Whole exome sequencing of a family revealed a novel variant in the CHM gene, c.22delG p.(Glu8Serfs*4), which co-segregated with choroideremia

Choroideremia (OMIM 303100) is a rare retinal degeneration that is characterized by the loss of retinal pigment epithelium (RPE), with subsequent loss of photoreceptors and choroid as the disease progresses [1]. The degeneration gradually progresses in a centripetal pattern with a pale fundus [2]. Although phenotypes may vary among patients, most typically complain of nyctalopia and constriction of the visual field at early stages, which eventually lead to complete blindness at late stages [3–5]. Although the structure and function of the central retina are altered at early stages, most patients retain intact visual acuity in the fifth decade of life [6,7]. The reported prevalence is approximately 1:50000 in the general population and inheritance mainly occurs in an X-linked recessive manner [8].

So far, variants affecting the CHM gene (OMIM 300390) are the sole known cause for choroideremia [9,10]. The CHM gene is located on chromosome Xq21.2 and contains 15 exons spanning 186382 bp; it encodes component A of the Rab geranylgeranyl transferase holoenzyme, Rab-escort protein 1 (REP-1). Rab is a GTP-binding protein that regulates post-translational isoprenyl modification and intracellular vesicular trafficking [11,12]. The Rab geranylgeranyl transferase holoenzyme activates the Rab GGTase subunit for catalysis of the geranylgeranyl transfer reaction. REP-1 is involved in prenylation of Rabs in rods and RPE [13–15] and binds unprenylated Rab GTPases. Rab GTPases need to be geranylgeranylated on either one or two cysteine residues in their C-terminus to localize to the correct intracellular membrane. REP-1 is a key mediator of membrane trafficking and an essential component of the catalytic Rab geranylgeranyl transferase II complex in retina. Any disease-causing variant of the CHM gene can lead to a loss of geranylgeranyl transferase function and insufficient transfer of geranylgeranyl pyrophosphate groups on to Rab proteins. Variants in the CHM gene can lead to a deficiency of REP-1 and the onset of chorioretinal atrophy [16].

In the present study, we analyzed three generations of a Chinese family; we performed routine ophthalmological examinations and whole exome sequencing to investigate the pathogenic variant and molecular etiology associated with choroideremia in the family. The proband was diagnosed with choroideremia and had a novel hemizygous variant, c.22delG p. (Glu8Serfs*4), in the CHM gene.

All participants were recruited at Renmin Hospital of Wuhan University. The present study was authorized by the Institutional Review Board of Renmin Hospital of Wuhan University. Prior to the study, written informed consent was obtained from each participant or legal guardian, in accordance with the tenets of the Declaration of Helsinki.

Exhaustive family histories, clinical data, and peripheral blood samples were obtained from all available family members. Routine ophthalmological examinations were performed, including assessment of best-corrected visual acuity, refractive error, intraocular pressure, slit-lamp microscopy, and ophthalmoscopy. Selected family members underwent additional ophthalmic examinations including high-resolution fundus photography, spectral-domain optical coherence tomography (SD-OCT) and full-field electroretinography (ERG). Fundus photographs were obtained with a digital fundus camera VISUCAM 200 (Carl Zeiss Meditec AG, Jena, Thuringia, Germany). SD-OCT was performed using an AngioVue® Imaging System (Optovue, Fremont, California, United States). Full-field ERG was recorded using an Espion System (Diagnosys, Westford, Massachusetts, United States), in accordance with the standards and methodology of the International Society for Clinical Electrophysiology of Vision [17].

Genomic DNA of family members was isolated from leukocytes in venous blood samples by using the TIANamp Blood DNA Midi Kit (TIANGEN Biotech, Beijing, China), in accordance with the manufacturer’s instructions. Paired-end multiplex libraries of the proband were constructed using xGen Exome Research Panel v1.0 (Integrated DNA Technologies, Coralville, Iowa, United States), in accordance with the manufacturer’s instructions. Sequencing enrichment was performed on a HiSeq platform (Illumina, San Diego, California, United States) to generate 150–200-bp paired-end reads, in accordance with the manufacturer’s protocol [18,19]. Genomic DNA data were analyzed via whole exome sequencing. Sequence alignments were performed using the Burrows–Wheeler Aligner (http://bio–bwa.sourceforge.net/) [20]. Variant calling and annotation were conducted in accordance with previously reported protocols [21].

Analysis-ready alignment data were acquired by filtering raw reads to remove duplicates and performing local alignment to the hg19 (GRCh37) human reference sequence. Base quality was recalibrated by Picard Mark Duplicates (http://sourceforge.net/projects/picard/), Genome Analysis Toolkit (https://gatk.broadinstitute.org/hc/en-us), and SAM tools (http://samtools.sourceforge.net/) [22]. Single nucleotide variations, insertions, and deletions were identified using the Genome Analysis Toolkit. Copy number variants were detected using the Weaver algorithm by comparing average depth between the proband and normal human samples [23]. An allele frequency chart was used to calculate the allele frequency of each locus in the target region. Variants were preferentially selected for further analysis and validation if they met all of the following criteria: (i) minor allele frequency < 0.01 in the 1000 Genomes Project database (http://www.internationalgenome.org/), Exome Aggregation Consortium database (ExAC, http://exac.broadinstitute.org/), and Genome Aggregation database (gnomAD, http://gnomad.broadinstitute.org/); (ii) occurrence in exon regions or canonical splicing sites that affected RNA splicing; (iii) potential functional effects of nonsynonymous single nucleotide variants were predicted to be damaging or deleterious using multiple lines of computational prediction; (iv) candidate gene variants related to ophthalmic hereditary disease, especially for inherited retinal disease; (v) other reported potential pathogenic variants that did not met the above criteria (e.g., high minor allele frequency variants, deep-intronic variants, and synonymous single nucleotide variants). Variant nomenclature complied with the recommendations of the Human Genome Variation Society (HGVS, http://www.hgvs.org/) [24]. Variant annotation complied with the guidelines of the American College of Medical Genetics (ACMG, https://www.acmg.net/) [25,26].

The preferentially selected variants were validated and co-segregation was analyzed by Sanger sequencing. The polymerase chain reaction was used to amplify CHM gene fragments that included variants with the forward primer CCCAAAACTCGCCACTGACAGA and reverse primer CACAGAGCAAACCGCCTTCAATT using 3500xL Dx Genetic Analyser (Applied Biosystems, Foster City, California, United States) with the ABI BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, California, United States). Primers were designed with Primer 3 (http://primer3.ut.ee/). Participants’ sequences and corresponding consensus sequences (obtained from the NCBI Human Genome Database: https://www.ncbi.nlm.nih.gov/) were analyzed using SeqMan II software in the Lasergene software package (DNASTAR, Madison, Wisconsin, United States).

In total, seven family members were enrolled in the study. Patients II:3 and III:2 complained of night blindness, constricted visual field, and impaired vision in the first decade of life. They exhibited fundus signs typical of choroideremia including confluent RPE atrophy, choriocapillaris loss, widespread depigmentation around mid-peripheral or centripetal retina, and choroidal vessels in exposed choroid areas. SD-OCT revealed thinning of the retina and choroid at the macular fovea, increased signal transmission posterior to RPE and Bruch’s membrane, and absence of photoreceptor-attributable reflectivity bands. Full-field ERG demonstrated extinguished dark-adapted and light-adapted responses. The proband’s mother (II:2) complained of mild nyctalopia with normal visual acuity; fundus examination showed numerous scattered pigmentary changes, patchy areas of chorioretinal degeneration, and crystal-like retina appearance. SD-OCT and full-field ERG findings were normal. The proband’s daughter (IV:1) did not complain of any symptoms; her fundus, SD-OCT, and full-field ERG examinations revealed normal findings. The proband’s father (II:1) did not complain of any symptoms; his fundus examination revealed normal findings. Typical fundus, SD-OCT, and ERG findings of family members are shown in Figure 1. Clinical features and CHM variant information among the family members are summarized in Table 1. Based on the clinical manifestations, the proband was diagnosed with choroideremia.

Average sequencing depth and coverages of the whole exome were computed on all target regions designed to cover 19396 genes of human genome according to product information provided by the manufacturer. It showed that the average depth is 103, with 99.03 and 98.99% of the entire target region covered over 20× and 30×, respectively. In total, 45.7 million reads mapped to the human reference genome; 85738 variants were called in the proband sample, consisting of 71724 single nucleotide variants and 14014 insertions and deletions. No copy number variants associated with the phenotype were found in the proband. The copy number and allele frequency charts are shown in Figure 2A.

Based on a comparative analysis of whole exome sequencing data and the use of a prioritization program, we determined that the proband (III:2) had four variants that were likely pathogenic or of uncertain significance. These variants were as follows: c.22delG p.(Glu8Serfs*4) in the CHM gene, c.8502A>T p.(Glu2834Asp) in the USH2A gene, c.950T>C p.(Ile317Thr) in the NHLRC2 gene, and c.2975-6G>A in the CACNA2D4 gene. Potential functional effects of c.8502A>T in the USH2A gene and c.950T>C in the NHLRC2 gene were predicted to be tolerated or benign using multiple lines of computational prediction. The variants of c.22delG in the CHM gene and c.2975-6G>A in the CACNA2D4 gene are not frequently found in an ethnically matched population from East Asia, whereas the variants of c.950T>C in the NHLRC2 gene and c.2975-6G>A in the CACNA2D4 gene are frequently found in that population. Predictive functional effects and population distribution frequencies of these variants are summarized in Table 2. So, the hemizygous variant of c.22delG p. (Glu8Serfs*4) in the CHM gene is the most preferentially selected variant for further analysis. The variant is located in the first exon of the CHM gene. A schematic representation of the genomic structure of the CHM gene is shown in Figure 2B. The variant replaced glutamic acid with serine at the eighth amino acid, which may lead to generate a premature termination codon at residue 12. The effect of the variant on the generation of a predicted premature termination codon is shown in Figure 2C. Co-segregation analysis by Sanger sequencing showed that the proband's mother (II:2), uncle (II:3), and daughter (IV:1) carried the hemizygous variant, c.22delG p. (Glu8Serfs*4); in contrast, the proband's father (II:1), aunt (II:5), and cousin (III:3) did not carry the c.22delG p. (Glu8Serfs*4) variant. The Sanger sequencing chromatographs of seven family members are shown in Figure 2D. The variant was completely co-segregated with the choroideremia in this family. The pedigree of the family is shown in Figure 2E. In addition, the variant was a null variant that was predicted to disrupt gene function by causing complete absence of the gene product. Based on the above findings, we concluded that the variant c.22delG p. (Glu8Serfs*4) was likely to be pathogenic. The variant may lead to formation of a premature termination codon and presumably nonsense-mediated mRNA decay; this is predicted to produce a truncated REP-1 protein that exhibits loss of function. Thus, the variant c.22delG p. (Glu8Serfs*4) in the CHM gene is the genetic etiological factor for this family.

Patients with choroideremia typically present with early nyctalopia, but the majority retain excellent visual acuity until the end stages of disease. This preservation of visual acuity is potentially the result of continuing Müller cell function, which contributes to the cone visual cycle in the absence of RPE [27,28]. Female carriers of choroideremia typically exhibit a clinically heterogeneous phenotype and display variable severity of the disease due to random X-chromosome inactivation. In this state, cells carrying variant alleles are intermixed with cells expressing normal X chromosomes, thereby leading to a mosaic pattern of the disease [29]. Although both the proband’s mother and daughter carried the hemizygous variant, c.22delG, the proband’s mother exhibited a mild abnormal fundus appearance, whereas the proband’s daughter exhibited a normal fundus appearance. This discrepancy is presumably because retinopathy was below the range of detection in the proband’s young daughter; in addition, some female carriers may exhibit normal clinical manifestations throughout life.

In the study, we detected a novel hemizygous variant, c.22delG p. (Glu8Serfs*4) in the CHM gene. The frameshift variant was predicted to create a premature stop codon and lack of full-length REP-1. This would presumably cause a deficiency of geranylgeranyl transferase function and insufficient transfer of geranylgeranyl pyrophosphate groups on to Rab proteins, leading to abnormal protein function and prevention of their participation in intracellular vesicular transport. The ultimate outcomes of these changes would include several clinical manifestations, as reported in the study.

Thus far, 309 disease-associated variants in the CHM gene have been identified in patients with choroideremia, including 84 missense/nonsense variants, 51 splicing variants, 71 small deletion variants, 26 small insertion variants, 10 small indel variants, 55 gross deletion variants, 3 gross insertion variants, 4 regulatory variants, and 5 complex rearrangement variants (https://www.hgmd.cf.ac.uk, updated 2019.3). The majority of the reported variants are null variants that can facilitate gene replacement therapy for patients [30,31]. Most nonsynonymous variants involved C to T transition, which produces a higher number of CpG dinucleotides [32]. This transition can result in a premature stop codon due to arginine residue substitution, thereby considerably truncating the REP-1 protein. This transition is a well-known trigger of variants in the human genome and has occurred with high frequency throughout evolution because of inherent instability via methylation [33]. Exons 5, 6, and 11 are the most commonly affected exons [34]; in this study, the variant was located in exon 1.

Traditionally, choroideremia is typically diagnosed on the basis of family history and the results of clinical examination. Most patients with choroideremia have a characteristic fundus phenotype, which includes RPE atrophy, choriocapillaris loss, retina depigmentation, and choroidal vessels in exposed choroid areas [35]; this phenotype differs from the phenotype observed in patients with retinitis pigmentosa, which includes bone spicule pigmentation, retinal vascular stenosis, and a waxy-pale optic disc [36]. Both patients with retinitis pigmentosa and patients with choroideremia often complain of nyctalopia, visual field restriction, and reduced visual acuity, including blindness in the late stages of disease [37,38]; in addition, some female carriers show mild phenotypes due to a mosaic X-linked inheritance pattern. Therefore, it is difficult to achieve an accurate diagnosis for some patients with atypical choroideremia [39] and choroideremia is often misdiagnosed as retinitis pigmentosa [40]. Exome sequencing is a powerful tool to identify causative variants and distinguish between choroideremia and retinitis pigmentosa [41,42]. In a study by Li et al., exome sequencing helped to achieve a revised diagnosis for six probands with atypical choroideremia, all of whom were initially misdiagnosed with retinitis pigmentosa [43]. Genetic sequencing can clarify the diagnosis and identify up to 94% of disease-causing variants in patients with choroideremia [44]. Whole exome sequencing is a non-invasive and non-biased diagnostic method that can accurately diagnose inherited eye diseases with complex clinical manifestations [45,46]. This approach is more efficient and cost-effective than Sanger sequencing and whole-genome sequencing [47–49]. Therefore, whole exome sequencing is extremely useful for the identification of genetic diseases by clinicians and researchers.

It was reasonable to conclude that the likely pathogenic variant c.22delG p. (Glu8Serfs*4) was the disease-causing variant of the family on the basis of co-segregation analysis, multiple lines of computational prediction, and absence of the variant in a control population; however, there was a notable limitation in the present study. We enrolled a small family, rather than a large, multigenerational family. Because of the complex inheritance, atypical clinical presentation of patients, and mild phenotypes of female carriers, a small pedigree is not sufficient to disentangle the genetic etiology of choroideremia. Thus, we plan to enroll multiple families in future research studies.

In conclusion, we analyzed a Chinese family with night blindness, constricted vision field, and impaired vision. The proband was diagnosed with choroideremia based on the findings of ophthalmological examinations. Through whole exome sequencing, we found that the proband had a hemizygous variant in the CHM gene, c.22delG p. (Glu8Serfs*4), which has not previously been reported. Sanger sequencing showed that the variant co-segregated with choroideremia within the family. These results expand the spectrum of variants in the CHM gene, thus enriching the understanding of the molecular basis, family genetic counseling, and clinical management of choroideremia.

The authors declare that there are no competing interests associated with the manuscript.

This work was supported by the National Key R&D Program of China [grant number 2017YFE0103400].

Study design, Yin Shen. Data curation, Xin Huang. Formal analysis, Yiqiao Xing. Funding acquisition, Yin Shen. Investigation, Xin Huang. Methodology, Handong Dan. Project administration, Yiqiao Xing. Resources, Tuo Li. Software, Xin Huang. Supervision, Yin Shen. Validation, Tuo Li. Visualization, Yiqiao Xing. Writing – original draft, Handong Dan. Writing – review and editing, Yin Shen.

They are available on special request.

The authors would like to thank all the subjects who participated in the present study.

ERG

electroretinography

REP-1

Rab-escort protein 1

RPE

retinal pigment epithelium

SD-OCT

spectral-domain optical coherence tomography