Neuromuscular Disorders 18 (2008) 137–145 www.elsevier.com/locate/nmd
LAMA2 mRNA processing alterations generate a complete deficiency of laminin-a2 protein and a severe congenital muscular dystrophy q Olfa Siala a
a,b
, Nacim Louhichi a, Chahnez Triki c, Madeleine Morinie`re b, Faiza Fakhfakh a,*,1, Faouzi Baklouti b,*,1
Laboratoire de Ge´ne´tique Mole´culaire Humaine, Faculte´ de Me´decine de Sfax, Avenue Majida Baklouti-Boulila 3029 Sfax, Tunisia b Universite´ Lyon 1, CNRS, UMR5534, Centre de ge´ne´tique mole´culaire et cellulaire, Villeurbanne, F-69622, France c Service de Neurologie, C H U Habib Bourguiba, 3029 Sfax, Tunisia Received 29 May 2007; received in revised form 31 August 2007; accepted 6 September 2007
Abstract An increasing number of genomic variations are no more regarded as harmless changes in protein coding sequences or as genetic polymorphisms. Studying the impact of these variations on mRNA metabolism became a central issue to better understand the biological significance of disease. We describe here a severe congenital muscular dystrophy (CMD) with lumbar scoliosis and respiratory complications in a patient, who died at the age of 10. Despite a poor linkage to any form of CMD, total deficiency of laminin-a2 rather suggested the occurrence of an MDC1A form. Extensive analysis of LAMA2 gene revealed two novel mutations: a (8007delT) frameshift deletion in exon 57, and a de novo 7 nt deletion in intron 17. Using an ex vivo approach, we provided strong evidence that the intron mutation is responsible for complete exon 17 skipping. The mutations are in trans and they each generate a nonsense mRNA potentially elicited to degradation by NMD. We further discuss the impact of mRNA alterations on the subtle phenotypic discrepancies. 2007 Elsevier B.V. All rights reserved. Keywords: Laminin; MDC1A; Congenital muscular dystrophy; mRNA splicing; NMD
1. Introduction Abbreviations: CMD, congenital muscular dystrophy; NMD, nonsensemediated mRNA decay; SSCP, Single strand conformation polymorphism. q OS performed the ex vivo splicing analysis, the molecular genetic studies and the immunohistochemical analyses, and she drafted the manuscript. NL carried out the linkage analyses and participated in the molecular identification of the mutations. CT provided the clinical explorations, and participated in the immunohistochemical studies. MM provided assistance and training to perform part of the molecular analyses and ex vivo splicing assays. FF supervised and conceived the genetic studies, and drafted the manuscript. FB supervised and conceived the molecular studies, and drafted the manuscript. All authors read and approved the final manuscript. * Corresponding authors. Tel.: +216 98 460 571; fax: +216 74 461 403 (F. Fakhfakh); tel.: +33 4 72 43 29 28; fax: +33 4 72 43 26 85 (F. Baklouti). E-mail addresses:
[email protected] (F. Fakhfakh),
[email protected] (F. Baklouti). 1 The last two authors should be regarded as joint last authors. 0960-8966/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.nmd.2007.09.003
Autosomal recessive congenital muscular dystrophy linked to chromosome 6 (MDC1A) is caused by mutations in LAMA2 gene encoding the laminin a2 chain (formerly named merosin) [1]. MDC1A represents approximately 50% of congenital muscular dystrophy (CMD) cases leading to a clinically homogeneous subgroup, generally characterized by a total deficiency of laminin-a2 and leading to severe phenotype [2]. A partial laminin-a2 chain deficiency was also reported in a subgroup of patients with a primary laminin-a2 deficiency caused by LAMA2 gene mutations. Overall, the clinical phenotype in these patients was milder than in patients with complete deficiency of laminin a2 [3]. The most severe disease courses in MDC1A are marked by neonatal hypotonia and weakness, no independent ambulation due to severe contractures, an
138
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
elevated level of serum creatine kinase (CK) and a striking increased signal of white matter on T2 weighted magnetic resonance imaging (MRI), generally without mental retardation or structural brain involvement [4]. Diagnosis of MDC1A form is made by the contribution of clinical features, immunohistochemistry and genetic analyses; it is thought that the complete laminin-a2 deficiency is a well-defined entity without phenotypic heterogeneity (see Section 4). Laminin-211 (named according to the recent nomenclature for laminins [5]) is the most abundant laminin in muscle. It is composed of a2, b1 and c1 forms of laminin chains. Laminin-211 is also expressed in schwann cells, synaptic basal lamina of peripheral nerves, heart, trophoblasts and skin [6]. The supramolecule structure of a2 chain allows its interaction with cell receptors, essentially a-dystroglycan in the dystrophin–glycoprotein complex [7]. Several LAMA2 gene mutations were reported in MDC1A patients, including nonsense, frameshift and splice site mutations [8]. These mutations affect the entire LAMA2 gene, with no particular ‘‘hot spot’’ sequences. A growing body of evidence has accumulated regarding the functional consequences of splicing defects in human pathology ([9,10]; see also [11,12] and included references). Considering the increasing number of diseases derived from mRNA processing alterations, and the increasingly complex interactions that occur co-transcriptionally, it became inevitable to address the connections between mRNA metabolism and functional repercussions to unravel the pathophysiology of human diseases, including cancers [11,13]. We and others have developed minigene-based technology to address how a genomic variation can alter pre-mRNA processing, and generate different qualitative and quantitative effects ([14,15], for review [16]). These approaches meet their full credit in the present work where we were dealing with post-mortem analysis of a de novo mutation, that turned to be an unusual splicing mutation. The patient displayed a severe MDC1A form with lumbar scoliosis and respiratory complications. Genetic analyses revealed two new mutations in LAMA2 gene, present in compound heterozygous state. The first is a de novo mutation at the vicinity of the 5 0 splice site (5 0 ss) of intron17 (c.2450+5_c.2450+11del); while the second mutation is a frameshift deletion in exon 57 (c.8007delT) of the maternal allele. Inheritance of the 2 altered genes led to total deficiency of laminin-a2 in skeletal muscle and to the patient’s death at the age of 10. We further discuss the connections between the defective mRNA metabolism and the clinical severity of the disease. 2. Methods Blood samples were collected from all family members and controls after informed consent. Biopsies from deltoid muscle were obtained after the consent of both the propositus’ family and the control individual, and then frozen at 80 C, awaiting investigation.
2.1. DNA extraction Genomic DNA was isolated from whole blood of the patient, her family members and 100 normal subjects originated from the same region. DNA extraction from blood leukocytes was performed according to a previously described protocol [17]. 2.2. Immunohistochemistry Frozen 5 lm sections of deltoid muscle biopsies from the patient and the control were thawed on glass slides. Immunohistochemical analyses were performed to study the expression of a2, b1 and c1 chains of laminin-211; laminin-a5; a dystroglycan and dystrophin, using the following primary antibodies: anti-laminin-a2 mAb, raised against the C-terminal fragment of the human 80 kDa laminin-a2 (mAb 1922 Chemicon, Temacula CA); anti-laminin-a2 mAb, raised against the N-terminal fragment of the human 300 kDa laminin-a2 (NCL-merosin Novo-Castra); anti-b1 laminin chain antibody (mAb 1928, chemicon); anti-c1 laminin chain antibody (mAb 1914, Chemicon); anti-a5 chain antibody (mAb 1924, Chemicon); anti-a dystroglycan antibody (VIA4-1-05298, Euromedex); and 2 anti-dystrophin antibodies (NCL Dys1 and Dys2, Novocastra). Muscle sections were incubated with primary antibodies for 1 h. All dilutions were made in phosphate buffered saline (PBS1X, pH 7.4). After washing with PBS, the samples were incubated with the appropriate fluorescein isothiocyanate (FITC) conjugated secondary antibody. 2.3. Linkage analysis Genetic linkage was performed on genomic DNA using microsatellite markers spanning the loci of the differents CMD forms (MEB, WWS, RSMD, FCMD and MDC1A forms). Additional 4 intragenic SNP(s) in LAMA2 gene were also used: 1905G > A, 2848A > G, 5551A > G, and 6286G > A. PCR amplification of microsatellite markers was carried out in 50 ll with 60 ng of genomic DNA, 20 pmol of each primer, 125 lM dNTPs, 1.5 mM MgCl2, 5 mM KCl, 10 mM Tris–HCl, pH 8.8, and 1 U of Taq DNA polymerase. PCR products were analyzed on 6% denaturing polyacrylamide gel, transferred onto N+Hybond membrane (Amersham Pharmacia Biotech), and hybridized with a poly (AC) probe labelled with a32P isotope. According to the notion of consanguinity in families, linkage analyses were performed on the basis of homozygosity in affected children. Additional genetic tests were also performed, when needed, using a panel of 11 highly polymorphic markers including (D3S1358, R01, D13S317, D16S539, D2S1338, D5S818, FGA, D8S1179, D21S11, D7S820 and CSF1PO). 2.4. PCR amplification and sequencing PCR amplification of the 65 exons of LAMA2 gene was performed using appropriate primers chosen so that at
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
least 30 to 50 bp of flanking intronic sequences are readable. Amplification was performed in a thermal cycler (Gene Amp PCR system 9700, Applied Biosystem) using touchdown method as previously described [18]. The direct sequencing of PCR products was performed with the ABI Prism Big Dye terminator cycle sequencing Ready Reaction Kit (ABI Prism/PE Biosystems), and the products were resolved on ABI Prism 3100-Avant. The blast homology searches were performed using softwares available at the National Center for Biotechnology Information Web site.
139
2.7. Cell culture and transfection Hela cells were cultured in DMEM (Dulbecco’s modified Eagle’s medium (Invitrogen) with 10% of fetal calf serum (FCS) in four well plates. Cells were transfected with 1 lg of wildtype or mutated minigene constructs using FuGENE 6 Transfection Reagent (Roche) according to manufacturer’s procedures. Stably transfected cells were selected for 10–12 days in the same medium containing 600–800 lg G 418/ml (Geneticin, invitrogen). 2.8. RT-PCR analysis
2.5. SSCP analysis Single strand conformation polymorphism (SSCP) was used as an alternate approach to test the genomic mutations in other members of the families. Two microliters of PCR products were diluted (1:4) in formamide buffer, denatured for 5 min at 95 C, and chilled on ice. DNA samples were loaded onto 10% polyacrylamide:bisacrylamide (50:1; Bio-Rad Laboratories Inc., Hercules, California, USA) gel in 1· TBE buffer pH 7.7, and run at 45 V for 16 hours at 4 C. The bands were stained with SYBR Green I (Amersham) and SSCP pattern was visualized on a FluoroImager 595 system (Molecular Dynamics, Sunnyvale, CA). 2.6. Minigene constructs Amplification of LAMA2 exon 17 and intron boundaries in the patient was performed using a forward primer (F2:5 0 -ACAG/GTAACCACGCAATATCATGTATTTTC TTAAGG-3 0 ) and a reverse primer (R2: 5 0 -TGTGCTAG/ CATGCACATGTATGTGTTTGTATA-3 0 ). Sequence mismatches were introduced as to create BstEII and NheI restriction sites (bolded letters), respectively. These primers lead to a 760 bp PCR product containing 320 bp of intron 16 and 314 bp of intron 17. PCR was performed in 50 ll with 0.1 lg of genomic DNA, 5 ll of 10· buffer (50 mM Tris–HCl, pH 9.2, 160 mM (NH4)2SO4, 22.5 mM MgCl2, 2% DMSO, and 1% Tween 20), 10 mM dNTP, 20 pmol of each primer and 2 U of Taq DNA polymerase. PCR conditions were as follows: 5 min at 96 C followed by 35 cycles, each consisting of: 40 s at 95 C, 45 s at 62 C and 50 s at 72 C, then a final elongation at 72 C for 10 min was required. The PCR products were digested with BstEII and NheI restriction endonucleases, and inserted at the BstEII/NheI site of the splicing cassette p(13,17)/CMV, that has been designed to contain the 2 adjacent constitutive exons 13 and 17 of human 4.1R gene with their downstream and upstream flanking intron sequences, respectively [14]. Intron 17 deletion abolishes a MaeIII restriction site. We used this feature to characterize the recombinant plasmids, and to distinguish between the patient’s alleles. The constructs were further sequenced to ascertain the absence of additional sequence changes.
Cells were washed twice with PBS1X, collected by trypsinization, and centrifuged. Total RNA extraction was performed using TRIZOL Reagents (Invitrogen), according to manufacturer’s protocol. RT-PCR amplification was carried out using forward and reverse primers within the cassette’s upstream (UE, 4.1R exon 13) and downstream (DE, 4.1R exon 17) exons, as previously described [14]. RT-PCR products were gel purified and directly sequenced. 3. Results 3.1. Complete deficiency of laminin-211 The patient is the sister of two unaffected brothers, born after an uneventful pregnancy and term deliverancy from consanguineous parents from the south of Tunisia. Clinical explorations showed neonatal and generalized muscular hypotonia and weakness, severe contractures and mild mental retardation, without calf hypertrophy and respiratory complications. The deep tendon reflex was abolished and the foot exhibited equinovarus deformity. The patient stood unaided at 36 months but developed lumbar scoliosis (Fig. 1a); she spoke at 4 years of age, and she never walked. The level of serum CK was elevated (1000 UI/L) and brain T2-weighted MRI revealed a striking increased signal of the periventricular white matter without structural brain involvement (Fig. 1b). At the age of 6, the proband showed respiratory complications and the lumbar scoliosis had become more pronounced. At the age of 10, the patient died after respiratory complications and severe amyotrophy. Immunostaining of muscle sections in the patient revealed total deficiency of laminin-a2 using two antibodies against the N-terminal and the C-terminal region of the protein (Fig. 2a–d). A total absence of immunoreactivity was also observed with antibodies directed against laminin-b1 (Fig. 2e and f) and c1 (Fig. 2g and h). However, there was a normal expression of dystrophin around the muscle fibers (not shown), whereas a dystroglycan was virtually absent (Fig. 2k and l). Moreover, the use of a laminin a5-specific antibody showed a slight overexpression, if any, of laminin-a5 in blood vessels comparing to the control (Fig. 2i and j).
140
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
Fig. 1. Clinical features. (a) The patient suffered severe muscle weakness, hypotonia, contractures, marked lumbar scoliosis, and mild mental retardation. (b) T2-weighted MRI showing white matter abnormalities.
Fig. 2. Immunohistochemistry analysis. Immunostaining of skeletal muscle biopsies from control (upper panels) and studied patient (lower panels), with the antibodies depicted on the bottom. Note the total deficiency of laminin-a2, b1 and c1 in the patient comparing to the control.
3.2. Inconclusive genetic linkage to LAMA2 gene Linkage analyses using markers flanking different loci of CMD forms (MEB, WWS, RSMD, and FCMD) showed no homozygosity in the patient suggesting the exclusion of all these forms (not shown). The total deficiency of laminin-a2 showed in immunohistochemistry, prompted us to further analyze the LAMA2 locus. Four additional intragenic SNP(s) were used, together with the microsatellite markers flanking LAMA2 locus. Unexpectedly, the haplotyping analysis revealed no homozygous linkage to MDC1A form (Fig. 3). Although the parents are consanguineous, they display 4 different haplotypes at the LAMA2 locus. These findings suggest either the exclusion of the MDC1A form, or the presence of two compound heterozygous mutations within LAMA2 gene. 3.3. Mutation analysis Despite the absence of linkage to LAMA2 gene, clinical features, and total deficiency of laminin-a2, described so far only in MDC1A form, both suggest the occurrence of
two compound heterozygous mutations within LAMA2 gene. The 65 exons of LAMA2 gene and their intron boundaries were amplified and sequenced: sequencing of exon 57 in the patient revealed a single base deletion at position 8007 (c.8007delT). The mutation is novel (submitted in LAMA2db-ID under an accession number: LAMA2_00140) and occurs at the heterozygous state in the patient, the mother and also in both unaffected brothers (Fig. 4). PCR amplification and sequencing of exon 57 consistently showed the absence of the mutation in 200 control chromosomes. The (8007delT) mutation induces a frameshift in the reading frame leading to the occurrence of a PTC at position (+103) of exon 58. No other nucleotide change has been detected within the coding sequence. However, PCR amplification and sequencing revealed a deletion of 7 bp (gtaacaa) from position +5 to position +11 of intron 17 (c.2450+5_c.2450+11del) (submitted in LAMA2db-ID under an Accession No. LAMA2_00183). Surprisingly, the deletion was present in the patient, but absent in the parents and all other family members (Fig. 4). A panel of 11 high polymorphic markers were
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
141
not able to analyze the endogenous mRNA. Therefore, we used a minigene approach to ask whether the small intronic deletion affects exon 17 splicing. A genomic fragment containing exon 17 and 300 bp of flanking intron 16 and intron 17 sequences was cloned into an efficient splicing cassette (see Section 2). Control and mutated minigene constructs were stably transfected in HeLa cells. RT-PCR analysis performed on total RNA extracted from cells transfected with the normal construct revealed a 374 bp fragment, corresponding to the expected splicing product containing exon 17 of LAMA2 gene (Fig. 6). Contrarily, RT-PCR on RNA from cells transfected with the mutated construct displayed a shorter band of 246 bp; direct sequencing of this PCR product revealed a total lack of exon 17 (Fig. 6). These results suggest that the 7 nt intronic deletion alters LAMA2 pre-mRNA splicing, and results in the total skipping of exon 17 (Fig. 6). Consequently, exon 17 skipping leads to a translation frameshift, and the occurrence of a PTC at position +4 within exon 18. 4. Discussion Fig. 3. Linkage analysis to LAMA2 gene. The affected child is represented by a black circle and her unaffected brothers by open squares. Their father and mother are represented by the open square and the circle respectively. Genotyping of five microsatellite markers (D6S1705, D6S407, D6S1572, D6S1620 and D6S1715) and five intragenic markers (WI2405, 1847 G > A, 1905 G > A, 2848 A > G, and 5551 A > G) showed no linkage to MDC1A form.
analyzed, and all of them ruled out a false paternity. On the other hand, SSCP analysis revealed the absence of the mutation in 100 unrelated controls originated from the same region (Fig. 5). Altogether, these results suggest that the (c.2450+5_c.2450+11del) is a de novo mutation. Whether this mutation is silent or could affect LAMA2 gene expression remains to be addressed.
3.4. The 7 nucleotide intronic deletion abolishes exon 17 splicing Mammalian 5 0 ss vary considerably but generally conform to the consensus AG/GURAGU (R=purine). The 7 nt deletion found in intron 17 is adjacent and overlaps the 5 0 ss consensus. The effect of (c.2450+5_c.2450+11del) mutation was first assessed using the consensus score calculation method to quantify the influence of the mutation on the formation of splicing loops [19] using the corresponding software available online at http:// ast.bioinfo.tau.aci/splicesiteframe.htm.expasy.ch. The calculated values of the numerical score for the altered sequence (TAA/gtaaact) were 57.36% as compared to 87.58% for the wild type splice donor site (TAA/gtaagta). A muscle biopsy was obtained from the patient, and it was dedicated to the immunostaining experiments. She was then in a critical stage of the disease, and we were
4.1. A very severe form of CMD due to complete absence of laminin-a2 In this study, we describe a severe MDC1A form in a Tunisian patient with total deficiency of laminin-a2. She died at the age of 10 after pronounced lumbar scoliosis and respiratory complications. Genetic and molecular studies revealed the occurrence of two novel compound heterozygous mutations in LAMA2 gene, consisting of a (8007delT) frameshift mutation in exon 57 arising from the maternal allele, and a de novo 7 nt deletion within intron 17 and overlapping the 5 0 ss consensus. We were unable to undoubtedly demonstrate that the de novo mutation occurs on the paternal chromosome. However, the total absence of laminin-a2 and the severe clinical picture strongly suggest that the two mutations are in trans, and therefore, the splicing mutation must arise from the paternal germ cells. It represents so far the second evidence for a de novo mutation in LAMA2 gene after the (5702delA) deletion in exon 38 [20]. A higher incidence of a paternal origin of de novo mutations is thought to derive from the high number of germ line cell divisions in the life history of a sperm relative to that of the egg, leading to DNA replication errors [21]. We believe the mutations described here both lead to massive mRNA degradation: the nucleotide deletion in exon 57 results in a frameshift and the occurrence of a PTC within exon 58. To test whether intron 17 mutation abolishes completely or only partially the 5 0 ss, we performed ex vivo functional assay that provided a compelling argument that the 5 0 ss is completely abolished. Skipping of exon 17 also generate a translation frameshift and the occurrence of a PTC. Both PTCs appear far upstream of the last exon-exon junction, and therefore they
142
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
Fig. 4. Mutational analysis of LAMA2 gene. Electrophoregrams showing in the patient and her unaffected brothers a single base deletion (8007delT) in exon 57 of LAMA2 gene inherited from the maternal allele, and indicated by an arrow. Sequencing of exon 17 and downstream intron of LAMA2 gene showed a heterozygous de novo (c.2450+5_2450+11del) mutation. This mutation was present in the patient but absent in all her relatives.
M
1
2
3 UE
17
DE WT
UE
17
UE
DE
DE
Mut
UE
17
DE
Fig. 6. Ex vivo splicing assay. Wildtype (WT) and mutant (Mut) constructs were stably transfected in HeLa cells, and exon 17 splicing pattern was analyzed by RT-PCR. Unaltered intron 17 led to the efficient and total inclusion of exon 17 (lane 2), whereas the 7 nt deletion within intron 17 (d) resulted in total skipping of the exon (lane 3). Sequencing analysis has confirmed these data, and allowed to deduce the splicing pathways, depicted on the right. M, size marker; lane 1, negative control (no reverse transcriptase); UE and DE, upstream and downstream exons of the cassette, respectively. Fig. 5. SSCP analysis of intron 17 mutation. LAMA2 exon 17 and its intron boundaries were amplified (lane 1) and several controls (lanes 2–8). Note the abnormal conformation with additional single strand bands (*). M, size marker; ss, single strands; ds, double strands.
would likely trigger rapid mRNA decay by NMD [22]. Translation of a small fraction of mRNA that would resist degradation, would result in a severely truncated laminin-211 lacking at least 2/3 of the functional domains (splicing mutation) and at best the G domain at the
C-terminus (frameshift mutation). The G domain serves as a ligand to a-dystroglycan, the extracellular component of the dystroglycan complex. The dystroglycans establish a bridge across the sarcolemma from the cytoskeleton and dystrophin within the cell to laminin-211 in the basal lamina [23]. Altogether, data obtained from mRNA analysis, consistently with the total absence of laminin-211 at the
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
muscle fibers upon IF experiments, suggest that translation is impaired because the mature mRNA molecules derived from both alleles are nonsense mRNAs that must be targeted to degradation. Moreover, laminin-a5 immunostaining showed a slight overexpression in blood vessels compared to the control; this result has already been reported in MDC1A form. It is thought that laminin a5 chain compensates for total laminin-a2 deficiency [8,24]. However, we were not able to observe a clear-cut expression at the basement membrane of muscle fibers.
4.2. Is the clinical heterogeneity of laminin-a2 deficiency correlated with translatable mRNA amount ? Several homozygous LAMA2 mutations have been reported. These mutations result in partial or virtually complete laminin-211 deficiency. Close comparison between clinical, histological and molecular features, suggest a subtle heterogeneity among the homozygous patients. In the present case, IF experiments showed a total deficiency of laminin-a2, together with a complete absence of b1 and c1 chains, probably due to the inability to form polymers in the extracellular matrix, in the absence of a2 chain. The fatal combination of these LAMA2 mutations led to death in early life. This very severe picture contrasts with earlier reports by Pegoraro et al. [25] and Prandini et al. [24]. In the first homozygous case, the authors described a complete deficiency of laminin-a2. However, laminin-b1 showed only a markedly decreased immunostaining around myofibers, whereas laminin-c1 appeared just slightly reduced relative to controls [25]. This form of CMD is due to nonsense mutation within exon 31. In fact, it occurs within an alternative motif consisting of a large part of exon 31. This leads to complete absence of the protein isoform translated from the nonsense exon-containing transcript, whereas the isoform lacking that motif is ‘‘normally’’ processed and translated. As a result, the small proportion of truncated protein is active and probably helps to reduce the severity. Similar feature has been encountered in a 4.1R variant resulting from a PTC within alternative exon 16. In this study, only cells expressing the nonsense exon 16 displayed a 4.1R mRNA deficiency [26]. The second case is a girl who carries a homozygous outof-frame deletion that encompasses the last two nucleotides of exon 56 of the LAMA2 gene but has a mild phenotype. She is still ambulant at age 13 years and shows white matter abnormalities on MRI [24]. laminin-a2 protein study revealed traces of the protein in her muscle biopsy using an antibody directed against domain IVa, suggesting that the mutated transcript is translated into a detectable amount of protein. The possibility remains that the out-of-frame deletion affects at least partially exon 56 splicing; skipping of this exon would rescue the reading frame, and would allow the production of a shorter but functional laminin-a2. In fact, the patient’s muscle biopsy
143
showed only a decrease in LAMA2 RNA relative to control subjects: 20% vs 100% [24]. 4.3. The intron deletion may affect the recruitment of spliceosomal components Functional studies by ex vivo splicing assays demonstrated that the (c.2450+5_c.2450+11del) mutation disclosed complete inactivation of the wildtype splicing donor site and caused a total skipping of exon 17. In LAMA2 gene, all exons are constitutive except exon 31, which is alternatively spliced leading to an isoform that is unevenly distributed in muscle fibers [25]. Exon skipping has been described in several occasions in association with laminin-211 deficiency. These include skipping of exon 15 [27], exon 25 [28], exon 31 [1] and exon 37 [3]. Of note, skipping of exon 15 occurs as a consequence of a (C > T) transition at position 2230 outside of the consensus splice site [27]. All these homozygous mutations causing exon skipping occurred in the 5 0 region of LAMA2 gene in patients who are mildly affected with MDC1A form. The clinical discrepancies between these patients and the patient we describe here, although dealing with mutations affecting the mRNA metabolism, is somehow puzzling. Within the 5 0 ss consensus, the fifth nucleotide of the intron is a (G) in 84% of 5 0 ss from about 400 vertebrate genes. A (G) at position +5 is also present in 70% of the introns in LAMA2 gene. The (c.2450+5_c.2450+11del) mutation disrupts the 5 0 consensus splice site of intron 17 of LAMA2 gene leading to total exon skipping. In fact, the deletion results in a AC dinucleotide at positions +5 and +6 of the consensus, instead of the original GT; and despite the conservation of the nucleotide nature at position +5 and +6 (purine–pyrimidine), the 5 0 ss was disrupted and caused total skipping of exon 17. A similar intronic deletion of 7 nt overlapping the 5 0 ss, has been described to cause the skipping of exon H of albumin, resulting in a deficiency of serum albumin in Nagase analbuminemic rats (NAR) [29,30]. The 5 0 ss is complementary to the 5 0 end of the U1 snRNA and recognition of the 5 0 ss involves base pairing with the U1 snRNA [31–33]. In fact, compensatory base-substitution in U1snRNA was found to suppress inactivation of the mutated 5 0 ss in several cases, including the intronic 7nt deletion causing albumin exon H skipping [34]. Consistent with these observations, in vitro studies have shown that a minimal base pairing between the 5 0 ss and U1snRNA is sufficient for its correct recognition by U1snRNP [35]. It also appears that a higher base pair match between U1 snRNA and the intronic portion of the 5 0 ss compensates for weak interaction of U1 snRNA and the exonic portion of the 5 0 ss, and vice versa [19]. 5. Conclusions It is well established that a splicing mutation that alters the highly conserved dinucleotide at either the 5 0 or the 3 0
144
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145
splice site, leads in virtually all cases to total exon skipping or intron inclusion, or/and use of cryptic splice sites [12]. Mutations elsewhere in the consensus sequences or in splicing enhancers or silencers, affect splicing with various degrees, the characterization of which became crucial for a better diagnosis of genetic disorders. Use of minigene ex vivo assays turned to be a powerful tool to address this issue. Such an approach exemplifies the importance to determine the actual impact of a genomic mutation on mRNA metabolism to better understand the pathophysiology of disease. The 7nt deletion alters the 3 0 end of the 5 0 ss consensus, and the adjacent sequence, which might be important for U6snRNP recruitment. The replacement of the U1 snRNA with the U6 snRNA at the 5 0 ss is believed to activate the spliceosome, bringing the catalytic site in close proximity of the 5 0 ss. However, controversy remains as to whether higher complementarity to the U1 snRNA stabilizes U1 snRNA binding to the 5 0 ss, but concurrently inhibits the assembly of the full spliceosome [35], or rather improves efficiency and increases 5 0 ss recognition without interference with the potential base pairing interaction between U6 snRNA and the 5 0 ss [36].
[10]
[11]
[12]
[13] [14]
[15]
[16] [17]
[18]
Acknowledgements This work was supported by funds from the Ministe`re de l’Enseignement Supe´rieur et de la Recherche, Tunisia, the CNRS (DGRST/CNRS), the Re´gion Rhoˆne-Alpes (MIRA), the Association Franc¸aise contre les Myopathies, and the Fondation pour la Recherche Me´dicale.
[19] [20]
[21]
References [22] [1] Helbling-Leclerc A, Zhang X, Topaloglu H, et al. Mutations in the laminin alpha 2-chain gene (LAMA2) cause merosin- deficient congenital muscular dystrophy. Nat Genet 1995;11(2):216–8. [2] Tome FM, Evangelista T, Leclerc A, et al. Congenital muscular dystrophy with merosin deficiency. C R Acad Sci III 1994;317(4):351–7. [3] Tezak Z, Prandini P, Boscaro M, et al. Clinical and molecular study in congenital muscular dystrophy with partial laminin alpha 2 (LAMA2) deficiency. Hum Mutat 2003;21(2):103–11. [4] Philpot J, Sewry C, Pennock J, Dubowitz V. Clinical phenotype in congenital muscular dystrophy: correlation with expression of merosin in skeletal muscle. Neuromuscul Disord 1995;5(4):301–5. [5] Aumailley M, Bruckner-Tuderman L, Carter WG, et al. A simplified laminin nomenclature. Matrix Biol 2005;24(5):326–32. [6] Leivo I, Engvall E. Merosin, a protein specific for basement membranes of Schwann cells, striated muscle, and trophoblast, is expressed late in nerve and muscle development. Proc Natl Acad Sci USA 1988;85(5):1544–8. [7] He Y, Jones KJ, Vignier N, et al. Congenital muscular dystrophy with primary partial laminin alpha2 chain deficiency: molecular study. Neurology 2001;57(7):1319–22. [8] Allamand V, Guicheney P. Merosin-deficient congenital muscular dystrophy, autosomal recessive (MDC1A, MIM#156225, LAMA2 gene coding for alpha2 chain of laminin). Eur J Hum Genet 2002;10(2):91–4. [9] Delhommeau F, Vasseur-Godbillon C, Leclerc P, et al. A splicing alteration of 4.1R pre-mRNA generates 2 protein isoforms with
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
distinct assembly to spindle poles in mitotic cells. Blood 2002;100(7):2629–36. Morinie`re M, Ribeiro L, Dalla Venezia N, et al. Elliptocytosis in patients with C-terminal domain mutations of protein 4.1 correlates with encoded messenger RNA levels rather than with alterations in primary protein structure. Blood 2000;95(5):1834–41. Buratti E, Baralle M, Baralle FE. Defective splicing, disease and therapy: searching for master checkpoints in exon definition. Nucleic Acids Res 2006;34(12):3494–510. Krawczak M, Thomas NS, Hundrieser B, et al. Single base-pair substitutions in exon-intron junctions of human genes: nature, distribution, and consequences for mRNA splicing. Hum Mutat 2007;28(2):150–8. Venables JP. Unbalanced alternative splicing and its significance in cancer. Bioessays 2006;28(4):378–86. Deguillien M, Huang SC, Morinie`re M, et al. Multiple cis elements regulate an alternative splicing event at 4.1R pre-mRNA during erythroid differentiation. Blood 2001;98(13):3809–16. Dreumont N, Poudrier JA, Bergeron A, et al. A missense mutation (Q279R) in the fumarylacetoacetate hydrolase gene, responsible for hereditary tyrosinemia, acts as a splicing mutation. BMC Genet 2001;2(1):9. Cooper TA. Use of minigene systems to dissect alternative splicing elements. Methods 2005;37(4):331–40. Kawasaki E. Sample preparation from blood cells and other fluids. In: Innis M, Gelfand D, Snisky J, White T, editors. PCR protocols. A guide to methods and applications. San Diego: Academic Press; 1990. Guicheney P, Vignier N, Zhang X, et al. PCR based mutation screening of the laminin alpha2 chain gene (LAMA2): application to prenatal diagnosis and search for founder effects in congenital muscular dystrophy. J Med Genet 1998;35(3):211–7. Carmel I, Tal S, Vig I, Ast G. Comparative analysis detects dependencies among the 5 0 splice-site positions. RNA 2004;10(5):828–40. Naom I, D’Alessandro M, Sewry CA, et al. Mutations in the laminin alpha2-chain gene in two children with early-onset muscular dystrophy. Brain 2000;123(Pt 1):31–41. Aretz S, Uhlhaas S, Caspari R, et al. Frequency and parental origin of de novo APC mutations in familial adenomatous polyposis. Eur J Hum Genet 2004;12(1):52–8. Nagy E, Maquat LE. A rule for termination-codon position within intron-containing genes: when nonsense affects RNA abundance. Trends Biochem Sci 1998;23(6):198–9. Michele DE, Campbell KP. Dystrophin–glycoprotein complex: posttranslational processing and dystroglycan function. J Biol Chem 2003;278(18):15457–60. Prandini P, Berardinelli A, Fanin M, et al. LAMA2 loss-of-function mutation in a girl with a mild congenital muscular dystrophy. Neurology 2004;63(6):1118–21. Pegoraro E, Fanin M, Trevisan CP, et al. A novel laminin alpha2 isoform in severe laminin alpha2 deficient congenital muscular dystrophy. Neurology 2000;55(8):1128–34. Maillet P, Dalla Venezia N, Lorenzo F, et al. A premature termination codon within an alternative exon affecting only the metabolism of transcripts that retain this exon. Hum Mutat 1999;14(2):145–55. Di Blasi C, He Y, Morandi L, et al. Mild muscular dystrophy due to a nonsense mutation in the LAMA2 gene resulting in exon skipping. Brain 2001;124(Pt 4):698–704. Allamand V, Sunada Y, Salih MA, et al. Mild congenital muscular dystrophy in two patients with an internally deleted laminin alpha2chain. Hum Mol Genet 1997;6(5):747–52. Esumi H, Takahashi Y, Sato S, et al. A seven-base-pair deletion in an intron of the albumin gene of analbuminemic rats. Proc Natl Acad Sci USA 1983;80(1):95–9. Shalaby F, Shafritz DA. Exon skipping during splicing of albumin mRNA precursors in Nagase analbuminemic rats. Proc Natl Acad Sci USA 1990;87(7):2652–6.
O. Siala et al. / Neuromuscular Disorders 18 (2008) 137–145 [31] Seraphin B, Kretzner L, Rosbash M. A U1 snRNA:pre-mRNA base pairing interaction is required early in yeast spliceosome assembly but does not uniquely define the 5 0 cleavage site. EMBO J 1988;7(8):2533–8. [32] Siliciano PG, Guthrie C. 5 0 splice site selection in yeast: genetic alterations in base-pairing with U1 reveal additional requirements. Genes Dev 1988;2(10):1258–67. [33] Zhuang Y, Weiner AM. A compensatory base change in U1 snRNA suppresses a 5 0 splice site mutation. Cell 1986;46(6):827–35.
145
[34] Hitomi Y, Sugiyama K, Esumi H. Suppression of the 5 0 splice site mutation in the Nagase analbuminemic rat with mutated U1snRNA. Biochem Biophys Res Commun 1998;251(1):11–6. [35] Lund M, Kjems J. Defining a 5 0 splice site by functional selection in the presence and absence of U1 snRNA 5 0 end. RNA 2002;8(2):166–79. [36] Freund M, Hicks MJ, Konermann C, et al. Extended base pair complementarity between U1 snRNA and the 5 0 splice site does not inhibit splicing in higher eukaryotes, but rather increases 5 0 splice site recognition. Nucleic Acids Res 2005;33(16):5112–9.
