|Home | About | Journals | Submit | Contact Us | Français|
Inherited retinopathies are a genetically and phenotypically heterogeneous group of diseases affecting approximately one in 2000 individuals worldwide. For the past 10 years, the Laboratory for Molecular Diagnosis of Inherited Eye Diseases (LMDIED) at the University of Texas-Houston Health Science Center has screened subjects ascertained in the United States and Canada for mutations in genes causing dominant and recessive autosomal retinopathies. A combination of single strand conformational analysis (SSCA) and direct sequencing of five genes (rhodopsin, peripherin/RDS, RP1, CRX, and AIPL1) identified the disease-causing mutation in approximately one-third of subjects with autosomal dominant retinitis pigmentosa (adRP) or with autosomal dominant cone-rod dystrophy (adCORD). In addition, the causative mutation was identified in 15% of subjects with Leber congenital amaurosis (LCA). Overall, we report identification of the causative mutation in 105 of 506 (21%) of unrelated subjects (probands) tested; we report five previously unreported mutations in rhodopsin, two in peripherin/RDS, and one previously unreported mutation in the cone-rod homeobox gene, CRX. Based on this large survey, the prevalence of disease-causing mutations in each of these genes within specific disease categories is estimated. These data are useful in estimating the frequency of specific mutations and in selecting individuals and families for mutation-specific studies.
To date, over 120 loci for inherited retinal degeneration have been identified, and 56 of the associated genes have been cloned (RetNet, http://www.sph.uth.tmc.edu/RetNet/). Retinal disorders are also phenotypically heterogeneous, such that a single mutation may be associated with substantially different phenotypes within a family or between families, and different mutations within the same gene can cause substantially different retinal disorders. For example, mutations in one gene, peripherin/RDS (RDS; MIM# 179605), have been associated with several forms of inherited retinal degeneration, including cone-rod dystrophy, cone dystrophy, foveal dystrophy, and retinitis pigmentosa [Weleber et al., 1993; Wells et al., 1993; Keen et al., 1994; Nakazawa et al., 1994, 1996]. This clinical heterogeneity has made the molecular diagnosis of inherited eye diseases complicated; moreover, it has made identification of the underlying molecular cause of retinopathy within individual families an essential component of clinical research. If the gene defect causing the inherited retinopathy of a patient is identified, then prenatal diagnosis, a more accurate prognosis, and more accurate genetic counseling may be offered. In addition, several gene-specific and mutation-specific treatments for inherited retinal disorders have been proposed [Lewin et al., 1998; Frasson et al., 1999].
For the past decade, the Laboratory for Molecular Diagnosis of Inherited Eye Diseases (LMDIED, CLIA#45D0935007) at The University of Texas-Houston Health Science Center has tested over 500 unrelated subjects (probands) with inherited retinal disorders who were ascertained within Canada and the United States. The complete coding sequences of rhodopsin (RHO; MIM# 180380), peripherin/RDS, CRX (MIM# 602225), and AIPL1 (MIM# 604392) were assayed in all samples by SSCA, followed by sequencing of variants. In addition, because disease-causing mutations in the RP1 gene (MIM# 603937) appear to cluster in one region of exon 4 [Bowne et al., 1999], that region was assayed by SSCA in all patient samples, and all variants were sequenced. We report identification of the disease-causing mutation in roughly one-third of subjects with autosomal dominant retinitis pigmentosa (adRP) or with autosomal dominant cone-rod dystrophy (adCORD) (Table 1). In addition, mutations were identified in approximately 15% of subjects with Leber congenital amaurosis (LCA).
Informed consent was obtained from all adult subjects, who were ascertained in the United States and Canada. Ethnicity of subjects could not be determined in most cases, as all samples were obtained under institutional rules of confidentiality. However, the population origin for these subjects is largely Caucasian of European ancestry. All tested individuals with disease-causing mutations received clinical evaluations by at least one of the co-authors, and in some instances, by two co-authors jointly.
DNA was isolated from peripheral blood using the Puregene DNA extraction kit (Gentra, Minneapolis, MN), in accordance with the manufacturer’s instructions. A working stock for mutation analysis was stored at 4°C, and the primary sample was stored separately at −70°C. All mutations identified by SSCA and direct sequencing of the working stock were confirmed by direct sequencing of the primary DNA sample.
Genomic DNA samples from subjects were screened by SSCA, using published PCR primer sequences and conditions: rhodopsin [Daiger et al., 1997], peripherin/RDS [Daiger et al., 1997], CRX [Sohocki et al., 1998], RP1 [Bowne et al., 1999], and AIPL1 [Sohocki et al., 2000].
After an SSCA variant was identified, a PCR reaction was performed under the same conditions as for SSCA. The amplified fragment was then treated with shrimp alkaline phosphatase and exonuclease I (USB, Cleveland, OH). Direct sequencing was performed with either the AmpliCycle sequencing kit (PE Biosystems) and a primer end-labeled with 33P-ATP followed by autoradiography, or the BigDye Terminator Sequencing Kit (PE Biosystems) on an ABI Prism 310 automated sequencer according to the manufacturer’s protocols.
For haplotype identification of the peripherin/RDS exon 3 polymorphisms, DNA was amplified incorporating 32P-dCTP with a biotinylated forward primer (5′-TTGGGCTGCTACCTACAG-3′) and an unbiotinylated reverse primer (5′-AGACTTTCGGAGTTGGATGAG-3′) and standard cycling conditions. Individual strands were separated with Dynabeads (DYNAL, Lake Success, NY) according to manufacturer’s protocols and were separated overnight on 0.4X MDE (Biowhittaker, Rock-land, ME) gels.
Likely disease-causing mutations were identified in 105 of the 506 total probands tested, or in approximately 21% of samples. The frequency of mutations identified in autosomal dominant retinopathies was considerably higher—approximately one-third of adRP and one-third of autosomal dominant retinal degeneration with cone involvement were explained by mutations identified in this study.
Rhodopsin mutations were identified in 53 probands from 206 adRP families (26%), two from isolated probands with RP, and four from individuals with RP of unknown inheritance pattern (Table 2A). As expected, the Pro23His rhodopsin mutation was the most frequent, representing 41% of rhodopsin mutations and causing adRP in 22 of 206 (11%) of subjects. The second most common mutation was Arg135Trp, found in 5 (3%) of adRP families. In addition to published mutations, five previously unreported disease-causing rhodopsin mutations were identified, each confirmed by direct genomic sequencing in at least one independently collected DNA sample.
The Pro170Arg mutation segregated with RP in three affected individuals of RFS153, and was not observed in any of the other 1,000 chromosomes tested. This nonconservative substitution replaces the neutral, hydrophobic, rigid ring of proline with a large, basic arginine residue within the fourth transmembrane region of rhodopsin.
The Cys185Arg mutation segregated with the retinitis pigmentosa of two adRP families, RFS075 and RFS088, and was not observed in any other probands tested. The cysteine at this position is conserved in lamprey, chicken, bovine, and human rhodopsin [Okano et al., 1992].
A Pro210Leu mutation was identified in the affected proband of UTAD088 and was not observed in the other individuals tested. This mutation is nonconservative, substituting leucine for the rigid ring of proline at this position within the fifth transmembrane region of rhodopsin. As proline is known to be important in folding the peptide chain, this mutation is likely to alter the secondary structure of the protein.
A 3 base pair deletion was identified in an isolated individual with RP, UTAD338. The deletion alters amino acids 318 and 319, deleting the leucine at position 318, and causing the threonine at position 319 to be replaced by a proline. The threonine at position 319 is highly conserved in lamprey, chicken, bovine, and human rhodopsin [Okano et al., 1992].
Disease-causing mutations in peripherin/RDS were identified in approximately 8% (17 of 206) of the adRP probands. In addition, mutations were identified in two isolated RP cases (Table 2B). Two previously unreported mutations were also identified.
A Tyr141Cys mutation segregated with the retinitis pigmentosa in affected individuals of family BCMAD033 and in none of the other 1,000 chromosomes tested. The tyrosine at this position is conserved in human, bovine, and murine peripherin/rds [Connell et al., 1991].
The Gln178Arg mutation was observed only in the affected proband of family UTAD185. The glutamine at residue 178 of peripherin/RDS is conserved in human, bovine, and murine peripherin/rds [Connell et al., 1991].
The majority of disease-causing RP1 mutations detected to date cluster in a region of exon 4 [Bowne et al., 1999; Grimsby, 2000]; therefore, 150 of the probands were assayed by SSCA in this region only. The remaining 56 probands are of large adRP families and were tested for mutations in the complete RP1 gene by SSCA. Five previously reported RP1 mutations were identified by LMDIED in eight of the 206 adRP probands, or in about 4% of adRP tested (Table 2C).
CRX mutations were identified in two families diagnosed with adRP, RFS087, and UTAD341. The proband of RFS087 was heterozygous for the Arg41Gln mutation reported previously [Swain et al., 1997] in the proband of a family with autosomal dominant cone-rod dystrophy. Following identification of the mutation, detailed clinical evaluations in this and other family members indicated that the phenotype in RFS087 is more accurately described as a late-onset, slowly progressing, mild form of cone-rod dystrophy [Tzekov et al., 2000].
Another previously unreported mutation, Arg115Gln (G344A, CGG→CAG), was identified in the affected proband of UTAD341. This mutation was not identified in 50 normal control individuals or in any of the other 1,000 chromosomes tested. The arginine at this position is conserved among human, bovine, and murine Crx, and in human CRX and OTX-1, and OTX-2, two members of the same protein family [Chen et al., 1997]. Additional clinical evaluations are necessary to determine if the affected individuals of UTAD341 have a retinal degeneration phenotype similar to the family with the Arg41Gln mutation and to confirm the phenotype in the UTAD341 family as “retinitis pigmentosa.”
The retinal disorders in this category include those with clinical diagnoses of cone-rod dystrophy, cone degeneration, macular degeneration, and Bardet-Biedl disease (Table 3). Six previously reported mutations were identified as the cause of retinal degeneration in 12 families whose retinal disorder included cone degeneration. No rhodopsin mutations were identified in this category.
Homozygous or compound-heterozygous mutations were identified in three of 25 (~12%) probands with recessive or isolated LCA. In addition, a heterozygous CRX mutation was identified as the cause of autosomal dominant LCA in one family (Table 4). Therefore, the causative mutation was identified in approximately 15% (four of 27) of all subjects with a diagnosis of Leber congenital amaurosis.
Several apparently benign coding sequence variants were identified in the current survey (Table 5). In addition, frequencies were determined for the haplotypes of peripherin/RDS codon 304/310/338 polymorphisms. The respective frequencies for the G/A/G, C/G/A, and C/ A/A haplotypes were 0.77, 0.09, and 0.13. Two unique haplotypes, C/A/G and G/A/A, were also identified in one individual each.
The purpose of this project was to determine the prevalence of disease-causing mutations in five genes known to cause inherited retinopathy: rhodopsin, peripherin/RDS, RP1, CRX, and AIPL1. Mutation testing was performed in over 500 subjects in the broad diagnostic categories of either retinitis pigmentosa without apparent cone involvement or retinal degeneration with cone involvement. Patients were further classified by mode of inheritance and detailed clinical findings. Patients in the study were from families ascertained in the United States or Canada. In addition to providing useful clinical information for the patients and families involved in the study, mutation screening in this population offers several further benefits. First, this survey permits estimation of the relative prevalences (relative to each disease category) for mutations within each gene and for each observed mutation. This information is useful in determining the gene-specific and mutation-specific diagnostic and treatment needs within this population. Second, we report eight novel disease-causing mutations which contribute to the overall allelic variability at these loci. Third, we report several benign sequence variants within these genes, and their respective frequencies. These data will assist other diagnostic laboratories in distinguishing disease-causing mutations from benign variants. Finally, families and patients with specific mutations, identified in this study by confidential numbers, may be available for further study. Access to these subjects, with appropriate consent and maintenance of confidentiality, is available through the clinicians participating in the survey.
These data reinforce the heterogeneity of inherited retinopathies and the need for molecular diagnostics. For example, a single peripherin/RDS splice-site mutation, IVS2+3A>T, was identified as the cause of retinopathy in eight families; the phenotype in these families ranges from retinitis pigmentosa to macular degeneration. Further, although our assays may have missed some mutations, our data indicate that the inherited retinal degeneration genes that were not screened or have not yet been identified may account for approximately two-thirds of inherited retinopathy in these diagnostic categories. The combination of the heterogeneity of inherited retinal disorders and the developing studies involving gene-specific or mutation-specific treatments for these disorders makes identification of the remaining genes and accurate molecular diagnosis of these disorders even more important.
We express our sincere gratitude to all of the families who participated in our survey. We thank the following individuals for expert technical assistance: Noelle Agan, Melanie An-drews-Casal, Heather Ferguson, Alex Gannon, Sharon Guilford, Patricia Hamilton, Dennis Hoffman, Dianna Hughbanks-Wheaton, Jill Overseer, Jill Sawyer, and Myla Tuazon.
Contract grant sponsors: Foundation Fighting Blindness; George Gund Foundation; William Stamps Farish Fund; M.D. Anderson Foundation; John S. Dunn Research Foundation; Contract grant sponsor: National Eye Institute-National Institutes of Health; Contract grant numbers: EY07142; EY05235.
RHO – OMIM: 180380; GDB:120347; GenBank: U49742 (cDNA); HGMD: RHO
RDS – OMIM: 179605; GDB:118863; GenBank: NM_000322 (cDNA); HGMD: RDS
CRX – OMIM: 602225, 120970 (CORD2); GDB:9836957, GenBank: AF024711 (cDNA); HGMD: CRX
AIPL1 – OMIM: 604392, 604393 (LCA4); GDB:10029257; GenBank AF148864 (cDNA)
RP1 – OMIM: 603937, 180100; GDB:120352; GenBank; AF143222 (cDNA); HGMD: RP1; http://www.sph.uth.tmc.edu/RetNet/ (RetNet)