Mutations in the ABCA3 Gene Are Associated With Cataract-Microcornea Syndrome

Abstract

Purpose.: Cataract-microcornea syndrome (CCMC) is an autosomal dominant inherited disease characterized by the association of congenital cataract and microcornea without any other systemic anomaly or dysmorphism. Although mutations of several genes have been shown to cause dominant CCMC, in many patients the causative gene has not yet been identified. Our aim was to identify the disease-associated gene in Chinese patients with CCMC.

Methods.: The CCMC patients from two unrelated Chinese families and 26 sporadic patients were enrolled. All the patients were screened by Sanger sequencing with no identified mutations. Genetic variations were screened by whole-exome sequencing and then validated using Sanger sequencing.

Results.: By sequencing the whole exome of three patients in a Chinese four-generation dominant CCMC family (Family A), three heterozygous missense mutation (c.115C>G, c.277G>A, and c.4393G>A) were identified in ATP-binding cassette protein A3 (ABCA3). At highly conserved positions, changes (c.115C>G and c.4393G>A) were predicted to have functional impacts and completely cosegregated with the phenotype. We further confirmed our finding by identifying another heterozygous missense mutation, c.2408C>T, in ABCA3 in an additional dominant CCMC family (Family B), which also cosegregated with the phenotype. Moreover, four heterozygous mutations, two missense mutations (c.4253A>T, c.2069A>T) and two splice site mutations (c.4053+2T>C, c.2765-1G>T) were identified from the sporadic patients. The ABCA3 protein was expressed in human lens capsule, choroid-retinal pigment epithelium and retinal pigment epithelial cells.

Conclusions.: Mutations in the human ABCA3 gene were associated with lethal respiratory distress. Our study showed, for the first time to our knowledge, that mutations in ABCA3 were associated with CCMC, warranting further investigations on the pathogenesis of this disorder.

Introduction

Congenital cataract is a leading cause of childhood blindness, accounting for approximately 10% to 38%,¹ with a prevalence of approximately 0.006% to 0.06% in live births.^2,3 It may occur alone or in association with other ocular or systemic abnormalities. Microcornea, one of the most frequent abnormalities associated with congenital cataract, results from secondary damage of the lens maldevelopment or from mutations in some growth or transcription factors.⁴ 

The combination of congenital cataract and microcornea, cataract-microcornea syndrome (CCMC; OMIM 116200), appears as a distinct phenotype affecting 12% to 18% of heritable congenital cataract patients.⁵ Genetically, CCMC is a heterogeneous condition. To date, approximately 200 genes and loci have been associated with cataracts.^4,6 Of these genes, mutations in at least nine genes were reported to be responsible for congenital cataract associated with microcornea, including genes encoding crystallins (crystallin alpha-A [CRYAA], OMIM 123580; crystallin beta-A4 [CRYBA4], OMIM 123631; crystallin beta-B1 [CRYBB1], OMIM 600929; crystallin beta-B2 [CRYBB2], OMIM 123620; crystalline gamma-C [CRYGC], OMIM 123680; and crystallin gamma-D [CRYGD], OMIM 123690),^5,7–14 gap junction protein alpha 8 (GJA8, OMIM 600897),^5,15 v-maf avian musculoaponeurotic fibrosarcoma oncogene homolog (MAF, OMIM 177075),^16,17 and solute carrier family 16 member 12 (SLC16A12, OMIM 611910).¹⁸ Only the transcription factor, MAF, has exclusively been associated with CCMC, whereas mutations in the other genes were reported in congenital cataract without additional malformations.⁵ 

Whether the CCMC phenotype is related to specific cataract gene alleles or closely linked modifiers remains to be clarified. Fortunately, the technique of exome sequencing has come to the aid, enabling the identification of disease-associated mutations by sequencing the whole exome of a small number of affected individuals.^19–21 In the present study, disease-associated mutations were identified in a Chinese family with dominant CCMC using the exome sequencing technique. 

Methods

Ethics Statement

The study was performed in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Shandong Eye Institute (Qingdao, China). Written informed consent was obtained from all participants (or guardians). 

Study Population

Two Han Chinese families (designated as Family A and Family B) with dominant CCMC, 26 sporadic patients with CCMC and their parents, and 200 matched, normal controls (27.20 ± 7.13 years old, 117 males) were included. The 26 sporadic patients and their parents, and the 200 controls of Han Chinese ethnicity were nonrelated Qingdao locals recruited at the Qingdao eye hospital of Shandong Eye Institute (Qingdao, China). All participants underwent an extensive, standardized examination by ophthalmologists. The diagnosis was confirmed with ophthalmic examinations, including visual acuity, slit-lamp microscopy, tonometry, keratometry, specular microscopy, ultrasonic A/B scan, and a history of cataract extraction. Ocular photographs were taken by slit-lamp photography. There was no family history of other systemic abnormalities in Family A, Family B, and the 26 sporadic patients. 

Mutation Screening in Reported Causative Genes of CCMC

All exons and exon/intron border regions were directionally sequenced and aligned to the GenBank reference sequences. The following genes were included in the DNA sequencing analysis: CRYAA, NM_000394; CRYBA1, NM_005208; CRYBB1, NM_001887; CRYGC, NM_020989; CRYGD, NM_006891; GJA8, NM_005267; MAF, NM_005360. Primers used to amplify the coding exons and adjacent intronic regions of the seven genes were determined according to previous reports.^16,22 

Individual exon was amplified by PCR and analyzed on an ABI 3730XL Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Sequencing data were compared pair-wisely with the Human Genome database. 

Targeted Capture and Exome Sequencing

The exome sequencing approach was used to identify the disease-causing genetic variant for the dominant CCMC Family A. Exome sequencing was performed on three patients (II2, III1, and III4) at the BGI, Inc., Shenzhen, China. Venous blood (5 mL) was collected from the participants, and total human genomic DNA was isolated with the DNA isolation kit for mammalian blood (Tiangen, Beijing, China). Venous blood and genomic DNA samples were stored at −80°C before use. NimbleGen (44 Mb) target enrichment system (Roche NimbleGen, Inc., Madison, WI, USA) was used to collect the protein coding regions of human genome DNA. 

The array was able to capture 18,283 (99.6%) of the 18,357 genes. The gene sequences for this array were available in the Consensus Coding Sequence Region (CCDS) database (available in the public domain at http://www.ncbi.nlm.nih.gov/projects/CCDS/). The exon-enriched DNA libraries then were subjected to a second library construction in preparation for Illumina GA sequencing and were sequenced using the Illumina Genome Analyzer II platform, following the manufacturer's instructions (Illumina, San Diego, CA, USA). 

Variant Analysis

The sequencing reads were aligned to the human reference genome (NCBI Build 36.3). Alignment of the sequences from the three affected individuals was performed using SOAPaligner after the duplicated reads were removed,²³ and single nucleotide polymorphisms (SNPs) were called using SOAPsnp set with the default parameters.²⁴ Indels affecting coding sequence or splicing sites were identified as described previously.²⁵ 

Data were provided as lists of sequence variants (SNPs and short indels) relative to the reference genome. Identified variants were filtered against the Single Nucleotide Polymorphism database (dbSNP 129, available in the public domain at http://www.ncbi.nlm.nih.gov/projects/SNP/snp_summary.cgi/), 1000 genome project (February 28, 2011 releases for SNPs, and February 16, 2011 releases for indels, available in the public domain at http://www. 1000genome.org/), HapMap 8 (available in the public domain at http://hapmap.ncbi.nlm.nih.gov/) database, and YH database.²⁶ 

Verification of Variants

Sanger sequencing was used to determine whether any of the remaining variants cosegregated with the disease phenotype in Family A. Primers flanking the candidate loci were designed based on genomic sequences of Human Genome (hg18/build36.3) and synthesized at the BGI, Inc. All shared variants of the three affected individuals after filtering then were confirmed by PCR and analyzed on an ABI 3730XL Genetic Analyzer. Sequencing data were compared pair-wisely with the Human Genome database. 

Afterwards, we sequenced all the exons and flanking introns of the ABCA3 gene (NM_001089) in patients of Family B and 26 sporadic patients to detect other mutation sites using the Sanger sequencing method. As an additional step, the detected variants were sequenced in 200 normal control subjects. 

Human Ocular Tissues and Cell Culture

Expression of the ABCA3 gene in the human cornea, sclera, conjunctiva, iris-ciliary body (ICB), retina, choroid-RPE, lens capsule, and human retinal pigment epithelial cell line (ARPE-19, catalog No.CRL-2302; ATCC, Manassas, VA, USA) were evaluated. 

Human donor ocular tissues were provided by the Eye Bank of Shandong Eye Institute. The whole globes, enucleated within 10 hours of death from human donors, were immediately dissected, and the cornea, sclera, conjunctiva, ICB, retina, choroid-RPE, lens capsule, and RPE were collected. 

The ARPE-19 cells were cultured in 1:1 of Dulbecco's modified Eagle medium/nutrient mixture F12 (DMEM/F12; Invitrogen, Carlsbad, CA, USA), containing 10% fetal bovine serum (FBS; Invitrogen). The cells were incubated at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. The culture medium was changed every 3 days. 

Reverse-Transcription PCR

Total RNA was prepared from venous blood (0.2 mL) of the family members, using the RNA isolation kit for mammalian blood (Tiangen). Total RNA was prepared from each human ocular tissue, using the NucleospinRNA kit (BD Biosciences, Palo Alto, CA, USA), and reversely transcribed into first-strand cDNA, using Primescript First-Strand cDNA Synthesis kit (TaKaRa, Dalian, China). Gene-specific cDNA fragments were amplified with DNA polymerase (Tiangen). The expression of genes was normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primer sequences used for ABCA3 were forward-5′-GAGGAGAGCCTCACTTCTGG-3′, reverse-5′-CGTACATGACCAGCATCTCC-3′ and for GAPDH were forward-5′- ACCACAGTCCATGCCATCAC-3′, reverse-5′-TCCACCACCCTGTTGCTGTA-3′. The PCR amplification products were analyzed by agarose gel electrophoresis. 

Western Blotting and Antibodies

Total protein was prepared from each tissue using radioimmunoprecipitation assay (RIPA) buffer (50 mmol/L Tris, 150 mmol/L NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, sodium orthovanadate, and sodium fluoride, pH 7.4; Galen, Beijing, China) and quantified. Protein (50 μg in 15 μL loading buffer) was resolved in 10% SDS-PAGE gel before transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). The blots were blocked in 5% nonfat dry milk dissolved in Tris-buffered saline Tween-20 (TBST; 20 mmol/L Tris, pH 7.5, 0.5 mmol/L NaCl, 0.05% Tween-20) for 1 hour and then incubated with the primary antibody in TBST for 1 hour, followed by incubation with horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences, Uppsala, Sweden) for 1 hour. All incubations were conducted at 25°C, and three washes with 10 mL TBST were performed between each step. The membranes then were developed with SuperSignal West Femto Maximum Sensitivity substrate (Pierce Biotechnology, Rockford, IL, USA) and exposed to X-ray film (Kodak, Rochester, NY, USA). The immunoreactive bands were quantified using National Institutes of Health (NIH) Image 1.62 software (NIH, Bethesda, MD, USA). All the experiments reported in this study were performed three times, and the results were reproducible. For each sample, the levels of proteins of interest were normalized to that of GAPDH. Primary antibodies included anti-ABCA3 antibody (Abcam, Cambridge, MA, USA) and anti-GAPDH antibody (Kangchen, Shanghai, China). 

Immunocytochemical Staining

Cells were plated in a glass culture dish (Biousing Biotech Co, Wuxi, China) at a density of approximately 4.0 × 10⁴ cells/35-mm dish in complete medium and grown to subconfluence. After the medium was removed, the cells were fixed in 4% paraformaldehyde in PBS for 15 minutes, followed by three PBS washes. The fixed cells were incubated with 0.2% Triton X-100 in PBS for 5 minutes, blocked in 5% BSA, and incubated with anti-ABCA3 antibody (Abcam) for 30 minutes, before incubation with the fluorescence-conjugated secondary antibody for 1 hour. Images were obtained using an Eclipse TE2000-U confocal laser scanning microscope (Nikon, Tokyo, Japan). 

Results

Clinical Assessment and Findings

Herein we described two Han Chinese families from Qingdao that had monogenic CCMC with a dominant inheritance model (Fig. 1A). By ophthalmic examinations, three of 24 members in Family A, a four-generation family, were identified to be affected with CCMC (Figs. 1A, 2A–D). Another patient in this family was deceased. Five affected individuals (Figs. 1B, 2E–K) were found among the 48 examined family members in the five-generation Family B, which also had three deceased patients. 

Figure 1

 

Pedigrees of the two Chinese families with dominant cataract-microcornea syndrome. (A) Pedigree of Family A. (B) Pedigree of Family B. Affected men and women are indicated by filled squares and circles, respectively. Normal individuals are shown as empty symbols. Deceased individuals are indicated with slashes (/).

Figure 2

 

Slit-lamp photographs of the affected individuals in Family A (A–D) and Family B (E–K). The phenotypes are described and summarized in Table 1. OD, right eye; OS, left eye.

Table 1

 

The Clinical Features of Patients With Cataract-Microcornea Syndrome in the Two Chinese Families

Patient	Age, y/Sex	Eye	Visual Acuity	Best Corrected Visual Acuity	Lens	Nystagmus	IOP, mm Hg	Axial Length, mm
Family A:II2	59/F	OD	FC	FC	Aphakia	Yes	23	25.38
		OS	FC	FC	Aphakia	Yes	22	26.13
Family A:III1	35/F	OD	FC	FC	Cataract	Yes	16	22.95
		OS	HM	HM	Cataract	Yes	18	23.26
Family A:III4	22/F	OD	FC	FC	Aphakia	Yes	20	23.35
		OS	20/200	20/200	Aphakia	Yes	19	23.5
Family B:III2	68/F	OD	10/200	10/200	Aphakia, posterior subcapsular cataract	Yes	18	32.44
		OS	HM	HM	Cataract	Yes	22	32.37
Family B:III3	65/M	OD	FC	FC	Aphakia, posterior subcapsular cataract	Yes	16	26.97
		OS	10/200	10/200	Aphakia, posterior subcapsular cataract	Yes	18	26.98
Family B:III7	49/M	OD	0.02	0.05	Aphakia	Yes	16	22.22
		OS	0.05	0.05	Aphakia	Yes	17	21.45
Family B:IV2	48/F	OD	0.02	0.02	Cataract	Yes	14	28.44
		OS	0.05	0.2	IOL, posterior subcapsular cataract	Yes	15	25.38
Family B:IV5	44/M	OD	0.3	0.5	Aphakia	Yes	14	30.78
		OS	0.05	0.05	IOL	Yes	14	30.53

 FC, finger counting; F, female; IOL, intraocular lens; M, male; NA, not available; OD, right eye; OS, left eye.

Table 1

 

Extended

Corneal Curvature, D	Corneal Diameter, mm	Endothelial Cells Density, mm	B-Scanning	Surgery and Trauma History
Min:53.1	Transverse:8.0 Vertical:8.5	2197	Aphakia, Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification
MAX:53.25
Min:52.73	Transverse:8.0 Vertical:8.0	2341	Aphakia, Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification
MAX:53.12
Min:45.6	Transverse:8.0 Vertical:8.2	2419	NA	No
MAX:47.5
Min:47.8	Transverse:8.0 Vertical:8.0	2395	NA	No
MAX:49.3
Min:47.5	Transverse:9.5 Vertical:9.0	2572	Aphakia	Phacoemusification
MAX:51.9
Min:45.9	Transverse:9.0 Vertical:9.5	2216	Aphakia	Phacoemusification
MAX:50.75
Min:47.03	Transverse:9.8 Vertical:9.5	2637	Aphakia, Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification
MAX:48.4
Min:50.5	Transverse:9.6 Vertical:9.5	2919	Vitreous bodies opaque, posterior scleral staphyloma,cataract	No
MAX:52.6
Min:40.90	Transverse:9.0 Vertical:9.0	2176	Vitreous bodies opaque, posterior scleral staphyloma	No
MAX:42.48
Min:41.10	Transverse:9.0 Vertical:9.5	2305	Vitreous bodies opaque, posterior scleral staphyloma	No
MAX:42.04
Min:44.12	Transverse:8.0 Vertical:8.0	2693	Aphakia, Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification
MAX:47.82
Min:45.91	Transverse:8.0 Vertical:8.5	2512	Aphakia, Vitreous bodies opaque	Phacoemusification
MAX:47.15
Min:45.25	Transverse:9.5 Vertical:9.8	2692	Vitreous bodies opaque, posterior scleral staphyloma	No
MAX:45.38
Min:43.96	Transverse:9.6 Vertical:9.8	2247	Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification+IOL
MAX:46.27
Min:45.10	Transverse:9.8 Vertical:9.5	2350	Aphakia, Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification
MAX:45.60
Min:45.10	Transverse:9.5 Vertical:9.5	2587	Vitreous bodies opaque, posterior scleral staphyloma	Phacoemusification+IOL
MAX:46.00

The two families in our study had some common features, such as congenital cataract and nystagmus. The corneal diameters ranged from 8.0 to 9.8 mm (8.89 ± 0.76 mm in the transverse meridian and 8.99 ± 0.68 mm in the vertical meridian). The axial length ranged from 21.45 to 32.44 mm (26.38 ± 3.62 mm). The B-scan ultrasonagraphy showed posterior scleral staphyloma in patient II:2 of Family A (Figs. 3A, 3B) and patients III:2, III:7, IV:2, and IV:5 of Family B (Figs. 3E–L). There was no family history of other systemic abnormalities. Detailed clinical characteristics of the patients from the two families are summarized in Table 1. 

Figure 3

 

B-scan ultrasonagraphic photographs of the affected individuals in Family A (A–D) and Family B (E–L). The phenotypes are described and summarized in Table 1.

Mutation Screening in Reported Causative Genes of CCMC

Direct sequencing of the CRYAA, CRYBA1, CRYBB1, CRYGC, CRYGD, GJA8, and MAF exons showed no pathogenic mutations in any of the affected individuals in Family A and Family B. 

Exome Sequencing Identified ABCA3 as the Associated Gene

To identify the causative gene of CCMC, exome sequencing was performed on DNA samples obtained from three affected members of Family A (II2, III1, and III4). At the depth of coverage, approximately 94.4% of each targeted exome was sufficiently covered to pass our thresholds for variant calling. 

After identification of variants, we focused only on nonsynonymous (NS) variants, splicing site (SS) mutations, and short, frame-shift coding insertions or deletions (indel) that were more likely to be pathogenic mutations than other variants. The total NS/SS/indel variants in patients II2, III1, and III4 of Family A are shown in Table 2. 

Table 2

 

Genetic Variants Identified in the Three Patients (II2, III1, and III4) of Family A Through Exome Resequencing

Filter	Genetic Variants
	Nonsynonymous SNP	Splicing Site	Indel	Total
II2	12881	2612	7095	22588
III1	12967	2638	7088	22693
III4	12538	2574	6992	22104
II2+III1+III4+Not in 1000 Genomes Project, the dbSNP129, HapMap 8, YH database	116	33	105	254

Because CCMC is a rare disorder, but has a clear phenotype, there was a very low likelihood of the causal mutation in these patients being shared with a wider healthy population. We compared the shared variants in these patients with the dbSNP 129, 1000 genome project, HapMap 8 database, and YH database. This left a total of 254 variants that were shared among these three patients. Of these, 116 nonsynonymous SNPs, 33 SSs, and 105 indels were predicted to potentially have a functional impact on the gene (Table 2). 

Segregation then was analyzed by Sanger sequencing on the 116 nonsynonymous SNPs, 33 SSs, and 105 indels, with the 24 members of the Family A. Three variants in ABCA3 gene cosegregated with the disease phenotype in Family A: c.115C>G resulting in an L39V amino acid change, c.277G>A resulting in an V93I amino acid change, and c.4393G>A resulting in an D1465N amino acid change (Table 3, Fig. 4). 

Figure 4

 

Sequence chromatograms of all ABCA3 mutations identified in the study. The left chromatogram represents the sequences of the affected individuals. The right chromatogram represents the sequence of a normal family member. The arrow indicates the location of the mutations.

Table 3

 

Genetic Variants Identified in ABCA3 in the Two Chinese Families and the 5 Sporadic Patients With Cataract-Microcornea Syndrome

Patient	Chromosome/ Position/Gene Name	dbSNP rs# Cluster ID	Mutation Type	Codons	Substitution	Prediction From SIFT	Prediction From PolyPhen–2	Score of Prediction From PolyPhen–2
II2, III1, III4 (family A)	chr16/2329098/ABCA3	rs201955122	Missense	GAT4393AAT	D1465N	DAMAGING	PROBABLY DAMAGING	0.994
II2, III1, III4 (family A)	chr16/2376215/ABCA3	rs200090198	Missense	CTC115GTC	L39V	TOLERATED	POSSIBLY DAMAGING	0.857
II2, III1, III4 (family A)	chr16/2376053/ABCA3	rs199840288	Missense	GTC277ATC	V93I	TOLERATED	BENIGN	0.006
III2, III3, III7, IV2, IV5 (family B)	chr16/2345597/ABCA3	Novel	Missense	ACG2408ATG	T803M	DAMAGING	POSSIBLY DAMAGING	0.801
Sporadic 4	chr16/2331134/ABCA3	Novel	Missense	AAT4253ATT	N1418I	DAMAGING	PROBABLY DAMAGING	0.996
Sporadic 13	chr16/2347524/ABCA3	Novel	Missense	GAG2069GTG	E690V	DAMAGING	PROBABLY DAMAGING	1.0
Sporadic13	chr16/2347541/ABCA3	Novel	Splice site	–	–	–	–	–
Sporadic12, 15, 17	chr16/2333185/ABCA3	Novel	Splice site	–	–	–	–	–

Mutations in ABCA3 Gene

To confirm ABCA3 as the associated gene of CCMC, we used Sanger sequencing to screen all members of Family B. All five clinically affected subjects, but none of those who were unaffected in Family B, carried the heterozygous c.2408C>T mutation resulting in a T803M amino acid change (Table 3; Fig. 4). Thus, the heterozygous missense mutation T803M in ABCA3 completely cosegregated with the dominant CCMC phenotype within Family B. 

We further carried out direct Sanger sequencing of the ABCA3 exons in the 26 unrelated (based on their self-identified geographical ancestry), sporadic patients with CCMC. A total of 43 eyes of 26 patients (12.6 ± 17.41 years old, 10 males) had congenital cataract and nystagmus. The corneal diameter of the 43 eyes ranged from 5.0 to 9.5 mm (7.84 ± 1.39 mm in the transverse meridian and 7.90 ± 1.34 mm in the vertical meridian). 

In the 26 sporadic patients, we identified a total of 4 variants when we sequenced the ABCA3 exons, and all the 4 variants (present in 5 unrelated individuals) were filtered against the 1000 Genomes Project, the dbSNP129 database, HapMap 8, YH database, and 200 controls. Detailed clinical characteristics of the five patients are summarized in Table 4 and Figure 5. There were no ocular abnormalities and no family history of other systemic abnormalities in the parents of the 26 sporadic patients. 

Figure 5

 

Slit-lamp and B-scan ultrasonagraphic photographs of sporadic patients with CCMC. (A, B) Slit-lamp photographs of sporadic patient 13. (C, D) Slit-lamp photographs of sporadic patient 15. (E, F) B-scan ultrasonagraphic photographs of sporadic patient 12. (G, H) B-scan ultrasonagraphic photographs of sporadic patient 15. (I, J) B-scan ultrasonagraphic photographs of sporadic patient 17. The phenotypes are described and summarized in Table 4.

Table 4

 

The Clinical Features of the 5 Patients With ABCA3 Gene Mutation in the 26 Sporadic Patients With Cataract-Microcornea Syndrome

Patient	Age, y/Sex	Eye	Visual Acuity	Best Corrected Visual Acuity	Lens	Nystagmus	IOP, mm Hg	Axial Length, mm	Corneal Curvature, D	Corneal Diameter, mm	B-Scanning
Sporadic 4	33/M	OD	FC	FC	Cataract	Yes	14	21.37	Min:43.90	Transverse:8.0 Vertical:8.0	Vitreous bodies opaque
									MAX:44.90
		OS	HM	HM	Cataract	Yes	14	22.18	Min:43.60	Transverse:8.0 Vertical:8.0	Vitreous bodies opaque
									MAX:44.70
Sporadic 12	6/F	OD	0.5	0.8	Normal	No	14	21.03	Min:44.05	Transverse:10.8 Vertical:10.5	Normal
									MAX:45.25
		OS	HM	HM	Cataract	Yes	15	18.33	Min:43.50	Transverse:8.5 Vertical:8.5	Short axial length
									MAX:46.50
Sporadic 13	7/M	OD	HM	HM	Cataract	Yes	16	25.8	NA	Transverse:5.0 Vertical:5.0	Vitreous bodies opaque, posterior scleral staphyloma
		OS	0.4	0.8	Normal	No	17	25.05	Min:40.05	Transverse:11.2 Vertical:11.0	Vitreous bodies opaque, posterior scleral staphyloma
									MAX:41.6
Sporadic 15	7/F	OD	0.08	0.08	Cataract	Yes	23	19.95	NA	Transverse:7.0 Vertical:7.0	Short axial length, posterior scleral staphyloma
		OS	FC	FC	Cataract	Yes	16	13.3	NA	Transverse:5.0 Vertical:5.0	Short axial length
Sporadic 17	0.5/F	OD	NA	NA	Cataract	Yes	16	17.86	Min:43.60	Transverse:7.0 Vertical:7.0	Posterior scleral staphyloma
									MAX:46.60
		OS	NA	NA	Normal	No	18	19.73	Min:45.36	Transverse:10.6 Vertical:11.0	Normal
									MAX:46.70

These four heterozygous mutations identified from the sporadic patients were c.4253A>T resulting in an N1418I amino acid change, c.2069A>T resulting in an E690V amino acid change, 4035+2T>C, and 2765-1G>T (Table 3, Figs. 4I–P). In the parents of the 5 sporadic patients (sporadic 4, 12, 13, 15, and 17), no variant in ABCA3 gene was detected when we sequenced the ABCA3 exons. 

The c.4393G>A generated a heterozygous missense mutation (p.Asp1465Asn). The c.2408C>T generated a heterozygous missense mutation (p.Thr803Met). The c.2069A>T generated a heterozygous missense mutation (p.Gly690Val). The c.4253A>T generated a heterozygous missense mutation (p.Asn1418Ile). The p.Gly690Val mutation was located in the first nucleotide binding domain. The p.Asn1418Ile and p.Asp1465Asn mutations were located in the second nucleotide binding domain. The p.Thr803Met mutation was located in the loop between the first nucleotide binding domain and the seventh membrane-spanning domain (Fig. 6A). 

Figure 6

 

(A) Model of ABCA3 protein structure. The protein is embedded in the membrane, and has two similar domains, each consisting of six membrane spanning domains (cylinders) and a nucleotide binding domain (NBD). The locations of the identified mutations associated with CCMC are indicated. ECD1 and ECD2, extracellular domains; NBD1 and NBD2, nucleotide binding domains. Numbers refer to amino acid position. (B) Alignment of sequences surrounding the L39V, V93I, E690V, T803M, N1418I, and D1465N mutation in human, chimpanzee, monkey, pig, rat, and mouse. The five mutations (L39V, E690V, T803M, N1418I, and D1465N) in ABCA3 are highly conserved among different species. The various species included Homo sapiens (NP_001080.2), Pan troglodytes (chimpanzee, XP_510744.2), Macaca mulatta (Rhesus monkey, XP_001085237.2), Sus scrofa (pig, XP_003124787.2), Rattus norvegicus (Norway rat. XP_001054650.1), Mus musculus (house mouse, NP_001034670.1), and Danio rerio (zebrafish, XP_002661144.3).

Multiple alignments of Asp1465, Thr803, Gly690, and Asn1418 of the human ABCA3 protein (Homo sapiens, NP_001080) from different species revealed 100% identity, which suggested that they were highly conserved during evolution (Fig. 6B). The Sorting Intolerant From Tolerant (SIFT) and Polymorphism Phenotype (PolyPhen) tool analysis revealed a score for each of the four mutations and predicted that the replaced amino acid was “damaging” to protein function. 

The c.115 C>G generated a missense mutation (p.Leu39Val). Multiple alignments of Leu39of the human ABCA3 protein from different species revealed 100% identity, which suggested that it was highly conserved during evolution (Fig. 6B). The SIFT tool analysis predicted that the replaced amino acid was “tolerated” to protein function. The PolyPhen-2 tool analysis predicted that the replaced amino acid was “possibly damaging” to protein function. 

ABCA3 Expression

To get an insight on the expression of ABCA3 expression in eye development, we performed an investigation of mouse embryo in the Eurexpress database (available in the public domain at http://www.eurexpress.org/ee/intro.html). The ABCA3 was expressed in the eye of mouse embryo. To get an insight on the expression of ABCA3 in human eye tissues, we performed an extensive examination of human expressed sequences located in the NCBI UniGene database (available in the public domain at http://www.ncbi.nlm.nih.gov/unigene/) and in the Eyebrowse site (available in the public domain at http://eyebrowse.cit.nih.gov/), which displayed expressed sequence tags (ESTs) obtained from complementary DNA clones from eye tissues derived from NEIBank and other sources. We identified a number of ESTs matching to the ABCA3 gene in different parts of the eye, such as fetal eyes, lens, eye anterior segment, optic nerve, retina, retinal fovea and macula, RPE, and choroid (Table 5). 

Table 5

 

Results of Nucleotide BLAST (blastn) Searches of ABCA3 Expressed Sequence Tags Located in the Unigene Database

GenBank EST	Sequence Length, bp	Eye Tissue
CK299203.1	679	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
CK301091.1	493	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
BE251915.1	618	Retinoblastoma
BG471039.1	664	Retinoblastoma
BG474383.1	906	Retinoblastoma
BM465915.1	1009	Retinoblastoma
BM711374.1	317	Fetal eyes
BM716567.1	726	Fetal eyes
BM717900.1	506	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
BQ186032.1	689	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
BQ187006.1	590	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
BQ189474.1	741	Fetal eyes, lens, eye anterior segment, optic nerve, retina, retina foveal and macular, RPE and choroid
BU153899.1	870	Retinoblastoma
BU176836.1	907	Retinoblastoma
BU184378.1	894	Retinoblastoma
BU738127.1	645	Fetal eyes
CA393439.1	484	RPE/choroid

To make sure that the ABCA3 was expressed in the eye, we examined ABCA3 expression in different human ocular tissues using RT-PCR and Western blotting. The ABCA3 gene and protein were expressed in the human lens capsule, choroid-RPE, and ARPE-19 cells (Fig. 7). In ARPE-19, ABCA3 was detected in the cytoplasm and on the plasma membrane by immunofluorescence microscopy (Fig. 8). 

Figure 7

 

Expression of ABCA3 in human ocular tissues and ARPE-19. (A) The mRNA expression of ABCA3 in the human cornea, conjunctiva, ICB, sclera, retina, choroid-REP, lens capsule, and ARPE-19 cells. (B) Expression of ABCA3 in the human cornea, conjunctiva, ICB, retina, choroid-REP, lens capsule, and ARPE-19 cells by Western blotting. ARPE-19, human retinal pigment epithelial cell; ICB, iris-ciliary body.

Figure 8

 

Representative figures of immunolabeling of ABCA3 in ARPE-19 cells. The arrowhead indicates the expression of ABCA3. The asterisk indicates the location of nucleus. Magnification: 1600×.

To confirm the change of the ABCA3 expression in the patients with CCMC and unaffected family members in Families A and B, total RNA and protein were prepared from venous blood. The ABCA3 mRNA level (normalized to GAPDH) was approximately 55% lower in patients with CCMC than in unaffected family members (*P < 0.01; Figs. 9A, 9C). The ABCA3 protein level (normalized to GAPDH) was approximately 70% lower in patients with CCMC than in unaffected family members (*P < 0.01; Figs. 9B, 9C). 

Figure 9

 

Confirmation of the change of ABCA3 expression in the patients with CCMC and unaffected members of Family A and B. (A) The ABCA3 mRNA expression in the patients with CCMC and unaffected family members. (B) ABCA3 protein expression in the patients with CCMC and unaffected family members. (C) Analysis of ABCA3 expression as mean ± SD. *P < 0.01, as compared with unaffected family members.

Discussion

Microcornea-cataract syndrome is an autosomal dominant inherited disease characterized by the association of congenital cataract and microcornea without any other systemic anomaly or dysmorphism. Clinical findings include a corneal diameter inferior to 10 mm in both meridians, and an inherited cataract. Other rare ocular manifestations include myopia, iris coloboma, sclerocornea, and Peters anomaly. Transmission seems to be autosomal dominant, sometimes with a high degree of penetrance. Although mutations of several genes have been shown to cause dominant CCMC, in many patients the causative gene has not yet been identified. 

In this study, we have identified a novel dominant CCMC associated gene, ABCA3, which had different heterozygous missense mutations in two autosomal dominant CCMC families (c.115C>G, c.277G>A, c.4393G>A, and c.2408C>T). Another four heterozygous mutations, 2 missense (c.4253A>T, N1418I; c.2069A>T, E690V), and 2 splice site mutations (c. 4053+2T>C, c.2765-1G>T) were identified from the sporadic patients. Parental clinical information of the 26 sporadic patients had been gathered. There were no ocular abnormalities and no family history of other systemic abnormalities. In the parents of the 5 sporadic patients (sporadic 4, 12, 13, 15, and 17), no variants in ABCA3 gene were detected when we sequenced the ABCA3 exons. 

We provided several lines of evidence to support the contention that the ABCA3 gene was the pathogenic causative gene: (1) three heterozygous missense variant in ABCA3 gene was identified in Family A with dominant CCMC via exome sequencing, (2) a different heterozygous missense mutation was identified in another family with dominant CCMC (Family B), (3) these variants completely cosegregated with the disease phenotype in the two families, (4) two heterozygous missense mutations and two heterozygous mutations in SSs were identified in 5 sporadic patients with CCMC, (5) five ABCA3 mutations (c.115C>G, c.4393G>A, c.2408C>T, c.4253A>T, and c.2069 A>T) were present at a conserved site among different vertebrates, (6) four missense mutations (c.4393G>A, c.2408C>T, c.4253A>T, and c.2069 A>T) were predicted to be functionally damaging by SIFT and PolyPhen-2 tools, (7) the ABCA3 mRNA and protein levels were significantly lower in patients with CCMC than in unaffected family members, (8) neither mutation was seen in the 200 unaffected healthy controls. 

The ABC family of transporters is a large family of related transmembrane proteins that bind and hydrolyze ATP to translocate a wide variety of substrates across biological membranes.^27,28 These proteins share a common structure, with half-transporters having 6 membrane spanning domains and a cytoplasmic ATP-binding domain with conserved motifs and full transporters containing 12 transmembrane regions and 2 nucleotide-binding domains. The ABCA3 protein is encoded by a single gene, located on human chromosome 16 which contains 33 exons, with the first 3 exons being untranslated, and the gene is referred to as ABCA3.^29–35 The ABCA3 gene spans over 80,000 nucleotide bases and is transcribed into an approximately 6500–base pair (bp) mRNA, which directs the synthesis of a 1704–amino acid protein.³³ Although the cDNA for ABCA3 was first isolated from thyroid tissue, it is most highly expressed in lung tissue. Its expression is low in a wide range of other human tissues, including heart, brain, and kidney, and platelets.^33,35 

Recently, mutations in the human ABCA3 gene were associated with lethal respiratory distress.^36,37 Genetic variants identified in ABCA3 in patients with surfactant metabolism dysfunction-3 (SMDP3) were shown in Table 6. 

Table 6

 

Genetic Variants Identified in ABCA3 in Patients With Surfactant Metabolism Dysfunction-3 (SMDP3)

dbSNP rs# Cluster ID	Codons	Substitution	Mutation Type	Mutation Mode
rs121909181	c.3426G>A	W1142X	Missense	Homozygosity
rs121909182	c.301T>C	L101P	Missense	Homozygosity
rs121909183	c.4657T>C	L1553P	Missense	Homozygosity
rs28936691	c.4772A>C	Q1591P	Missense	Heterozygosity
rs121909184	c.1702G>A	N568D	Missense	Heterozygosity
–	c.4909+1G>A	–	Splice site	Homozygosity
rs121909185	c.977T>C	L326P	Missense	Homozygosity

It may be that a specific interaction between an unknown protein and ABCA3 during eye development is disrupted by the specific mutations identified in congenital cataract patients; these would rather not result from a decrease in the amount of ABCA3 proteins but in conformational properties of the mutant proteins. There were now some well documented examples of severe bronchopulmonary dysplasia related to heterozygous mutations in the ABCA3 gene.^38,39 The existence of an increasing number of instances where the heterozygous state carries significant risk burdens is a compelling argument for further study and clarification of the risk and the processes by which these disorders are manifested. 

All the mutations that we identified in the two dominant CCMC families and 26 sporadic patients affected by CCMC were heterozygous, and the mutations in Table 3 were entirely different from that in ABCA3 gene, which was associated with lethal respiratory distress. The p.Gly690Val mutation was located in the first nucleotide binding domain. The p.Asn1418Ile and p.Asp1465Asn mutations were located in the second nucleotide binding domain. The p.Thr803Met mutation was located in the loop between the first nucleotide binding domain and the seventh membrane-spanning domain. Two transmembrane domains consisting of multiple membrane-spanning α-helices (typically six α-helices per domain) form the conduit through which substrate crosses the membrane. These domains also contain a substrate-binding site (or sites), which contributes to transport specificity. Two nucleotide binding domains couple the energy of ATP hydrolysis for substrate translocation. Multiple alignments of Asp1465, Thr803, Gly690, and Asn1418 of the human ABCA3 protein were highly conserved during evolution. 

Given that ABCA3 is predicted to be a glycoprotein that may hydrolyze ATP to provide energy for substrate transport involved in eye development, a mutant ABCA3 protein may impact the normal eye development. However, the exact mechanism of ABCA3 action and its role in dominant CCMC pathogenesis remain unclear, and future functional studies will be important. To date, there have been no documented studies on the ABCA3 gene in eye development, and the data in our study indicating its involvement in a devastating eye disease provided excellent motivation for future investigation of the ABCA3 gene, which in turn should enable dissection of its relationship with dominant CCMC pathogenesis. 

Acknowledgments

The authors thank the participating families and sporadic patients, and the Single Nucleotide Polymorphism database, 1000 genome project, HapMap 8 database, and YH database for the data set used to filter variants. 

Supported by the National Natural Science Foundation of China (81070759, 81300742), and the Young and Middle-Aged Scientists Research Awards Fund of Shangdong Province, China (BS2013YY013). The authors alone are responsible for the content and writing of the paper. 

Disclosure: P. Chen, None; Y. Dai, None; X. Wu, None; Y. Wang, None; S. Sun, None; J. Xiao, None; Q. Zhang, None; L. Guan, None; X. Zhao, None; X. Hao, None; R. Wu, None; L. Xie, None 

Footnotes

 PC and YD contributed equally to the work presented here and should therefore be regarded as equivalent authors.