CC BY
18
Section 4. General biology
https://doi.org/10.29013/ELBLS -21-1.2-52-70
Hubert Chen, Ivymind Academy, New Jersey, USA E-mail: [email protected]
SYSTEMATIC ANALYSIS OF GENETIC VARIATION OF DUCHENNE MUSCULAR DYSTROPHY AND IMPLICATION FOR CANCER
Abstract. Duchenne muscular dystrophy (DMD) is a rare, severe, progressive genetic disorder causing disability and premature death. Mutations in the DMD gene encoding the dystrophin protein lead to the dystrophinopathies DMD. Phenotypic variations in DMD may also occur in patients with the same primary mutation due to secondary genetic modifiers. Recent advances in molecular therapies for DMD have need of precise genetic diagnosis, since a large number of therapeutic approaches are mutation specific [1]. Interestingly, it has also been reported that muscular dystrophy patients may be at increased risk of malignancy [2; 3]. Although mutations in the DMD genes have been widely studied, to our knowledge, a systematic genetic analysis of all variants, especially single nucleotide polymorphisms (SNPs), of the gene in human have not been reported. In this study, we performed a systematic analysis of the DMD genetic variants via the Single Nucleotide Polymorphism Database (dbSNP), and annotated the functions of the variants with wANNOVAR [4]. The protein-protein interaction (PPI) network for genetic modifiers identified in DMD patients was explored. DMD genetic alternations in different tumors have also been investigated via cBioPortal [5].
We examined a total of 3,627 exonic SNPs in the DMD gene. SNPs distributed across all exons. The largest category was nonsynonymous account for nearly 64% of all mutations. Exon 19 appeared to have most density of pathogenic SNP distribution. Nonsense mutation (i. e. stopgain) or frameshift mutation likely lead to more pathogenic. Among the genetic modifiers identified in DMD patients, THBS1 has higher network topological parameters, followed by SPP1, ACTN3 and LTBP4. Network analysis highlighted non-random interconnectivity between the genetic modifiers identified in DMD patients, and potentially shed light on new genetic modifiers by their functional coupling to these known genes. Gene therapy is a promising experimental approach to treat genetic disease such as DMD [6]. In addition, our results also suggest DMD gene may serve as a diagnostic and therapeutic target for certain types of cancer.
Keywords: Duchenne muscular dystrophy (DMD), data mining, genetic, protein-protein interaction (PPI), single nucleotide polymorphisms (SNPs), tumor.
1. Introduction
Duchenne muscular dystrophy (DMD), a rare X-linked disorder, is caused by a genetic mutation that prevents the body from producing dystrophin [7], a protein that enables muscles to work properly. It is one of the most common types of muscular dystrophy.
DMD symptom onset is in early childhood, usually between ages 3 and 5 years. Over time, children with Duchenne will have difficulty walking and breathing, then lead to disability, dependence, and premature death. The disease primarily affects boys, but in rare cases it also can affect girls. The prevalence of DMD is approximately 1 in 3500 to 5000 male births worldwide [8]. Muscle weakness is the principal symptom of DMD and worsen over time, first affecting the proximal muscles and later affecting the distal limb muscles. Patients with DMD progressively lose the ability to perform activities independently and often require a wheelchair by their early teens. As the disease progresses, life-threatening heart and respiratory conditions can occur [9]. In general, patients succumb to the disease in their 20s or 30 s [9]; however, disease severity and life expectancy can vary.
DMD is caused by mutations in the DMD gene, which encodes the protein product called dystrophin. DMD gene is one of the largest of the identified human genes, spanning 2.4 Mb of a genomic sequence and corresponding to about 0.1% of the total human genome [10]. The gene consists of 79 exons encoding a 14,000 bp messenger RNA transcript that is translated into dystrophin [11]. The most common mutation responsible for DMD is a deletion spanning one or multiple exons. Such deletions account for 60-70% of all DMD cases. Point mutations are responsible for around 26% of DMD cases [12]. Exonic duplications account for 10 to 15% of all DMD case. Mutations in the DMD gene disrupt the protein's reading frame causing premature stop codons, leaving little or no functional dystrophin protein produced in cells. In addition, phenotypic variations in DMD may also occur in patients with
the same primary mutation. A wide range of clinical manifestations suggest that genetic modifiers can influence the severity of DMD disease [13].
Dystrophin protein consists of3,685 amino acids with molecular weight of427 kDa and plays a crucial role in muscle function. Cytoplasmic face of the sar-colemma is where dystrophin is located[14]. Based on sequence homology, dystrophin is divided into four distinct structural domains. (i) amino terminal actin binding domain that contains two calponin homology domain which directly interacts with cellular actin cytoskeleton[15]; (ii) a long central rod shaped domain is composed of 24 structurally similar spectrin-type repeats[16]; (iii) cysteine rich domain binds to the intrinsic membrane protein (3-dystroglycan; (iv) the carboxy-terminal domain binds to dystrobrevin and syntrophins[17]. Dystro-phin is a cytoskeletal protein that binds actin and associates with the dystrophin-glycoprotein complex (DGC) to link the cytoskeleton to the extracellular matrix (ECM)[18, 19]. The DGC consists of integral and peripheral proteins: dystroglycans, sarcoglycans, and syntrophins [18, 19]. Defects in dystrophin and/or other components of the DGC are responsible for several phenotypes of muscular dystrophy including DMD.
The DMD gene produces different transcripts encoding at least seven protein isoforms of various sizes and tissue distributions, each translated from a unique message initiated from one of the seven distinct promoters in the gene. Three full-length, large molecular weight isoforms (427 kDa), expressed primarily in skeletal muscle cells and cardiomyo-cytes, cortical neurons and the hippocampus of the brain and cerebellar Purkinje cells [20]. The shorter Dp260 and Dp116 isoforms are expressed primarily in the retina and the peripheral nerve, respectively [21]. Dp 140 is expressed in the central nervous system (CNS) and kidney. There have been studies linking the lack of Dp140 expression in some DMD patients to cognitive impairment [22]. Finally, the Dp71 isoform is ubiquitously expressed, with higher
levels in the CNS [23]. The literatures usually only describe seven protein coding isoforms. However, the Ensembl.org automatic annotation software predicts the presence of more protein coding splice variants of the dystrophin gene [24].
Although mutations in DMD genes have been widely studied, to our knowledge, a systematic genetic analysis of all variants, especially single nucleotide polymorphisms (SNPs), of the gene in human have not been reported. In this study, we carried out a systematic analysis of the DMD genetic variants via the dbSNP database, and annotated functions of these variants with wANNOVAR[4]. Protein-protein interactions for genetic modifiers identified in DMD patients were explored. In addition, potential
Open Data Source:
DbSNP contains SNP Human Var VCF Format (GRCh37 (h<
Filtering:
Duchenne muscular dystrophys
f
Annotation: Upload DMD.VCF to web server \ Download DMD Ensembl variants Transcript ENSG0000019.
1 1
1 1
Amino acid change,X Chrom position___ - R Coding Find the longest Transcript- -
In brief, all the genetic variants of DMD genes were extracted from the open access Single Nucleotide Polymorphism Database (dbSNP) database with variant call format (VCF). Variants using the online wANNOVAR server to generate output results files. These output files were annotated with function, distribution and disease etc [4]. Data mining and visualization were further performed based on the annotation.
2.2 Source dbSNP database and SNPs extraction The source data for the DMD gene comes from the dbSNP, which was established by National Center for Biotechnology Information (NCBI) in collab-
relationships of genetic alternations in the DMD gene with cancer were also investigated.
2. Materials and Methods
2.1 Research workflow
The DMD genetic research mainly focus on using public-domain human genome databases to perform a systematic analysis of genetic variants especially SNPs in human populations, explore protein-protein interactions for genetic modifiers identified in DMD patients and examine relationships of genetic alternations in DMD gene and types of cancer. No statements of approval or informed consent were required for our study as we obtained data from an open access database. A research workflow was presented in (Figure 1).
oration with the National Human Genome Research Institute (NHGRI) in 1998 for GenBank of publicly available nucleic acid and protein sequences [32]. In order to extract specific human DMD gene data, build-in Unix compress tool "MacCompress" under a Mac environment was selected to unzip the large size vcf.gz file. To extract DMD variants, we used the command 'grep DMD00-All.vcf > DMD.vcf', then added the header lines of 00-All.vcf into the file.
2.3 Functional annotation with wANNOVAR
Online server wANNOVAR was used to annotate functional consequences of genetic variation [26]. Online annotating single nucleotide variants,
Figure 1. Research Workflow
insertions and deletions functions effects on genes, ClinVar database information(Clinvar_DIS, Clin-var_ID), calculating predicted importance scores (SIFT_pred,Polyphen2) of each variant, retrieving allele frequencies in public databases (the 1000 Genomes Project and Genome Aggregation Database (gnomAD) of 15,708 genomes)[27], and implementing a 'variants reduction' protocol to identify a subset of potentially deleterious variants/genes[26].
2.4 Retrieve ensembl transcript
The dystrophin gene ENSG00000198947 has different transcripts which described at Ensembl. org [28]. According to wANNOVAR instruction, the most popular approach was to use the longest transcript which provided the most genetic annotation information if multiple transcripts were available [24]. The full-length transcript variant of the dystrophin gene was ensembl transcript (ENST) ID ENST00000357033.9, with13992 length in base pairs (bp) [28]. The R program (R-4.0.2) identified the longest transcript start and end, then split long characters string for AAChange.ensGene. It also picked the exon number and variant type, protein coding information (R scripts in appendix 6.1).
2.5 Protein-protein interaction map
Protein-protein interaction map for genetic modifiers identified in DMD patients was constructed using
search tool for the retrieval of interacting genes/proteins (STRING v11) [29]. Active interaction sources were restricted to "Textmining," "Experiments," and "Databases" Only interactions with confidence score over 0.9 were mapped to the network. The network obtained from STRING was subsequently analyzed using Cytoscape 3.8.1 plugin Network Analyzer [30]. The nodes in the PPI network represented the genes/ proteins, and the edges between the nodes represented the interactions between them. Nodes with high degree and betweenness centrality (BC) value were considered as key parameters to analyze the network. The node with a high degree was deemed with an essential biological function.
2.6 Integrative analysis genetic alternations in the DMD gene with cancer
Genetic alternations in the DMD gene with cancer was examined by using Cancer Genomics Portal (cBioPortal) [5]. This web portal provides query interface combined with customized data enabled us to interactively explore genetic alterations across DMD samples, genes mutation site and frequency. Kaplan-Meier curves stratified by genotype were plotted and comparisons were tested using the Logrank test.
3. Results
3.1 Variant prioritization
Figure 2. Catalog of genomic regions in the DMD gene
Online wANNOVAR submission for DMD.vcf 3.2 DMD gene annotation output
processed a total of 382160 variants. Variants located DMD exome summary contains a total of 3627
mainly in intronic, exonic, and UTR3 variant regions rows and 140 columns, which includes information
(Figure 2). In this study, genetic variants in exons rather such as function class of exonic variant. We presented
than introns were focused on, because variants in exon- some selected key information (Table 1) in the
ic (coding) region often can alter the protein function. annotated file that is crucial for our research.
Table 1. - Example of selected key columns in the annotated file
Columns Name Description Annotation
Start Start nucleotide number on X chromosome. 31140001
End End nucleotide number on X chromosome. 31140003
Ref Original nucleotide(s) present before mutation. C
Alt Alternative nucleotide(s) present after mutation. G
Gene.ensGene Gene Variant number. ENSG00000198947
Func.ensGene Regions (e.g., exonic, intronic, non-coding RNA)) that one variant hit
ExonicFunc.ens-Gene Exonic variant function, e.g., synonymous, non-synonymous, frameshift insertion. nonsynonymous SNV
ENST00000357033.9:
AAChange.ens-Gene Amino Acid Change exon45: c.G6502C: p.E2168Q ENST00000378677: exon45: c.G6490C: p.E2164Q
COSMIC_DIS Distribution in Catalogue of Somatic Mutations in Cancer 1(prostate)
ClinVar uses standard terms for clinical signifi-
ClinVar_SIG cance recommended by an authoritative source when available. These standards include five terms for diseases [33] Pathogenic
ClinVar DIS Clinical significance parameter Duchenne muscular dystrophy
SIFT score SIFT_pred SIFT score (deleterious or tolerated) [31] D: Deleterious (sift<=0.05); T: Tolerated (sift>0.05) 0.239 D
Polyphen2 HDIV_ score Impact on amino acid sequence and protein function. 0.291
1000 Genomes Project dataset with allele fre-
1000_exome_ALL quencies in six populations including ALL, AFR (African), AMR (Admixed American), EAS (East Asian), EUR (European), SAS (South Asian). These are whole-genome variants[31]. 0.005
3.3 Exonic SNPs frequency in the DMD gene deletions, insertions or substitutions for frameshift; We choose wANNOVAR's default definitions of stop gain; start loss; deletions, insertions, or exonic functional categories in order of precedence: substitutions for nonframeshift; nonsynonymous
and synonymous (Table 2). A total of 3627 exonic SNPs in the DMD gene has been examined. The largest category was nonsynonymous account for
Table 2.- Variants Type and
nearly 64% of all mutations. Synonymous accounted for nearly 43% and followed by stop gain account for nearly 6.7% of all mutations.
Frequency in the DMD Gene
Variants Definition Variants Type Count Frequency
Insertion, deletion or substitution of one or more nucleotides that cause frameshift changes in protein coding sequence. frameshift insertion 47 1.30%
frameshift deletion 116 3.20%
frameshift substitution 3 0.08%
Variant that leads to the immediate creation of stop codon at the variant site; stopgain 242 6.67%
Variant that leads to the immediate elimination of stop codon at the variant site startloss 1 0.03%
Insertion, deletion or substitution of 3 or multiples of 3 nucleotides that do not cause frameshift changes in protein coding sequence. nonframeshift insertion 4 0.11%
nonframeshift deletion 27 0.74%
nonframeshift substitution 1 0.03%
A single nucleotide change that cause an amino acid change nonsynonymous 2322 64.02%
A single nucleotide change that does not cause an amino acid change synonymous 864 23.82%
Total variants 3627
High frequency nucleotide substitutions have followed by Cytosine (C) to Thymine (T) with 15% been observed in the DMD gene. The largest cat- and Thymine (T) to Cytosine (C) with 15%, respec-egory was Guanine (G) to Adenine (A) with 17%, tively (Figure 3).
Figure 3. Summary of amino acid change in the DMD gene
3.4 DMD gene variants in Human 1000 Genomes Project
DMD ensemble in this project contains the 1000 from several different continents were showed in Table
Genomes project data in annotation. 1000 Genomes 3 and Figure 4. Approximately 7% nonsynonymous
project was designed to provide a comprehensive de- mutations were observed from 1000 Genomes project,
scription ofhuman genetic variation through sequenc- with relatively higher alteration frequency in African
ing multiple individuals. Populations were classified (~3%). Similar frequency distributions were observed
into 5 major continental groups: Africa (AFR), Ameri- among America, Europe, East Asia, and South Asia.
ca (AMR), Europe (EUR), East Asia (EAS), and South No frameshift and nonframeshift mutations and start-
Asia (SAS)[34] .The DMD gene variants frequency loss were reported in 1000 Genomes project.
Table 3.- DMD gene variants frequency in Human 1000 Genomes Project
Variants Type All DMD 1000G 1000G 1000G 1000G 1000G 1000G
Exome ALL AFR AMR EAS EUR SAS
nonsynonymous 2320 165(7.1%) 67(2.9%) 34(1.6%) 47(2%) 35(1.5%) 41(1.8%)
stopgain 242 2(0.8%) 0 1(0.4%) 0 0 0
synonymous 735 83(11.3%) 33(4.5%) 19(2.6%) 19(2.6%) 16(2.2%) 21(2.9%)
frameshift deletion 95 0 0 0 0 0 0
frameshift insertion 37 0 0 0 0 0 0
frameshift substitution 2 0 0 0 0 0 0
nonframeshift deletion 22 0 0 0 0 0 0
nonframeshift insertion 2 0 0 0 0 0 0
nonframeshift substitution 1 0 0 0 0 0 0
startloss 1 0 0 0 0 0 0
Total 3297 250 (7.6%) 100 (3%) 154 (4.7%) 66 (2%) 41 (1.2%) 63 (1.9%)
Figure 4. DMD gene variants frequency from different continents. Different colors represent for different levels of frequency values by R Heatmap Script in the appendix 6.2.
3.6 SNPs distribution by exons
In the DMD gene, SNPs were distributed across all the exons. Exon 79 is the longest with 2703 bp in length. Exon 78 is the shortest with 32 bp. To a better understanding of SNPs distribution density, we normalized the length of the exon, and as such high-frequency SNPs were observed in exon 19, 13, 71, respectively (Figure 5). Interestingly, although Exon 79 is the longest exon, there is a low
100 so 60 40 20
Raw coLint I Cbunt normalized by exon length
distribution of genetic variants in this gene region. A further investigation revealed that the majority region of the exons belongs in the 3'untranslated region (3'-UTR). With consideration for the low SNP frequency in the 5'-UTR, demonstrates that the genetic variation of DMD gene mainly occurs in the protein-coding region. This implies a potential genetic interaction between variation and protein function.
The x-axis only shows exon odd-number due to space Figure 5. The frequency and distribution density of genetic variants in the exons of DMD gene
3.7 SNPs distribution by ACMG classification
Variants according to ACMG Standards and Guidelines for clinical laboratories were classified as: 'Benign', 'Likely benign', 'Pathogenic', and 'Likely pathogenic [35].
Distribution of SNPs by ACMG Classifications in the DMD Gene presented in (Figure 6). Stop-gain caused 57% pathogenic, followed by frameshift (22%). In general, most of synonymous mutation have no functional consequence. Interestingly, we observed a few cases with synonymous mutation
(2%) also associated with pathogenic DMD in the current analysis.
Distribution of SNPs by exonic region and ACMG classification in the DMD gene presented in (Figure 7). In general, pathogenic variants were distributed across almost all exons. In many exons (i.e. exon 19), most variants were pathogenic. However, in some other exons (i.e. exon 37), benign variants occurred more frequently. Exon 19 has most density ofpathogenic SNP distribution and followed by exon 20.
MONFRAMESHIFT DELETION FRAMESHIFT SUBSTITUTION SYNONYMOUS SNV FRAMESHIFT INSERTION PdONSYNONYMOUS SNV FRAMESHIFT DELETION
STOPGAIN ■ Pathogenic ■ Likely Pathogenic
Benign
Likely Benign
U ncerta insignificance
Stopgain Frameshift Nonsynonymous Frameshift Synonymous Fr a mesh ¡ft Nonframeshift
deletion SNV Insertion SNV Substitution Deletion
Pathogenic 151 57 26 20 5 2 1
Likely Pathogenic 1 1 0 0 0 0 0
Benign 0 0 47 0 24 0 0
Likely Benign 0 0 16 0 16 0 0
Uncertain Significance 1 0 35 0 0 0 0
Figure 6. Distribution of SNPs by ACMG-AMP classifications in the DMD gene
The x-axis only shows every third exon number due to space Figure 7. Distribution of SNPs by exonic region and ACMG classification in the DMD gene
3.8 Network analysis highlights non-random interconnectivity between the genetic modifiers identified in DMD patient
Genetic modifiers have been associated with variability in human DMD sub-phenotypes[13]: SPP1, LTBP4, and THBS1 are mapped to the
TGF-^ pathway, and implicated in several interconnected molecular pathways regulating inflammatory response to muscle damage, regeneration, and fibrosis. ACTN3 deficiency triggered an increase in oxidative muscle metabolism through activation of calcineurin, and then ameliorated the progression of
dystrophic pathology. We considered SPP1, LTBP4, ACTN3 and THBS1 as the "seed" genes/proteins to construct the PPI network associated with DMD (Figure 8) [36]. Fifteen additional interactors were allowed in the network to identify the most significant interactions and achieve a meaningful size for network analysis. PPI analyses showed that ACTN3 had direct interaction with DMD. SPP1 may interact with DMD through ITGB1, which has the highest node degree and BC values in the network. ITGB1 is one of the most common forms in muscle. Disruptions of integrins are responsible for a further class of muscular dystrophies [37]. Among the "seed"
genetic modifiers, THBS1 has higher network topological parameters, followed by SPP1, ACTN3 and LTBP4. The network enrichment p-value was < 3.09e-08, meaning that this connected network has significantly more interactions than expected at random, and that the genetic modifiers have more interactions among themselves than what would be expected for a random set of genes/proteins of similar size. Such enrichment also indicates that these genetic modifiers are, at least partially, biologically connected. In addition, the PPI network may be possible to shed light on new genetic modifiers by their functional coupling to these known "seed" genes.
Figure 8. Interaction network resulting from the
3.9 Exon DMD genetic alternations in different tutm>rs
After examining SNPs distribution by ACMG Classification from DMD functional annotation output file, it appeared genetic alternations in the DMD gene associated with different types of cancers. It has also been reported that DMD gene is involved in tumor development and progression [38-41].
genetic modifiers identified in DMD patients
Therefore, we further analyzed data from 25 published cancer studies from The Cancer Genome Atlas (TCGA) and 4 pediatric cancer studies that included a minimum of 100 samples in the cBioPor-tal database. Since people with DMD could develop rare and aggressive type of muscle cancer (i.e. rhab-domyosarcoma), we also included one study that analyzed 43 rabdomyosarcomas cases. Total 11927
patients (age from ~ 3 years to 90 years; ~ 48% male and ~ 46% female; ~ 60% white, ~7% black or africa america and ~ 5% asian) and 11977 samples from 30 studies have been included in the analysis. Ap-
The majority of DMD genetic alterations corresponded to mutations and deep deletions, and a low frequency of gene amplifications. The occurrence of DMD alterations varied across the studies/tumor
proximately 10% of the cases have an alteration in the DND gene, consisting mainly of missense mutation and truncating mutation. The studies analyzed were listed in the appendix 6.4.
types (Figure 9). The highest ratio of genetic alternations was approximately 31% (Esophagus, TCGA PanCan), and followed by 25% (Ut erine TCGA Pan-Can 2018). There were approximately 967 missense
Table 4. - Example of DMD gene mutations with cancers
Cancer Type Functional Impact Mutation Type Variant Type HGVSc Exon
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: impact: deleterious, score: 0; Polyphen-2: impact: probably damaging, score: 0.994 Missense Mutation SNP ENST00000357033.4: c.1940T>C 16/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: impact: tolerated, score: 0.76; Polyphen-2: impact: probably damaging, score: 0.985 Missense Mutation SNP ENST00000357033.4: c.2215G>A 18/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: NA; Polyphen-2: NA Nonsense Mutation SNP ENST00000357033.4: c.4222C>T 30/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: impact: tolerated, score: 0.25; Polyphen-2: impact: benign, score: 0.219 Missense Mutation SNP ENST00000357033.4: c.4117C>A 30/79
Rectal Adenocar-cinoma MutationAssessor: NA; SIFT: impact: tolerated, score: 0.19; Polyphen-2: impact: possibly damaging, score: 0.475 Missense Mutation SNP ENST00000357033.4: c.5824G>A 41/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: NA; Polyphen-2: NA Missense Mutation SNP ENST00000357033.4: c.7213G>T 50/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: impact: deleterious, score: 0.03; Polyphen-2: impact: probably damaging, score: 0.926 Missense Mutation SNP ENST00000357033.4: c.7891C>T 54/79
Adrenocortical Carcinoma MutationAssessor: NA; SIFT: impact: tolerated, score: 0.4; Polyphen-2: impact: benign, score: 0.164 Missense Mutation SNP ENST00000357033.4: c.8464C>A 57/79
Adrenocortical Carcinoma MutationAssessor: NA; SIFT: NA; Polyphen-2: NA Frame Shift Ins INS ENST00000357033.4: c.8528dup 57/79
Bladder Urothelial Carcinoma MutationAssessor: NA; SIFT: impact: tolerated, score: 0.35; Polyphen-2: impact: benign, score: 0.005 Missense Mutation SNP ENST00000357033.4: c.8641C>G 58/79
Adrenocortical Carcinoma MutationAssessor: NA; SIFT: impact: deleterious, score: 0; Polyphen-2: impact: probably damaging, score: 1 Missense Mutation SNP ENST00000357033.4: c.9766G>A 67/79
mutations and 246 truncating mutation in the tumor samples, with highest frequency (8 of nonsense mutation) being R1459*/Q,in the kinase domain. There appeared to be no hot spots of alteration in the gene region (Figure 10). Interestingly, DMD alterations were not found in rabdomyosarcomas samples.
3.10 Patients with DMD alterations have poorer overall survival
To study the overall survival (OS) of patients with and without DMD genetic alterations, Kaplan-
Meier curves stratified by genotype were plotted and the comparisons were tested using the Log-rank test [5]. OS analysis has been initially conducted using cBioPortal with pooled all patient data from 30 different cancer studies as mentioned in previous section (appendix 6.4). We observed that patients with genetic alterations in DMD had significantly shorter OS (63.8 months) compared to patients with wild-type DMD (82.9 months) (Figure 11).
The x axis shows the types of cancer (color coded), availability of mutation and copy number variation data, and the study abbreviation Figure 9. Frequency of genetic alternations in the DMD gene in different types of tumors
Figure 10. Mutation diagram of DMD gene in the tumor samples
■ Wild type DMD
■ Mutated DMD
1 1 II Log rank P-value: 2.172e-4
0 20 40 60 SO 100 120 140 160 180 200 220 240 260 280 300 320 340 360 Months Overall
Figure 11. Overall survival analyses using c BioPortal data
Additional OS analysis has been performed by various tumor types with greater than 5% DMD genetic alteration frequency[5]. Invasive breast carcinoma and esophageal carcinoma that revealed significant differences between by wild-type DMD or mutated DMD groups (Figure 12). Others such as ovarian carcinomas showed similar trend, but not statistically significant (Figures not presented). OS analysis of breast cancer patients with low or high ex-
Invasive Breast Carcinoma
pression of DMD gene has been further conducted by using Kaplan-Meier Plotter (www.kmplot.com/ analysis) [42]. Likewise, low DMD expression is associated with poorer survival in breast cancer patients. These results demonstrate that the relationship between DMD genetic status and prognosis may be tumor-type specific. Moreover, its biological function in tumorigenesis as well as prognosis is complicated and co-regulated with other factors.
Esophageal Carcinoma
Figure 12. Overall survival for invasive breast carcinoma and esophageal carcinoma using cBioPortal data
4. Discussions
DMD is a rare, severe, progressive genetic disorder causing disability and premature death. DMD caused by mutations in the dystrophin-encoding DMD gene, which is one of the largest of the identified human genes. There are currently no curative therapies for DMD. Gene therapy is a promising experimental method that uses genes (the fundamental units of heredity) to treat disorders that result from genetic mutations. Currently, several kinds of gene-based therapies including exon skipping are being developed to treat DMD.
Although mutations in DMD genes have been widely studied, to our knowledge, a systematic genetic analysis of all variants, especially SNPs, of the gene in human have not been reported. In general, Variants in exonic (coding) region can alter the protein function. Therefore, we focused on a total of 3627 exonic SNPs. In the DMD gene, SNPs distributed across all exons. The largest category was nonsynonymous account for nearly 64% of all mutations. Exon 19 appeared to be most pathogenic region. As expected, nonsense mutation (i.e. stopgain) or frameshift mutation likely lead to more pathogenic DMD. Of note, the limitation of current database may not perfectly represent the DMD patient population, since the database collects all genetic variations from various individuals, but not merely DMD patients. Therefore, the result in the study reveals richer variations in human populations than in DMD patients. Interestingly, some pathogenetic variants were also observed in healthy individuals. For example, a few nonsynonymous and stop gain cases were observed in the 1000 Genomes project, although the allele frequency was very low. Healthy individuals may also carry DMD related genetic alternations. However, for each gene, there are homologous alleles from the mother and father, so the abnormality of one of the alleles will not lead to disease. This may explain why normal individuals carry DMD mutations without clinic symptoms.
Phenotypic variations in DMD may also occur in patients with the same primary mutation. A wide range of clinical manifestations suggest that genetic modifiers such as SPP1, LTBP4, ACTN3, and THBS1 can modify the clinical severity of DMD disease [13]. As expected, our PPI analysis highlighted non-random interconnectivity between the genetic modifiers identified in DMD patients, and potentially shed light on new genetic modifiers by their functional coupling to these known genes.
The genomic, functional and clinicopathological evidence demonstrate dystrophin tumor suppressor roles in human cancers. People with DMD may increase risks to develop aggressive and rare muscle cancer such as rhabdomyosarcoma[38, 43]. In the current study, we have used the cBioPortal for cancer genomics as a tool for visualizing, exploring and analyzing the biological and clinical characteristics of DMD genetic alterations with cancer. Total 11927 patients and 11977 samples from 30 studies with various types of tumors have been included in the analysis. Approximately 10% of the cases had an alteration in the DMD gene, consisting mainly of missense mutation and truncating mutation. We observed DMD genetic alterations varied across the studies/tumor types. Patients with DMD genetic alterations appeared to have shorter overall survival, particularly with esophageal carcinoma and invasive breast carcinoma. Similarly, Stephens et al. also reported poorer survival for patients with upper gastrointestinal cancer and low expression of DMD[39]. However, there were no significant correlations between DMD genetic alterations and overall survival in some other types ofhuman cancers. These results demonstrate that the relationship between DMD genetic status and prognosis may be tumor-type specific. Moreover, its biological function in tumorigenesis as well as prognosis is complicated and co-regulated with other factors.
Leonela N. Luce et al conducted paired tumor/ normal tissues showed that majority of tumor specimens had lower DMD expression compared to the normal adjacent tissue [44], in concordance to the
other reports suggesting the tumor suppressor role of DMD[38]. Contrastingly, overexpression of the DMD gene has also been reported in leukemias, renal carcinomas, ependymomas, and astrocytomas [44]. The function and molecular mechanism involved in altered DMD gene expression in different cancer types remains to be further investigated.
5. Conclusions
In conclusion, to our knowledge, this is first data mining study with a systematic genetic analysis of all variants, especially SNPs, of the DMD gene in human. SNPs distributed across all Exons. The largest category was nonsynonymous account for nearly 64% of all mutations. Exon 19 appeared to have most density of pathogenic SNP distribution. Nonsense mutation (i.e. stopgain) or frameshift mutation likely lead to more pathogenic. Network analysis highlighted non-random interconnectivity between the genetic modifiers identified in DMD patients, and potentially shed light on new genetic modifiers by their functional coupling to these known genes. In addition, our results also suggest DMD gene may serve as a diagnostic and therapeutic target for certain types of cancer.
6. Appendix
6.1 R Script of "Extract the longest length transcript with Exon Number, amino acid change from DMD gene exome file"
# String work library(stringr) library(dplyr)
# Read Path where your CSV file is located DMD<- read.csv(file = 'Read Path where CSV file located\\DMD_query.output.exome_summary. csv')
head(DMD) library(stringr)
location<-str _locate(DMD$AAChange.
ensGene,"ENST00000357033")
location
startpos<-location[,1] startpos endpos<-location[,2] endpos str_sub(DMD$AAChange.ensGene, startpos,
endpos+25)
#ExportENST00000357033 segment CSV DMD_ENST_Output<-str_ sub(DMD$AAChange.ensGene, startpos, endpos+25)
write.csv(DMD_ENST_putput, file = "DMD_ ENST_Output.csv")
6.2 R Script for Heatmap of DMD variants frequency from different population in 1000 Genomes project
# Read DMD variants frequency VS1000 gene population file from ENSG0000019894 DMDgeneExp <- read.csv'Read Path where CSV file located\\heatmap\\1000genomeproject.csv") install.packages('caTools')
library(gplots) head(DMDgeneExp)
rownames(DMDgeneExp) <- make.unique(as. character(DMDgeneExp $Variants.Type)) DMDgeneExp_matrix <- as.matrix(DMDgeneExp [2:6])
head(DMDgeneExp_matrix) dist_no_na <- function(mat) { edist <- dist(mat)
edist[which(is.na(edist))] <- max(edist, na.rm=TRUE) * 1.1
return(edist) }
colors = c(seq(-3,-2, length=100), seq(-2,0.5, length=100), seq(0.5,6, length=100))
my_palette <- colorRampPalette(c("grey","blue","d arkred"))(n = 100)
# Heatmap plotted
heatmap.2(DMDgeneExp_matrix, distfun=dist_
no_na, col=my_palette,
key=TRUE, symkey=FALSE,
cexRow=0.01, cexCol=1.2,
offsetRow = 0,
densadj = 0.15,
margins=c(4,16),
lwid = c(5,20))
6.3 Amino acid descriptions
One letter code Three letter code Amino acid Possible codons
A Ala Alanine GCA, GCC, GCG, GCT
B Asx Asparagine or Aspartic acid AAC, AAT, GAC, GAT
C Cys Cysteine TGC, TGT
D Asp Aspartic acid GAC, GAT
E Glu Glutamic acid GAA, GAG
F Phe Phenylalanine TTC, TTT
G Gly Glycine GGA, GGC, GGG, GGT
H His Histidine CAC, CAT
I Ile Isoleucine ATA, ATC, ATT
K Lys Lysine AAA, AAG
L Leu Leucine CTA, CTC, CTG, CTT, TTA, TTG
M Met Methionine ATG
N Asn Asparagine AAC, AAT
P Pro Proline CCA, CCC, CCG, CCT
Q Gln Glutamine CAA, CAG
R Arg Arginine AGA, AGG, CGA, CGC, CGG, CGT
S Ser Serine AGC, AGT, TCA, TCC, TCG, TCT
T Thr Threonine ACA, ACC, ACG, ACT
V Val Valine GTA, GTC, GTG, GTT
W Trp Tryptophan TGG
X X any codon NNN
Y it Tyrosine TAC, TAT
Z Glx Glutamine or Glutamic acid CAA, CAG, GAA, GAG
* * stop codon TAA, TAG, TGA
6.4 Cancer studies analyzed from cBioPortal
Type of cancer - study abbreviation Number
1 2
Bladder Urothelial Carcinoma (TCGA, PanCancer Atlas) 411
Colorectal Adenocarcinoma (TCGA, PanCancer Atlas) 594
Breast Invasive Carcinoma (TCGA, PanCancer Atlas) 1084
Brain Lower Grade Glioma (TCGA, PanCancer Atlas) 514
Glioblastoma Multiforme (TCGA, PanCancer Atlas) 592
Cervical Squamous Cell Carcinoma (TCGA, PanCancer Atlas) 297
Esophageal Adenocarcinoma (TCGA, PanCancer Atlas) 182
Stomach Adenocarcinoma (TCGA, PanCancer Atlas) 440
Head and Neck Squamous Cell Carcinoma (TCGA, PanCancer Atlas) 523
Kidney Renal Clear Cell Carcinoma (TCGA, PanCancer Atlas) 512
Kidney Renal Papillary Cell Carcinoma (TCGA, PanCancer Atlas) 283
1 2
Liver Hepatocellular Carcinoma (TCGA, PanCancer Atlas) 372
Lung Adenocarcinoma (TCGA, PanCancer Atlas) 566
Lung Squamous Cell Carcinoma (TCGA, PanCancer Atlas) 487
Acute Myeloid Leukemia (TCGA, PanCancer Atlas) 200
Ovarian Serous Cystadenocarcinoma (TCGA, PanCancer Atlas) 585
Pancreatic Adenocarcinoma (TCGA, PanCancer Atlas) 184
Prostate Adenocarcinoma (TCGA, PanCancer Atlas) 494
Skin Cutaneous Melanoma (TCGA, PanCancer Atlas) 448
Pheochromocytoma and Paraganglioma (TCGA, PanCancer Atlas) 178
Sarcoma (TCGA, PanCancer Atlas) 255
Testicular Germ Cell Tumors (TCGA, PanCancer Atlas) 149
Thymoma (TCGA, PanCancer Atlas) 123
Thyroid Carcinoma (TCGA, PanCancer Atlas) 500
Uterine Corpus Endometrial Carcinoma (TCGA, PanCancer Atlas) 529
Pediatric Pan-Cancer (DKFZ, Nature 2017) 961
Pediatric Ewing Sarcoma (DFCI, Cancer Discov 2014) 107
Pediatric Preclinical Testing Consortium (CHOP, Cell Rep 2019) 261
Pediatric Pan-cancer (Columbia U, Genome Med 2016) 103
Rhabdomyosarcoma (NIH, Cancer Discov 2014) 43
References
1. Mohammed F., et al., Mutation spectrum analysis of Duchenne/Becker muscular dystrophy in 68 families in Kuwait: The era of personalized medicine. PLoS One, 2018.- 13(5).- e0197205 p.
2. Gadalla S. M., et al. Cancer risk among patients with myotonic muscular dystrophy. Jama, 2011.-306(22).- P. 2480-6.
3. Win A. K., et al. Increased cancer risks in myotonic dystrophy. Mayo Clin Proc, 2012.- 87(2).- P. 130-5.
4. Lab W. G. wANNOVAR. 2010-2020. Available from: URL: http://wannovar.wglab.org/
5. cBio Portal for Cancer Genomics. 2020. Available from: URL: https://www.cbioportal.org/
6. Nelson S. F., et al. Emerging genetic therapies to treat Duchenne muscular dystrophy. Current opinion in neurology, 2009.- 22(5).- 532 p.
7. Salmaninejad A., et al. Duchenne muscular dystrophy: an updated review of common available therapies. Int J Neurosci, 2018.- 128(9).- P. 854-864.
8. Wang R. T. and Nelson S. F. What can Duchenne Connect teach us about treating Duchenne muscular dystrophy? Curr Opin Neurol, 2015.- 28(5).- P. 535-41.
9. BioIncept L. Potential to effectively protect muscle function, combatting Duchenne muscular dystrophy. 2020. Available from: URL: https://bioincept.com/product-development/duchenne-muscular-dystrophy/
10. Van Belzen D. J., et al. Mechanism of Deletion Removing All Dystrophin Exons in a Canine Model for DMD Implicates Concerted Evolution of X Chromosome Pseudogenes. Mol Ther Methods Clin Dev, 2017.-4.- P. 62-71.
11. Lee B. L., et al. Genetic analysis of dystrophin gene for affected male and female carriers with Duchenne/ Becker muscular dystrophy in Korea. J Korean Med Sci, 2012.- 27(3).- P. 274-80.
12. Gao Q. Q. and E. M. Mc Nally. The Dystrophin Complex: Structure, Function, and Implications for Therapy. Compr Physiol, 2015.- 5(3).- P. 1223-39.
13. Bello L., Pegoraro E. The "Usual Suspects": Genes for Inflammation, Fibrosis, Regeneration, and Muscle Strength Modify Duchenne Muscular Dystrophy. J Clin Med. 2019 May 10; - 8(5).- 649 p.
14. Porter G. A., et al. Dystrophin colocalizes with beta-spectrin in distinct subsarcolemmal domains in mammalian skeletal muscle. J Cell Biol, 1992.- 117(5).- P. 997-1005.
15. Norwood F. L., et al. The structure of the N-terminal actin-binding domain of human dystrophin and how mutations in this domain may cause Duchenne or Becker muscular dystrophy. Structure, 2000.-8(5).- P. 481-91.
16. Muthu M., Richardson K. A. and Sutherland-Smith A.J. The crystal structures of dystrophin and utrophin spectrin repeats: implications for domain boundaries. PLoS One, 2012.- 7(7).- e40066 p.
17. Broderick M.J. and Winder S.J. Spectrin, alpha-actinin, and dystrophin. Adv Protein Chem, 2005.- 70.-P. 203-46.
18. Henry M. D. and Campbell K. P. Dystroglycan: an extracellular matrix receptor linked to the cytoskel-eton. Curr Opin Cell Biol, 1996.- 8(5).- P. 625-31.
19. Blake D.J., Tinsley J. M. and Davies K. E. Utrophin: a structural and functional comparison to dystrophin. Brain Pathol, 1996.- 6(1).- P. 37-47.
20. Heikoop J. C., Hogervorst F. B.L., Meershoek E.J. et al. Expression of the Human Dp 71 (Apo-Dys-trophin-1) Gene from a 760-kb DMD-YAC Transferred to Mouse Cells. Eur J Hum Genet 3, 1995.-P. 168-179.
21. D'Souza V.N. et al. A novel dystrophin isoform is required for normal retinal electrophysiology. Human molecular genetics - Vol. 4,5. 1995.- P. 837-42.
22. Taylor P.J., et al. Dystrophin gene mutation location and the risk of cognitive impairment in Duchenne muscular dystrophy. PLoS One, 2010.- 5(1).- e8803 p.
23. Lederfein D. et al. A 71-kilodalton protein is a major product of the Duchenne muscular dystrophy gene in brain and other nonmuscle tissues. Proceedings of the National Academy of Sciences of the United States ofAmerica - Vol. 89.- 12. 1992.- P. 5346-50.
24. Yates A., et al. Ensembl 2016. Nucleic Acids Res, 2016.- 44(D1).- P. D710-6.
25. dbSNP. 2020. Available from: URL: https://www.ncbi.nlm.nih.gov/snp
26. Chang X. and Wang K. wANNOVAR: annotating genetic variants for personal genomes via the web. J Med Genet, 2012.- 49(7).- P. 433-6.
27. Genomics Lab. 2010-2020. Available from: URL: http://wannovar.wglab.org
28. Institute, E.M.B.L.s.E.B. DMD ENSG00000198947 Transcripts. August 2020. Available from: URL: http://Aug2020.archive.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000198947; r=X:31097677-33339441
29. Szklarczyk Damian et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic acids research -Vol. 47.- D1 (2019).- D607-D613.
30. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Shannon P., Markiel A., Ozier O., Baliga N. S., Wang J. T., Ramage D., Amin N., Schwikowski B., Ideker T. Genome Res. 2003.- Nov,- 13(11).- P. 2498-504.
31. ANNOVAR Documentation. Utilize update-to-date information to functionally annotate genetic variants detected from diverse genomes, wANNOVAR supports only human genome annotation 2010-2018. Available from: URL: https://doc-openbio.readthedocs.io/projects/annovar/en/latest
32. Gao J., et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal, 2013.- 6(269).- l1 p.
33. Information N. C. f.B. Clinical significance on ClinVar submitted records. 2020-03. Available from: URL: https://www.ncbi.nlm.nih.gov/clinvar/docs/clinsig
34. Belsare S., et al. Evaluating the quality of the 1000 genomes project data. BMC Genomics, 2019.- 20(1).-620 p.
35. New ACMG Guidelines. September 30, 2015. Available from: URL: https://www.genedx.com/whats-new/new-acmg-guidelines
36. Garton Fleur C. et al. The Effect of ACTN3 Gene Doping on Skeletal Muscle Performance. American journal of human genetics - Vol. 102.- 5. 2018.- P. 845-857.
37. Smith Lucas R. et al. Systems analysis of biological networks in skeletal muscle function. Wiley interdisciplinary reviews. Systems biology and medicine - Vol. 5,1. 2013.- P. 55-71.
38. Wang Y., et al. Dystrophin is a tumor suppressor in human cancers with myogenic programs. Nat Genet, 2014.-46(6).- P. 601-6.
39. Sgambato A., et al. Dystroglycan expression is frequently reduced in human breast and colon cancers and is associated with tumor progression. Am J Pathol, 2003.- 162(3).- P. 849-60.
40. Hosur V., et al. Dystrophin and dysferlin double mutant mice: a novel model for rhabdomyosarcoma. Cancer Genet, 2012.- 205(5).- P. 232-41.
41. Korner H., et al. Digital karyotyping reveals frequent inactivation of the dystrophin/DMD gene in malignant melanoma. Cell Cycle, 2007.- 6(2).- P. 189-98.
42. Kaplan-Meier Plotter. 2009-2020. Available from: URL: http://www.kmplot.com/analysis
43. Boscolo Sesillo F., Fox D. and Sacco A. Muscle Stem Cells Give Rise to Rhabdomyosarcomas in a Severe Mouse Model of Duchenne Muscular Dystrophy. Cell reports, 2019.- 26(3).- P. 689-701.e6.
44. Luce L. N., et al. Non-myogenic tumors display altered expression of dystrophin (DMD) and a high frequency of genetic alterations. Oncotarget, 2017.- 8(1).- P. 145-155.
45. Rania Horaitis N. b. J. T. D. D. Codons and amino acids. October, 2009. Available from: URL: https://www. hgvs.org/mutnomen/codon.html
