Background
Limb-girdle muscular dystrophies (LGMD) are characterised by wide clinical and genetic heterogeneity. Considering this factor, achieving a precise diagnosis can be difficult and requires a comprehensive clinical and laboratory approach. LGMD are classified on the basis of an inheritance pattern and a genetic cause. Limb-girdle muscular dystrophies of type 1 (LGMD1) include forms of the disorder that have an autosomal dominant inheritance associated with a mutation in one of at least 8 genes (LGMD1A-1H). Limb-girdle muscular dystrophies of type 2 (LGMD2) include forms with an autosomal recessive inheritance and mutations in one of at least 16 genes (LGMD2A-Q).
It is helpful to take into account the geographical and ethnic origins of patients in differential diagnosis, since the relative local frequency of the different forms of LGMD varies considerably [
1]–[
4]. The clinical course of LGMD ranges from severe to milder forms with different age of onset and progression even within the same family [
5]. Diagnosis of LGMD relies on a combination of clinical findings, results of histopathological examination of muscle biopsies (including the analysis of muscle proteins using immunohistochemistry and/or immunoblotting), followed by DNA sequencing to identify the primary mutation, which is essential for the provision of genetic and prognostic counselling.
The most frequent form of LGMD2 in Europe is probably LGMD2A, caused by mutations in the calpain-3 gene (
CAPN3). This gene encodes a muscle-specific member of the family of Ca
2+-activated neutral proteases that is important for muscle remodelling [
6]–[
8]. A marked clinical heterogeneity is observed in calpainopathies. While null-type gene mutations are usually associated with absence of calpain-3 protein in muscle and a severe phenotype, missense-type mutations are associated with unpredictable effects at both the protein and the phenotype levels [
9]–[
11].
Mutations in the fukutin-related protein gene (
FKRP) result in two distinct allelic diseases: severe congenital muscular dystrophy and LGMD2I [
12]. This gene encodes a putative Golgi-resident glycosyltransferase that is involved in posttranslational glycosylation of α-dystroglycan. The missense mutation c.826C > A, p.Leu276Ile is particularly common in LGMD2I patients and has been reported to confer a relatively mild phenotype when present in the homozygous state [
13]–[
15].
LGMD2C-2F is due to mutations in the genes encoding the components of the sarcoglycan (SG) complex. This complex, composed of 5 glycoproteins (α-, β-, γ-, δ-, ε-SG), is a member of the dystrophin-associated glycoprotein (DAG) complex localised to the sarcolemma of muscle fibres, which acts as a link between the extracellular matrix and the cytoskeleton, confers structural stability, and protects the sarcolemma from mechanical stress developed during muscle contraction. The clinical phenotype of sarcoglycanopathies ranges from a severe Duchenne-like dystrophy to a relatively mild LGMD [
16]. Although the relative frequency of mutations in the different sarcoglycan genes varies from population to population, α-sarcoglycanopathy (LGMD2D) appears to be the most frequent [
17].
Recently, recessive mutations in the anoctamin 5 gene (
ANO5) were identified as a cause of LGMD2L and non-dysferlin Miyoshi myopathy [
18],[
19].
ANO5 encodes a member of the anoctamin family of proteins which contains eight transmembrane domains. Dominant mutations in the
ANO5 gene are associated with the skeletal disorder gnathodiaphyseal dysplasia [
20]. While the role of ANO5 is unknown, ANO1 and ANO2, which share significant sequence homology with ANO5, are known to be calcium-activated chloride channels [
21]–[
23]. In one study [
24], analysis of
ANO5 was performed in a group of 59 British and German probands, and the mutation c.191dupA was found in 15 patients, homozygously in 11 and in compound heterozygosity with another
ANO5 variant in the rest. An intragenic SNP and an extragenic microsatellite marker were in linkage disequilibrium with the mutation, suggesting a founder effect in the populations analysed.
In this study, we determined the frequency of LGMD subtypes within a cohort of Czech LGMD2 patients using mutation analysis of the CAPN3, FKRP, SGCA, and ANO5 genes. We also describe two patients with mutations in the gene encoding dysferlin (DYSF) and a patient with mutations in the gene encoding β-sarcoglycan (SGCB). These mutations were identified using Sequence capture and targeted resequencing (SeqCap-TR), a method for the capture and enrichment of selected genomic regions from full genomic DNA in a single step which, in association with targeted resequencing, allows one to focus on genomic regions of interest to discover causative mutations.
Results and discussion
Considering the wide clinical and genetic variability of LGMD, determination of particular types is a comprehensive process. Muscle biopsies still represent an important and economical step in diagnosis, but protein changes documented by immunohistochemistry can be secondary or not pronounced and therefore a definitive diagnosis requires genetic analysis. Mutations of the
CAPN3 gene are the most common cause of LGMD2, and in the set of 218 Czech probands with a suspicion of LGMD2 two mutations in this gene were identified in 67 patients. In 4 patients only one
CAPN3 mutation was determined, and sequence analysis was completed by MLPA but no gene deletions/duplications were found (Table
1). In total we identified 37 different mutations of which 12 have been described only in Czech LGMD2A patients. The most frequent mutation among our LGMD2A probands is c.550delA which was detected in 65 mutant alleles from the total number of 138 (47.1%). The patients’ clinical and pathological findings (when available) are presented in (Additional file
1: Table S1).
Table 1
Mutations identified in Czech LGMD2A probands
1 - 23 | c.550delA/c.550delA | p.(Thr184Argfs*36)/p.(Thr184Argfs*36) |
24, 25 | c.550delA/c.245C > T | p.(Thr184Argfs*36)/p.(Pro82Leu) |
26 | c.550delA/c.328C > T | p.(Thr184Argfs*36)/p.(Arg110*) |
27 | c.550delA/c.509A > G | p.(Thr184Argfs*36)/p.(Tyr170Cys) |
28 | c.550delA/c.598_612del | p.Thr184Argfs*36/p.Phe200_Leu204del |
29 | c.550delA/c.1043delG | p.(Thr184Argfs*36)/p.(Gly348Valfs*4) |
30 | c.550delA/c.1069C > T | p.(Thr184Argfs*36)/p.(Arg357Trp) |
31 | c.550delA/c.1451T > C
| p.(Thr184Argfs*36)/p.(Leu484Pro)
|
32 | c.550delA/c.1465C > T | p.(Thr184Argfs*36)/p.(Arg489Trp) |
33, 34 | c.550delA/c.1468C > T | p.Thr184Argfs*36/p.(Arg490Trp) |
35 | c.550delA/c.1470delG
| p.(Thr184Argfs*36)/p.(Arg490Argfs*6)
|
36, 37 | c.550delA/c.1722delC
| p.Thr184Argfs*36/p.Ser575Leufs*20
|
38 | c.550delA/c.1823G > A | p.(Thr184Argfs*36)/p.(Arg608Lys) |
39 | c.550delA/c.1981delA | p.Thr184Argfs*36/p.Ile661* |
40 | c.550delA/c.2245A > C
| p.(Thr184Argfs*36)/p.(Asn749His)
|
41 |
c.1A > G/c.865C > T |
p.(Met1Val)/p.(Arg289Trp) |
42 | c.133G > A/c.133G > A | p.Ala45Thr/p.Ala45Thr |
43 | c.146G > A/c.1069C > T | p.(Arg49His)/p.(Arg357Trp) |
44 |
c.224A > G/c.224A > G
|
p.Tyr75Cys/p.Tyr75Cys
|
45 | c.245C > T/c.245C > T | p.Pro82Leu/p.Pro82Leu |
46 | c.245C > T/c.1800 + 1G > A
| p.(Pro82Leu)/splicing
|
47 | c.245C > T/c.1855C > T
| p.(Pro82Leu)/p.(Gln619*)
|
48 | c.245C > T/c.2314_2317del | p.Pro82Leu/p.Asp772Asnfs*3 |
49 | c.509A > G/c.509A > G | p.(Tyr170Cys)/p.(Tyr170Cys) |
50 | c.598_612del/c.598_612del | p.(Phe200_Leu204del)/p.(Phe200_Leu204del) |
51 | c.598_612del/c.640G > A | p.(Phe200_Leu204del)/p.(Gly214Ser) |
52 | c.598_612del/c.2245A > C
| p.Phe200_Leu204del/p.Asn749His
|
53 | c.1043delG/c.1094G > A
| p.(Gly348Valfs*4)/p.(Trp365*)
|
54 | c.1043delG/c.1343G > A | p.(Gly348Valfs*4)/p.(Arg448His) |
55 | c.1194-9A > G/c.1800 + 1G > A
| splicing/splicing
|
56 | c.1194-9A > G/c.2393C > A | splicing/p.(Ala798Glu) |
57 | c.1250C > T/c.1250C > T | p.(Thr417Met)/p.(Thr417Met) |
58 | c.1322G > A/c.1322G > A | p.Gly441Asn/p.Gly441Asn |
59 | c.1322delG/c.2114A > G | p.(Gly441Valfs*22)/p.(Asp705Gly) |
60 | c.1343G > A/c.2093G > A
| p.(Arg448His)/p.(Arg698His)
|
61 | c.1468C > T/c.2314_2317del | p.(Arg490Trp)/p.(Asp772Asnfs*3) |
62 |
c.1788_1793del/c.2242C > T |
p.Lys597_Lys598del/p.Arg748* |
63 - 66 | c.598_612del/c.1746-20C > G | p.(Phe200_Leu204del)/splicing |
67 |
c.614T > C/c.1746-20C > G |
p.(Leu205Pro)/splicing |
68, 69 | c.550delA/- | p.(Thr184Argfs*36)/- |
70, 71 | c.598_612del/- | p.(Phe200_Leu204del)/- |
Probands negative for
CAPN3 mutations (151) were analysed for the most common mutation in the
FKRP gene, p.Leu276Ile. The homozygous occurrence of this mutation was identified in 7 patients. In two patients heterozygous for p.(Leu276Ile), we were able to identify a second mutation: patient 79 carries the mutation p.(Pro316Arg) described in [
12] and patient 80 has the new mutation p.(Trp359Ser) (Table
2).
Table 2
Mutations identified in Czech LGMD2I, 2D, 2 L, 2B, and 2E probands
72 - 78 |
FKRP
| c.826C > A/c.826C > A | p.(Leu276Ile)/p.(Leu276Ile) |
79 |
FKRP
| c.826C > A/c.947C > G | p.(Leu276Ile)/p.(Pro316Arg) |
80 |
FKRP
| c.826C > A/c.1076G > C
| p.(Leu276Ile)/p.(Trp359Ser)
|
81 |
SGCA
| c.229C > T/c.229C > T | p.(Arg77Cys)/p.(Arg77Cys) |
82, 83 |
SGCA
| c.157 + 1G > A/c.850C > T | splicing/p.(Arg284Cys) |
84 |
SGCA
| c.229C > T/c.739G > A | p.(Arg77Cys)/p.(Val247Met) |
85 |
SGCA
| c.290A > G/c.303dupA
| p.(Asp97Gly)/p.(Gln101Glnfs*4)
|
86 |
SGCA
| c.229C > T/c.308 T > C | p.(Arg77Cys)/p .(Ile103Thr) |
87 |
ANO5
| c.191dupA/c.966A > T
| p.(Asn64Lysfs*15)/p.(Leu322Phe)
|
88, 89 |
ANO5
| c.191dupA/c.2272C > T | p.(Asn64Lysfs*15)/p.(Arg758Cys) |
90 |
DYSF
| c.3832C > T/c.5509G > T
| p.(Gln1278*)/p.(Asp1837Tyr)
|
91 |
DYSF
| c.509C > A/c.5907G > C/c.610C > T/c.1120G > C/ | p.(Ala170Glu)/p.(Trp1969Cys)/p.(Arg204*)/p.(Val374Leu)/ |
92 |
SGCB
| c.341C > T/c.341C > T | p.(Ser114Phe)/p.(Ser114Phe) |
In 142 probands without mutations in the
CAPN3 or
FKRP genes, we performed analysis of the
SGCA gene. Mutations were detected in 6 patients (Table
2). The most common mutation among our LGMD2D patients is p.(Arg77Cys), which was present in 4 mutant alleles. We also identified one new
SGCA mutation, c.303dupA (patient 85).
Probands negative for
CAPN3,
FKRP, and
SGCA mutations (136) were screened for the most common mutation in the
ANO5 gene, c.191dupA. This mutation was found heterozygously in 3 patients, and subsequent sequencing analysis of all exons and adjacent intron regions detected the mutation p.(Arg758Lys) in two of them (Table
2). This mutation was described in combination with c.191dupA in other studies [
24], [
31] and the associated phenotypes corresponded to distal non-dysferlin Miyoshi myopathy, unlike our patient 88 whose phenotype matches rather LGMD2L (a detailed description of the phenotype of patient 89 was not available). Analysis for this mutation was subsequently performed in all LGMD2 patients with unconfirmed diagnosis at the DNA level, but it was not detected. In patient 87 carrying c.191dupA on one allele, we did not identify a second mutation but only the polymorphism p.(Leu322Phe) described in LMDP. The patient’s clinical and pathological findings are presented in (Additional file
2: Table S2).
The aim of this study was to evaluate the relative proportion of the most frequent types of LGMD2 identified in other European countries in Czech probands with a suspicion of LGMD2. The results show that the frequency of the forms of LGMD2 which were analysed is 32.6% for LGMD2A (71 probands), 4.1% for LGMD2I (9 probands), 2.8% for LGMD2D (6 probands), and 1.4% for LGMD2L (3 probands). These results indicate that there is good agreement between the frequency of particular forms of LGMD2 in the Czech Republic and in Italy (respectively LGMD2A 31.1%, LGMD2I 7.4%, LGMD2D 8.4%, LGMD2L 2.1% [
28] and LGMD2A 28.4%, LGMD2I 6.4%, LGMD2D 8.3% [
2]), in contrast to studies from Denmark where the frequency of LGMD2A is significantly lower (12.1%) and that of LGMD2I significantly higher (38.4%) [
29] (Table
3).
Table 3
Frequency of LGMD2A, 2I, 2D, 2 L in European countries
This study | Czech | 218 LGMD2 probands | 71 (67 had two mutations, 4 one); 32.6%
| 9; 4.1%
| 6; 2.8%
| 3; 1.4%
|
| Italy | 190 LGMD probands | 59; 31.1%
| 14; 7.4%
| 16; 8.4%
| 4; 2.1%
|
| Italy | 155 LGMD probands | 44 (30 had two mutations, 14 one); 28.4%
| 10; 6.4%
| 13; 8.4%
| NI |
| Italy | 550 patients with LGMD, myopathy, or asymptomatic hyperCKemia | 102; 18.5%
| 16; 2.8%
| 37; 6.7%
| NI |
| Italy | 519 patients with LGMD, myopathy, or asymptomatic hyperCKemia | 94 (76 had two mutations, 18 one); 18.1%
| NI | NI | NI |
| Italy | 530 patients with muscular dytrophy | 141 (104 had two mutations and 37 one); 26.6%
| NI | NI | NI |
| Italy | 214 probands with muscular dystrophy or myopathy | NI | 13 (9 had two mutations, 4 one); 6.1%
| NI | NI |
| Germany | 98 probands with LGMD2 and 102 probands with asymptomatic or minimally symptomatic hyperCKemia | NI | 7; 3.5%
| NI | NI |
| Denmark | 99 LGMD2 patients | 12; 12.1%
| 38; 38.4%
| NI | NI |
| Dutch | 67 LGMD probands | 14; 20.9%
| 5; 7.5%
| NI | NI |
| Dutch | 68 LGMD probands | NI | NI | NI | 11; 16.2%
|
Histopathological and immunohistochemical analysis of muscle tissue performed in 42 of the 71 patients with identified mutation(s) in the
CAPN3 gene showed that 34 patients had a dystrophic and 8 had a myopathic pattern of muscle tissue. Immunoblotting of the CAPN3 protein was implemented in 33 patients. In patients with a mutation creating a protein termination codon (PTC) on both
CAPN3 alleles, immunoblotting showed the absence of 94 kDa CAPN3 (16 cases); in patients with a combination PTC/missense mutation we noted the absence (4 patients) or weak labelling (2 patients) of CAPN3; and in cases with a combination missense/missense mutation we detected the absence (2 patients), weak labelling (1 patient), or normal labelling (1 patient) of CAPN3. Other cases encompassing combinations of in-frame deletions, splicing mutations, or with a mutation on one
CAPN3 allele occurred exceptionally (Additional file
1).
Analysis of muscle tissue performed in 6 patients with mutations in the
FKRP gene showed a dystrophic pattern of muscle tissue. In two cases (patients 72 and 78), immunohistochemical analysis of alpha-dystroglycan was performed and both patients had a deficit of this protein. Three patients with mutations in the
SGCA gene showed absence or weak labelling of alpha, beta, and gamma-sarcoglycans (patients 82, 85), but in contrast immunodetection of these proteins was normal in patient 84 (Additional file
2).
In 2013, we introduced the SeqCap-TR method for genetic diagnosis of neuromuscular disorders and this approach was applied in 20 patients, 4 of whom had the LGMD2 phenotype and were negative for
CAPN3,
FKRP,
SGCA, and
ANO5 mutations. We identified two patients with LGMD2B (mutations in the
DYSF gene), one patient with LGMD2E (mutations in the
SGCB gene), and one patient is without identified causal mutations as yet. Patient 90 is a compound heterozygote for two
DYSF mutations (genotype p.[Gln1278*]; [Asp1837Tyr]). Compound heterozygous and homozygous occurrence of p.(Gln1278*) was described in [
36], in the case of a homozygous patient an initial phenotype corresponded to Miyoshi myopathy. The mutation c.5509G > T, p.(Asp1837Tyr) was not reported so far, but c.5509G > A, p.(Asp1837Asn) was described in several studies. In [
37] a patient carried the genotype p.[Asp1837Asn]; [Trp1968*] and the phenotype of LGMD2B, and in [
38] and [
39] patients with the genotype p.[Asp1837Asn]; [Tyr522*] had Miyoshi myopathy. In cases of homozygous occurrence of this mutation, phenotypes of Miyoshi myopathy [
40]–[
42] and also of LGMD2B [
42] were observed. The phenotype of our patient 90 corresponded rather with LGMD2B. Patient 91 carries 4 mutations in the
DYSF gene, p.(Ala170Glu), p.(Arg204*), p.(Val374Leu), and the novel mutation p.(Trp1969Cys). The mutation p.(Ala170Glu) is described in the LMDP as unknown pathogenic, probably pathogenic, or pathogenic and p.(Val374Leu) as probably not pathogenic or pathogenic. The evaluation of these missense mutations using
in silico tools is described in (Additional file
3: Table S3) together with the evaluation of selected missense mutations identified in the genes analysed. DNA analysis was performed in patient’s parents; her father carries p.(Arg204*) and p.(Val374Leu), and her mother carries p.(Ala170Glu) and p.(Trp1969Cys). On the basis of the
in silico analyses and inheritance in the patient’s family, we suppose that p.(Arg204*) and p.(Trp1969Cys) are causal mutations. Patient 92 is homozygous for the
SGCB mutation p.(Ser114Phe) described in the LMDP in association with LGMD2E. The patients’ clinical and pathological findings are shown in (Additional file
2: Table S2). None of the novel missense mutations were detected in 150 control DNA samples.
Acknowledgements
This work was funded by projects of the Internal Grant Agency of the Czech Ministry of Health (NT/14574-3), the Czech Ministry of Education “CEITEC – Central European Institute of Technology” (CZ.1.05/1.1.00/02.0068) and SuPReMMe (CZ.1.07/2.3.00/20.0045), and the Czech Ministry of Health for conceptual development of research organization 65269705 (University Hospital Brno, Brno, Czech Republic) and MH – DRO, University Hospital Motol, Prague, Czech Republic 00064203.
We would like to thank the physicians from Departments of Neurology and Medical Genetics in the Czech Republic (Drs. P. Doležalová, M. Forgáč, R. Gaillyová, D. Grečmalová, A. Gřegořová, M. Havlová, T Honzík, P. Ješina, Z. Kalina, K. Kalous, V. Křivková, R. Kutějová, M. Langová, T. Maříková, Š. Prášilová, D. Polendová, M. Soukupová, J. Staněk, M. Ševčíková, E. Šilhánová, S. Širůčková, D. Tenora, J. Zvolská) for providing us with their patients’ blood samples.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KS, DS, JZ carried out the molecular genetic studies. LM, PV, RM, SV, JH, HO, ND, PS performed clinical diagnostics of LGMD patients. MH, JZ, OS were involved in pathological analysis of muscle tissue. LF participated in design study and manuscript preparation. All authors read and approved the final manuscript.