Choroideremia Research Foundation Inc.
Affected males suffer progressive loss of vision (reduction of central vision, constriction of visual fields, night blindness) beginning at an early age, and the choroid and retina undergo complete atrophy. Heterozygous females show no visual defect but often show striking funduscopic changes such as irregular pigmentation and atrophy around the optic disc. Fully affected females have been reported (Fraser and Friedmann, 1967; Shapira and Sitney, 1943). These raise the usual questions of X-chromosomal berration, unfortunate lyonization in a heterozygote, homozygosity, etc. An extensive study in Holland was conducted by urstjens (1965). The term choroideremia, which is comparable to irideremia and means absence of choroid, is, strictly speaking, inappropriate since there is no congenital absence of the choroid. The condition is an abiotrophy beginning shortly after birth and rogressing gradually. Waardenburg favored the alternative designation 'tapetochoroidal dystrophy' (Pameyer et al., 1960). Harris and Miller (1968) observed visual impairment in a heterozygote in the family reported earlier by McCulloch and cCulloch (1948). In Finland, about 58 cases had been identified by 1980 (Forsius et al., 1980).
Almost all of them come from the northern part of the country. Karna (1986) traced 111 choroideremia patients and 188 carriers in 4 kindreds from northern Finland and 1 from the Savo district. A large proportion of both groups, 80 patients and 126 carriers, were examined hthalmologically. The largest of the kindreds, from the Salla area of Finland, had 80 cases and 146 carriers in 8 generations among the more than 3000 descendants of an ancestral female. The clinical picture proved unexpectedly variable with some males already virtually blind under age 30 years and others over age 50 who were symptom-free.
By history only 7 of 105 carriers had symptoms but 21 of 52 carriers examined had changes in the visual fields and defective dark adaptation. Decline in the latter function over a 3-year period was observed in 1 heterozygote and the changes, including funduscopic alterations, were most marked in older carriers. It was often difficult to be sure of the diagnosis before the person was 10 years of age, but the diagnosis was made in 2 boys aged 3 months and 10 months.
To test directly the question of whether the choroideremia gene is subject to inactivation, Carrel and Willard (1993) examined inactive X-chromosome expression of the CHM gene in a lymphoblastoid cell line derived from a female with a translocation that disrupted the gene. The normal X chromosome in this t(X;13) cell line was nonrandomly inactivated as shown by late-replication studies and by methylation analysis at the DXS255 and FMR1 (309550) loci. Using PCR of reverse transcribed RNA (RT-PCR) from this cell line, Carrel and Willard (1993) identified CHM transcripts that crossed the translocation breakpoint, indicating that CHM is expressed from the normal, inactivated X chromosome.
Quantitative comparison of RT-PCR products from the inactive X in the t(X;13) cell line with that from cell lines from a normal male and a 49,XXXXX female indicated that there was significant CHM transcription from the inactive X, at levels that were at least 50% of those seen in the active X. Confirming these results, CHM expression was also seen in RT-PCR in 3 active and 5 inactive human X-containing somatic cell hybrids. CHM is the first gene that is distal to the X-inactivation center on Xq, i.e., on the 'ancestral X chromosome,' to be shown to escape inactivation.
Close linkage of choroideremia with the Xg locus was excluded by Bell and McCulloch (1971), who found 3 recombinants out of 6. Nussbaum et al. (1985) found that the polymorphic DNA probe DXYS1, located at Xq13-q21, shows no recombination with choroideremia (lod = 5.78). This result indicates that, with 90% probability, choroideremia maps within 9 cM of DXYS1. Lesko et al. (1985) had a lod score of 12 for 0.0 recombination with DXYS1. Gal et al. (1986) suggested the following order: Xcen DXYS1, DXS3, TCD, DXS11, and Xqter. DXS3 maps to Xq21.3-q22 and DXS11 to Xq24-q26 (HGM8). Sankila et al. (1986, 1987) urged a historical-genealogical approach to linkage analysis. In studies of isolated populations, they found that all of 36 patients and 48 carriers with choroideremia in Finland had the same haplotype: TCD/DXYS1, 11 kb/DXYS12, 1.6 kb. The DXYS1 locus and the DXYS12 locus are located at Xq13-q21 and Xq13-q22, respectively. Given that at least 105 female meioses transmitting TCD occurred in these kindreds since 1650, extremely close linkage between the 3 loci is suggested. This method is comparable to the use of recombinant inbred strains in mice (Bailey, 1971; Taylor, 1978) and to homozygosity mapping as discussed by Lander and Botstein (1987). Also, it is, of course, the same phenomenon as linkage disequilibrium. Sankila et al. (1987) also reported haplotype data on Finnish TCD using multiple DNA probes and presented evidence on the order of the several marker loci and TCD. Haplotyping suggested that the large northern Finnish choroideremia pedigrees, comprising over 80 living patients representing more than a fifth of all TCD patients described worldwide, carry the same mutation (Sankila et al., 1989). Using DNA markers in 3 Danish families, Schwartz et al. (1986) provided further evidence for assignment for the choroideremia locus to Xq13-q21. Hodgson et al. (1987) described a family in which an X-chromosome deletion was segregating with choroideremia. The affected male was also mentally retarded. The loss of 2 RFLPs in the affected male indicated the localization of the choroideremia locus to Xq13-q21 and placed the loci for anhidrotic ectodermal dysplasia (305100) and the X-linked immunodeficiencies (e.g., 300300) outside this region. Lesko et al. (1987) found recombination frequencies of 0 to 4% between TCD and 5 markers located in Xq13-q22. Uhlhaas et al. (1987) also provided data on linkage to multiple DNA markers. MacDonald et al. (1987) found a maximum lod score of 3.98 at theta = 0.14 for linkage with DXS3 and a total maximum lod score from all studies of 5.23 at theta = 0.05 for DXYS1. These findings represented looser linkage than had previously been reported. DXS3 maps to Xq21.3-q22. Multipoint linkage analysis by Sankila et al. (1989) placed TCD distal to PGK (311800) and DXS72, very close to DXYS1 and DXYS5 (maximum lod = 24 at theta = 0) and proximal to DXYS4. In a study of 14 families, Wright et al. (1990) found linkage to 3 markers on Xq21, giving a 4 point lod score of 8.25. No evidence of submicroscopic deletion was observed using 2 DNA markers thought to lie within 1 Mb of the TCD gene. Schwartz et al. (1987) mapped the TCD locus by demonstrating deletion of 3 XY-probes in 2 males with choroideremia and X-chromosomal deletion.
Nussbaum et al. (1987) studied 2 families in which males with choroideremia also had mental retardation and deafness. In 1 family an interstitial deletion in Xq21 was visible by cytogenetic analysis, and 2 DNA markers, DXYS1 and DXS72, were deleted. In the second family, an interstitial deletion was suspected on phenotypic grounds but could not be confirmed by high-resolution cytogenetic analysis. The authors used phenol-enhanced reassociation of 48,XXXX DNA in competition with excess DNA from the second family to generate a library of cloned DNA enriched for sequences that might be deleted. Two of the first 83 sequences characterized from the library were found to be deleted in probands from both families. Isolation of these sequences proved that the second family indeed carried a submicroscopic deletion and provided a starting point for identifying overlapping genomic sequences that span the deletion and may contain exons from the choroideremia locus. Using subtractive hybridization, Lesko et al. (1987) isolated and characterized the sequences deleted from an individual with choroideremia and a visible deletion at Xq21. In 1 patient with no visible deletion, submicroscopic deletion was indicated by the absence of 2 single-copy sequences. Using these sequences as probes for in situ hybridization, Lesko et al. (1987) localized the choroideremia gene to Xq21. In 2 of 8 unrelated male patients with choroideremia, Cremers et al. (1987) found deletion of DXS165, which maps to Xq12-q21.3. Two other closely linked and probably flanking TCD markers, DXYS1 and DXS72, were not deleted, which may indicate that the physical distance between DXS165 and TCD is small. Schwartz et al. (1988) used DNA from 2 unrelated males who had choroideremia and an interstitial deletion on the proximal long arm of the X chromosome. In one the deletion was Xq21.1-q21.33; in the other the deletion was Xq21.2-q21.31. By Southern blot analysis, the authors mapped 10 different polymorphic DNA loci to the deletion and to the TCD locus. One probe was shown to cover one of the breakpoints of the smaller deletion.
In his Atlas of the Fundus Oculi, Wilmer (1934) showed the fundus of a 35-year-old man with 'choroidal sclerosis' whose maternal grandfather was also affected. Furthermore, 2 brothers and the maternal grandfather of the proband's maternal grandfather were also affected, i.e., the proband had inherited the disorder from his great-great-grandfather through the intermediacy of a carrier mother and great-grandmother. When I followed up on this family by letter in 1962, no further information was available. Stankovic (1958) reported a similar family, which was of further interest because female carriers showed partial expression. Sorsby (in Franceschetti et al., 1963) was of the opinion that the cases reported by Sorsby and Savory (1956) as X-linked choroidal sclerosis were instances of choroideremia. Krill and Archer (1971) were of the same view. From study of affected members of 1 kindred, Shapiro and Gorlin (1974) concluded that choroidal sclerosis is a stage in the evolution of choroideremia.
Siu et al. (1988, 1990) found a de novo balanced translocation, t(X;13)(q21.2;p12), in a patient with choroideremia. They found evidence that the DXYS1 locus is distal to the choroideremia gene; the derivative chromosome 13 carried a RFLP allele of this locus. Using chromosome walking and jumping techniques in a study of 4 deletions associated with choroideremia and a de novo X/13 translocation in a female with choroideremia, Cremers et al. (1989) narrowed the assignment of the TCD gene, or part of it, to a DNA segment of only 15 to 20 kb. Cremers et al. (1989) reported another case of a female with TCD and a de novo X-autosome translocation. Merry et al. (1990) also studied the X;13 translocation described by Siu et al. (1988). Cremers et al. (1990) identified new DNA markers around the TCD locus which they used to define the minimal region of overlap from 4 deletions found in male patients with TCD and to isolate a 45-kb genomic DNA segment corresponding to this region of overlap. cDNA clones from a human retinal library were isolated using an evolutionarily conserved sequence from this DNA segment as a probe. cDNA subclones detected a transcript of 5,400 basepairs in choroid, retinal pigment epithelium, and other cells. The consensus cDNA of approximately 4.5 kb contained an open reading frame of 948 bp encoding a polypeptide of 316 amino acids. This open reading frame was partially deleted or disrupted in 9 male TCD patients with deletions and in a female patient with a balanced translocation involving the Xq21 band, strongly arguing for a causal role of this gene in TCD. Fodor et al. (1991) pointed to homology between the predicted sequence of this protein and a protein involved in GTP metabolism, p25A-GDI. Cremers et al. (1990) found that deletions in TCD cases varied in size from 45 kb to several megabases.
Merry et al. (1992) isolated cDNAs from a human retinal library with a genomic probe located at the X-chromosomal breakpoint in a female with choroideremia and an X;13 translocation. This cDNA spanned the breakpoint in the translocation female and was deleted in males with choroideremia as part of a complex phenotype including mental retardation and deafness (303110). However, this cDNA detected no alterations in the DNA of 34 maleswith isolated choroideremia. Nonetheless, the cDNA detected reduced or absent levels of mRNA in 25 of 34 unrelated male patients with an apparently intact gene. The gene was very similar but not identical to that reported by Cremers et al. (1990).
Seabra et al. (1992) purified component A of RAB geranylgeranyl transferase, a single 95-kD polypeptide. The holoenzyme, which consists of components A and B (179080), attaches (3)H-geranylgeranyl to cysteines in 2 GTP-binding proteins, RAB3A (179490) and RAB1A (179508). The reaction is abolished when both cysteines in the COOH-terminal cys-cys sequence of RAB1A are mutated to serines. Six peptides from rat component A showed striking similarity to the products of the gene defective in choroideremia. The choroideremia protein resembles RAB3A GDI, which binds RAB3A. Seabra et al. (1992) suggested that component A binds conserved sequences in RAB and that component B transfers geranylgeranyl. A defect in this reaction may cause choroideremia. Seabra et al. (1993) established this to be the case by demonstrating that lymphoblasts from choroideremia subjects have a marked deficiency in the activity of component A, but not component B, of RAB GG transferase. The deficiency was more pronounced when the substrate was RAB3A, a synaptic vesicle protein, than it was when the substrate was RAB1A, a protein of the endoplasmic reticulum. Their studies suggested the existence of multiple component A proteins, one of which is missing in choroideremia. The multiplicity and functional redundancy of component A genes creates a situation in which defects in one of them might cause a degenerative disease of the organ in which that particular form of component A is most essential. Seabra et al. (1993) suggested that it will be of interest to isolate the genes encoding the other component A proteins and to determine whether mutations in these genes underlie degenerative diseases of the nervous system or other organs.
Isolation and characterization of the complete open reading frame of the CHM gene and its exon-intron structure was reported by van Bokhoven et al. (1994). The CHM gene encodes a protein of 653 amino acids. The open reading frame comprises 15 exons, spanning at least 150 kb, and there may be an additional exon corresponding to the 5-prime noncoding region of the gene. Van Bokhoven et al. (1994) demonstrated that the breakpoint on the X chromosome in a CHM female with an X;7 translocation lay between exons 3 and 4. In a companion paper, van Bokhoven et al. (1994) reviewed the spectrum of mutations in the CHM gene in patients from 15 Danish and Swedish families. They noted that all CHM gene mutations detected to that time gave rise to the introduction of a premature stop codon.
Waldherr et al. (1993) presented evidence based on sequence that MRS6 of S. cerevisiae is the yeast homolog of the CHM gene. Rab escort protein-1 (REP1) was the designation used by van den Hurk et al. (1997) in a review of mutations involving this gene in choroideremia. In 18 patients, REP1 gene deletions of different sizes were found. Two females suffering from CHM were reported to have translocations that disrupted the REP1 gene. In 22 patients, small mutations were identified. The authors noted that these were all nonsense, frameshift, or splice site mutations; with one possible exception, missense mutations were not found.
A gene targeting approach was used by van den Hurk et al. (1997) to disrupt the mouse chm/rep-1 gene. Chimeric males transmitted the mutated gene to their carrier daughters but, surprisingly, these heterozygous females had neither affected male nor carrier female offspring. The targeted rep-1 allele was detectable, however, in male as well as female blastocyst stage embryos isolated from a heterozygous mother. Thus, disruption of the rep-1 gene gives rise to lethality in male embryos; in females embryos, it is lethal only if the mutation is of maternal origin. This observation could be explained by preferential inactivation of the paternal X chromosome in murine extraembryonic membranes, suggesting that expression of rep-1 is essential in these tissues. In both heterozygous females and chimeras, the rep-1 mutation caused photoreceptor cell degeneration. Consequently, conditional rescue of the embryonic lethal phenotype of the rep-1 mutation may provide a faithful mouse model for choroideremia.
The finding of van den Hurk et al. (1997) and that of Skuse et al. (1997), who found evidence of an imprinted X-linked locus affecting cognitive function (CGF1; 300082), expanded the list of imprinted X-linked genes from 1 (XIST; 314670) to 3. Naumova et al. (1998) analyzed the transmission of maternal alleles at loci spanning the length of the X chromosome in 47 normal, genetic disease-free families. They found a significant deviation from the expected mendelian 1:1 ratio of grandparental:grandmaternal alleles at loci in Xp21.1-p11.4. The distortion in the inheritance ratio was found only among male offspring and was manifested as a strong bias in favor of inheritance of the alleles of the maternal grandfather. No evidence for significant heterogeneity among the families was found, which implies that the major determinant involved in the generation of the nonmendelian ratio is epigenetic. The analysis of recombinant chromosomes inherited by male offspring indicated that an 11.6-cM interval on the short arm of the X chromosome, bounded by DXS538 and XS7, contains an imprinted gene that affects the survival of male embryos.
Sankila et al. (1992) described a point mutation that is responsible for choroideremia in the large Salla pedigree from northeastern Finnish Lapland that accounts for one-fifth of the world's choroideremia patients. They showed that the mutation is unique in that it is not responsible for choroideremia in any of the other Finnish pedigrees. The mutation was detected by single-strand conformation polymorphism (SSCP) analysis with subsequent sequencing of the relevant DNA segment. Sequencing showed insertion of a T within the splice donor site of the intron downstream of exon C, changing the normal sequence of AGgtaag to AGgttaag. A new restriction site for MseI was created by the mutation, thus permitting screening. Although the CHM gene is mainly expressed in the retina, choroid, and retinal pigment epithelium, low levels of transcripts are also found in lymphoblasts by means of polymerase chain reaction (PCR). This illegitimate transcription provides a convenient means of screening and analyzing the transcript. Lymphoblast-derived mRNA from a patient with what the authors referred to as the CHM*SAL mutation showed 2 aberrantly spliced mRNAs and no normal transcript.
Using PCR-SSCP analysis and direct DNA sequencing, van den Hurk et al. (1992) detected and characterized different point mutations in 5 patients with CHM. Each of these mutations introduced a termination codon into the open reading frame of the CHM candidate gene, thereby predicting a distinct truncated protein product. One of the mutations was a TCC-to-TGA change in codon 116 in exon B3 leading to the change of a serine codon to a stop codon. Codons 116 and 117 in exon B3 are TCC (ser) and AGG (arg). The mutation in this case involved the replacement of CC by G, so that codons 116 and 117 became TGA (stop) and GGG.
A second mutation found by van den Hurk et al. (1992) involved a C-to-A transversion in exon B4 converting serine (TCA) to a stop codon (TAA) at position 158.
Athird mutation found by van den Hurk et al. (1992) was a G-to-T transversion changing codon 154 from glutamic acid (GAG) to stop (TAG).
The mutation identified in one case of choroideremia by van den Hurk et al. (1992) consisted of deletion of 1 bp, a G, converting glycine GGA) to glutamic acid (GAA) at position 146 and causing a frameshift with premature termination at codon 159.
One of the 5 mutations identified by van den Hurk et al. (1992) by PCR-SSCP and direct DNA sequencing involved a deletion of 4 nucleotides, changing codons 191-193 from TTT GTT CCA to TTC CAT ATA. A frameshift with premature termination at codon 198 resulted.
In 12 Danish families with choroideremia, Schwartz et al. (1993) identified a point mutation in 6 patients by use of PCR, SSCP analysis, and direct DNA sequencing. One of these was a TGC (cys) to TGA (stop) mutation in codon 162.
In a mutation screening of patients from 15 Danish and Swedish families with choroideremia, van Bokhoven et al. (1994) found mainly deletions or insertions. There were, however, 4 single nucleotide substitutions of which 2 were missense mutations and 2 were splice errors. One of the missense mutations (in patient LN) was a C-to-T transition at nucleotide 907 resulting in a change of arg294 to a termination codon.
In patient the with choroideremia, van Bokhoven et al. (1994) found a C-to-A transversion at nucleotide 1527 of the CHM gene resulting in a substitution of a termination codon for cys500.