Comprehensive Mutational Screening in a Cohort of Danish Families with Hereditary Congenital Cataract

Lars Hansen,1,2 Annemette Mikkelsen,2 Peter Nurnberg,3,4 Gudrun Nurnberg,3 Iram Anjum,1,2 Hans Eiberg,2 and Thomas Rosenberg5
PURPOSE. Identication of the causal mutations in 28 unrelated families and individuals with hereditary congenital cataract identied from a national Danish register of hereditary eye diseases. Seven families have been published previously, and the data of the remaining 21 families are presented together with an overview of the results in all families. METHODS. A combined screening approach of linkage analysis and sequencing of 17 cataract genes were applied to mutation analyses of total 28 families. RESULTS. The study revealed a disease locus in seven of eight families that were amenable to linkage analysis. All loci represented known genes, and subsequent sequencing identied the mutations. Mutations were found in eight genes, among them crystallins (36%), connexins (22%), and the transcription factors HSF4 and MAF (15%). One family carried a complex CRYBB2 allele of three DNA variants, and a gene conversion is the most likely mutational event causing this variant. Ten families had microcornea cataract, and a mutation was identied in eight of those. Most families displayed mixed phenotypes with nuclear, lamellar, and polar opacities and no apparent genotypephenotype correlation emerged. CONCLUSIONS. In total, 28 families were analyzed, and mutations were identied in 20 (71%) of them. Despite considerable locus heterogeneity, a high mutation identication rate was achieved by sequencing a limited number of major cataract genes. Provided these results are representative of Western European populations, the applied sequencing strategy seems to be suitable for the exploration of the large group of isolated cataracts with unknown etiology. (Invest Ophthalmol Vis Sci. 2009;50:32913303) DOI:10.1167/iovs.08-3149
From 1The Wilhelm Johannsen Centre for Functional Genome Research, Institute of Cellular and Molecular Medicine, and the 2Institute of Cellular and Molecular Medicine, Section IV, Panum Institute, University of Copenhagen, Copenhagen, Denmark; the 3Cologne Center for Genomics (CCG) and Institute for Genetics, and the 4Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany; and the 5 Gordon Norrie Centre for Genetic Eye Diseases, Kennedy Center, Hellerup, Denmark. Supported by grants from The Danish Association of the Blind and The Danish Eye Health Society. RC-Link is supported by The Danish Medical Research Council. The Wilhelm Johannsen Centre for Functional Genomics and the Genome Group/RC-LINK hosted the project; the Danish National Research Foundation funds the Wilhelm Johannsen Centre for Functional Genomics. Submitted for publication November 12, 2008; revised January 1, 2009; accepted April 16, 2009. Disclosure: L. Hansen, None; A. Mikkelsen, None; P. Nurn berg, None; G. Nurnberg, None; I. Anjum, None; H. Eiberg, None; T. Rosenberg, None The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked advertisement in accordance with 18 U.S.C. 1734 solely to indicate this fact. Corresponding author: Lars Hansen, Section IV, ICMM, Panum Institute, University of Copenhagen, Blegdamsvej 3b, DK-2200 Copenhagen N, Denmark; lah@sund.ku.dk.
Investigative Ophthalmology & Visual Science, July 2009, Vol. 50, No. 7 Copyright Association for Research in Vision and Ophthalmology

ongenital cataract (CC) is among the most common developmental anomalies of the eye. It occurs as an isolated trait, in association with other ocular dysmorphology as well as systemic malformations. The etiology of isolated CC is unknown in approximately 50% of cases, and approximately 30% is monogenetic, with autosomal dominant transmission as the most common mode of inheritance. The knowledge of the genetic background has increased considerably during the past decennia, mainly based on linkage strategies in large families (for review, see Refs. 1 and 2 and references therein). Extensive locus heterogeneity has been documented, and more than 40 cataract-associated loci are known, of which 25 represent identied genes, and the number of mutations exceeds more than 100.1,2 Mutations causing developmental cataracts mainly involve proteins with structural and chaperone functions, including -, -, and -crystallins. Another group includes the lens-specic transmembrane gap junction protein genes GJA3 and GJA8, and the membrane protein genes MIP and LIM2. A third group of genes represents the lens-associated transcription factors HSF4, PITX3, MAF, PAX6, and FOXE3. Mutations in HSF4 have mainly been associated with nonsyndromic cataract, whereas MAF mutations often involve the microcornea cataract phenotype. Structural proteins as the lens-specic beaded lament protein genes BFSP1 and BFSP2 represent an additional group of proteins that may have mutations leading to cataract formation. For most of these genes, cataract is the only disease phenotype observed.1,2 Dominantly inherited mutations are mainly missense mutations that lead to amino acid substitutions. Only a few examples of nonsense mutations or frame shift mutations have been described1,2 (see dbCCM, http://www.wjc.ku.dk/ccmd1.html/ Congenital Cataract Mutation Database, provided in the public domain by the Panum Institute, University of Copenhagen, Copenhagen, Denmark). With a few exceptions, such as the hyperferritinemia-cataract syndrome,3,4 no consistent genotypephenotype relations have become evident to facilitate the identication of the involved gene. We initiated an investigation of hereditary isolated CC and CC with microcornea in the Danish population, to trace the genetic background in selected families.57

MATERIAL
Patients

AND

METHODS

Families and patients were recruited from The National Danish Register of Hereditary Eye Diseases at the former National Eye Clinic for the Visually Impaired, now The Kennedy Center (www.kennedy.dk/). Of 97 families with congenital or infantile cataract with or without microcornea we contacted members of 11 families suited for linkage analyses. A sufcient number of participants from eight families attended the study. DNA from an additional 20 unrelated individuals belonging to smaller families collected during clinical examinations was retrieved from the DNA bank of the Eye Clinic. Except microcornea cataract, families with syndromic cataract including congenital cataract and mental retardation and cataract in aniridia syndrome were

3291

3292

Hansen et al.
TABLE 1. Cataract Disease Loci and Genes A. Disease Loci Chromosomal Band 1pter-p36.13 1q21.1 2q33.3 3q22.1 10q24.32 11q23.1 12q13.3 13q12.11 16q22.1 17q11.2 21q22.3 22q11.23-12.1 B. Genes Gene Symbol CRYAA CRYAB CRYBB1 CRYBB2 CRYBB3 CRYBA4 CRYBA1 CRYGC CRYGD GJA3 GJA8 HSF4 MIP BFSP1 BFSP2 MAF PITX3 GenBank Ref. Seq. NM_000394.2 NM_001885.1 NM_001887.3 NM_000496.2 NM_004076.3 NM_001886.1 NM_005208.3 NM_020989.2 NM_006891.2 NM_021954.3 NM_005267.3 NM_001538.2 NM_012064.2 NM_001195.2 NM_003571.2 NM_005360.3 NM_005029.3 Gene Name STS Marker D1S243 D1S2612 D2S2208 D3S1290 D10S1697 D11S4192 D12S1691 D13S175 D16S3086 D17S841 D21S1890 D22S421 Locus

IOVS, July 2009, Vol. 50, No. 7

CCV, Volkmann cataract GJA8 CRYGA, CRYGB, CRYGC, CRYGD BFSP2 PITX3 CRYAB MIP GJA3 HSF4 CRYBA1 CRYAA CRYBB2, CRYBA4, CRYBB3, CRYBB1

Crystallin -A Crystallin -B Crystallin -B1 Crystallin -B2 Crystallin -B3 Crystallin -A4 Crystallin -A1 Crystallin -C Crystallin -D Gap junction protein, 3 Gap junction protein, 8 Heat shock factor 4 Major intrinsic protein of lens ber Beaded lament structural protein 1, Filensin Beaded lament structural protein 2, Phakinin v-Maf musculoaponeurotic brosarcoma Paired-like homeodomain transcription factor

excluded. The results of mutational analysis of patients with aniridia syndrome have been reported earlier.8 A single two-generation family with possible X-linked transmission and clinical signs of Nance-Horan syndrome, and two families with probable autosomal recessive inheritance did not participate in the present project. Genomic DNA was extracted from whole blood by standard procedures. Mutation controls in the background population were performed with DNA from unrelated normal individuals retrieved from the Copenhagen Family Bank.9 The study adhered to the tenets of the Declaration of Helsinki and was approved by the Copenhagen Scientic Ethics Committee. After being informed, all subjects gave written consent to participate in the study.

from New England Bio Laboratories (Ipswich, MA), and PCR primers were purchased from TAG Copenhagen A/S (Copenhagen, Denmark). A complete genomic scan (data not shown) was performed with a 10K SNP array analysis in two families (CC00116 and CC00162) in which the STS marker analyses failed to show linkage either due to incomplete penetrance (CC00162) or inconclusive results (CC00116).

Direct Genomic DNA Sequencing

Direct DNA sequencing of amplied PCR products of all exons, exon intron borders, and parts of the 5 and 3 UTR was performed in 17 known cataract disease genes (Table 1B; BigDye version 1.1 sequencing technology and 3130xl sequencing apparatus; Applied Biosystems, Inc. [ABI], Foster City, CA). PCR and sequencing primers were designed by Primer 3 and purchased from TAG Copenhagen A/S (Copenhagen, Denmark). (PCR primer sequences are found in Supplementary Table S1, online at http://www.iovs.org/cgi/content/full/50/7/3291/ DC1.) The sequence data for the coding regions were aligned to GenBank reference sequences, and genomic intron sequences were aligned to human reference assembly hg17 (NCBI Build 35 from UCSC, http://genome.ucsc.edu/ provided in the public domain by UCSC Genome Bioinformatics, University of California at Santa Cruz, Santa Cruz, CA).11 Taq DNA polymerases were purchased from three sources (New England Biolabs; HotStartTaq DNA Polymerase from Qiagen, Hilden, Germany; and Platinum Taq DNA polymerase from Invitrogen, Carlsbad, CA). PCR and sequencing were performed according to standard protocols and analyzed (Chromas software; Technelysium Pty. Ltd., Tewantin, Australia).

Linkage Analysis
Four families were analyzed using one or several locus-specic STS markers close ( 1 cM) to known cataract disease genes. Haplotypes were drawn for each locus (Cyrillic ver. 2.1.3; Cherwell Scientic, Oxford, UK) and checked by visual examination. Initial locus screening was performed with markers for 12 different known cataract loci (Table 1A). Additional STS markers were included for ne mapping of candidate loci (Table 2) and two-point LOD scores for initial exclusion were calculated with the program LIPED.10 The STS marker analyses were performed using P-33 radioactive labeled oligonucleotide primers followed by PCR. Endlabeling of oligonucleotides was performed according to a standard protocol with T4-DNA-polynucletide kinase (Fermentas, Vilnius, Lithuania) and -33P-ATP (Hartmann Analytic, Braunschweig, Germany) and the labeled oligonucleotides were used for PCR without further purication. Taq-DNA polymerase was purchased

IOVS, July 2009, Vol. 50, No. 7

TABLE 2. Two-Point LOD Score Calculations

Hereditary Congenital Cataract Mutations

3293

Z( ) Mb* CC00105CRYAA D21S1411 c.61C T D21S1890 CC00124CRYAA c.61C T D21S1890 CC00128HSF4 D16S3086 c.341T C D16S3095 CC00145GJA8 D1S442 c.218C T D1S3466 CC00162GJA3 c.227G A 0.00 0.01 0.05 0.10 0.20 0.30 0.40

43.03 43.46 43.67 43.46 43.67 65.49 65.75 68.50 143.12 144.61 147.00

1.73 3.38 1.29 3.01 0.36 2.77 5.61 3.86 2.25 5.34 3.43 2.50

1.77 3.38 1.32 2.96 0.34 2.70 5.52 3.78 2.21 4.99 3.39 2.51

1.82 3.29 1.37 2.74 0.30 2.43 5.14 3.43 2.04 4.55 3.22 2.47

1.76 3.07 1.35 2.46 0.26 2.09 4.66 2.99 1.82 3.58 2.98 2.32

1.41 2.46 1.13 1.85 0.22 1.44 3.62 2.10 1.39 2.51 2.40 1.88

0.91 1.69 0.76 1.16 0.17 0.83 2.50 1.25 0.95 1.32 1.72 1.32

0.36 0.79 0.29 0.43 0.09 0.29 1.28 0.46 0.50 5.42 0.93 0.69

* Physical distances in mega base pairs according to UCSC Marts 2006, NCBI Build 36.1. Penetrance in the calculations for Z( ) is 0.95. Penetrance in the calculations for Z( ) is 1.00.

Restriction Enzyme Digests

Identied mutations were analyzed by restriction enzymes digests in accordance with the manufacturers protocol (New England Biolabs) in a 20- L volume with 2- to 4- L PCR product and 5 to 10 units enzyme. The cleaved PCR products were analyzed by 2% agarose or 20% acrylamide gel electrophoresis with 1 TBE, and the DNA was visualized by staining with ethidium bromide.

GenotypePhenotype Relations
Most patients had their cataracts surgically removed, and the phenotypes were therefore retrieved retrospectively from the les of various ophthalmology departments. Many of the notes, however, dated 30 to 50 years ago and the available information was often too insufcient for an adequate classication resulting in only fragmentary data. Cases demanding early surgical intervention were frequently accompanied by nystagmus, which persisted after surgery. The study included six families with CRYAA mutations, CC00105, CC00124, CC00174, CCMC0101, CCMC0106, and CCMC0106. Among these identical mutations, p.Arg21Trp, were found in three families, CC00105, CC00124, and CCMC0108. Affected members of CC00105 and CC00124 had anterior polar cataracts and various nuclear and lamellar cortical involvements with highly variable impact on visual function. Furthermore, one member of CC00105 had a congenital, unilateral, inferior iris coloboma, and another individual in the same family had reduced corneal diameters of 9 mm. A third mutation carrier had completely clear lenses when examined at 24 years of age. Four of six individuals in family CC00124 still had their lenses left at the ages of 5, 13, 45, and 59 the latter with optical iridectomies in both eyes. The cataracts were mainly of the anterior polar type with different involvement of nucleus, posterior pole, and cortex in six individuals. (See the Discussion section for further details.) Information was available for two families with a GJA3 mutation (CC00162) and a GJA8 mutation (CC00145). Three members of family CC00162 showed a lamellar cataract type with moderate opacity of the fetal nucleus and Y-shaped condensations in the anterior suture. In three members of family CC00145 the lens morphology was described as dense and star-shaped with various locations in the nucleus or the poles. One individual with an anterior polar opacity developed a nuclear opacity during early infancy. Family CC00128 harbored a HSF4 mutation. In this family, four individuals had lamellar cataracts involving faint opacities of the fetal nucleus with condensation along the anterior Y-suture and various, partly progressive opacities in

Subcloning of PCR Products

PCR products were subcloned by TA cloning (pCR-XL-TOPOII vector; Topo XL PCR cloning procedure; Invitrogen). The cloned PCR fragments were screened by direct colony PCR in conditions identical with those used for the genomic DNA PCR.

RESULTS
The present study includes the remaining 21 families from an investigation of 28 families of Northern European decent with hereditary congenital cataract. The results from seven of these families have been presented in previous publications.57 Overall, likely pathogenic mutations were identied in 20 (71%) of the 28 families. The afrmative results implicated a total of eight genes involving four crystallin genes, CRYAA (six families), CRYBB2 (two families), and CRYBB3 and CRYGD (one family each); two connexin genes, GJA3 and GJA8 (three families each); and two transcription factors, HSF4 and MAF (two families each; Table 3). In addition to the identied mutations a genome-wide scan of a large family (CC00116) revealed two provisional novel loci with equal LOD scores (Z 2.7, 0). The two regions, 2q32.2-33.3 and 17q11.2q21.2, are bordered by the SNPs rs952242 and rs1551443 for chromosome 2 and the SNPs rs952581 and rs1846043 for chromosome 17 (data not shown). The remaining seven inconclusive families underwent sequencing of the 17 examined cataract genes without identifying any pathogenic mutation. Figures 1 to 5 illuminate the hitherto unpublished results according to the implicated genes. Seven novel polymorphisms were identied and are presented in Table 4.

3294

Hansen et al.

IOVS, July 2009, Vol. 50, No. 7

TABLE 3. Accumulated Results of the Danish Congenital Cataract Mutation Study Family CCMC0101 CC00105, CC00124, CCMC0108 CC00174 CCMC0106 CC00156 CC00133 CCMC0102 CCMC0109 CC00103 CC00129 CC00162 CC00145 CCMC0103 CC00110 CC00128 CC00171 CCMC0112 CCMC0113 CCMC0107 CCMC0110 CC00109 CC00117 CC00155 CC00159 CC00805 CC00116 Gene CRYAA CRYAA CRYAA CRYAA CRYBB2 CRYBB2 Nucleotide Change* c.34C T c.61C T c.155C T c.337G A c.498C A c.[433C T; 440A G; 449C T] c.224G A c.418C A c.32T C c.176C T c.227G A c.218C T c.565C T c.836C A c.341T C c.355C T c.895C A c.958A G No mutation No mutation No mutation No mutation No mutation No mutation No mutation chr2q32.233.3 or chr17q11.2q21.2 Amino Acid Change* p.Arg12Cys p.Arg21Trp p.Arg49Cys p.Arg116His p.Tyr159X p.[Arg145Trp; Gln147Arg; Thr150Met] p.Arg75His p.Tyr134X p.Leu11Ser p.Pro59Leu p.Arg76His p.Ser73Phe p.Pro189Leu p.Ser259Tyr p.Leu114Pro p.Arg119Cys p.Arg299Ser p.Lys320Glu Number of Analyzed Control Persons 170 170 None 170 None 100 238 None 60 None None 60 170 170 None None 52 173 References This study, 6 This study, 6 This study, 12 This study, 6 Novel Novel (rs2330991, rs2330992, rs4049504) Novel This study, 6 This study, 5 This study, 13 This study, 14 Novel This study, 6 Novel This study, 15 This study, 15 This study, 7 Novel Novel

CRYBB3 CRYGD GJA3 GJA3 GJA3 GJA8 GJA8 GJA8 HSF4 HSF4 MAF MAF Locus

* DNA and protein variation nomenclature is according to Human Genome Variation Society recommendations (http://www.hgvs.org/).16

the cortices. One of these patients also developed a dense anterior polar cataract. Two families with MAF mutations (CCMC0102 and CCMC0113; Table 3) had small corneae. One member of the latter family was reported to be unaffected but showed up to carry the mutation. An examination, however, showed microcorneae with 10-mm horizontal diameters and only scattered punctuate stromal opacities, which otherwise would not have been noted. His visual acuity was 1.0 on both eyes. Another member of the same family showed dense nuclear opacities with a clear periphery as documented by a photograph from 1967 when the patient was 26 years of age (Fig. 6).

DISCUSSION
Apart from mutation screening projects in South Indian families1719 and Australian families,20 our study is among the rst to report on the mutation spectrum in a relative large cohort of patients with CC. Although there is a sizeable and still expanding genetic heterogeneity, we show that a relative high success rate would have been obtained with a sequencing strategy involving rather few major genes. Although this is true of the present family collection, it is not necessarily true of new unselected families. However, it suggests that these genes may be particularly useful for screening in samples of similar origin. In family CC00162 the initial linkage analysis failed to point out a locus. A whole-genome screening, however, led to the identication of a known cataract gene, GJA3. The family investigations disclosed the presence of two instances of nonpenetrance (CC00105 and CC00162) that explained the failed linkage analysis. Without the whole-genome screening, the reduced penetrance in this family supposedly would have

passed unnoticed. Many of the included families were too small to decide about the hereditary mode. Except for one family (CC00116) linkage analyses identied only the already known loci representing known genes, which implies that the same results would have been obtained by the sequencing of a single affected individual. SNP-array analysis in family CC00116 identied two different loci representing a 15.6-Mbp region on chromosome 2 and a 9.3-Mbp region on chromosome 17. The results suggest at least one novel cataract-associated locus. Known cataract genes are located close to both regions, on chromosome 2 the -crystallin cluster (CRYGA to CRYGD) is found 1.4 Mbp distal to the linkage region and on chromosome 17, the CRYBA1 gene is located 2.4 Mbp proximal to the linkage region. Both these loci are outside the mapped regions and the genes have in addition been sequenced before the genome scan. Whether one of the two linkage regions harbors a mutation in a longdistance regulatory element or a new cataract gene awaits disclosure and is under investigation. In total, 20 mutations were identied among the 28 families included in the study, and it is noteworthy that ve of these mutations were reported earlier. We document that a founder effect seems very unlikely with one exception (CC00124 and CCMC0108). The DNA sequence analyses included 17 of the known at least 25 cataract-associated genes. The remaining genes were excluded due to a weighing of workload against the possible relevance for our samples. The newly identied gene EPHA221 was published after the conclusion of this study, but will be included in future analyses. In our study, most of the involved genetic variants were missense mutations; only two nonsense mutations were en-

IOVS, July 2009, Vol. 50, No. 7

Hereditary Congenital Cataract Mutations

3295

FIGURE 1. Pedigrees and restriction digests of three families with CRYAA mutations. (A) The MspI restriction enzyme digest of exon 1 identied all affected individuals as carriers of the mutation c.61C T in family CC00105. Note that person IV:7a was a healthy carrier. Wild-type allele, 205, 117, and 107 bp; mutant allele, 321 bp. (B) The mutation and the sizes of the MspI restriction enzyme digest in family CC00124 were identical with the digest of family CC00105. (C) The pedigree and the AciI restriction enzyme digest of family CC00174 showed the two affected individuals to be carriers of the mutation CRYAA c.155C T. Wild-type allele: 15, 98, 124, and 191 bp; mutant allele: 98, 124, and 206 bp (only the 191- and 206-bp fragments are shown). Filled symbols: affected individuals; open symbols: unaffected individuals; circles: females; squares: males; M: 50-bp DNA ladder; (C) Digest of a normal unrelated individual. U: uncut PCR products. (D) The DNA sequence and the translation of the rst 70 nucleotides of the CRYAA coding region show the SNP rs872331 (c.6C T) and the CRYAA mutation (c.61C T) together with the haplotypes of the families CC00105 and CC00124 and family CCMC0108.6 All three families carried the CRYAA mutation c.61C T, and the haplotypes excluded a common founder for CC00105 and CC00124.

countered and no insertions or deletions were found, which is in accordance with the nding of other investigators. The majority of genetic analyses of congenital cataract include AD-CC families and very few cases of AR-CC forms or sporadic cases are reported. These results infer that mainly missense and nonsense mutations are found for AD-CC, whereas AR-CC also is due to frameshift mutations. The nding of only one mutation in the Danish families indirectly supports an assumption of autosomal dominant transmission. The present paper includes the remaining un-

published mutational results from 21 families and includes 12 mutations in 13 families.

Mutations in the Lens-Specic Crystallins

Eight mutations were found in 10 families corresponding to 36% of the analyzed families, which is in the same magnitude as the percentage of crystallin mutations in South India ( 30%) when corrected for the share of 53% autosomal dominant families.19

3296

Hansen et al.

IOVS, July 2009, Vol. 50, No. 7

IOVS, July 2009, Vol. 50, No. 7

Hereditary Congenital Cataract Mutations

3297

FIGURE 3. Sequence analyses of families carrying CRYBB3 mutations. (A) The CRYBB3 mutation c.224G A was found in individual II:1 in family CCMC0102 and conrmed by a SacII restriction enzyme digest. Only individual II:1 was available for analyses. Wild-type allele (C): 130 and 270 bp; mutant allele and uncut (U): 400 bp; M: 100-bp DNA ladder. (B) The -B crystallin proteins share a common secondary and tertiary structure of two crystallin domains, each composed of two Greek key motifs. A Greek key motif is inserted in the top left corner. The mutations found in the families CC00133 and CC00156 are denoted. The amino acid numbers for intron positions are shown. (C) The protein sequence alignment of the third and the fourth Greek key motif for the human -B crystallin-1, -2, and -3 showed conservation or semiconservation for p.Arg145, p.Gln147, p.Thr150, p.Gln155, and p.Tyr159 (positions highlighted in yellow). Protein Ref. Seq.: NP_001878, NP_000487, and NP_004067. (D) The CRYBB3 c.224G A position was highly conserved among several mammalian genomes as shown by the DNA sequence alignment (UCSC Human genome browser).11

FIGURE 2. Pedigrees and analyses of two families with CRYBB2 mutations. (A) In family CC00156, only individual II:2 was available for analysis. The DNA chromatogram shows the nonsense mutation CRYBB2 c.498C A, which changes the tyrosine codon TAC into TAA. (B) Alignment of the DNA sequence for exon 6 for the wild-type CRYBB2 (NM_000496) and the corresponding sequence for individual II:2 and for the homologous pseudogene CRYBB2P1 (BC037884) shows variation for c.475C (green), for position c.483C (blue), and position c.489C (yellow). The most abundant CRYBB2 mutation c.475C T may be due to a gene conversion, whereas the alignment of positions c.483 and c.489 excludes a gene conversion for the novel mutation c.489C A. Redundant nucleotide M: C and A. (C) The pedigree and the haplotypes of the two affected individuals and the inferred haplotype of one healthy individual of family CC00133. The DNA chromatogram shows the three DNA polymorphisms for individual I:1. (D) Alignment of the genomic sequences of exon 5 and the border to intron 5 shows wild-type CRYBB2 (top), the analyzed sequence of individual II:1 (middle), and the CRYBB2P1 pseudogene (bottom) suggesting that the three nonsynonymous changes in the disease allele was a result of a gene conversion. The converted region was a minimum of 80 bp long (green); the maximum length (gray) could not be predicted due to missing sequence information. Redundant nucleotides Y: C and T; R: A and G; and S: C and G. Exons are shown in capital letters, introns in lower case; the SNP rs57112959 (G A) refers to the wild-type CRYBB2 gene.

3298

Hansen et al.

IOVS, July 2009, Vol. 50, No. 7

IOVS, July 2009, Vol. 50, No. 7 Among 32 families with autosomal dominant inheritance and four families of uncertain inheritance from southeastern Australia, only two (5%) crystallin mutations were identied.20 We have no explanation for this difference. However, there were few tissue samples, and selection bias for larger families may inuence the results. It is also noteworthy that the Australian and the Danish source populations are of similar magnitude, while the total number of ascertained families was more than twofold in the Danish population. The mutations in our cohort were found in four of the nine lens-expressed crystallin genes that we have analyzed (Table 3). One previously published mutation6 was found in two families (CRYAA, c.61C T, p.Arg21Trp; Fig. 1) and a different mutation in the same codon (p.Arg21Leu) has been reported before in association with cataract,22 which suggests the Arg21 residue to be of crucial importance for the protein function. Sequencing of exon 1 of CRYAA using the SNP rs872331 located 55 nt upstream of the mutant nucleotide c.61T demonstrated the haplotype c.[6T;61C] for family CC00105 and the haplotype c.[6C;61C] for family CC00124 (Fig. 1D).This suggests that the mutation arose independently in the two families. The third family (CCMC0108) with cataract and microcornea carried the same mutation and sequence analysis of subcloned separated PCR products from one affected individual from CCMC0108 revealed the haplotype c.[6T;61C] identical with family CC00105 (Fig. 1D). Subsequent genealogical studies conrmed a common ancestral founder for the two families. According to our knowledge, incomplete penetrance as documented in family CC00105 (Fig. 1A, IV:7) has not been reported before in families with a CRYAA mutation. The novel CRYBB2 mutation p.Tyr159X (Table 3, Figs. 2A, 2B, 3B, 3C) presumably terminates the reading frame of exon 6 before the authentic stop codon. The mutant mRNA will presumably avoid the nonsense-mediated RNA decay pathway and be translated into a truncated protein. Another nonsense mutation (p.Gln155X, Fig. 2B, 3B, 3C) has been reported in ve unrelated families with dominantly inherited cataract.20,2327 Two of these mutations25,27 has been shown to be a consequence of gene conversions between a region of 9 to 104 bp surrounding the mutation and the homologous region in the CRYBB2P1 pseudogene. The remaining three mutations seem to have occurred by point substitutions.23,24,26 By alignment of the wild-type CRYBB2, the corresponding sequence from II:2CC00156, and the CRYBB2P1 exon 6 sequence (Fig. 2B), it is obvious that the p.Tyr159X mutation found in family CC00156 is a point mutation and not a result of gene conversion, which is further shown by the chromatogram that demonstrates the sequence [CCCCGGCTAC/A], which should have been [CCCC/TGGTAC/A] if both a point mutation and gene conversion have taken place. Identication of the two nonsense mutations in the same fourth Greek key motif of CRYBB2 suggests a crucial region for cataract-associated mutations, and the pathogenic mechanism is presumably the same for the two mutations.

Hereditary Congenital Cataract Mutations

3299

A complex CRYBB2 allele with three nucleotide changes in exon 5 was detected in both affected individuals of family CC00133 (Table 3, Fig. 2C). The DNA variations rs2330991, rs2330992, and rs4049504 (dbSNP, http://www.ncbi.nlm.nih.gov/, provided in the public domain by the National Center for Biotechnology Information [NCBI]National Institute of Health, Bethesda, MD) are afrmed as nonsynonymous polymorphisms. PCR products from individual I:1 were subcloned, and a probable cis position of the rare SNP variants was conrmed by sequencing of the subclones. Sequence analysis of the SNPs in CRYBB2 exon 5 for 100 normal unrelated individuals of matching ethnic background only detected the wild-type allele c.[433C; 440A; 449C] (data not shown) and none of the SNPs was found to be polymorphic in any of the other sequenced cataract family members. The SNP rs2330992 (p.Gln147Arg) is nonpolymorphic and represented only by the A-allele (p.Gln147) in the HapMap program (dbSNP, ss3282510, Build 129, NCBI, http://www.ncbi.nlm.nih.gov/). This strongly supports the unusual character of the G-allele for rs2330992 and from the above mentioned observations we consider it reasonable to assume that the occurrence of the three mutations in cis are pathogenic, possibly by transforming the secondary structure of the -crystallin protein (Fig. 3B). A gene conversion between wild-type CRYBB2 and the pseudogene CRYBB2P1 has been shown to be the most likely mechanism for the p.Gln155X mutation.25 A similar mechanism is the most likely cause for the complex allele, as shown by alignment of the homologous DNA sequences for the wild-type CRYBB2 and the CRYBB2P1 pseudogene with the sequence of individual II:1-CC00133 (Fig. 2D). The mutation found in CRYBB3 (c.224G A, p.Arg75His; Table 3) is the rst report of a cataract-associated CRYBB3 mutation with a dominant effect. The mutant genotype was not detected in 238 normal individuals of matching ethnic background or among the other cataract families. The mutation is in the second Greek key motif (Fig. 3B) and destroys a highly conserved amino acid (Fig. 3D) and is therefore most likely pathogenic. Unfortunately, only a single affected family member, individual I:1 (Fig. 3A) with a microcornea cataract was available for investigation. One other CRYBB3 mutation (p.Gly165Arg) has been reported in a consanguineous Pakistani family with recessively inherited cataract.28

Mutations in the Gap Junction Proteins

The six mutations, among which two were novel (Table 3), represent 22% of the families in our sample (Fig. 4). Three of the mutations were located in the rst extracellular loop of the two gap junction proteins, and all three amino acid positions are highly conserved in humans (Fig. 4E). This nding implies that the primary structure of transmembrane regions and the extracellular loops are crucial for the assembly of gap junction proteins into connexons. The GJA8 mutation affecting amino acid p.Ser259 in the carboxyl terminus was novel. The muta-

FIGURE 4. Pedigrees and restriction digests of four families with mutations in the gap junction proteins. (A) Pedigree of family CC00145 shows the haplotypes for all analyzed persons. The STS marker D1SGJA5-GJA8 was located between two genes, GJA5 and GJA8. The EarII restriction enzyme digest showed cosegregation of the mutation GJA8 c.218C T with the disease trait in the family. Wild-type allele: 199 bp; mutant allele: 165 bp. (B) Pedigree of family CC00162. The AciI restriction enzyme digest illustrated the segregation of the mutation GJA3 c.227G A. Note the healthy carrier V:2a. Wild-type allele: 151, 91, and 79 bp; mutant allele: 191, 91, and 79 bp; M: 50 bp DNA ladder. (C) Pedigree of family CC00110. The BseRI restriction enzyme digest showed segregation of the mutation GJA8 c.836C A in both affected individuals. Wild-type allele (C): 114, 135, 154, and 225 bp; mutant allele: 114, 225, and 289 bp. (D) The pedigree of family CC00129. The AluI restriction enzyme digest illustrates the mutation in individual II:1. Wild-type allele (C): 129 and 328 bp; mutant allele: 129, 160, and 168 bp; U: undigested PCR products; M: 50 bp DNA ladder. (E) Graphic representation of the two lens-specic gap junction proteins and the mutations found in the Danish cohort. CP, cytoplasmic domain; TM, transmembrane domain; EC, extracellular domain. Alignment of the group of human gap junction proteins demonstrated conservation of the mutant positions except for the C-terminal mutation. Protein Ref. Seq.: Gja1, NP_000156; Gja3, NP_068773; Gja4, NP_002051; Gja5, NP_005257; Gja7, NP_005488; Gja8, P_005258; Gja10, NP_115991; and Gja12, NP_065168.

3300

Hansen et al.

IOVS, July 2009, Vol. 50, No. 7

IOVS, July 2009, Vol. 50, No. 7

TABLE 4. Novel Polymorphisms Identied in the Danish Cataract Study Gene CRYBA1 CRYBA1 CRYGD CRYGD HSF4 GJA8 MIP MIP Exon Ex2 Ivs3 Promoter Ex3 Ex7 Ex2 Ex1 Ex1 Variation c.74C T c.215 16C T c. 16_37del c.376G A c.636G T c.658A G c.141A G c.319G A

Hereditary Congenital Cataract Mutations

3301

Amino Acid Change p.Pro25Leu p.Val126Met p.Met212Ile p.Asn220Asp p.Ala45Ala p.Val107Ile

Allele Frequency* 6/110 (1/38) (2/30) 0/340 (1/34) 2/240 (1/34) 0/340 (1/34) 1/340 (1/38) (4/30) 5/76 (1/30)

Family (dbLaH) CC00159 (391) CC00805 (395) CC00805 (395) CC00171 (393) CC00109 (387) CC00159 (391) Several families CC00110 (388)

* Allele frequencies are given for number of chromosomes analyzed in the normal persons and in brackets the number chromosomes representing persons from the cataract cohort.

tion segregated with the phenotype in all members of the small family (Fig. 4C), which supports the interpretation as a pathogenic mutation, although the amino acid position is less conserved among the human gap junction proteins. All GJA3 mutations have been reported previously in association with isolated congenital cataract (Table 3). Of interest, the mutation p.Arg76His is characterized by incomplete penetrance in family CC00162 which also was observed in the rst report of the mutation.14 A third mutation affecting the same arginine residue (p.Arg76Gly) has been described in association with a fully penetrant total cataract.17

This observation points toward a common regulatory mechanism of corneal and lens crystallins in humans. Experimental evidence in mice seems to support the presence of such a mechanism. Recently Davis et al.34 showed activation of the specic mouse corneal crystallin Aldh3a1 by different transcription factors as Pax6, Oct1, and p300. Conrmation of a possible dual mechanism involving corneal and lens development induced by the novel MAF zipper domain mutation awaits experimental elucidation.

Polymorphisms
Several DNA variations of the major cataract-associated genes were found by the sequence analyses. Among eight novel polymorphisms, we encountered one synonymous base-exchange, ve missense mutations, one intronic substitution, and one promoter deletion (Table 4). All variations except two were present in normal control samples. The nonsynonymous HSF4 change p.Met212Ile in the C-terminal part of the protein (Table 4) was initially considered to be causal in family CC00109, because of the absence among 170 normal unrelated individual of Danish origin. The variation, however, was also absent in two affected relatives and therefore was classied as a rare polymorphism. A CRYGD promoter deletion was detected in one affected individual of family CC00805 (Table 4). Analyses of 170 unrelated normal individuals of same ethnic background failed to identify the deletion, but it was not found in an affected daughter of the proband and therefore considered to be a rare DNA variant without pathogenic effects. A nonsynonymous GJA8 mutation was found in family CC00159 (Table 4) in which no pathogenic mutation has been identied so far (Table 3). The mutation, p.Asn220Asp involves an amino acid position that is highly conserved among human and mammalian gap junction proteins (data not shown). Surprisingly, the mutation was found in one allele among 170 normal persons with same ethnic background, which led to a classication as nonpathogenic.

Mutations in HSF4
Both HSF4 mutations have previously been described by Bu et al.15 The mutation p.Leu114Pro was reported in a large Chinese family with cataract and p.Arg119Cys in the large Danish family with cataract rst described by Marner in 1949 (Table 3, Fig. 5, MIM 116800).29 31 The repetition of the latter mutation in another Danish family suggests a common founder, which could not be documented by genealogical studies. The nomenclature for the two mutations has been corrected (Fig. 5C) according to the recommendation from the Human Genome Variation Society (http://www.hgvs.org/).16

MAF Mutations
Previously, two MAF mutations have been published in association with microcornea cataract.32,33 The recurrent mutation p.Arg299Ser was ascertained in family CCMC0112 (Table 3) and is predicted to modify the conserved DNA-binding region (Fig. 5D).7 The novel mutation p.Lys320Glu detected in family CCMC0113 affects the leucine zipper region and is the rst pathogenic cataract MAF mutation outside the DNA binding domain. Of note, an apparent case of incomplete expression was observed in individual III:2 who had microcornea, but no cataract. This shows that isolated microcornea may be caused by a mutation in a cataract-associated transcription factor gene.

FIGURE 5. Pedigrees and restriction digests of families with HSF4 and MAF mutations. (A) Pedigree with haplotypes for family CC00128. The BsrI restriction enzyme digests showed that the mutation HSF4 c.341T C cosegregates with the disease. Wild-type allele: 49, 57, 60, 74, and 252 bp; mutant allele: 49, 57, 74, and 312 bp; U: uncut PCR products; M: 50 bp DNA ladder. (B) Pedigree of family CC00171. The mutation HSF4 c.355C T was conrmed by HpyCH4V restriction enzyme digest in both affected individuals. Wild-type allele: 492 bp; mutant allele: 48 bp and 444 bp; U: uncut PCR products; M: 100 bp DNA ladder. (C) The mutations are identical with two mutations described in a Chinese and a Danish family. Bu et al.15 named the mutations c.348T C L115P and c.362C T, R120C, respectively, according to the GenBank cDNA clone, accession number D87673, that starts at the 4 position from the rst ATG codon. The translation of the D87673 results in a HSF4 protein that includes an additional valine residue at position 2. This results in discrepancies between the mutation names published by Bu et al. and the nomenclature used herein (Fig. 4C). The HSF4 isoform A (NM_001538) has been used for the systematic nomenclature (http://www.hgvs.org/mutnomen/).16 (D) Pedigree of family CCMC0113. The half-lled symbol of individual III:3 refers to a case of no cataract with microcornea. The restriction enzyme MboII digest showed segregation of the mutation in the family. Wild-type allele: 84, 92 (seen as one band), and 231 bp; mutant allele: 92 and 315 bp; U: uncut PCR products; M: 100-bp DNA ladder. The graphic representation of the MAF protein shows the known mutations6,32,33 and the novel cataract mutation.

3302

Hansen et al.

IOVS, July 2009, Vol. 50, No. 7

References
1. Shiels A, Hejtmancik JF. Genetic origins of cataract. Arch Ophthalmol. 2008;125:165173. 2. Hejtmancik JF. Congenital cataracts and their molecular genetics. Semin Cell Dev Biol. 2007;19:134 149. 3. Bonneau D, Winter-Fuseau I, Loiseau MN, et al. Bilateral cataract and high serum ferritin: a new dominant genetic disorder? J Med Genet. 1995;32:778 779. 4. Beaumont C, Leneuve P, Devaux I, et al. Mutation in the iron responsive element of the L ferritin mRNA in a family with dominant hyperferritinaemia and cataract. Nat Genet. 1995;11:444 446. 5. Hansen L, Yao W, Eiberg H, et al. The congenital ant-egg cataract phenotype is caused by a missense mutation in connexin46. Mol Vis. 2006;12:10331039. 6. Hansen L, Yao W, Eiberg H, et al. Genetic heterogeneity in microcornea-cataract: ve novel mutations in CRYAA, CRYGD, and GJA8. Invest Ophthalmol Vis Sci. 2007;48:39373944. 7. Hansen L, Eiberg H, Rosenberg T. Novel MAF mutation in a family with congenital cataract-microcornea syndrome. Mol Vis. 2007;13: 2019 2022. 8. Grnskov K, Rosenberg T, Sand A, Brndum-Nielsen K. Mutational analysis of PAX6: 16 novel mutations including 5 missense mutations with a mild aniridia phenotype. Eur J Hum Genet. 1999;7: 274 286. 9. Eiberg H, Nielsen LS, Klausen J, et al. Linkage between serum cholinesterase 2 (CHE2) and crystalline gene cluster (CRYG): assignment to chromosome 2. Clin Genet. 1989;35:313321. 10. Ott J. A computer program for linkage analysis of general human pedigrees. Am J Hum Genet. 1976;28:528 529. 11. Kent WJ, Sugnet CW, Furey TS, et al. The human browser at UCSC. Genome Res. 2002;12:996 1006. 12. Mackay DS, Andley UP, Shiels A. Cell death triggered by a novel mutation in the alphaA-crystallin gene underlies autosomal dominant cataract linked to chromosome 21q. Eur J Hum Genet. 2003;11:784 793. 13. Bennett TM, Mackay DS, Knopf HL, Shiels A. A novel missense mutation in the gene for gap-junction protein alpha3 (GJA3) associated with autosomal dominant nuclear punctate cataracts linked to chromosome 13q. Mol Vis. 2004;10:376 382. 14. Burdon KP, Wirth MG, Mackey DA, et al. A novel mutation in the Connexin 46 gene causes autosomal dominant congenital cataract with incomplete penetrance. J Med Genet. 2004;41:e106. 15. Bu L, Jin Y, Shi Y, et al. Mutant DNA-binding domain of HSF4 is associated with autosomal dominant lamellar and Marner cataract. Nat Genet. 2002;31:276 278. 16. den Dunnen JT, Antonarakis SE. Mutation nomenclature extensions and suggestions to describe complex mutations: a discussion. Hum Mutat. 2000;15:712. 17. Devi RR, Reena C, Vijayalakshmi P. Novel mutations in GJA3 associated with autosomal dominant congenital cataract in the Indian population. Mol Vis. 2005;11:846 852. 18. Devi RR, Vijayalakshmi P. Novel mutations in GJA8 associated with autosomal dominant congenital cataract and microcornea. Mol Vis. 2006;12:190 195. 19. Devi RR, Yao W, Vijayalakshmi P, Sergeev YV, Sundaresan P, Hejtmancik JF. Crystallin gene mutations in Indian families with inherited pediatric cataract. Mol Vis. 2008;14:11571170. 20. Burdon KP, Wirth MG, Mackey DA, et al. Investigation of crystallin genes in familial cataract, and report of two disease associated mutations. Br J Ophthalmol. 2004;88:79 83. 21. Shiels A, Bennett TM, Knopf HL, et al. The EPHA2 gene is associated with cataracts linked to chromosome 1p. Mol Vis. 2008;14: 20422055. 22. Graw J, Klopp N, Illig T, Preising MN, Lorenz B. Congenital cataract and macular hypoplasia in humans associated with a de novo mutation in CRYAA and compound heterozygous mutations in P. Graefes Arch Clin Exp Ophthalmol. 2006;244:912919.

FIGURE 6. Photograph of the right eye of an individual with nuclear cataract. The 26 year-old patient III:5 belonged to family CCMC0113 with microcornea cataract and a MAF p.Lys320Glu mutation (Table 3). The corneal diameter was not measured but she had steep corneas (K-readings 6.7 6.85 mm of curvature), which indirectly indicates a reduced overall corneal size. The cataract consisted of a circular dense nuclear opacity with condensations in a triangular conguration according to the fetal Y-suture. Faintly seen triangular extensions outside the central opacity were present. The cortex and the anterior polar zones were clear. The left eye had identical ndings. The patient underwent surgery with insertion of articial lenses at the age of 43 years.

Phenotypes
Most of our patients showed a composite morphology with regard to size, density, and localization of the lens opacities showing mixtures of nuclear, cortical, and polar cataracts. This nding was further accentuated by considerable intrafamilial differences in phenotypes and additional congenital dysmorphology (microcornea and coloboma). In addition, some cataracts were progressive, leading to changing morphology during infancy. The highly heterogenic phenotypes preclude sound genotypephenotype predictions based on this study. It should be kept in mind, however, that our phenotype data were historical. The descriptive terminology probably differed among clinicians, and centers and routine examination before surgery may have been cursory. A qualied phenotype description should rely on photographic documentation and be based on a descriptive standard for infantile cataracts.

CONCLUSION
A mutational analysis strategy involving direct sequencing of 17 cataract genes identied nearly three fourths of the mutations in a cohort of hereditary congenital cataracts of Northern European descent. We propose the application of the strategy to investigate cases of isolated congenital cataract and unknown etiology.

Acknowledgments
The authors thank the families for their participation, Jeanette Andreasen and Maria Jrgensen for excellent technical assistance, and Erik Kann for the genealogical studies.

IOVS, July 2009, Vol. 50, No. 7

23. Litt M, Carrero-Valenzuela R, LaMorticella DM, et al. Autosomal dominant cerulean cataract is associated with a chain termination mutation in the human beta-crystallin gene CRYBB2. Hum Mol Genet. 1997;6:665 668. 24. Gill D, Klose R, Munier FL, et al. Genetic heterogeneity of the Coppock-like cataract: a mutation in CRYBB2 on chromosome 22q11.2. Invest Ophthalmol Vis Sci. 2000;41:159 165. 25. Vanita V, Sarhadi V, Reis A, et al. A unique form of autosomal dominant cataract explained by gene conversion between betacrystallin B2 and its pseudogene. J Med Genet. 2001;38:392396. 26. Yao K, Tang X, Shentu X, Wang K, Rao H, Xia K. Progressive polymorphic congenital cataract caused by a CRYBB2 mutation in a Chinese family. Mol Vis. 2005;11:758 763. 27. Bateman JB, von-Bischhoffshaunsen FR, Richter L, Flodman P, Burch D, Spence MA. Gene conversion mutation in crystallin, beta-B2 (CRYBB2) in a Chilean family with autosomal dominant cataract. Ophthalmology. 2007;114:425 432. 28. Riazuddin SA, Yasmeen A, Yao W, et al. Mutations in betaB3crystallin associated with autosomal recessive cataract in two Pakistani families. Invest Ophthalmol Vis Sci. 2005;46:2100 2106.

Hereditary Congenital Cataract Mutations

3303

29. Marner E. A family with eight generations of hereditary cataract. Acta Ophthalmol. 1949;27:537551. 30. Eiberg H, Marner E, Rosenberg T, Mohr J. Marners cataract (CAM) assigned to chromosome 16: linkage to haptoglobin. Clin Genet. 1988;34:272275. 31. Marner E, Rosenberg T, Eiberg H. Autosomal dominant congenital cataract: morphology and genetic mapping. Acta Ophthalmol. 1989;67:151158. 32. Jamieson RV, Munier F, Balmer A, Farrar N, Perveen R, Black GC. Pulverulent cataract with variably associated microcornea and iris coloboma in a MAF mutation family. Br J Ophthalmol. 2003;87: 411 412. 33. Vanita V, Singh D, Robinson PN, Sperling K, Singh JR. A novel mutation in the DNA-binding domain of MAF at 16q23.1 associated with autosomal dominant cerulean cataract in an Indian family. Am J Med Genet A. 2006;140:558 666. 34. Davis J, Davis D, Norman B, Platigorsky J. Gene expression of the mouse corneal crystalline Aldh3a1: activation by Pax6, Oct1, and p300. Invest Ophthalmol Vis Sci. 2008;49:1814 1826.