Abstract
Introduction: Recessive mutations in the CAPN3 gene can lead to limb-girdle muscular dystrophy recessive 1 (LGMD R1). Targeted next-generation sequencing facilitates the discovery of new mutations linked with disease, owing to its ability to selectively enrich specific genomic regions. Methods: We performed targeted next-generation sequencing of all exons of the CAPN3 gene in 4 patients with sporadic limb-girdle muscular dystrophy (LGMD) and further analyzed the effects of the novel identified variant using various software tools. Results: We found 5 variants in CAPN3 gene in 4 patients, c.82_83insC (insertion mutation) and c.1115+2T>C (splicing mutation) are reported for the first time in CAPN3 (NM_000070.2). The bioinformatics analysis indicated that these two novel variants affected CAPN3 transcription as well as translation. Discussion: Our findings reveal previously unreported splicing mutation and insertion mutation in CAPN3 gene, further expanding the pathogenic gene profile of LGMD.
Introduction
Limb-girdle muscular dystrophy (LGMD) is a group of hereditary muscular diseases characterized by weakness and muscle atrophy in the pelvic, shoulder girdle muscles, and proximal limb muscles, which were proposed by Walton and Nattrass in 1954 [1]. According to the 229th ENMC international workshop, LGMD is classified into autosomal dominant forms (LGMD D) and autosomal recessive forms (LGMD R), and the latter forms predominate in terms of both number of subtypes and individual prevalence [2]. LGMD R1 (previously known as LGMD2A) was first identified in a population on the Réunion Island by Richard et al. [3], which usually occurs in the second decade of life but can vary from 2.5 to 50 years old [4, 5]. The phenotypic manifestations of LGMD R1 are similar to other LGMDs with symmetrical and progressive weakness of girdle and proximal limb muscles [6]. According to the distribution of affected muscle and the age at onset, LGMD R1 could be divided into three subtypes: pelvi-femoral phenotype, scapula-humeral phenotype, and hyperCKemia [7, 8].
LGMD R1 is caused by mutations in CAPN3 gene, which is located in the chromosomal region 15q15.1-q21.1 and contains 24 exons [9]. It encodes the calpains 3 protein, a skeletal-muscle-specific calpain that processes substrates through limited and specific proteolysis to facilitate Ca2+ signal transduction and regulate various protein functions in cells [10, 11]. The development of targeted next-generation sequencing greatly advanced the identification of CAPN3 variants in recent years. To date, more than 500 unique pathogenic or likely pathogenic variants of CAPN3 have been reported [12]. Any types of mutation of CAPN3 can lead to the occurrence of LGMDR1, such as missense mutation, frameshift mutation, nonsense mutation, deletions/insertion, splice site mutation. Among these variants, missense mutations were the most common [13, 14]. Moreover, different types of CAPN3 gene mutations in LGMD R1 patients have been extensively studied across various countries, highlighting substantial regional and national differences. For instance, while the variant c.2120A>G is predominantly seen in China, c.550delA is more common among European populations [15].
In the present study, we applied target next-generation sequencing in four non-consanguineous Chinese families with preliminary clinical diagnoses of calpainopathy. This study expands the genotypic and phenotypic spectrum of LGMD R1.
Methods
Patients
This retrospective study consisted of 4 non-consanguineous Chinese LGMD patients who were referred to the neurology department of the First Affiliated Hospital of Soochow University. The main symptoms of these patients were proximal limb weakness and atrophy, the initial diagnosis depends on sufficient clinical assessment evidence (family history, age of onset, typical signs, muscle MRI, blood creatine kinase), and the final diagnosis depends on genetic test results. It is noteworthy that their family members did not present any similar signs or symptoms.
Genetic Analysis
Genomic DNA samples were isolated from peripheral blood using DNA Isolation Kit (CW2553, CWBio, Taizhou, China) and their concentrations were determined by Qubit dsDNA HS Assay Kit (Q32851, Invitrogen, Carlsbad, CA, USA). The quality control of samples was performed by agarose gel (1%) electrophoresis. Hybridization was performed using the IDT xGen Exome Research Panel v1.0 (Integrated DNA Technologies, Inc., USA). Targeted DNA samples were sequenced by the Illumina NovaSeq 6000 System (Illumina, USA), and quality control was applied to the raw data (stored in FASTQ format) by Illumina Sequence Control Software (SCS). High-quality data were aligned to the human reference genome sequence hg19 using Burrows-Wheeler Aligner (v0.7.15) [16]. Consensus single nucleotide variants and insertion/deletions were filtered via Genome Analysis Toolkit v3.6 [17]. All single nucleotide variants and insertion/deletions information was annotated by the software ANNOVAR4. The pathogenicity of candidate mutations was analyzed based on the American College Medical Genetics and Genomics (ACMG) guidelines [18]. The CAPN3 mutations were finally selected according to their clinical relevance and pathogenicity.
Bioinformatic Analysis of Variant
Then, we use Splice AI (https://spliceailookup.broadinstitute.org/) [19] and Netgene2 (https://services.healthtech.dtu.dk/services/NetGene2-2.42/) [20] to predict the change of splice site of the variant. We utilized varSEAK Online (https://varseak.bio/) [21] to examine the sequence alterations following mutation and employed Adobe Illustrator to depict the structural diagram post-mutation. We used PyMOL software (http://www.pymol.org) to analyze the structure of calpain 3 in wild type and variants of c.82_83insC (p. Q28Pfs*6).
Results
Clinical Presentation
All patients presented with proximal symmetric muscle weakness in upper or lower extremities in limb-girdle, atrophy, scapular winging, and hyperCKemia. The clinical presentation of all probands is summarized in Table 1. None of the individuals appeared to be dependent on a wheelchair or unable to walk. Additionally, none of them exhibited any cardiac or respiratory symptoms. Muscle MRI studied showed diffuse muscle atrophy at different extent, and fatty replacement of lower limbs (P1 with missing information, shown in Fig. 1). Laboratory test results showed elevated serum creatine kinase levels (shown in Table 1).
Clinical and biological observations for the patients of the four families harboring the CAPN3 variant
Patients . | P1 . | P2 . | P3 . | P4 . |
---|---|---|---|---|
Sex | Male | Male | Female | Male |
Family history | Yes | No | No | No |
Age at onset, years | 10 | 25 | 35 | 29 |
Age at exfrom 4 unrelated amination, years | 22 | 65 | 48 | 33 |
Weakness | Low limbs | Low limbs | Proximal limbs | Low limbs |
Atrophy | Proximal limbs | Low limbs | Proximal limbs | Low limbs |
Scapular winging | Yes | No | Yes | No |
Walking aids | Yes | Yes | Yes | Yes |
Serum CK, U/L | 2,769 | 1,154 | 549.9 | 12,563 |
Patients . | P1 . | P2 . | P3 . | P4 . |
---|---|---|---|---|
Sex | Male | Male | Female | Male |
Family history | Yes | No | No | No |
Age at onset, years | 10 | 25 | 35 | 29 |
Age at exfrom 4 unrelated amination, years | 22 | 65 | 48 | 33 |
Weakness | Low limbs | Low limbs | Proximal limbs | Low limbs |
Atrophy | Proximal limbs | Low limbs | Proximal limbs | Low limbs |
Scapular winging | Yes | No | Yes | No |
Walking aids | Yes | Yes | Yes | Yes |
Serum CK, U/L | 2,769 | 1,154 | 549.9 | 12,563 |
Genotypes of Patients
Targeted next-generation sequencing was performed in all 4 probands. P1 carried a homozygous pathogenic variant, whereas P2, 3, and 4 had a compound heterozygous genotype as shown in Table 2. Totally five CAPN3 mutations have been identified (1 nonsense mutation, 1 missense mutation, 1 insertion, and 2 splice site mutations) (shown in Fig. 2). In these mutations, c.2120A>G, c.1194–9A>G, and c.734dup have been reported in previous studies [15, 22], while c.1115+2T>C and c.82_83insC were discovered and reported for the first time. Variant c.1115+2T>C result in a change in the splice site of intron 8, and c.82_83insC in exon 1 of the CAPN3 gene results in a translational frameshift of 6 amino acids and formation of a premature termination codon (shown in Table 2). In silico analysis and the ACMG Guidelines revealed that both c.1115+2T>C and c.82_83insC can be rated as pathogenic variants.
List of CAPN3variants in the proband
Patients . | Chr-start . | cDNA . | Amino acid . | Exon . | Zygosity . | Frequency . | Classification . | ACMG evidence . | ClinVar . |
---|---|---|---|---|---|---|---|---|---|
P1 | chr15-42686541 | c.1115+2T>C | Splice-5 | Intron 8 | Hom | - | Pathogenic | PSV1, PM2, PP3 | Yes |
P2 | chr15-42681225 | c.734dup | p.S246* | Exon 5 | Het | - | Pathogenic | PSV1, PM2, PM5, PP3 | No |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes | |
P3 | chr15-42652082 | c.82_83insC | p. Q28Pfs*6 | Exon 1 | Het | - | Pathogenic | PSV1, PM2, PP3 | No |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes | |
P4 | chr15-42691681 | c.1194–9A>G | Splice | Intron 9 | Het | 0.00001 | Pathogenic | PSV, PS1, PM2, PP3 | Yes |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes |
Patients . | Chr-start . | cDNA . | Amino acid . | Exon . | Zygosity . | Frequency . | Classification . | ACMG evidence . | ClinVar . |
---|---|---|---|---|---|---|---|---|---|
P1 | chr15-42686541 | c.1115+2T>C | Splice-5 | Intron 8 | Hom | - | Pathogenic | PSV1, PM2, PP3 | Yes |
P2 | chr15-42681225 | c.734dup | p.S246* | Exon 5 | Het | - | Pathogenic | PSV1, PM2, PM5, PP3 | No |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes | |
P3 | chr15-42652082 | c.82_83insC | p. Q28Pfs*6 | Exon 1 | Het | - | Pathogenic | PSV1, PM2, PP3 | No |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes | |
P4 | chr15-42691681 | c.1194–9A>G | Splice | Intron 9 | Het | 0.00001 | Pathogenic | PSV, PS1, PM2, PP3 | Yes |
chr15-42702630 | c.2120A>G | p. D707G | Exon 20 | Het | 0.00015 | Pathogenic | PSV1, PM2, PP3 | Yes |
CAPN3 gene reference genome version: GRCh37/hg19, NM_000070.2. The Genome Aggregation Database (GnomAD) version: v2.1.1. Hom, homozygous; het, heterozygous. Frequency results are based on genome aggregation database. Variation Classification refers to the American College of Medical Genetics and Genomics guidelines. ACMG evidence: PVS1-null variant (nonsense, frameshift, canonical ± 1 or 2 splice sites, initiation codon, single or multiexon deletion) in a gene where loss of function is a known mechanism of disease; PS2 de novo (both maternity and paternity confirmed) in a patient with the disease and no family history; PS3, well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product (a variant performed <40% in a minimum of 2 functional assays in our study; moderate PS3 well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product (a variant performed <40% in only 1 of our assays); PM1, variants located in a mutational hotspot and/or critical and well-established functional domain; PM2, absent from controls (or at an extremely low frequency if recessive) as shown in the Exome Sequencing Project; PM5, a novel missense change at an amino acid residue where a different missense change was determined to be pathogenic and has been seen before; PP1, cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease; PP3, multiple lines of computational evidence that support a deleterious effect on the gene or gene product (evolutionary impact, splicing, etc.); BS2, observed in a healthy, adult individual for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) disorder with full penetrance expected at an early age; BS3, well-established in vitro or in vivo functional studies show no damaging effect on protein function or splicing; BS4, the lack of segregation in affected family members.
Prediction of Novel CAPN3 Variants c.1115+2T>C and c.82_83insC
For c.1115+2T>C, the results of Splice AI showed that the donor splice site of the intron 8 has moved 2bp backward compared with the reference sequence (shown in Fig. 3a). The results of Netgene2 showed that the neural network donor site score of the original splice site decreased after the mutation, while the neural network donor site score of “GT” sequence in intron 8 increased (shown in Table 3). This is consistent with the results of Splice AI and varSEAK Online. In addition, the observation that the neural network coding score at the splice donor site decreased from 0.945 to 0.1 after the mutation strongly suggests that the mutation has disrupted the ability of the splicing enzyme to accurately cut the intron at the correct location and facilitate the correct connection of the exons. This disruption in splicing can have significant consequences for the formation of mature mRNA molecules, potentially leading to aberrant protein products or even loss of gene function (shown in Fig. 3b). The structural analysis of calpain 3 in wild type and variants of c.82_83insC (p. Q28Pfs*6) showed that the mutation is a novel frameshift variant, which caused early termination of protein synthesis and loss of protein function after the catastrophe point (shown in Fig. 4).
Software prediction results of splice site mutation. a Top: reference and variant of partial sequence of CAPN3gene exons 8, 9, and intron 8 (purple: splice site). Middle: schematic of CAPN3 gene exon 8, 9 and intron 8. Bottom: schematic of CAPN3 gene structure. b The figure “Coding” is the activity of an ensemble of coding predicting networks, values close to 0.0 indicate intron region, while values close to 1.0 indicate exon. Black: reference; red: variant (results from NetGene2-2.42).
Software prediction results of splice site mutation. a Top: reference and variant of partial sequence of CAPN3gene exons 8, 9, and intron 8 (purple: splice site). Middle: schematic of CAPN3 gene exon 8, 9 and intron 8. Bottom: schematic of CAPN3 gene structure. b The figure “Coding” is the activity of an ensemble of coding predicting networks, values close to 0.0 indicate intron region, while values close to 1.0 indicate exon. Black: reference; red: variant (results from NetGene2-2.42).
Prediction results of NetGene2 of c.1115+2T>C
. | 34842G . | 34826G . | ||
---|---|---|---|---|
normal . | mutation . | normal . | mutation . | |
Neural network donor site score | 0.945 | 0.100 | 0.964 | 0.986 |
Neural network acceptor site score | 0 | 0 | 0 | 0 |
Neural network coding score | 0.059 | 0.126 | 0.010 | 0.015 |
Neural network frame score | 3 | 3 | 1 | 1 |
90% sensitivity level cutoff value for donor site predictions | 0.689 | 0.650 | 0.744 | 0.704 |
90% sensitivity level cutoff value for acceptor site predictions | 0.521 | 0.544 | 0.487 | 0.511 |
Confidence of the donor site prediction | 0.846 | 0.079 | 0.832 | 0.895 |
Confidence of the acceptor site prediction | 0 | 0 | 0 | 0 |
. | 34842G . | 34826G . | ||
---|---|---|---|---|
normal . | mutation . | normal . | mutation . | |
Neural network donor site score | 0.945 | 0.100 | 0.964 | 0.986 |
Neural network acceptor site score | 0 | 0 | 0 | 0 |
Neural network coding score | 0.059 | 0.126 | 0.010 | 0.015 |
Neural network frame score | 3 | 3 | 1 | 1 |
90% sensitivity level cutoff value for donor site predictions | 0.689 | 0.650 | 0.744 | 0.704 |
90% sensitivity level cutoff value for acceptor site predictions | 0.521 | 0.544 | 0.487 | 0.511 |
Confidence of the donor site prediction | 0.846 | 0.079 | 0.832 | 0.895 |
Confidence of the acceptor site prediction | 0 | 0 | 0 | 0 |
Predicted mutational impact of p.Q28Pfs*6 on the CAPN3 protein structure. Image representation of CAPN3 structures in its wild type (residues 1–821) and the p.Q28Pfs*6 (residues 1–821) mutants (AlphaFold code: AF-P20807-F1). Red, yellow, and green are marked helices, sheets, and loops, respectively. The translation of loop structure after mutation is colored blue. Residues Q28 and their mutation (P28) are shown as purple sticks. P26 and K30, which make contact with Q28 or P28, are shown as orange sticks. Lower panel, the mutation of Q28 to P28 mutations disrupts the hydrogen bond between amino acids 26 and 30, destabilizing the original hydrogen bonding framework and interfering with normal conformational transmission.
Predicted mutational impact of p.Q28Pfs*6 on the CAPN3 protein structure. Image representation of CAPN3 structures in its wild type (residues 1–821) and the p.Q28Pfs*6 (residues 1–821) mutants (AlphaFold code: AF-P20807-F1). Red, yellow, and green are marked helices, sheets, and loops, respectively. The translation of loop structure after mutation is colored blue. Residues Q28 and their mutation (P28) are shown as purple sticks. P26 and K30, which make contact with Q28 or P28, are shown as orange sticks. Lower panel, the mutation of Q28 to P28 mutations disrupts the hydrogen bond between amino acids 26 and 30, destabilizing the original hydrogen bonding framework and interfering with normal conformational transmission.
Discussion
In this study, we identified 4 patients with LGMD R1 and found a total of five variants in the CAPN3 gene. Among these variants, c.2120A>G is the most common variant in the Chinese population [15]. In our report, the c.734dup variant also occurs on exon 5 of the CAPN3 gene but turns Serine residue in position 246 of the protein to ochre termination codon, producing a truncated protein (p.S246*). This is similar to the previous report (c.734dupC, p.S246Ter) [22]. This suggests that c.734dup is equally likely to be pathogenic. The c.82_83insC is not recorded in either gnomAD or ClinVar. Although C.1115+2T>C has been recorded as likely pathogenic in ClinVar and pathogenic in LOVD. However, clinical cases of the C.1115+2T>C variant have not yet been published to our knowledge.
As a group of rare muscle genetic diseases, LGMD has high genetic heterogeneity [23]. The phenotypes of LGMDR1 demonstrate variability in some degree related to gender and mutation type, the nonsense mutations lead to a more severe phenotype, while compound heterozygous patients are the least affected [24]. In our study, the 3 probands carrying compound heterozygous mutations all presented with mature adult disease onset and mild clinical manifestation. Compared with these 3 cases, both the age at onset and the appearance time of scapular winging are earlier in case 1 who carry the homozygous mutation. This could be attributed to the complex heterozygote scenario where two different mutated proteins are produced, and their interaction facilitates compensation for each other [25].
Donor and acceptor splice site variants typically disrupt protein function [26], and dysfunction variants in CAPN3are known to be pathogenic [27]. mRNA splicing analysis showed the c.1115+2T>C variant in the CAPN3gene disrupts the canonical splice donor site in intron 8, resulting in the exonization of part of intron sequences, which likely leads to an abnormal protein product. Besides, the retained sequences also can serve as abnormal information mediating mRNA decay. Although we did not perform transcriptional studies to confirm this prediction, the data we surmised suggested that this mutation is pathogenic.
When a number of nucleotides, not divisible by three, is inserted into the DNA, it results in the misreading of downstream codons. This kind of insertion mutation disrupts the reading frame, leading to potential loss of protein function [28]. Extra-base insertion leads to frameshift which causes amino acid translation to stop prematurely and produce nonfunctional calpains 3. The ClinVar database has recorded 65 insertional mutations in CAPN3, most of which are pathogenic mutations. Herein, we found a compound heterozygous mutation (c.82_83insC, p. Q28Pfs*6; c.2120A>G, p.D707G) of CAPN3 gene in Chinese adult with LGMDR1. CAPN3 is Ca2+-dependent cysteine protease composed of 821 amino acid residues, the 28th amino acid is located in the loop region of protein sequence [29]. Using the PyMOL software, we found the mutation of Gln-28 disrupts hydrogen bond formation between residues 26 and 30, leading to the instability of the original hydrogen bonding framework.
Although we reported two novel pathogenic mutations in LGMD patients for the first time, the shortcoming of this study is that the influence of mRNA level after mutation is not further verified through appropriate experiments. DNA sequence variants affect pre-mRNA processing through exon skipping, cryptic splicing, intron inclusion, leaky splicing, and other methods [30]. Therefore, it is difficult to accurately predict the specific cause of pathogenicity by artificial intelligence. Experimental mRNA and cDNA analysis are necessary for accurately assessing the effects of splicing mutation [31].
Conclusions
Totally 5 mutations have been found in our four LGMDR1 probands, three of the four probands carrying the hotspot c.2120A>G mutation, which is consistent with the previous reports. Another two mutations, c.734dup, p.S246* and c.1194-9A>G, were previously reported in Chinese LGMDR1 patients. More importantly, two novel mutations, c.1115+2T>C and c.82_83insC in the CAPN3 gene, have been identified in patients who were clinically diagnosed as LGMDR1. Although the mRNA and protein levels of calpain 3 were not further verified, bioinformatics analysis showed that these mutations resulted in abnormal CAPN3 protein synthesis, thereby providing valuable information on the diagnosis of LGMDR1 and expanding the spectrum of disease-causing CAPN3mutations.
Acknowledgments
Authors are grateful to all the families for their participation in this study.
Statement of Ethics
The study was conducted in accordance with the Declaration of Helsinki, and the protocol was reviewed and approved by the First Affiliated Hospital of Soochow University (NO. 2023082). All participants provided written informed consent to participate in this study.
Conflict of Interest Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Funding Sources
This study was supported by the National Natural Science Foundation of China (Nos. 82071300 to Q.F. and 82001219 to D.D.) and the Natural Science Foundation of Jiangsu Province (BK20190183).
Author Contributions
Lulu Zhang, Yi Zhang, and Chunru Han participated in the material preparation and medical data collection. Xiuying Cai and Liqiang Yu provided genetic information of patient. Huan Qi performed bioinformatics analysis of mutations with the support of Jianhua Jiang and Juean Jiang. The first draft of the manuscript was written by Lulu Zhang. Qi Fang and Dongxue Ding contributed to the interpretation of the findings and critical revision of the manuscript. All authors contributed to the study conception and design and read and approved the final manuscript.
Data Availability Statement
The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (D.X.D.) upon reasonable request.