouch M, Reghis A, et al. Genetic and allelic heterogeneity of LGMD in North Africa. Neuromusc Disord 1997;7:441. [19] Ben Hamida C, Zouari M, Amouri R, et al. Phenotype-genotype correlation in Tunisian patients with not linked to chromosome 13 autosomal recessive muscular dystrophy. MDA LGMD Genetics Workshop, Tunis, Tunisia, 1996. [20] Bo¨nnemann CG, Passos-Bueno R, McNally EM, et al. Genomic screening for b-sarcoglycan mutations: missense mutations may cause severe limb-girdle muscular dystrophy type 2E (LGMD 2E). Hum Mol Genet 1996;5:1953–1961. [21] Little S. ARMS analysis of point mutations, in: Dracopoli NC, Haines DL, Korf BR, et al., editors. Current Protocols in Human Genetics, New York: Wiley, 1995;Suppl. 3:9.8.1–9.8.12. [22] Ferrie RM, Schwarz MJ, Robertson NH, et al. Development, multi-
[23]
[24]
[25]
[26]
[27]
[28]
197
plexing, and application of ARMS tests for common mutations in the CFTR gene. Am J Hum Genet 1992;51:251–262. Hentati F, Ben Hamida C, Belal S, et al. Immunohistochemical analysis of dystrophin-associated glycoproteins in Duchenne-like autosomal recessive muscular dystrophies in Tunisia. Vth European Congress of Neuropathology, Paris, France, 1996. McNally EM, Duggan DJ, Gorospe JR, et al. Mutations in the carboxyl-terminus of g-sarcoglycan cause muscular dystrophy. Hum Mol Genet 1996;5:1841–1847. Sewry CA, Taylor J, Anderson LV, et al. Abnormalities in alpha-, beta- and gamma-sarcoglycan in patients with limb-girdle muscular dystrophy. Neuromusc Disord 1996;6:467–474. Vainzof M, Passos-Bueno MR, Moreira ES, et al. The sarcoglycan complex in the six autosomal recessive limb-girdle muscular dystrophies. Hum Mol Genet 1996;5:1963–1969. Duggan DJ, Gorospe JR, Fanin M, Hoffman EP, Angelini C. Mutations in the sarcoglycan genes in patients with myopathy. N Engl J Med 1997;336:618–624. dos Santos MR, Vieira EM, Ribeiro MG, Santos MM, Gomes R, Guimara¨es A. Known and novel sarcoglycan gene mutations in Portoguese patients. Neuromusc Disord 1997;7:439,GP1B.4.
LGMD 2E in Tunisia is caused by a homozygous missense mutation in b-sarcoglycan exon 3 C.G. Bo¨nnemann a , b, J. Wong a, Ch. Ben Hamida c, M. Ben Hamida c, F. Hentati c, L.M. Kunkel a ,* a Division of Genetics, Howard Hughes Medical Institute, Boston, MA, USA Department of Neurology, Children’s Hospital and Harvard Medical School, Boston, MA, USA c Institut National de Neurologie, Tunis, Tunisia
b
Received 15 December 1997; accepted 4 February 1998
Abstract Four of the currently recognized autosomal recessive limb-girdle muscular dystrophies (LGMD type 2C–F) are caused by mutations in the genes encoding components of the sarcoglycan complex. LGMD 2C, caused by mutations in g-sarcoglycan, is prevalent in northern Africa, especially in Tunisia, where this type of muscular dystrophy was originally described. Although the disease initially was assumed to be genetically homogeneous in this region, linkage to the a-sarcoglycan locus (LGMD 2D) has also been found. We have now identified the first Tunisian family with b-sarcoglycanopathy (LGMD 2E), further adding to the genetic heterogeneity of autosomal recessive LGMD in this population. Direct sequencing of the b-sarcoglycan gene revealed a homozygous mutation (G272 → T, Arg91Leu) in exon 3. This change affects the same arginine residue in the immediate extracellular domain of the protein that was mutated to a proline (G272 → C, Arg91Pro) in a Brazilian family with a severe form of the disease. Immunohistochemical analysis for the sarcoglycan complex demonstrates absence of the known components of the complex in both of these families. We postulate that the immediate extracellular domain of b-sarcoglycan may be important for the assembly and/or maintenance of this complex, potentially mediated by disulfide-bond formation to another sarcoglycan via the single cysteine residue in that domain. 1998 Elsevier Science B.V. Keywords: Limb-girdle muscular dystrophy; b-Sarcoglycan; Sarcoglycanopathy; LGMD 2E
1. Introduction The disorders of the sarcoglycan complex (sarcoglycanopathies) are a genetically heterogeneous subgroup within the autosomal recessive limb-girdle muscular dystrophies (LGMD) [1,2]. Although severe childhood presentations appear to be more common amongst the group of sarcoglycanopathies as a whole, milder disease manifestations are part of the spectrum [3–5] (C.G. Bo¨nnemann and L.M. Kunkel, unpublished data). With the identification of the four currently-known sarcoglycan genes encoding for a-, b-, g-, and d-sarcoglycan [6–10], mutation analysis, and * Corresponding author. Division of Genetics and Howard Hughes Medical Institute, Enders 570, Children’s Hospital, 300 Longwood Avenue, Boston, MA 02115, USA. Tel.: +1 617 3557576; fax: +1 617 3557588; e-mail: [email protected]
0960-8966/98/$19.00 1998 Elsevier Science B.V. All rights reserved PII S0960-8966 (98 )0 0014-5
thereby precise genetic diagnosis, has become possible. However, genotype/phenotype correlations have been difficult to achieve. These difficulties are partly due to the fact that only homozygous missense mutations are useful for structure/function predictions within the individual proteins. An additional difficulty is the possible variability in clinical severity, even with identical underlying mutations [3,11]. Autosomal recessive LGMD is prevalent in northern African countries [12], and in that population most of the families with this LGMD phenotype showed genetic linkage to chromosome 13q12 (LGMD 2C) [13–15] and are now known to be caused by a common single base-pair deletion in exon 6 of the g-sarcoglycan gene (del521T) [9]. However, even in this population, the disease appears to be genetically heterogeneous, with a low proportion of families linked to the a-sarcoglycan locus on chromosome 17q21 [16–18]. b-Sarcoglycan mutations (LGMD 2E) have not
194
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
yet been detected in northern Africa. Here we present the first family with LGMD 2E in the Tunisian population caused by a novel homozygous missense mutation. The phenotype and location of the mutation is compared with the previously-described b-sarcoglycan mutations.
extension at 72°C for 1 min for 36 cycles in a total reaction volume of 25 ml. A Taq polymerase allowing for antibodymediated hotstart of the reaction (Platinum Taq, Gibco, Gaithersburg, MD) was used. Products were separated by electrophoresis in 3% agarose and visualized with ethidium bromide for gel photography.
2. Methods 3. Results Linkage analysis was performed by standard techniques [13]; the polymorphic dinucleotide repeat markers for chromosome 4 used in this study were D4S405, D4S399, D4S395 [19]. The mutation was detected by direct sequencing of PCR products generated from genomic DNA with the primer sets and conditions described in [20]. Sequencing was performed by dRhodamine Terminator Cycle Sequencing on an ABI 377 automated sequencer (PE Applied Biosystems, Foster City, CA). The co-inheritance of the mutation with the disease was confirmed using the amplification resistant mutation system (ARMS) essentially as described [21,22]. A wild type-specific primer (5′ ACT CCA TAC TAT CAC AGC CAT TTG GTC CAA TAC 3′) or a mutation-specific primer (5′ ACT CCA TAC TAT CAC AGC CAT TTG GTC CAA TAA 3′) was used in conjunction with a common primer located in intron 2, (5′ TGG TGA TAA TAT TTT CTA CTT GTT TTC CAA TTA C 3′). Control primers were those given in [21,22] for a 360-bp fragment from exon III of the a1-antitrypsin gene. Amplification conditions included denaturation at 94°C for 1 min, annealing at 60°C for 1 min and
Of 20 Tunisian families with severe childhood onset autosomal recessive LGMD large enough for linkage analysis, 16 are linked to the g-sarcoglycan locus on chromosome 13q12, LGMD 2C, and carry the common northern African mutation del521T. In one family, linkage analysis indicated the a-sarcoglycan locus (LGMD 2D) on chromosome 17q21, and in another family there was linkage to the LGMD 2A locus on chromosome 15q15 (data not shown). One family was unlinked to any of the known loci for autosomal recessive LGMD. The remaining family (A) was a consanguineous marriage with three affected and four unaffected offspring. Segregation analysis with the markers D4S405, D4S399, D4S395 suggested linkage to proximal chromosome 4q, as shown by homozygosity at these loci in the affected individuals. Age of onset in patients 1 and 2 (22 and 24 years old) was dated to about 8 years of age in both, when difficulties in walking were first noted. These two patients were wheelchair-dependent at ages 17 (patient 1) and 12 (patient 2). Patient 3, now 14 years old, noted first symptoms at 9 years of age and is still ambulatory with
Fig. 1. Amplification resistant mutation system (ARMS) analysis of the G272 → T transversion in b-sarcoglycan exon 3 and pedigree structure of Tunisian family A. The affected offspring 1, 2 and 3 are depicted next to each other in this pedigree to facilitate the presentation of results, and does not correspond to the actual succession of offspring in this family. The band depicted with an asterisk (*) corresponds to the control amplification form from exon III of the a1antitrypsin gene. In patient 2 there was no amplifiable DNA template indicated by the failure of the control amplification. The bands indicated by the arrow correspond to the allele specific amplification products, with N specifying the wildtype and M the mutant reaction for each proband in the pedigree. As expected from the first cousin union in this pedigree, the two patients for whom there was amplifiable DNA (1 and 3) are homozygous for the mutant allele, whereas the parents and siblings are all heterozygous. A control specimen from a normal individual is homozygous for the wildtype allele.
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
significant Achilles tendon contractures and lordosis. CK values were significantly elevated in all affected siblings. A muscle biopsy from patient 2 revealed changes consistent with muscular dystrophy and clear progression of the histologic changes over an 8-year period. Immunohistochemistry on this biopsy specimen from patient 2 demonstrated normal dystrophin immunoreactivity, but absent immunoreactivity for a-, b-, and g-sarcoglycan (not shown and [23]). Evidence of linkage to the b-sarcoglycan locus on chromosome 4q12 suggested a primary b-sarcoglycan mutation in this family. b-Sarcoglycan exons were amplified from genomic DNA from patient 1 and sequenced directly. A homozygous G272 → T transversion was found in exon 3, which resulted in a predicted arginine to leucine missense change at position 91 in the immediate extracellular domain of the protein. Co-inheritance of the mutation with the disease in other family members was confirmed using an amplification-resistant mutation system (ARMS) based on the design of allele-specific (mutant versus wild type) primers. The parents and all unaffected siblings were found to be heterozygous for the mutation (Fig. 1). Both patients for whom the control amplification was positive were homozygous for the mutant allele (Fig. 1).
195
4. Discussion Here we present a not-previously-described b-sarcoglycan mutation (LGMD 2E) in northern Africa, further contributing to the genetic heterogeneity of autosomal recessive LGMD even within this fairly genetically-homogeneous appearing population [13–15]. Clinically, the Tunisian bsarcoglycanopathy patients were not recognizably different from the much more prevalent patients with LGMD 2C due to the del521T mutation in g-sarcoglycan. However, the absence of immunohistochemical detection of the sarcoglycan proteins in this family was different from the pattern usually observed in LGMD 2C in Tunisia, as well as in other populations with different g-sarcoglycan mutations, in which a partial preservation in a- and b-sarcoglycan immunoreactivity is often seen [11,17,19,24–26]. Truncating mutations in b-sarcoglycan (Fig. 2) are mostly associated with a severe phenotype and are as virtual null mutations less helpful for predictions about functional protein domains. Thus, the identification of homozygous missense mutations in b-sarcoglycan is particularly valuable for attempts at structure/function predictions. The novel mutation described here affects the same nucleotide as the one
Fig. 2. Schematic summary of b-sarcoglycan mutations published so far plus additional unpublished observations. The single cysteine residue is emphasized. Missense mutations are indicated by solid arrows above the schematic of the molecule, and truncating mutations by broken arrows below the molecule. The Tunisian Arg91Leu mutation is boxed. Missense mutations that appear in homozygous form are underlined. References to the mutations are given in brackets and follow the numbering in the reference section. Mutations designated by [*] refer to unpublished observations (CGB and LMK). Details are given in the main text.
196
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197
described by us in a Brazilian pedigree with LGMD 2E, where G272 → C results in an Arg91Pro change [20]. The phenotype in the Brazilian family appears to be somewhat more severe with onset at or before 6 years of age and loss of ambulation by 10 years in all three affected patients. There was also total absence of the sarcoglycan complex by immunohistochemistry. The Brazilian Arg91Pro change not only abolishes a positive charge in the immediate extracellular domain, it also interrupts the betasheet structure in that part of the molecule, whereas the Tunisian Arg91Leu change abolishes the charge, but does not interrupt the beta sheet. This difference may contribute to the somewhat milder course of the disease in the Tunisian family reported here. However, such assumptions have to be made with caution, since there is known inherent variability of severity within the sarcoglycan disorders even on the basis of an identical mutation [3,11]. Indeed, there also is a suggestion of variability of severity in our family, as shown by the age of loss of ambulation in patient 1 (17 years) compared with patient 2 (12 years). Given the vicinity of these mutations to the single extracellular cysteine residue in b-sarcoglycan (Fig. 2) we speculate further that the observed disintegra-tion of the sarcoglycan complex may possibly be related to interference with disulfidebond mediated cross-linking between b-sarcoglycan and another component of the complex. a-, g- and d-sarcoglycan also have single cysteine residues in the extracellular domain that could function as cross-linking partners to bsarcoglycan, thus participating in the assembly and/or maintenance of the sarcoglycan complex. It is of interest to note that there appears to be a small cluster of missense mutations in this immediate extracellular domain encoded on exon 3 of b-sarcoglycan (Fig. 2). In all cases examined by us, we also found a total absence of the complex by immunohistochemistry [20,26]. These mutations appear to lead to a rather severe phenotype [20] and may interfere with the assembly and/or maintenance of the complex in a way similar to the mutations at residue 91. On the other hand, the recurrent mutation Ser114Phe located slightly further-out in the extracellular domain is much more variable in clinical expression, allowing ambulation into adult life in some patients [5] (C.G. Bo¨nnemann and L.M. Kunkel, unpublished data), while presenting with a severe childhood muscular dystrophy in another case [27]. Clinical variability is also seen in the Thr151Ala mutation found in the Indiana Amish population [8]. Tyr184Cys likewise presented with a somewhat milder phenotype [28], whereas Thr182Ala was more severe [27]. The intracellular Gln11Glu mutation was associated with a clearly severe phenotype [27], pointing to another potentially important domain of the molecule. Although genotype/phenotype correlations in the sarcoglycan genes have to be qualified by the intrinsic variability of the disease itself, they nonetheless may guide biochemical research into structure/function correlations in the corresponding proteins and the sarcoglycan complex as a whole.
Acknowledgements We would like to thank Richard Bennet, Julie Scarfo and Ivan Guerrero for expert help with sequencing as well as the members of the Kunkel lab for helpful discussions. The project described was supported by grant number 5 RO1 NS 23740-09 of the NINDS to L.M.K. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NINDS. L.M.K. is an investigator of the Howard Hughes Medical Institute.
References [1] Bushby KMD. Diagnostic criteria for the limb-girdle muscular dystrophies: report of the ENMC consortium on limb-girdle dystrophies. Neuromusc Disord 1995;5:71–74. [2] Bo¨nnemann CG, McNally EM, Kunkel LM. Beyond dystrophin: current progress in the muscular dystrophies. Curr Opin Pediatr 1996;8:569–582. [3] McNally EM, Passos-Bueno R, Bo¨nnemann CG, et al. Mild and severe muscular dystrophy caused by a single g-sarcoglycan mutation. Am J Hum Genet 1996;59:1040–1047. [4] Eymard B, Romero NB, Leturcq F, et al. Primary adhalinopathy (alpha-sarcoglycanopathy):clinical, pathologic, and genetic correlation in 20 patients with autosomal recessive muscular dystrophy. Neurology 1997;48:1227–1234. [5] Ginjaar HB, vd Kooi A, Ceelie H, et al. Sarcoglycanopathies in Dutch patients with autosomal recessive limb girdle muscular dystrophies. Neuromusc Disord 1997;7:440,GP1B.6. [6] Roberds SL, Leturcq F, Allamand V, et al. Missense mutations in the adhalin gene linked to autosomal recessive muscular dystrophy. Cell 1994;78:625–633. [7] Bo¨nnemann CG, Modi R, Noguchi S, et al. b-Sarcoglycan (A3b) mutations cause autosomal recessive muscular dystrophy with loss of the sarcoglycan complex. Nat Genet 1995;11:266–273. [8] Lim LE, Duclos F, Broux O, et al. b-Sarcoglycan (43 DAG): characterization and involvement in a recessive form of limb-girdle muscular dystrophy linked to chromosome 4q12. Nat Genet 1995;11:257–265. [9] Noguchi S, McNally EM, Ben Othmane K, et al. Mutations in the dystrophin-associated protein g-sarcoglycan in chromosome 13 muscular dystrophy. Science 1995;270:819–822. [10] Nigro V, de Sa Moriera E, Piluso G, et al. The 5q autosomal recessive limb-girdle muscular dystrophy (LGMD 2F) is caused by a mutation in the d-sarcoglycan gene. Nat Genet 1996;14:195–198. [11] Ben Hamida M, Ben Hamida C, Zouari M, Belal S, Hentati F. Limbgirdle muscular dystrophy 2C: clinical aspects. Neuromusc Disord 1996;6:493–494. [12] Ben Hamida M, Fardeau M, Attia N. Severe childhood muscular dystrophy affecting both sexes and frequent in Tunisia. Muscle Nerve 1983;6:469–480. [13] Ben Othmane K, Ben Hamida M, Pericak-Vance MA, et al. Linkage of Tunisian autosomal recessive Duchenne-like muscular dystrophy to the pericentromeric region of chromosome 13q. Nat Genet 1992;2:315–317. [14] Azibi K, Bachner L, Beckmann JS, et al. Severe childhood autosomal recessive muscular dystrophy with the deficiency of the 50 kDa dystrophin-associated glycoprotein maps to chromosome 13q12. Hum Mol Genet 1993;2:1423–1428. [15] el Kerch F, Sefiani A, Azibi K, et al. Linkage analysis of families with severe childhood autosomal recessive muscular dystrophy in Morocco indicates genetic homogeneity of the disease in North Africa. J Med Genet 1994;31:342–343.
C.G. Bo¨nnemann et al. / Neuromuscular Disorders 8 (1998) 193–197 [16] Ben Hamida C, Belal S, Hchaichi O, Hentati F, Ben Hamida M. Linkage analysis of Tunisian Duchenne-like muscular dystrophies and other limb-girdle muscular dystrophies. 30/31 ENMC International workshop on LGMD/SCARMD. Naarden, The Netherlands, 1995. [17] Ben Hamida M, Ben Hamida C, MDA LGMD Workshop, Minneapolis, MA, 1995. [18] Azibi K, Chaouch M, Reghis A, et al. Genetic and allelic heterogeneity of LGMD in North Africa. Neuromusc Disord 1997;7:441. [19] Ben Hamida C, Zouari M, Amouri R, et al. Phenotype-genotype correlation in Tunisian patients with not linked to chromosome 13 autosomal recessive muscular dystrophy. MDA LGMD Genetics Workshop, Tunis, Tunisia, 1996. [20] Bo¨nnemann CG, Passos-Bueno R, McNally EM, et al. Genomic screening for b-sarcoglycan mutations: missense mutations may cause severe limb-girdle muscular dystrophy type 2E (LGMD 2E). Hum Mol Genet 1996;5:1953–1961. [21] Little S. ARMS analysis of point mutations, in: Dracopoli NC, Haines DL, Korf BR, et al., editors. Current Protocols in Human Genetics, New York: Wiley, 1995;Suppl. 3:9.8.1–9.8.12. [22] Ferrie RM, Schwarz MJ, Robertson NH, et al. Development, multi-
[23]
[24]
[25]
[26]
[27]
[28]
197
plexing, and application of ARMS tests for common mutations in the CFTR gene. Am J Hum Genet 1992;51:251–262. Hentati F, Ben Hamida C, Belal S, et al. Immunohistochemical analysis of dystrophin-associated glycoproteins in Duchenne-like autosomal recessive muscular dystrophies in Tunisia. Vth European Congress of Neuropathology, Paris, France, 1996. McNally EM, Duggan DJ, Gorospe JR, et al. Mutations in the carboxyl-terminus of g-sarcoglycan cause muscular dystrophy. Hum Mol Genet 1996;5:1841–1847. Sewry CA, Taylor J, Anderson LV, et al. Abnormalities in alpha-, beta- and gamma-sarcoglycan in patients with limb-girdle muscular dystrophy. Neuromusc Disord 1996;6:467–474. Vainzof M, Passos-Bueno MR, Moreira ES, et al. The sarcoglycan complex in the six autosomal recessive limb-girdle muscular dystrophies. Hum Mol Genet 1996;5:1963–1969. Duggan DJ, Gorospe JR, Fanin M, Hoffman EP, Angelini C. Mutations in the sarcoglycan genes in patients with myopathy. N Engl J Med 1997;336:618–624. dos Santos MR, Vieira EM, Ribeiro MG, Santos MM, Gomes R, Guimara¨es A. Known and novel sarcoglycan gene mutations in Portoguese patients. Neuromusc Disord 1997;7:439,GP1B.4.