New Report of a Different Clinical Presentation of CD151 Splicing Mutation (c.351+2T>C): Could TSPAN11 be Considered as a Potential Modifier Gene for CD151?

Abstract

CD151, a member of the tetraspanin family, is essential for normal development of skin and kidney. To date, only 2 pathogenic variants of the CD151 gene have been identified in a related disorder with recessive inheritance. Here, in the third study of CD151 mutations, we report 3 affected siblings presenting variable degrees of renal and dermal symptoms. Whole exome sequencing (WES) was performed on the proband, followed by data analysis and in silico assessments. Confirmation of the mutation in the other patients were carried out using Sanger sequencing. The consequence of the CD151 mutation was investigated by RNA extraction and Sanger sequencing of PCR products from cDNA. Multiple computational tools were applied for protein alignment, homology modeling, and molecular interaction analysis. WES revealed the variant c.351+2T>C, NM_139029 (GRCh37) in CD151, and this was confirmed by Sanger sequencing in all patients. This variant is the result of a substitution of nucleotide T with C that changes the position +2 of the donor splice site in intron 5, leading to total loss of exon 5 from the transcript. The mentioned variant was not found in population allele frequency databases, and prediction tools concurred in its damaging effect on the protein. Based on the criteria from ACMG guidelines, this variant is pathogenic. Interestingly, in terms of clinical findings, symptoms and severity of the disease in the patients in this study were different compared to the previous report of the mutation and the disease. In addition, in silico analysis in this study appears to suggest a candidate protein, Tetraspanin-11 (TSPAN11), that could partially modify CD151 functions. This study supports the pathogenic effect of the CD151 variant c.351+2T>C, highlights the extensive variable expressivity amongst patients, reinforces the contribution of genomic content to clinical characteristics of CD151 mutations, and accentuates the importance of modifier genes.

Established Facts

• To date, only 2 pathogenic variants of CD151 have been identified which are associated with a very rare disease, characterized primarily by pretibial or extended Epidermolysis bullosa type skin blisters in all patients and variable renal abnormalities mostly in the form of proteinuric nephropathy.

• The mutation c.351+2T>C has been reported only in 1 patient with extensive skin blisters, mild renal involvement, and a history of protein wasting.

Novel Insights

• Here, in the third research on CD151 patients, 3 siblings were studied who were affected by the CD151 mutation c.351+2T>C and had distinct clinical symptoms from what had been reported before. The siblings had very mild skin involvement, notably the proband who had only one very mild small lesion, but on the contrary to previous report, they had congenital renal anomalies, with kidney agenesis and end-stage renal disease.

• This study demonstrates extensive variable expressivity amongst patients with CD151 mutations, in which skin involvement could be mild accompanied with severe renal anomalies or it could manifest as mild renal anomalies along with severe skin involvement.

• Furthermore, in this study, for the first time we suggest TSPAN11, which is another member of the tetraspanin family, as a possible modifier gene with the genetic redundancy for CD151. Although to prove this hypothesis, more patients and other functional studies are needed.

Introduction

Cluster of differentiation 151 (CD151), also known as GP27, MER2, RAPH, SFA1, PETA-3, and TSPAN24, is a member of the tetraspanin family which are identified by 4 transmembrane domains and 2 different-sized extracellular loops, EC1 and EC2 [Maecker et al., 1997]. Association of tetraspanins with the same or different proteins in the plasma membrane forms compartments called tetraspanin-enriched microdomains (TEMs). Cell adhesion molecules, signaling receptors, and enzymes are the most important components of TEMs [Yáñez-Mó et al., 2009].

To the best of our knowledge, to date, only 4 patients and 2 pathogenic germline variants of CD151 have been reported in a very rare recessive disorder, which was mostly characterized by pretibial skin blisters and proteinuric nephropathy [Karamatic Crew et al., 2004; Vahidnezhad et al., 2018]. The disease was introduced by Kagan et al. in 1988, and skin biopsy from their patients showed detachment of epidermis from dermis, and kidney biopsy demonstrated splitting and thickening areas in GBM and TBM. Karamatic Crew et al. [2004] revealed the genetic defect behind the disease by identifying a frameshift mutation in CD151. In 2018, Vahidnezhad et al. [2018] reported a patient with homozygous splicing variant of CD151, c.351+2T>C, who also had features somehow similar to the previous patients.

Here, we report a third family of the disease with the same splicing mutation. But interestingly, symptoms and severity of the disease are different compared with both previous studies. This may be explained by modifier genes including other members of the tetraspanin family.

Case Presentation

The proband was referred to a genetic clinic showing congenital renal anomaly and end-stage renal disease (ESRD). Her 2 siblings, who were affected by renal abnormalities, were also recruited to participate in the study.

Whole Exome Sequencing, RT-PCR, and Sanger Sequencing

Genomic DNA from 2 affected siblings and the mother were also obtained in addition to the proband's DNA. Unfortunately, a sample from the father was not available. WES was performed on the proband's genomic DNA with requested depth of 100X. Exome enrichment was accomplished using Illumina TruSeq DNA Exome kit followed by paired-end sequencing of 151 bp reads on the Illumina NovaSeq 5000/6000 platform. FASTQ data was processed and analyzed by Biomedical Genomics Workbench 5.0.1 and CLC Genomics Workbench 12.0 softwares.

In brief, variants were called in VCF format and annotated using web-based tool, wANNOVAR [Yang and Wang, 2015]. After filtering out by allele frequencies in population databases, variants with MAF <0.05 were retained. The remaining variants were prioritized according to phenotype-relevant genes extracted from HPO (Human Phenotype Ontology) database. Regarding a recessive inheritance of the disease in the pedigree, homozygous variants were analyzed more intently.

Sanger sequencing was performed on an Applied Biosystems 3130XL Genetic Analyzer and the data were analyzed by Applied Biosystems Sequence Scanner Software v1.0 (for primers see online suppl. material 1A; for all online suppl. material, see www.karger.com/doi/10.1159/519633).

Total RNA was extracted from fresh peripheral blood received from the affected brother, using Invitrogen TRIzol reagent and was reverse transcribed using ExcelRT RP1300 kit. PCR was carried out on cDNA using gene specific primers followed by Sanger sequencing (for primers see online suppl. material 1B).

Protein Sequence Alignment, Homology Modeling, and Molecular Interaction Prediction

FASTA sequences of proteins were downloaded from UniProtKB database under entry codes P48509 (CD151 antigen), A1L157 (Tetraspanin-11), and P26006 (Integrin alpha-3). The sequence of the conserved domain CD151_like_LEL was obtained from NCBI Conserved Domains Database (CDD) v3.18–55570 PSSMs under accession number NP_004348.2. CD151 protein sequence NP_004348.2 alignment was performed using NCBI BLAST, BlastP suite web-based tool. Structure prediction and homology-modeling of proteins were carried out on web-based tool, SWISS-MODEL [Waterhouse et al., 2018]. Models were built on SMTL (SWISS-MODEL template library) based on templates of ID 6wvg.1.A for CD151 antigen or Tetraspanin-11, and 3ije.1.A for Integrin alpha-3. Molecular graphics and analysis were performed using UCSF ChimeraX software version 1.1 [Pettersen et al., 2021]. Web-based tool ClusPro 2.0 protein-protein docking was used for in silico simulation of protein interactions [Kozakov et al., 2017].

Clinical Features

Patient 1

The proband was a 16-year-old girl with unilateral renal agenesis, proteinuria, and chronic kidney disease (CKD) which led to ESRD and consequently renal transplantation. She was born at term from asymptomatic consanguineous parents and had a low birth weight (BW) of 2,200 g, indicating intrauterine growth restriction (IUGR). Following the genetic findings, dermatologic examination of the patient revealed dystrophic toenails and a single mild small pretibial lesion without any other skin involvements. Most of her teeth were decayed and a few of the molars were lost. Figure 1 depicts the familial pedigree.

Fig. 1.

Family pedigree of the patients. The proband (Sib 2), her older brother (Sib 1), and her younger sister (Sib 3) presented in this study are affected by various degrees of renal anomalies and skin involvements. Their parents are consanguineous and asymptomatic.

Patient 2

Her affected brother was 30 years old with a BW of 2,000 g (IUGR), unilateral renal agenesis, protein wasting kidneys, dystrophy of toenails and 2 of the fingernails, and thin hair. He was almost edentulous and had few blisters spread over the feet and the pretibial area (Fig. 2).

Fig. 2.

Pictures from the affected brother, patient 2. a, b Figures depict dystrophic toenails and a skin blister over the left foot. c Nail dystrophy presented in 2 of the fingers. d Another blister-like lesion on the distal pretibial area of the right leg.

Patient 3

Her sister was 14 years old with a BW of 2,000 gr (IUGR) and was affected by unilateral hypoplastic kidney and proteinuria, dystrophic toenails, pruritus, and few mild pretibial skin lesions. Her teeth decay was more severe than the proband's and all molars were lost.

Results

Genetic Findings

Table 1 depicts the quality control (QC) metrics and main characteristics of WES data analysis.

Table 1.

Information of WES data analysis and quality control (QC) metrics

Sequence statistics and QC metrics
Total number of reads	75,314,542
Average read length, bp	151
Target region, Mbp	45,327
Target Reads (post-alignment), %	73.75
Sequences with average base QV >30), %	98.24
Initial Mappable Reads), %	99.5
Duplicate reads (pre-alignment)), %	9.12
Coverage >25X, %	91.51
Coverage >50X, %	71.44
Data analysis
SNP, n	32,765
Stop gained, n	79
Indels, n	297
Transition to transversion (Ts/Tv) ratio	2.45
Het/Hom ratio	1.41

Data analysis revealed a homozygous pathogenic variant, c.351+2T>C, located in the splicing donor sequence of intron 5 in the gene CD151, NM_139029 (GRCh37). This substitution disrupts the 5′ canonical splice site by replacing the second nucleotide T of intron 5 with C (Fig. 3a). The variant was absent in population allele frequency databases and dbSNP, and in silico analysis affirmed its damaging effect. Literature search revealed a previous report of the variant in a patient of the same ethnicity but with some differences in symptoms and severity. Based on criteria from the American college of medical genetics and genomics and the association for molecular pathology (ACMG-AMP), this variant is classified as pathogenic [Richards et al., 2015].

Fig. 3.

Mutation c.351+2T>C of the gene CD151 and its consequence on splicing of the transcript. a Section of Sanger sequencing chromatogram of the DNA encompassing the exon 5-intron 5 boundary. The second nucleotide of intron 5, T, is replaced by nucleotide C, and its schematic location in the CD151 transcript (pre-mRNA) is shown. This substitution disrupts the conserved 5′ donor splice site of intron 5. Nucleotides have the same color as the respective intron or exon. b Aberrant splicing of the transcript using the 5′ donor site of intron 4 and the 3′ acceptor site of intron 5 (shown by scissors and intersecting lines) results in the loss of exon 5 from the mRNA and junction of exon 4–6. Sanger sequencing chromatogram displays the nucleotides sequence in the junction site.

Sanger sequencing of affected siblings confirmed their homozygosity for the variant, while their mother was heterozygous (Fig. 4). Sanger sequencing of PCR products from cDNA, encompassing exons 3–8 of the transcript, demonstrated the complete loss of the exon 5, containing 75 nucleotides, and junction of exon 4–6 (Fig. 3b).

Fig. 4.

a Visualization of the mapped reads track by Genome Browser view in CLC Genomics Workbench and the following tracks of Human reference sequences of genome, mRNA, coding sequence (CDS), and genes. The red column indicates the mutation c.351+2T>C. b–e Sanger sequencing confirmations of the variant in homozygous C/C state for the proband (b), affected brother (c) and affected sister (d), and T/C heterozygous state in the mother (e) are shown.

Protein in silico Findings

In order to investigate our conjecture about potential partial compensatory role between CD151 and other members of the tetraspanin superfamily, FASTA sequence of the CD151 protein was aligned with human reference proteins. The most similar protein sequence to CD151 was Tetraspanin-11 (TSPAN11) isoform 2, NP_001357230.1, which was 58% identical and had 78% positive residues. Aligning large extracellular loop (LEL) of CD151 revealed a region with 61% identity and 77% positives, encompassing the same residues 112–220 in TSPAN11 isoform 1, NP_001073978.1 (online suppl. material 2, Fig. 1). Structural prediction demonstrated similar conformations for TSPAN11 and CD151. It is noteworthy that TSPAN11 has 2 out of the 3 amino acids which are critical for strong associations between CD151 and integrins, namely, glutamine and arginine on residues 194 and 195, respectively [Kazarov et al., 2002] (Fig. 5).

Fig. 5.

Predicted structures of proteins CD151 (a) and TSPAN11 (b). Variable regions of large extracellular loops (LEL) in both proteins with identical or similar residues are shown in yellow and dissimilar residues on TSPAN11 are shown in red. A QRD^194-196 area is identified in which glutamine 194 and arginine 195 are the same in both proteins. Numbers indicate transmembrane domains (TD). SEL, small extracellular loop; CD, cytoplasmic domain.

Protein-protein interaction prediction showed a similar pattern for CD151 and TSPAN11 interactions with integrin α3. These simulations included interactions between cytoplasmic domains which can be considered in vivo. The most appropriate interaction, deduced from predictions, is between the large extracellular loop of CD151 or TSPAN11 and the proximal part of the extracellular domain of integrin α3 (online suppl. material 2, Fig. 2). In simulations, CD151 was skewed relative to an assumptive cell membrane which may be needed for making the interaction feasible. This pattern is in agreement with a recent study in which docking of CD151 to the C-terminal extracellular domain of integrin α6 showed that the transmembrane regions of CD151 were tilted with respect to the cell membrane plane [Jankovicova et al., 2020].

Discussion

CD151 mutations are associated with a very rare disease presented in the OMIM database as nephropathy with pretibial epidermolysis bullosa and deafness (MIM#609057); however, some characteristics such as deafness may not be found in all patients [Karamatic Crew et al., 2004; Vahidnezhad et al., 2018]. In this study, we presented 3 homozygous siblings for the known mutation c.351+2T>C, who exhibited varying degrees of symptoms among themselves and in comparison with previously reported patients (Table 2). Interestingly, Epidermolysis bullosa type pretibial skin blisters, the hallmark of the previous studies, are not present in 2 of the patients of this study (proband and patient 3, aged 14 and 16 years, respectively) and they had no history of this condition. Furthermore, discrepancy in renal abnormalities was remarkable, the proband was affected by unilateral renal agenesis and kidney failure in early adolescence, while the adult patient from the previous report of the same mutation had a history of protein wasting nephropathy.

Table 2.

Comparison of clinical features observed in patients with CD151 mutations

Abnormalities	c.351+2T>C				c.383insG
	This study			Vahidnezhad et al. [2018]	Karamtic Crew et al. [2004]
	Sib 1	Sib 2 (P)	Sib 3	N/A	Sib 1	Sib 2
Pretibial blistering	Fewa	−a	−a	Extended	+	+
Poikiloderma	−	−	−	+	N/A	N/A
Dystrophic nails	+	+	+	+	+	+
Alopecia	−	−	−	+	N/A	N/A
Unilateral kidney agenesis	+	+	−	−	+	−
Hypoplastic kidney	−	−	Unilateral	−	N/A	N/A
Protein wasting nephropathy	+	+	+	+	+	+
Endstage renal disease	−	+	−	−	+	+
Distal vaginal agenesis	N/A	−	−	N/A	N/A	+
Urinary incontinence	+	−	−	+	N/A	N/A
Dystrophic teeth	Edentulous	+	Severe	Edentulous	+	−
Oral erosions	−	−	−	+	N/A	N/A
Microstomia	−	−	−	+	N/A	N/A
Esophageal webbing and stricture	−	−	−	+	N/A	N/A
Gastroesophageal reflux	−	−	−	+	N/A	N/A
Malnutrition	+	+	+	+	N/A	N/A
Constipation	−	−	−	+	N/A	N/A
Ectropion	−	−	−	+	N/A	N/A
Lacrimal duct stenosis	−	−	−	+	+	+
Sensorineural deafness	−	−	−	−	+	+
Anemia	−	−	−	−	ΒT mi	BT mi
Cervical ribs	N/A	N/A	N/A	N/A	−	Bilateral
IUGR	+	+	+	N/A	N/A	N/A

BT mi, beta thalassemia minor; IUGR, intrauterine growth restriction; N/A: not applicable or available. ^a Sib 1 had only one pretibial and few toe blisters, Sib 2 just had a pretibial small mild lesion, and Sib 3 had a few small mild pretibial lesions, both of them without blistering.

The CD151 gene is located on chromosome 11p15.5, encoding a membrane protein with 253 amino acids. Aberrant splicing of pre-mRNA caused by the variant c.351+2T>C, leads to the deletion of the whole exon 5 from mRNA NM_139029. The lost region contains 25 amino acids of the third transmembrane domain and the large extracellular loop (LEL, EC2 domain) (Fig. 5; online suppl. material 2, Fig. 3). Although aberrant transcripts were not eliminated by nonsense-mediated mRNA decay (NMD), Vahidnezhad et al. [2018] have shown that the truncated protein was not present on the cell surface. It is probably because the imperfect product is unable to settle inside the plasma membrane.

CD151 is broadly expressed in various tissues such as skin basal keratinocytes and renal tubular and glomerular basement membranes (TBM and GBM), and it mostly interacts with laminin-binding integrins [Sincock et al., 1997; Yáñez-Mó et al., 2009].

The most abundant integrin in glomerular and tubular systems of kidney is α3β1 [Pozzi and Zent, 2013], and the supporting role of CD151-α3β1 interaction for strengthening podocyte-GBM adhesions has been established which may explain glomerulosclerosis and proteinuria observed in patients with CD151 mutations [Sachs et al., 2012].

Dermal involvement in CD151 mutations are also present in defects in α3 or α6β4 integrins which mediate adhesion of keratinocytes to their BM via hemidesmosomes (HDs) and focal adhesions (FAs), respectively [Rippa et al., 2013]. Enamel defects have been found in α6β4 mutations [Kambham et al., 2000; Dang et al., 2008; Has et al., 2012; Schumann et al., 2013]. Therefore, teeth decay associated with CD151 mutations may be caused by disturbed function of this integrin.

Extensive variable expressivity in patients with CD151 mutations may be explained in the context of different modifier genes such as variations in spliceosome complexes or protein families, namely tetraspanins and integrins for which functional redundancy and compensatory roles have been observed [Dunn et al., 2010; Viquez et al., 2017].

According to the results from protein alignment in the current study, CD151 has the most sequence similarity to TSPAN11, another member of the tetraspanin family, and it is also the case throughout the large extracellular loop. Furthermore, inspection of expression levels of TSPAN11 across human tissues involved in CD151 disease using The Human Protein Atlas and NCBI AceView, and expression data from NCBI Gene (online suppl. material 3; Fig. 1–3) showed that TSPAN11 has a low expression in kidney compared with skin. Interestingly, TSPAN11 is highly expressed in lungs. Considering the importance of interaction between CD151 and α3 integrin and the fact that CD151 mutations resemble ITGA3 mutations in kidneys and cutaneous phenotype but do not show abnormalities associated with ITGA3 mutations in lungs [Has et al., 2012; Pozzi and Zent, 2013; Colombo et al., 2016], it can be speculated that higher expression of TSPAN11 in lungs with its compensatory role has suppressed potential pulmonary effects in patients with CD151 deficiency. However, more studies on patients along with functional and in vitro analysis are needed in the future to elucidate this hypothesis.

Conclusion

This study is the third report of a very rare disease caused by CD151 mutations, highlighting the extensive variable expressivity amongst patients even in those having the same mutation. Notably, it must be considered that pleiotropic effects of the disease can be more severe in the renal system but very mild in skin, which may be overlooked in nephrological assessments. Proteinuric nephropathy and nail dystrophy are constant symptoms in all patients which can be indicative of the disease in clinical assessments. This broad spectrum of symptoms and severity may be modulated by other genes, and here, we proposed TSPAN11, another member of the tetraspanin family, to be a possible candidate for a partial modifying role in CD151 functions.

Statement of Ethics

This study has been approved by constituted Ethics Committee of Iran University of Medical Sciences under approval reference number 96.04.30.31916.

Informed consent forms have been received from the adult patient and the parent of the 2 adolescent patients. Permission for publication of pictures from the adult patient has been obtained.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study is supported by Research Deputy, Iran University of Medical Sciences under grant no. 96.04.30.31916.

Data Availability Statement

The data generated or analyzed during this study for protein in silico analysis are included in this article, its supplementary material files, or are publicly available from databases which are mentioned.

The patient's WES data are not publicly available due to information that could compromise the privacy of the research participants.

Author Contributions

N.R.: sample collection and WES data analysis, cooperated in genetic counselling, clinical data analysis, protein in silico analysis, and writing the manuscript.

S.T.: guidance and help with WES data analysis.

R.H.: patient management and clinical assessments.

N.A.K.: primer design and performing PCR.

A.S.: providing grant and supervision of the study, from sample collection and genetic counselling to data analysis and interpretation, and editing the manuscript.

Supplementary Material

Supplementary data

Supplementary data

Supplementary data