DSH is an autosomal dominant skin disorder characterized primarily by hyperpigmentation and hypopigmentation spots on the dorsal extremities [26]. It may also be associated with other complications, including dystonia, acromegaly, psoriasis, and depression [27]. DSH is characterized by a symmetrical distribution of pigmentation and pigmented spots on the dorsal sides of the hands and feet. This pigmentation occurs in infancy or early childhood and usually stops progressing before puberty. Histologically, a focal increase or decrease in the amount of melanin or the number of melanocytes in the basal layer of the skin is observed. In our study, all the probands developed freckle-like macules on the face with an irregular distribution of pigment spots on the dorsal aspects of his hands and feet. In addition, family 2 is part of a three-generation pedigree that includes five affected individuals (Fig. 1B). Mutations in ADAR1 comprise the genetic basis of DSH. However, the mechanism through which ADAR1 mutations cause DSH is still unclear.
ADAR1 gene, located at 1q21.1-q21.2, encodes an RNA-specific adenosine deaminase that represents a type of RNA-editing enzyme; it spans 30 kb, contains 15 exons, and the encoded protein contains 1226 amino acid residues (Fig. 3) [28]. Two adenosine deaminase domains, Zα, and three DSRMs that are spread over exon 2 and exons 2–7 are responsible for RNA editing by site-specific deamination of adenosine [29]. The tRNA-specific double-stranded RNA adenosine deaminase (ADEAMc) domain is encoded in exons 9–14. ADAR catalyzes the deamination of adenosine in double-stranded RNA to form inosine (A-to-I editing), which is a type of post-transcriptional modification widely present in mammals [30]. Every DBRM domain can enhance the efficiency of editing, and the ADEAMc domain is essential for editing [31]. Since inosine is recognized as guanosine in the process of translation, the editing of the A-to-I RNA-coding sequence may lead to a change in amino acids and protein function. Thus, A-to-I editing may change the splicing patterns, RNA structure, and stability [32]. Abnormal RNA editing may cause the differentiation of melanoblasts into hypoactive melanocytes, thus establishing an irregular colonization of the lesioned skin [33].
To date, a total of 267 clinically significant sequence variants of ADAR1 have been obtained from HGMD linked to dbSNP, consisting of 147 missense and nonsense mutations, 27 splicing mutations, 2 regulatory mutation, 61 small deletions, 25 small insertions, 4 small indels, and 1 gross insertion [4, 8, 9, 34–45]. Analysis based on the Clinvar, UniprotKB, HGMD, and DBSNP databases showed that the five novel ADAR1 mutations identified in this study had not been reported previously. To date, over half of the known missense mutations are located within the ADEAMc domain encompassing amino acids 839–1222 [46]. The two missense mutations identified in our study were also present in this domain. These findings suggest that the ADEAMc domain is critical for the function of this enzyme and may be defined as a potential mutational hotspot region within ADAR1 [36, 46, 47]. To date, 32 nonsense mutations have been reported in ADAR1, and they seem to be randomly spread throughout the genome having apparent unifying connections.
SWISS-MODEL analysis was used to evaluate homology modeling of the tertiary structure of wild-type and mutant ADAR1 and examine the impact of gene mutations on the tertiary structure of the protein [23]. Our SWISS-MODEL modeling showed that the nonsense mutations in exon 2 resulted in significant truncation of the protein, leading to complete deletion of the DSRM and ADEAMC domains of the ADAR1 protein that are considered key structural domains for its function. The other three mutations (p.R239Q, p.W1128C, and p.C1129G) resulted in structural alterations of DSRM and ADEAMC domains of ADAR1 protein without significant changes in spatial structure; however, the prediction tools, SIFT, Polyphen-2, and MutationTaster, showed that all of them were pathogenic mutations (Table 1).
To date, cumulative studies have shown that mutations in ADAR1 contribute to several hereditary diseases, including DSH, AGS, bilateral striatal necrosis, dystonia, athetosis, Leigh-like syndrome, and others, whose hereditary pattern may change [48, 49]. Table 3 summarizes the distribution of mutations in ADAR1 according to HGMD. DSH is characterized by a symmetrical distribution of pigmentation and pigmented spots on the dorsal sides of the hands and feet. This pigmentation occurs in infancy or early childhood and usually stops progressing before puberty. Histologically, it is characterized by a focal increase or decrease in the amount of melanin or the number of melanocytes in the basal layer of the skin. The data from HGMD showed that the majority of known mutations were related to DSH. For decades, most people have attributed DSH to haploinsufficiency and dominant-negative effects of mutant ADAR1. In addition, AGS, an autosomal recessive autoimmune disease that affects the nervous system (with complications such as intracranial calcification, leukystrophy, and severe developmental delay), is caused by ADAR1 mutation [9]. Kono et al. [50] reported a case of a Japanese patient with AGS and phenotypic characteristics of DSH (neurological symptoms and brain calcification), who was compound heterozygous for ADAR1 mutation, suggesting that homozygous ADAR1 or compound heterozygous ADAR1 mutations may lead to a combination of AGS and DSH in East Asian patients, but only in non-East Asian patients.
Depending upon the expression patterns, two isoforms of ADAR1, an interferon-inducible, cytoplasmic protein with a molecular mass of 150 kDa (p150) and a constitutively expressed nuclear protein with a molecular mass of 110 kDa (p110) are synthesized by translation initiation at alternative methionine-encoding AUG codons in mammalian cells [3, 51]. The p150 isoform that contains a nuclear export signal at the N-terminal of the Zα domain is the full-length protein predominant in the cytoplasm [52]. The p110 isoform consists of three DRBMs and a catalytic domain and is mainly localized in the nucleus owing to the presence of a nuclear localization signal [53]. So far, five mutations of ADAR1, including p.E171* identified in our study, have been reported at the 5′ side of codon 296 [54–56] [9]. However, the distinct roles of the p150 and p110 isoforms are still obscure, and the specific roles of these two isoforms in the pathogenesis of DSH are poorly understood. Recently, Zhang et al. [8] reported that a novel frameshift mutation (p.R91fsX123) could eliminate the expression of the p150 isoform through nonsense-mediated mRNA decay, although it did not change the expression of p110 isoform; this indicates that the ADAR1 p150 isoform might be a determinant of DSH.
The chromatic aberration in DSH is mainly confined to areas exposed to sunlight, which affects the appearance of the skin. Current treatments have not been successful in alleviating this problem. An increasing number of studies have shown that topical application of sunscreen can be successful in controlling hereditary color symmetry disorder [57]. Currently, there is no complete cure for DSH, and treatment is limited to supportive care, presenting many challenges in clinical practice [58]. Therefore, the establishment of a reliable diagnosis of DSH is crucial for the identification of DSH subtypes and development of therapies. Our study may contribute to the timely and improved clinical management of patients with DSH.