Detection of Mutations in LRPAP1, CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2 in 298 Families With Early-Onset High Myopia by Exome Sequencing

Abstract

Purpose.: To evaluate variants in the LRPAP1, CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2 genes in 298 unrelated patients with early-onset high myopia (eoHM).

Methods.: Genomic DNA from 298 patients with eoHM was analyzed by whole exome sequencing. Variants in LRPAP1, CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2 genes were selected and analyzed with bioinformatics. Potential candidate variants were confirmed by Sanger sequencing and then validated in available family members and 192 healthy controls.

Results.: A total of nine variants predicted to affect the functional residues were detected. The LRPAP1 gene showed a homozygous frameshift mutation (c.199delC, p.Q67Sfs*8) in a consanguineous family. The ZNF644 gene showed five heterozygous missense mutations (c.1106A>T, p.K369M; c.1648G>A, p.A550T; c.2014A>G, p.S672G; c.2048G>C, p.R683T, and c.2551G>C, p.D851H) in five families, but the c.1106A>T, (p.K369M) and c.1648G>A, (p.A550T) in ZNF644 did not co-segregated with high myopia in the families and should be excluded as causative mutations. The SLC39A5 gene showed a heterozygous missense variant (c.1238G>C, p.G413A) in a sporadic individual. The SCO2 gene showed two heterozygous missense variants (c.334C>T, p.R112W and c.358C>T, p.R120W) in two families. None of the variants was detected in 192 healthy controls and all were predicted to be damaging by both Polyphen-2 and SIFT, except for the previously reported p.S672G mutation in ZNF644, which was predicted to be damaging by SIFT but benign by Polyphen-2. No homozygous or compound heterozygous variants were found in CTSH and LEPREL1.

Conclusions.: Our results provide additional evidence to support the idea that mutation in LRPAP1 is associated with high myopia. Further studies are expected to evaluate the pathogenicity of the variants in CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2.

Introduction

High myopia, defined as refractive errors greater than −6 diopters (D)¹ or an ocular axial length (AL) longer than 26 mm,² is a leading cause of impaired vision due to its associated complications,³ such as macular degeneration,^4,5 retinal detachment,^6,7 and glaucoma.^8,9 Most high myopia is a complex trait, with both genetic and environmental factors involved.^1,10 Genome-wide association studies have identified a number of significant single nucleotide polymorphisms (SNPs) in several loci associated with high myopia, such as 11q24.1,¹¹ 5p15,¹² 21q22.3,¹³ 15q14,¹⁴ and 4q25,¹⁵ but the responsible genes have not been determined. On the other hand, transmission of high myopia has been reported as autosomal dominant,^16,17 autosomal recessive,^18,19 and X-linked traits^20,21 in a number of families. Linkage analysis has identified at least 13 loci for high myopia (MYP1-3, MYP5-6, MYP11-13, MYP15-17, MYP18-19).^{16–18,21–30} Many candidate genes in these loci have been studied extensively, including TGF-β1,^31–33 IGF,^34–37 LUM,^33,38–40 COL1A1,^25,41–43 and PAX6.^44–48 However, no causative mutation in these genes has been found for high myopia, nor have any SNPs been consistently reported to associate with high myopia. 

High myopia can be classified according to age of onset, as early-onset high myopia (eoHM, high myopia was observed before school age) and late-onset high myopia (loHM, high myopia developed during or after school age).⁴⁹ The loHM is much more common and is frequently associated with extensive near-work.⁵⁰ Compared with loHM, eoHM is more likely to be genetically determined with less environmental impact,⁴⁹ and therefore might represent an ideal source for a search for monogenic defects. 

Recent whole exome sequencing has revealed variations in a few genes reported to associate with high myopia. For example, Shi et al.⁵¹ reported a mutation A672G in ZNF644 that co-segregated with high myopia in a large Chinese family with autosomal dominant high myopia. Other mutations in ZNF644 were also detected in individuals (Chinese, Caucasian, and African American) with sporadic high myopia.^51,52 Co-segregation of two heterozygous mutations (Y47* and M304T) in SLC39A5 with autosomal dominant high myopia was identified in two Chinese families.⁵³ Tran-Viet et al.⁵⁴ reported that a heterozygous nonsense mutation in the SCO2 gene (c.157C>T, p. Gln53*) co-segregated with high myopia in a large European family with autosomal dominant extreme high myopia, and three other rare variants were detected in sporadic high myopia. Mutations in some genes also have been reported for autosomal recessive high myopia. Aldahmesh et al.¹⁹ identified two homozygous frameshift mutations (Asn202Thrfs*8 and Ile288Argfs∗118) in LRPAP1 and one homozygous mutation in CTSH in four Arabic consanguineous families with eoHM. Homozygous mutations (Gln5X and Gly508Val) in LEPREL1 also have been reported in families with autosomal recessive high myopia.^55,56 Variants in these genes were detected in a limited number of cases, and were expected to be extended to additional families. In the present study, we report additional mutations in these genes, based on whole exome sequencing of 298 unrelated Chinese patients with eoHM. 

Materials and Methods

This is a part of our ongoing project on identification of genetic defects for Mendelian high myopia by exome sequencing. The current study focused on exome sequencing results for the LRPAP1, CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2 genes from 298 unrelated patients with eoHM. The inclusion criteria for patients with eoHM included spherical refraction less than or equal to −6.00 D, high myopia observed before school age (<7 years), and no other ocular diseases or systemic diseases. Exome sequencing results of the four genes from patients with eoHM were filtered with following steps: (1) exclude noncoding variants without altering splicing sites predicted by the Berkeley Drosophila Genome Project (available in the public domain at http://www.fruitfly.org/seq_tools/splice.html); (2) exclude the synonymous variants without altering splicing sites in the genes; (3) exclude database SNPs with minor allele frequency (MAF) greater than or equal to 0.01 in the 1000Genome database; (4) exclude the variants with MAF greater than or equal to 0.01 observed in 316 patients with other conditions other than high myopia; (5) exclude missense variants predicted to be benign by both Polyphen-2 (available in the public domain at http://genetics.bwh.harvard.edu/pph/) and SIFT (available in the public domain at http://sift.jcvi.org); (6) heterozygous variants for ZNF644, SLC39A5, and SCO2, whereas homozygous or compound heterozygous variants for LRPAP1, CTSH, and LEPREL1 were finally selected for confirmation, as mutations in the latter three genes were reported to associate with autosomal recessive high myopia. The remaining variants were confirmed by Sanger sequencing and then evaluated in192 healthy control individuals, who met the following criteria: −0.50 D ≤ spherical refraction ≤ +1.00 D, bilateral visual acuity greater than or equal to 1.0, no family history of myopia, and no other ocular diseases or systemic diseases. Written informed consent was obtained from all the participants or their guardians before the collection of their genomic DNA and clinical data. This study confirmed to the tenets of the Declaration of Helsinki and was approved by the institutional review board of Zhongshan Ophthalmic Center, Sun Yat-Sen University. 

Exome sequencing was completed as a commercial service. The DNA sample from each patient was sequenced with an Agilent (Santa Clara, CA, USA) SureSelect Human All Exon Enrichment Kit V4 (51189318 base pairs) array on the Illumina (San Diego, CA, USA) HiSequation 2000 101PE. The average sequencing depth was 125-fold. All sequencing reads were mapped against UCSC hg19 (available in the public domain at http://genome.ucsc.edu/). The SNPs and Indels were detected by SAMTOOLS (available in the public domain at http://samtools.sourceforge.net/), and annotated according to the dbSNP135 and 1000Genome databases. 

Sanger sequencing was used to confirm the candidate variants, using the primers listed in Table 1 to amplify the DNA fragments harboring the variants and a touch-down PCR program as previously described.⁵⁷ The amplicons were sequenced with an ABI BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and electrophoresed on an ABI 3130 Genetic Analyzer (Applied Biosystems). The sequencing results were compared with the consensus DNA sequences obtained from the UCSC genome browser hg19 (available in the public domain at http://genome.ucsc.edu/cgi-bin/hgGateway). The confirmed variants were further validated in the available family members and 192 healthy controls. 

Table 1

 

Primers Used to Amplify the Sequences Harboring the Variants in This Study

Primer Name	Primer Sequence (5′–3′)
Chr1:91404360-F	AGACCCTCATAAGCCTGACG
Chr1:91404360-R	AGCATTTTCTCCAGGCTCCT
Chr1:91404863-91405263-F	TGTGGACGGACATTTCGAGA
Chr1:91404863-91405263-R	TTTTGGTCAACGCTGCTTTT
Chr1:91405805-F	ACCGAGGATTGCTTTAGTGAT
Chr1:91405805-R	TGTGGAGGTGCTTCTTCTCC
Chr4:3533943-F	TGGTTCTGTCCCTACGCTC
Chr4:3533943-R	GGAGGGTCAGGTCGTTTCTG
Chr12:56630745-F	CCATAGGTGTGAGGGGTGG
Chr12:56630745-R	CAATCCCAGGGCTCCAGAC
Chr22:50962483-F	GGGGTCCACAGTGATGAAGA
Chr22:50962483-R	CTGAGGGCTGAGAAGGAGAG
Chr22:50962507-F	AAACCAGGCTCTGCTTCCA
Chr22:50962507-R	TCTGAGGTCCTGGCTTTTGT

 The amplication condition was the same as described by Li et al.⁵⁷

Results

From the exome sequencing data of 298 individuals with high myopia, a total of 232 variants were detected in the six genes (Supplementary Table S1; 45 variants in LRPAP1, 60 variants in CTSH, 71 variants in LEPREL1, 21 variants in ZNF644, 20 variants in SLC39A5, and 15 variants in SCO2). After six filtering steps, nine variants remained. Sanger sequencing confirmed the nine variants (Fig.) and segregation analysis excluded two variants (c.1106A>T, p.K369M and c.1648G>A, p.A550T) in ZNF644. Segregation analysis could not be performed on the other seven variants because DNA samples were available from only one individual of each seven families. The seven potential causative variants included a homozygous frameshift mutation (c.199delC, p.Q67Sfs*8) in LRPAP1, three heterozygous mutations in ZNF644 (c.2014A>G, p.S672G; c.2048G>C, p.R683T; c.2551G>C, p.D851H), one heterozygous mutation (c.1238G>C, p.G413A) in SLC39A5, and two heterozygous mutations (c.334C>T, p.R112W and c.358C>T, p.R120W) in SCO2. Of the seven, p.Q67Sfs*8 in LRPAP1, p.D851H in ZNF644, and p.G413A in SLC39A5 were novel; p.S672G in ZNF644 was reported in one high myopia case; and the remaining three (p.R683 in ZNF644, p.R112W and p.R120W in SCO2) were known in a database (Table 2). Six of the seven were predicted to be damaging by both Polyphen-2 and SIFT, whereas the reported mutation p.S672G in ZNF644 was predicted as damaging by SIFT but benign in Polyphen-2 (Table 2). None of the seven variants was detected in 192 healthy control individuals. No homozygous or compound heterozygous variants were found in CTSH and LEPREL1. 

Figure

 

Nine variants confirmed in this study. From left to right: pedigrees with variants, the sequence traces harboring the variants, and the corresponding sequences in the healthy controls. Each pedigree was named with the ID of the proband, who is indicated by the black arrow. +, wild-type; M, mutation.

Table 2

 

Potential Pathogenic Variants in LRPAP1, CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2

Chr. Position	Gene	Exon	Variant	Status	Patient	1000G	EVS	Polyphen2	SIFT	Note
Chr4:3533943	LRPAP1	1	NM_002337: c.199delC, p.Q67Sfs*8	Hom	HM759	None	None	NA	NA	Novel
Chr1:91405805	ZNF644	3	NM_201269: c.1106A>T, p.K369M	Het	HM832	None	None	D (0.947)	D (0.02)	Novel*
Chr1:91405263	ZNF644	3	NM_201269:c.1648G>A, p.A550T	Het	HM877	None	None	D (0.996)	D (0.37)	Novel*
Chr1:91404897	ZNF644	3	NM_201269: c.2014A>G, p.S672G	Het	HM372	None	None	B (0.282)	D (0.001)	Shi et al., 2011
Chr1:91404863	ZNF644	3	NM_201269: c.2048G>C, p.R683T	Het	HM693	Yes	None	D (0.558)	D (0.034)	rs201546602
Chr1:91404360	ZNF644	3	NM_201269: c.2551G>C, p.D851H	Het	HM878	None	None	D (0.573)	D (0.000)	Novel
Chr12:56630745	SLC39A5	11	NM_173596:c.1238G>C, p.G413A	Het	HM382	None	None	D (1.000)	D (0.000)	Novel
Chr22:50962507	SCO2	2	NM_001169110: c.334C>T, p.R112W	Het	HM640	None	1/13006	D (0.567)	D (0.003)	rs370130010
Chr22:50962483	SCO2	2	NM_001169110: c.358C>T, p.R120W	Het	HM789	None	1/13006	D (0.881)	D (0.040)	rs375954523

 All the six variants were absent in 2010 alleles of controls from the same region (813 individuals with other conditions and 192 healthy controls). Chr, chromosome; 1000G, 1000Genome database; B, benign; D, damaging; Het, heterozygous; Hom, homozygous.

*   Variant was not co-segregated with high myopia.

The homozygous c.199delC (p.Q67Sfs*8) mutation in LRPAP1 was detected in a 5-year-old boy (HM759), who had bilateral high myopia of −8.00 D with an ocular AL of 32.59 mm for the right eye and 32.24 mm for the left eye. No other abnormalities were detected. No family history of high myopia was reported, but his parents had a consanguineous marriage (Fig.). The other five heterozygous mutations in ZNF644 and SCO2 were all identified in isolated cases. Clinical data of the six patients are summarized in Table 3. 

Table 3

 

Clinical Data of the Nine Patients With Mutations

Case ID	Gene	Mutation	Effect	Age, y	Sex	BVA		Refraction, D		AL, mm
						OD	OS	OD	OS	OD	OS
HM759	LRPAP1	c.[197delC];[197delC]	Frameshift	5	M	0.40	0.30	–8.00	–8.00	32.59	32.24
HM832	ZNF644	c.[1106A>T];[=]	Missense	40	F	0.70	0.10	–14.50	–16.50	NA	NA
HM877	ZNF644	c.[1648G>A];[=]	Missense	4	M	0.50	0.60	–11.25	–10.75	26.70	26.42
HM372	ZNF644	c.[2014A>G];[=]	Missense	27	F	1.00	1.00	–8.50	–9.00	NA	NA
HM693	ZNF644	c.[2048G>C];[=]	Missense	4	M	0.40	0.60	–8.00	–8.25	26.97	26.74
HM878	ZNF644	c.[2551G>C];[=]	Missense	12	M	0.80	0.90	–7.00	–7.00	NA	NA
HM382	SLC39A5	c.[1238G>C];[=]	Missense	54	F	NLP	0.02	NA	–16.75	NA	NA
HM640	SCO2	c.[334C>T];[=]	Missense	32	F	0.50	0.60	–11.00	–9.25	NA	NA
HM789	SCO2	c.[358C>T];[=]	Missense	6	F	0.20	0.20	–10.25	–11.25	25.40	25.56

 All patients had a tigroid fundus appearance of both eyes. BVA, best visual acuity; F, female; M, male; NLP, no light perception.

Discussion

This analysis of exome sequencing data from 298 patients with eoHM identified nine potential mutations in the LRPAP1, ZNF644, SLC39A5, and SCO2 genes in nine unrelated patients. These mutations are predicted to affect the function of coding residues and were absent in 2010 alleles of controls from the same region (813 individuals with other conditions and 192 healthy control individuals). 

Aldahmesh et al.¹⁹ previously identified two homozygous frameshift mutations (Asn202Thrfs*8 and Ile288Argfs∗118) in LRPAP1 in three Arabic consanguineous families with eoHM. In the current study, a novel homozygous frameshift mutation (Q67Sfs8*) in LRPAP1 was detected in a consanguineous Chinese family with eoHM. This novel mutation was located in the first exon of LRPAP1 and it would result in complete loss of the encoded protein if translated, but was more likely to undergo nonsense-mediated decay. This is the third homozygous truncation mutation in LRPAP1 associated with high myopia. Searching the Exome Variant Server (EVS) database (available in the public domain at http://evs.gs.washington.edu/EVS/), revealed two other rare heterozygous truncation variations (c.529delC (p.L177Cfs*17) in 1 of 12,519 alleles and c.202C>T (p.R68*) in 2 of 11,716 alleles). These findings all support homozygous truncation mutations in LRPAP1 as a likely cause of high myopia. 

As reported, LRPAP1 acted as a chaperone of LRP1 and inhibited ligand binding to LRP1, thereby inhibiting the degradation of LRP1.^58,59 Mice with homozygous deficiency in LRPAP1 had a reduced expression of LRP in the liver and brain,⁶⁰ which in turn activated TGF-β.⁶¹ Aldahmesh et al.¹⁹ also reported that the homozygous mutated LRPAP1 in the patients with high myopia resulted in downregulation of LRP but upregulation of TGF-β. Transforming growth factor-β has been proposed to regulate sclera metabolism in the development of myopia, through modulation of the extracellular matrix in the sclera.⁶² Thus, LRPAP1 may be associated with myopia through regulation of the expression of TGF-β. Further studies are expected to reveal the molecular pathway of LRPAP1 related to high myopia. 

The ZNF644 is a zinc finger transcription factor that is expressed in the retina and RPE,⁵¹ but its biological function has not been identified by any studies. The ZNF644 was proposed to have a role in the development of high myopia because mutations in ZNF644 were detected in the patients with high myopia.⁵¹ At present, eight heterozygous mutations in ZNF644 have been reported to associate with high myopia in Chinese and Caucasian subjects, including six missense variants (p.T242M, p.E274V, p.I587V, p.S672G, p.R680G, and p.C699Y) and two untranslated region variants (g.4138C>G, 3′UTR+12C>G; g.4718G>A, 3′UTR+592G>A).^51,52 Six of these missense variants are located in exon 3. Of the three heterozygous variants (p.S672G, p.R683T, and p.D851H) detected in the current study, p.S672G has been reported in a previous study⁵¹ and p.R683T is present in the 1000Genome data (frequency unknown) but not in EVS (0 of 12,998 alleles). The p.D851H is novel. These three missense mutations also occur in exon 3. 

The SLC39A5 is expressed in the sclera and retina⁵³ and interacts with the BMP/TGF-β pathway,⁵³ which has been proposed to participate in the development of high myopia by modulating extracellular matrix in the sclera. Guo et al.⁵³ reported a heterozygous nonsense mutation (Y47*) co-segregated with autosomal dominant high myopia in a Chinese family, and a heterozygous missense mutation (M304T) in a sporadic case. Our study detected another heterozygous missense variant (c.1238G>C, p.G413A) in a sporadic individual with high myopia. The possibility of a contribution of SLC39A5 to high myopia is expected to be confirmed in further studies. 

Previous exome sequencing and Sanger sequencing has identified four heterozygous mutations (p.Q53*, p.R114H, p.E140K, and p.A259V) in SCO2 in four families with high myopia.⁵⁴ These four variants are also present in the 1000Genome or EVS databases with a frequency of 0.002, 2 of 13,006, 3 of 13,006, and 114 of 13,006, respectively. The p.R120W and p.R112W mutations detected in the current study also are present in the EVS with a frequency of 1 of 13,006 alleles; our findings are comparable with the previous report. The SCO2 encodes a copper chaperone and is essential for the formation of cytochrome c oxidase (COX),⁶³ which is the terminal enzyme in the respiratory electron transport chain in the mitochondria and plays a critical role in producing aerobic ATP.⁶⁴ An overlapped mutation E140K in SCO2 mutations has been reported to cause cardioencephalomyopathy with COX deficiency,^65,66 as well as high myopia.⁵⁴ Further studies are expected to identify the specific mechanism of myopia development associated with the COX deficiency. 

In summary, nine mutations were detected in nine families with eoHM by exome sequencing, including a homozygous truncation mutation in LRPAP1 in one family, five heterozygous mutations in ZNF644 in five families, one heterozygous mutation in SLC39A5 in one family, and two heterozygous mutations in SCO2 in two families. Our results provide additional evidence to support the possibility that mutation in LRPAP1 is associated with high myopia. Further studies are expected to evaluate the pathogenicity of the variants in CTSH, LEPREL1, ZNF644, SLC39A5, and SCO2. 

Acknowledgments

The authors thank all of the patients and controls for their participation in this study. 

Supported by the National Natural Science Foundation of China (U1201221), Natural Science Foundation of Guangdong (S2013030012978), Guangdong Department of Science and Technology Translational Medicine Center (2011A080300002), and the Fundamental Research Funds of State Key Laboratory of Ophthalmology. 

Disclosure: D. Jiang, None; J. Li, None; X. Xiao, None; S. Li, None; X. Jia, None; W. Sun, None; X. Guo, None; Q. Zhang, None