New ZNF644 mutations identified in patients with high myopia

• Introduction
• Methods
• Results
• Discussion

Abstract

Purpose: Myopia, or near-sightedness, is one of the most common human visual impairments worldwide, and high myopia is one of the leading causes of blindness. In this study, we investigated the mutation spectrum of ZNF644, a causative gene for autosomal dominant high myopia, in a high-myopia cohort from a Chinese population.

Methods: DNA was isolated with the standard proteinase K digestion and phenol-chloroform method from a case cohort of 186 subjects diagnosed with high myopia (spherical refractive error equal or less than −6.00 diopters). Sanger sequencing was performed to find potential mutations in all coding exons, flanking splicing sites, and untranslated regions (UTRs) of ZNF644 (NM_201269). Identified novel variants were further screened in 526 ethnically matched normal controls. Functional prediction and conservation analysis were performed using ANNOVAR.

Results: Five novel variants were identified. Three are missense (c.1201A>G:p.T401A, c.2867C>G:p.T956S, c.3833A>G:p.E1278G), one is synonymous (c.2565A>G:p.T855T), and one (c.-219C>A) is located in the 5′ UTR. Functional prediction indicates that c.3833A>G:p.E1278G was predicted to be damaging by SIFT and Polyphen2. Conservation analysis using PhyloP and GERP++ indicate all of the missense variants are highly conserved. None of these novel mutations was identified in 526 normal controls.

Conclusions: ZNF644 is associated with high myopia in a cohort from a Chinese population. ZNF644 mutations have a minor contribution to the genetic etiology of high myopia.

Myopia, characterized by refraction error, is the most common ocular disorder in the world [1]. The prevalence varies across countries. Multiple studies have reported an approximate prevalence rate of 17% in Australia, 26% in the United States, and 27% in Western Europe [2]. The prevalence in Asian countries, such as China, Singapore, and Japan, is even higher, estimated at about 71% to 96% [3-5]. With high prevalence, myopia causes a serious social burden, and the economic impact is substantial [6,7].

Myopia is usually divided into two groups classified by the degree of refraction error. One is common myopia with low or moderate refractive error, and the other is high myopia [8]. Patients with high myopia have a spherical equivalent refractive error more than or equal to −6.0 diopter sphere (DS) and an axial length longer than or equal to 26.0 mm. High myopia may also present retinal pathological changes and ocular comorbidities, such as macular choroidal degeneration, retinal detachment, premature cataract, and glaucoma; therefore, high myopia is also called pathological or degenerative myopia.

Epidemiology studies have shown that genetic and environmental factors contribute the development of myopia [9]. Twin and family studies have demonstrated that myopia, especially high myopia, has a high heritability [10,11]. A dozen linkage regions and several genome-wide significant associated loci have been identified in families with high myopia and case-control cohorts [12-19]. Familial high myopia is usually inherited as a monogenic disorder, and three inheritances have been found, including autosomal dominant (AD), autosomal recessive (AR), and X-linked inheritance. AD inheritance is the most common. Recently, using whole exome sequencing, Shi et al. identified a causative gene in a Chinese family with AD high myopia, and replicated their results in a sporadic cohort [20]. Subsequently, Tran-Viet et al. performed mutation screening in an American cohort for ZNF644 (gene ID:64146, OMIM number: 614159), and identified a novel missense mutation, which supports that ZNF644 may be a causative gene for high myopia [21]. ZNF644 is a zinc finger protein that functions as a transcriptional factor. In this study, we attempted to replicate the results and enlarge the mutation spectrum of ZNF644 in a separate Chinese high myopia cohort.

All subjects with high myopia are from a Chinese population and received full ophthalmologic examinations before being included. The average refractive error of the 186 patients with high myopia was −13.10 DS for the right eye (OD) and −12.39DS for the left eye (OS), and ranged from −6.00 DS to −30.00 DS (OD) and −6.25 DS to −30.00 DS (OS). The average axial length was 28.18 mm for the right eye (OD) and 28.16 mm for the left eye (OS), and ranged from 26.17 mm to 44.38 mm (OD) and 26.2 mm to 44 mm (OS). Further detailed clinical information is summarized in Table 2.

Thirteen variants were identified (Table 3). Five are novel variants that are not reported in dbSNP137, 1000 Genomes, and NHLBI ESP6500 exome sequencing data (Table 3, Figure 1). All of these novel variants were also evaluated in 526 population-matched normal controls, and none were identified in the control individuals. Functional prediction using SIFT and Polyphen2 indicated that p.T401A and p.T956S were either tolerant or benign, whereas p.E1278G was predicted to be damaging by SIFT and Polyphen2 (Table 4). Although p.T401A and p.T956S were not predicted to be damaging, conservation analysis using PhyloP and GERP++ indicated that all are highly conserved (Table 4). The missense and synonymous mutations were identified in only one patient; however, the mutation located in the 5′ UTR was identified in three patients. The phenotypes of the cases with the novel variants are serious. All had refractive error more than −10 DS, except one case (M21787) whose refractive errors were −7.5 DS for right eye and −9 DS for left eye. Detailed clinical information for the patients with these novel variants is described in Table 5.

To test whether the identified common SNPs with minor allele frequency (MAF) larger than 1% (rs17131232, rs76101054, rs1188952, rs193167060, rs41286763, rs149597385, Table 3) are associated with high myopia, we performed association analysis using 197 Chinese subjects (CHB, CHS) who had no phenotype record from the 1000 Genome project as the controls. Unfortunately, we failed to find an association of these polymorphisms between the patients with high myopia and the controls (rs17131232: p=0.81; rs76101054: p=0.43; rs1188952: p=0.06; rs193167060: p=0.37; rs41286763: p=1.00; rs149597385: p=0.75).

We performed a mutation screening of ZNF644 in a separate high-myopia cohort from a Chinese population, and identified five novel variants that had not been reported in dbSNP137, 1000 Genomes, and NHLBI ESP exome in seven patients. All of the variants were also absent in 526 population-matched normal controls. These novel variants were all identified in sporadic cases.

Until now, three studies, including this study, have investigated the mutations of ZNF644 in patients with high myopia [20,21]. A total of 12 novel variants have been identified (Table 5). Three were identified in the 3′ UTR, nine were identified in coding regions (one synonymous and eight missense), and one was identified in the 5′ UTR. Functional prediction for the missense variants demonstrated that p.E1278G, p.R680G, p.S672G, p.E274V, and p.C699Y were predicted to be damaging by either SIFT or PolyPhen2 or both; however, p.T956S, p.I587L, and p.T401A were predicted to be either benign or tolerant. Conservation analysis revealed that p.E1278G, p.T956S, p.R680G, p.E274V, p.C699Y, and p.T401A are conserved, but p.S672G and p.I587L failed to survive the conservation threshold (Table 4). Variants p.I587, p.C699Y, c.+592G>A, and c.-219C>A were recurrently identified in more than one case. All variants were identified in sporadic cases except p.S672G, which was identified and cosegregated with the phenotype in a large family with high myopia [20].

Genetic studies have revealed that high myopia has an extremely high genetic heterogeneity. For example, 19 linkage peaks have been identified up to now, and most cannot be replicated in independent study; a genome-wide association study also revealed dozens of risk variants or susceptible genes [22]. Mutations of genes identified in Mendelian inheritance families and sporadic cases, such as ZNF644, SCO2 (gene ID:9997, OMIM number: 604272), LRPAP1 (gene ID:4043, OMIM number: 104225), and LEPREL1 (gene ID:55214, OMIM number: 610341), explain only a small proportion of the subjects with high myopia [23-25]. Therefore, ZNF644 mutations identified in sporadic patients with high myopia must be evaluated in a larger cohort of patients with well-characterized high myopia and normal controls to determine whether these variants are associated with the clinical outcome or not.

ZNF644 encodes zinc finger transcription factor, which is ubiquitously expressed in several tissues such as the eye, liver, and placenta. The biologic function and the mechanism of this gene in high myopia pathogenesis are still unclear, although this gene was revealed to be associated with high myopia three years ago. Further studies should be conducted to investigate the functional consequence of these mutations, or at least the mutation cosegregating with high myopia in the large family (p.S672G). An animal model study should also be conducted to analyze the phenocopy and molecular mechanism of ZNF644 in the development of myopia.