Two Novel and Two Recurrent Variants of the ADAR1 Gene in Three Chinese Families with Dyschromatosis Symmetrica Hereditaria

Introduction

Dyschromatosis symmetrica hereditaria (DSH, OMIM#127400), also called symmetric dyschromatosis of the extremities or reticulate acropigmentation of Dohi, is a rare autosomal dominant inherited pigmentary dermatosis.¹ It is characterized by intermingled hyper- and hypopigmented spots primarily located on the dorsal aspects of the extremities. Some affected individuals may also exhibit freckle-like lesions on the face.² This condition typically manifests in infancy or childhood, worsens during adolescence, and persists throughout life.³ DSH is caused by pathogenic mutations in the adenosine deaminase acting on RNA1 (ADAR1) gene, which encodes RNA-specific adenosine deaminase 1 composed of 1,226 amino acid residues.⁴ The ADAR1 protein acts as an RNA editing enzyme by catalyzing the deamination of adenosine (A) to inosine (I) in double-stranded RNA (dsRNA) substrate during transcription.⁵ ADAR1 activity is involved in various processes, including viral inactivation, structural modification of proteins, and cell survival.⁶

To investigate the pathogenic variant of the ADAR1 gene, we conducted whole-exome sequencing (WES) and Sanger sequencing on three Chinese families with DSH. Furthermore, we utilized bioinformatics tools to predict the pathogenicity of the variants.

Material and Methods

Patients

Three Chinese families with DSH were enrolled in this study. We collected detailed clinical data and pictures. Approximately 5 mL of peripheral blood or 1 mL of saliva was collected from the participants. All procedures were approved by the Ethics Committee of the Second Affiliated Hospital of Nanchang University. Informed consent was obtained from all individual participants included in the study.

Whole-Exome Sequencing

Genomic DNA was extracted from peripheral blood samples using a Puregene Blood Core Kit B (QIAGEN, Hilden, Germany) following standard procedures. WES was conducted using the SureSelectXT Human All Exon V6 kit (Agilent Technologies, Santa Clara, USA).

For each subject, 1.5 µg of genomic DNA was utilized to prepare a captured library, which was subsequently sequenced on a HiSeq X Ten platform (Illumina, San Diego, USA), generating 150 bp paired-end reads. The average depth of target region coverage was more than 133 reads per bp, and 94.61% of the bases had a target coverage of at least 30x. In terms of sequence quality, an average of 93% of the bases achieved a quality score of Q30 where Q30 denotes the accuracy of a base call to be 99.9%. Raw data of approximately 10 GB per exome were mapped to a human reference genome sequence (GRCh37/hg19) using the Burrows–Wheeler Alignment (BWA) tool. Variant calling was performed using the Genome Analysis Toolkit (GATK), and all variants were further annotated with ANNOVAR software.

Gene Filtration

Rare and novel protein-altering variants (missense, frameshift, nonsense, and essential splicing-site) were preferred in this study. Variant frequency was analyzed based on different ethnic subgroups from the Exome Aggregation Consortium (ExAC), the Genome Aggregation Database (gnomAD), and the 1000 Genomes Project. Coding sequence alterations (exonic) and part of noncoding sequence variants (exon-intron boundaries) that present an unknown frequency or minor allele frequency <1% in these databases were reserved.

Sanger Sequencing

Sanger sequencing confirmed the suspected pathogenic variants. Primer pairs were designed by Primer 5.0 software to amplify the exons, including the exon/intron boundaries (Table 1). Samples were subsequently amplified by PCR, and products were directly sequenced using a 3730xl Genetic Analyzer (Applied Biosystems). PolyPhred Analysis Software was used to compare and analyze the sequence.

Table 1 Amplification and Sequencing Primer Pairs

ADAR1 Protein Structure Visualization and Function Prediction

Three-dimensional structures of ADAR1 protein encoded by the wild-type and mutated ADAR1 gene were predicted using Swiss-Model (http://swissmodel.expasy.org). The variant effect was assessed by Mutation Taster (http://www.mutationtaster.org/), PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/), and Protein Variation Effect Analyzer (PROVEAN) (https://provean.jcvi.org/index.php). When using Mutation Taster for prediction, the closer the score is to 1, the more likely it is to cause disease. PolyPhen-2 prediction scores range from 0 to 1, and the prediction results are divided into “probably damaging”, “possibly damaging” and “benign”. If the score is close to 1, it is predicted that the pathogenicity will be greater after amino acid changes. When using the PROVEAN score, if the score is lower than −2.5, then the variation is predicted to be harmful.

Results

Clinical Manifestations

As shown in Figure 1, family 1 consists of five affected and four unaffected individuals. The proband in family 1 (III-1) was a 13-year-old boy. He had hyperpigmented and hypopigmented macules on the dorsal aspects of his hands and feet at the age of three and then gradually became prominent (Figure 2A). The skin lesions became more pronounced after exposure to sunlight. The other affected family members also exhibited similar skin lesions (Figure 2B–D). Except for his mother, the proband is not consanguineous to the above family members. Family 2 includes two affected and two unaffected individuals (Figure 1). The proband of family 2 (II-2), a 19-year-old male, displayed intermingled hyperpigmented and hypopigmented macules located on the dorsal aspects of the extremities (Figure 2E). His father also exhibited similar lesions in the same location (Figure 2F). Family 3 includes 9 affected and 26 unaffected individuals (Figure 1). The proband in family 3 was a 32-year-old woman (III-17). She had an asymptomatic mixture of hyperpigmented and hypopigmented small macules on the back of her hands and feet since she was 7 years old, and gradually progressed to both knees and the face (Figure 2G). All affected individuals in the pedigree showed similar eruptions (Figure 2H).

Figure 1 Pedigree charts of DSH in this study. Roman numerals (I, II, III, IV) indicate generations.Square indicates male and cycle indicates female. The filled symbols represent affected members, and the arrow indicates the proband.

Figure 2 Clinical phenotypes of partly affected individuals. (A) Family 1 III-1; (B) Family 1 II-3; (C) Family 1 II-2; (D) Family 1 III-2; (E) Family 2 II-2; (F) Family 2 I-2; (G) Family 3 III-17; (H) Family 3 IV-2.

Mutation Analysis of ADAR1 Gene

In this study, we identified four variants of the ADAR1 gene (NM_001111.4) among three Chinese families with DSH (Table 2). Two novel missense variants were identified in family 1. The novel variant c.2369G>C (p. Arg790Pro) in exon 7 of the ADAR1 gene was found in the proband, his mother, and two young sisters (Figure 3A). Another new missense variant c.503C>T (p. Pro168Leu) in exon 2 of the ADAR1 gene was found in his stepfather and asymptomatic young brother (Figure 3B). Additionally, two recurrent ADAR1 variants were identified in family 2 and family 3. In family 2, a missense variant c.3232C>T (R1078C) in exon 13 of the ADAR1 gene was identified in the proband and his father (Figure 3C). In family 3, the nonsense variant c.1472C>G (p. Ser491X) in exon 2 of the ADAR1 gene was identified (Figure 3D).

Table 2 Mutation Analysis of ADAR1 Gene in This Study

Figure 3 Genetic mutation of DSH in this study. (A and B) Two novel missense variants, c.2369G>C and c.503C>T were identified in family 1; (C) A missense variant c.3232C>T was identified in family 2; (D) A nonsense variant c.1472C>G was detected in family 3. The black arrow shows the mutation site.

Bioinformatics Analysis of the Mutation

Based on the comprehensive evaluations of Mutation Taster, PolyPhen-2, and PROVEAN, all mutations were unequivocally classified as disease-causing variants. The novel c.503C>T mutation was predicted as “deleterious” (score = −2.704) by PROVEAN, and “probably damaging” (score = 1) by PolyPhen2. Similarly, the other novel c.2369G>C was also predicted as “deleterious” (score = −4.167) by PROVEAN, “probably damaging” (score = 1) by PolyPhen2, and “disease-causing” (p = 0.999) by Mutation Taster. For novel mutations, we constructed 3D models of the wild-type proteins and mutant proteins c.503C>T: p. Pro168Leu (Figure 4A and B) and c.2369G>C: p. Arg790Pro (Figure 4C) using Swiss-Model.

Figure 4 Three-dimensional structure of wild-type proteins and the novel mutant proteins of DSH in this study. (A)The original 3D structure of ADAR1; (B) The 3D structure of the protein products of ADAR1 c.503C>T; (C) The 3D structure of the protein products of ADAR1 c.2369G>C.

Discussion

DSH was first discovered by Toyama in 1910 and formally named as a clinical entity in 1929.It has been reported primarily in Chinese and Japanese populations, although cases have also been observed among Europeans, Indians, and South Americans.^7–12 ADAR1 was mapped to chromosome 1q21.3 and identified as the gene responsible for DSH by Chinese and Japanese researchers in 2003.^13,14 The gene spans 30 kb and contains 15 exons.¹⁵ ADAR1 includes two adenosine deaminase Z-alpha (Zalpha) domains, three double-stranded RNA binding motifs (DSRMs), and a tRNA-specific and double-stranded RNA adenosine deaminase (ADEAMc) domain located in exons 2, 2–7, and 9–14, respectively.^10,16 To date, over 200 mutations in ADAR1 have been reported among DSH patients. These mutations are scattered within the gene with no specific hotspot, but most DSH-associated missense mutations identified so far are within the ADEAMc domain.⁴

In our study, we identified two new pathogenic variants in family 1: c.2369G>C (Arg790Pro) and c.503C>T (Pro168Leu). This is the first report of two different ADAR1 variants being identified in a DSH family. The novel ADAR 1 variant c.2369G>C is a missense variant that results in an arginine to proline substitution at codon 2369, located in the DSRMs domain and potentially affecting RNA editing efficiency. Another novel ADAR1 variant, 503C>T (Pro168Leu), is a missense variant that leads to a proline to leucine substitution at codon 168 in the highly evolutionarily conserved z-alpha adenosine deaminase domain, disrupting important atomic interactions between the protein and DNA/RNA. Furthermore, a previously reported missense variant, ADAR1 c.3232C>T (R1078C), was identified in family 2. This variant was originally identified in a Chinese family with DSH by Yang et al in 2004.¹⁴ Lastly, another previously reported nonsense variant, ADAR1 c.1472C>G (p.S491X), was found in family 3. This mutation was previously found in a Japanese DSH family by Kawaguchi et al in 2012.¹⁷ This nonsense variant results in a truncated protein lacking the deaminase domain, rendering the ADAR1 enzyme inactive.

The genotype-phenotype correlation for DSH remains unclear. Our study observed that family members with the same mutation can exhibit varying phenotypic severities within the three families. Similarly, Kawaguchi and Zhang et al reported the same mutation could lead to different phenotypes in one Japanese and one Chinese family.^17,18 Factors such as exposure to ultraviolet light, infection, and chilblains may influence phenotype expression. Further research is needed to explore the relationship between genotype and phenotype.

Conclusion

In summary, our study identified two novel and two recurrent ADAR1 variants in three Chinese families with DSH. This research has expanded the ADAR1 gene database for DSH, enhancing our comprehension of the underlying mechanisms.