A novel homozygous splice site variant in the CLCN7 causes osteopetrosis

2.1 Blood mononuclear cells isolation and study approval

Peripheral blood mononuclear cells (PBMCs) were gathered from the probands (II-1) from family A; (II-2; II-4) from family B and their normal parents and siblings (Fig. 1). According to the manufacturer's recommendations, PBMCs were separated using Ficoll-prefilled Leucosep® tubes by density gradient centrifugation. The trypan blue exclusion test was used to determine the present viability of PBMC (cells should be > 85 %-90 % viable). The King Abdullah International Medical Research Center (KAIMRC), Ministry of National Guard, Health Affairs, Riyadh, Saudi Arabia, served as the source of patients for this study. The Institutional Review Board provided their approval to the present study (approval# KAIMRCIRB- H-01-R-005) according to Helsinki protocols. Written informed consents were obtained from the parents to publish these cases, photographs and their molecular data.

Close

Fig. 1

Pedigrees and sequence analysis of the families with autosomal recessive osteopetrosis. (A), (B) Pedigree of the osteopetrosis family A indicating affected family members (fully shaded). The arrow shows the probands. (B) Images of Probands with CLCN7 variant. Representative images of the probands at age 18 and 12 show the patient's face and ears, respectively. (C) Sanger sequencing chromatograms show the reference sequence in the top panel. The middle panel represents the nucleotide substitution in the patient's homozygous sequence. The last one represents the carrier individual in heterozygous states, respectively.

Export to PPT

2.2 Genomic DNA extraction

High-quality DNA was extracted manually utilizing the QIAamp® DNA Mini kit method (QIAGEN®, Germany) from the whole blood according to the manufacturer's guidelines. Briefly, cellular membranes were ruptured by mechanical lysis and enzymatic activity (shaking and heating); DNA was bound to the silica-gel membrane of the QIAamp spin column; the impurities were washed across the membrane using buffers, and DNA was eluted. Quantification of gDNA (ng/μl) and purity (A260/A280) were assessed by a Nanodrop™ spectrophotometer (ND-1000, V3.8.1, NanoDrop Technologies, U.S.A.) following slandered methods.

2.3 Whole-exome sequencing (WES)

Whole-exome sequencing was achieved for each patient (II-1) from family A; (II-2; II-4) from family B described (Alfadhel et al., 2021a,b). The genomic DNA (g-DNA) was fractionated using a solicitor, and Illumina adapters were ligated and sequenced on the HiSeqX platform (Illumina) with an average ∼ 30X coverage depth. Further, an in-house pipeline base calling was performed. Diseases-causing variants were illustrated in ClinVar, HGMD®, CentoMD®, or PubMed and variants with minor allele frequency (MAF) of<1 % in the ExAC/gnomAD database were given preference.

All inheritance patterns were considered during the filtration of variants; however, homozygous and compound heterozygous variants of the pedigree recessive pattern were favored. The study focused primarily on pathogenic disease-causing non-synonymous (NS) variants. Moreover, clinical details and phenotypes provided from patients were used to demonstrate genotype-phenotype correlation.

2.4 In silico prediction

In silico prediction applications used to investigate the identified variant included MutPred Splice, NNSplice (Berkeley), Human Splice Finder, SKIPPY, MutationTaster, and Varsome.

2.5 Sanger sequencing

Sanger sequencing was performed on all the available family members to verify the results identified in the patients using WES. The primers for CLCN7 (NM_001287.5) were designed using Primer3 online software (https://frodo.wi.mit.edu/primer3/). The following primers sequence were used for PCR amplification forward primer 5‘-GGC ACA CTT GGT CTC AGG TT −3‘ and reverse primer 5‘- GTG CTG TTT TAC GGC TCA GG −3‘. The target sequences were amplified using PCR and then examined by agarose gel (2 %) electrophoresis. The sequencing was performed on a 3730 XI DNA Analyzer (Applied Biosystems®, HITACHI, Japan). The acquired sequence data were interpreted using Mutation Surveyor® software, version 5.1.2.

2.6 RNA extraction

Total RNA was extracted from the (1 × 10⁸) PBMCs samples and cultured fibroblasts (10⁶ cells) of patients using the QIAamp® RNA Mini kit method (QIAGEN®; Germany) following manufacturer instructions of blood and cells, respectively. Extracted RNA purity and quantified was assessed using a nanodrop™ spectrophotometer (ND-1000, V3.8.1, NanoDrop Technologies, U.S.A.) following standard procedures (Asiri et al., 2020).

2.7 Rt-PCR

To quantitatively monitor CLCN7 (NM 001287.5) mRNA expression to the internal control “house-keeping” gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH); DQ403057, total RNA was extracted, and the high-capacity cDNA reverse transcription kit (Applied Biosystems; Lithuania) was used to create the cDNA by PCR-recommended cycles in four steps: 25 °C for 7 min, 37 °C for 120 min, 85 °C for 5 min and at four °C for ∞. RT reactions were performed using 2 μg of RNA, previously subjected to DNase digestion and reverse transcription reagents, followed was subjected to thermocycler (applied biosystems®, Veriti; 96 Well Thermal Cycler, REF:4375786, Singapore). The primer sequences for the CLCN7 cDNA amplification and GAPDH will be shared upon request. The primers were designed using Online Primer3 software (https://frodo.mit.edu/primer3/).

The RT-PCR reaction was executed using an optimized Dream Taq Green PCR Master Mix (2X) (Thermo Scientific®, Lithuania) on a thermocycler PCR System (Applied Biosystems®, Singapore) by PCR recommended cycles. The initial denaturation step at 95 °C for 7 min; then 35 cycles at 95 °C for 30 sec (Denaturing); for 30 sec (Annealing) at the different temperatures depending on the primers that were used, and at 72 °C for 1 min (Extention); and a further extension at 72 °C for 7 min. All the RT-PCR reactions were carried out in triplicate and repeated independently. The results were analyzed on agarose gel (2 %) containing trace amounts (0.3 µg/ml) of ethidium bromide. The gel was run for 35 min, 120 V (BioRAD®, REF: 041BR 25589, USA) and visualized by ChemiDoc™ MP Imaging System with UV light (BioRAD®, REF: 731BR 03013, USA) (Asiri et al., 2020; Lien et al., 2022).

2.8 Protein extraction and western blotting

Whole-cell protein lysates of fibroblast cells (10⁶ cells) were performed on 100 µl of lysis buffer 6 (biotechne®, USA) and 1 µl of protease inhibitors (Thermo Scientific, USA). Protein Quantification was performed using Qubit Fluorometer 3.0 (Invitrogen, REF: Q33216, USA). Protein was loaded in equal amounts (100 µg) into the wells of a Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) gel, along with a molecular weight marker (BIO-RAD, Singapore). Proteins were transferred to the PVDF membrane to complete the steps of western blotting. The western blot signals were measured using an ODYSSEY CLx (LI-COR®, Model: 9140, USA). The Actin antibody was used as a reference for loading. The antibodies used in the present study include Rabbit Polyclonal anti-ClC-7 Antibody, 1:2500 (Thermo Scientific, Cat. No. PA5-60203, USA). Mouse Polyclonal anti-Actin Antibody (loading control), 1:1000 (Abcam, Cat. No. ab1801, USA). Goat anti-Rabbit 926–68071, IRDye® 680RD, 1:10000 (LI-COR, Lot NO. C40723-01, USA). Goat anti-Mouse 926–32210, IRDye® 680RD, 1:10000 (LI-COR, Lot NO. C50113-06, USA).

2.9 Statistical analysis

One-way ANOVA test was utilized to interpret the functional study result, including Western blot; all findings with a P-value of < 0.05 were regarded as significant (Al Mutairi et al., 2020).

3.1 Family A

4.1 Case description

An affected individual is an 18-year-boy (II-2) with a history of multiple fractures, bilateral optic atrophy with nystagmus, right lower motor neuron (LMN), facial palsy, and a similar history in his brother (II-4), 12-year-boy-old, were investigated in the present study (Fig. 1B).

Medical history explained the first concern of the affected individual at 12 months when his mother was noticed delayed walking and decreased eyesight. After two years, he developed fractures. Pregnancy was unremarkable, and the birth was full-term normal spontaneous vaginal delivery (FT- NSVD). The index had only jaundice postnatally.

Developmental history showed gross standing and walking at three years. However, now the patient is restricted to a wheelchair due to femur (thigh bone) fractures. Furthermore, he had poor vision and used the braille style that reflected fine motor development. The proband could say single words at three years; however, now he is partly talking. The affected individual (II-2) is socially up to his age; however, he suffers developmental delay (Fig. 1 B). Family history includes healthy first-cousin parents; they have a healthy-one brother and two affected brothers. The first one had a fracture and unilateral blindness (II-4), but the other died at ten months with the same thrombocytopenia (II-1) (Fig. 1 B). Their paternal aunt, 35 years, had hydrocephalus with shunt and bone disease, multiple recurrent fractures, and osteogenesis.

Physical examination revealed his height and weight were 152 cm [< 1 percentile (-3.1 SD)] and 52 kg [5 percentile], respectively. His Head circumference (cm) was 59.5 [> 99 percentile (+3.1 SD)]. The index's head and neck dysmorphic showed dolichocephaly, long face, prognathism, high forehead, prominent nose, and low set ears. CVS examination reflected well perfused. GI examination explained no distended organomegaly. The affected individual suffered from vitiligo. Radiographs of musculoskeletal showed a general greater bone density with multiple fractures.

4.2 WES and Sanger sequencing

The two families (A and B) were subjected to WES using standard methods (Alfadhel, 2019; Alfadhel et al., 2021a,b). Step-by-step filtering and validation of various compound heterozygous and homozygous variants revealed a splice site variant (c.739-18G > A); (p.Met250Argfs*6) in the intron 8 of the CLCN7 gene (NM_001287.5; NP_001278.1).

Using the Sanger sequencing method, the identified variant segregated perfectly with the disease phenotypes in both families (Fig. 1 C). The homozygous state of the variant was present in the patients. To exclude the non-pathogenic of the identified variant, it was screened within ExAC, gnomAD, and 2000 Saudi exome databases. The pathogenicity index was calculated using various online analysis applications [(FATHMM-MKL: Damaging, Mutation Taster: Disease-causing, Varsome: PM2, PP3, DANN: 0.9924)] and was predicted disease-causing variant. The CLCN7 variant (c.739-18G > A) is novel and not reported in any currently genetic variant databases such as the human gene mutation database (HGMD).

4.3 Rt-PCR

To independently validate and quantify the splice defect identified by WES and Sanger sequencing, we performed an RT-PCR assay using primers flanking intron 8 from exon 7 to exon 9. Interestingly, RT-PCR results confirm significant retentions of intron 8 in patients' samples (Fig. 2 A). These findings expand the understanding of the association between a synonymous variant of CLCN7 and the molecular pathogenesis in osteopetrosis patients.

Close

Fig. 2

(A) Gel image showing the RT-PCR results of the index, father, mother and control samples. Lane (L: 1) showing 100 bp ladder, lane (L:2) index [449 bp], lane (L: 3) father, lane (L: 4) mother, and lane (L: 5) represented the control [399 bp] (healthy individual). The control shows a clear single band size of 399 bp. While the index showed a 449 bp band, suggesting the insertion of 50 bp in the index, both parents showed double bands of 449 bp and 399 bp. (B) Schematic representation of CLCN7 exons and introns. Arrow represents the identified variant's position and retention of the 50 bp sequence of intron 8. As a result, this variant results in large exon 9 as compared to the wild type. (C) Representing exon 8, intron 8, and exon 9 in the CLCN7 gene. (D) Showing the cDNA-Sanger sequencing electrogram results of the control. (E) cDNA-Sanger sequencing electrograms of the index, analysis showing 50 bp of intron 8 and exon 9. (F) The green shaded area represents the nucleotides in the intron 8 confirms the presence of intron sequence in index cDNA.

Export to PPT

Retention of the intron may cause disease or regulate gene expression. That happens by introducing additional protein sequences that prohibit the transcript from leaving the nucleus if the retained intron introduces a premature stop codon or might lead to nonsense-mediated mRNA decay (NMD). The trend toward lower CLCN7 levels proposes that intron retention leads to NMD.

RT-PCR of the novel variant abnormality within the CLCN7 gene in osteopetrosis showed significantly increased retention in the homozygous condition of intron 8. The identified splice site variant resulted in the retention of 50 bp of intron 8, thus adding 50 bp extra sequence to exon 9 (Fig. 2 B). Retention of intron 8 results in a premature stop codon in the exon 8 that is 24 bp downstream and is thought to lead to NMD of the mutant transcript or a truncated protein (246 amino acids versus wild-type 805 amino acids).

4.4 Western blot analysis

Western blot was performed to assess the impact of the pathogenicity of the identified mutation at the protein level (Al Mutairi et al., 2020). The fibroblast lysates from the affected individual were subjected to Western blot analysis to detect ClC-7 (Fig. 3A). Chloride-proton exchangers and ClC-type chloride channels exhibit an evolutionarily, distinctive conserved double-barreled structure with two ion conduction pathways created by a single subunit. The functional protein of hetero-dimeric channels comprises a homogenous population of functionally distinct ClC subunits. Western blot analysis showed that the heteromeric form of ClC-7 disappeared in the patient’s fibroblasts. However, the monomeric form corresponding to the standard ClC-7 expression level revealed no significant change in the patient’s fibroblasts. Our findings suggest that the mutation may lead to an anomaly at the binding site, involved in inter-subunit interactions, which impedes the formation of heteromeric ClC channels. This outcome is consistent with the predicted loss-of-function effect of the deep intronic variant, which may be related to an impact of the p.Met250Argfs*6 variant on the protein conformation and stability (Fig. 3B). Thus, this protein expression level analysis further supported the implication of the CLCN7 variant (c.739-18G > A) on our patient’s phenotypic features.

Close

Fig. 3

(A) Representative western blot showing the protein expression levels of homomeric and heteromeric ClC-7 detected in fibroblast lysates isolated from healthy individuals (i.e., Control) and patient (i.e., Index). The heteromeric ClC-7 protein expression level fully disappeared in the index compared to the control fibroblasts, whereas the homomeric ClC-7 protein expression level did not change. (B) Bar graph indicating the relative homomeric and heteromeric ClC-7 protein expression levels related to actin, a housekeeping cytoskeletal protein used as a loading control. Data are shown as mean ± standard deviation (SD) (n = 3). **: p = 0.006528 (One-way ANOVA).

Export to PPT