Central areolar choroidal dystrophy (CACD) is a rare inherited retinal disease which causes progressive profound loss of vision in patients during their 4th decade. We have identified a Northern Irish family with 19 affected individuals in three living generations. We have performed a total genome search and established linkage of CACD in this family to chromosome 17p (multipoint Zmax = 5.65 at D17S938). The genes for phosphatidylinositol transfer protein (PITPN), retinal guanylate cyclase (GUC2D), [beta]-arrestin 2 (ARRB2), pigment epithelium-derived factor (PEDF) and recoverin (RCV1) map to this region and are candidate genes for retinal disease. Analysis of the coding region of the PITPN gene failed to reveal any mutation in this family.
Central areolar choroidal dystrophy affects the posterior pole of the eye (1 ). Early lesions consist of a non-specific area of granular hyperpigmentation at the fovea (2 ). It is only after several decades that the pathognomonic zone of atrophy, affecting retina, retinal pigment epithelium and choriocapillaris, develops in the macular region of the eye (3 ). Patients usually become aware of visual symptoms in the 3rd to 4th decade (4 ) with the development of absolute central visual scotomas, causing severe visual disability by the seventh decade. CACD may show autosomal recessive or dominant inheritance (5 ). The purpose of this study was to identify linkage for this condition in a large Northern Irish family with this disorder.
Previously mutations in the peripherin-RDS gene (6 -8 ) have been reported in patients with CACD. However, in our family linkage to known retinal candidate gene loci including peripherin-RDS, ROM-1, rhodopsin and the [beta]-subunit of cyclic GMP had been excluded (9 ). Therefore a total genome search was undertaken using microsatellite markers and standard linkage methodology.
This study involves a family with CACD affecting members of four generations (Fig. 1 ). A systematic search for linkage was undertaken in all available individuals over the age of 14 with an affected parent. More than 160 highly polymorphic microsatellite polymorphisms were typed with negative results before a cluster of markers from chromosome 17p showed evidence of linkage (Fig. 2 a, Table 1 ), with a maximum two-point lod score of 4.7 at [theta] = 0.00 with D17S796. No recombinants were identified in members of generation II or III (in whom diagnosis was unequivocal) with D17S938 or D17S796, or with D17S1298 which was relatively uninformative in this family. A haplotype of markers could be tracked with the disease in this family (Fig. 1 ). The maximum multipoint lod score of 5.65 was obtained at D17S938 (Fig. 2 b).
The gene which causes CACD in this family shows linkage to markers on chromosome 17p (11 -13 ). Figure 1 shows haplotypes with markers D17S849, D17S5, D17S938, D17S796, D17S786, D17S804 and D17S520, and the 2-1-3-1-2-3-4 haplotype segregates with the disease in most members of this family. Analysis of recombination in affected members of generations II and III places the disease gene in the interval of approximately 16 cM between D17S5 and D17S520. Additional evidence of recombination in IV6 who has a 90% risk of being affected suggests a narrower localisation of the gene responsible for CACD to between D17S5 and D17S804. Linkage analysis implies that the gene is unlikely to lie close to D17S786, however, this is based on apparent double recombination within an interval of approximately 7 cM flanking this marker in II6. Unfortunately the absence of DNA from members of generation I or an unaffected member of generation II restricts the number of identifiable haplotypes when linkage is found. Neither of the D17S786 alleles inherited by II6 is shared with any of his siblings, and it appears likely that he received a new mutation with a 2 base increase in size of the 2 allele at this microsatellite marker within the common disease haplotype.
Table 1
No mutation associated with CACD was identified within the coding region of the PITPN gene. All PCR products were of the predicted size, except for those primed with PIT1F. This primed cDNA correctly but extension with Taq polymerase skipped a section of 115 bases of exceptionally high GC content which included the first seven codons of the gene. This region was therefore sequenced from PCR products employing PIT3F as the forward primer.
Previously two separate mutations in the retinal degeneration slow (RDS) gene which codes for peripherin-RDS have been implicated as a cause of CACD. One mutation at codon 172 was found in one Spanish (8 ) and two British families (7 ) and another at codon 142 in a Dutch cohort (6 ). Our family was not linked to the peripherin-RDS gene (9 ). This is not surprising as genetic heterogeneity has been observed in several ophthalmic disorders. Peripherin-RDS mutations have been identified in both central and peripheral retinal dystrophies including autosomal dominant retinitis pigmentosa (14 ), macular dystrophies (15 ), adult vitelliform dystrophy (15 ), butterfly type of pattern dystrophy (16 ) and retinitis punctata albescens (17 ). A single peripherin-RDS mutation may produce a widely varying phenotype within the same family (18 ). Several retinal diseases have been mapped recently to chromosome 17p. These include autosomal dominant retinitis pigmentosa (19 ,20 ), Leber's congenital amaurosis (21 ) and an autosomal dominant cone degeneration (22 ).
It may be that, as is the case with the peripherin-RDS gene, one or more of these retinal dystrophies results from mutations in the same gene at the 17p13 locus. Candidate genes which have been localised to this region include PITPN, GUC2D, ARRB2, PEDF and RCV1.
The human gene for PITPN which maps to 17p13.3 (23 ) shares sequence and functional homology with part of the Drosophila retinal degeneration B gene (rdgB) (24 ). Flies carrying the rdgB mutation undergo light-enhanced retinal degeneration. It has been proposed that the rdgB mutation may alter lipid metabolism (23 ). Pathological changes in Bruch's membrane which may be related to altered lipid metabolism have been found in patients with CACD (25 ,26 ). PITPN is therefore an excellent candidate gene for inherited retinal dystrophies.
GUC2D maps to 17p13.1 (27 ). Its product, retinal guanylate cyclase, is responsible for the intracellular second messenger 3',5'-cyclic guanosine monophosphate (cGMP) which regulates phototransduction in mammals. No mutation, however, has yet been detected in this gene in inherited retinal disease.
The ARRB2 gene maps to 17p13 (28 ). It is part of a multigene family which includes [beta]-arrestin 1 and arrestin (also known as S-antigen). The primary structure of [beta]-arrestin 1 and [beta]-arrestin 2 represents a family of proteins that regulate receptor coupling to G proteins. A mutation in the arrestin gene has recently been shown to cause Oguchi disease (an autosomal recessive form of congenital stationary night blindness) (29 ).
The gene for PEDF maps to 17p13.1 (30 ). It is a member of the serine protease inhibitor gene family and is highly expressed in young adult retinal pigment epithelium cells (31 ). The PEDF protein possesses both neurotrophic and neuronal-survival activities and so its loss may produce cell senescence (31 ).
Recoverin mediates the recovery of the dark current after photoactivation in the retina and has been implicated as the self-antigen responsible for cancer associated retinopathy. One of our young affected patients (IV6) showed recombination at D17S945 (data not shown) which is within 50 kb of the RCV1 locus (32 ) and maps close to D17S804. RCV1 is therefore a less probable candidate.
Although CACD is a rare disorder, discovering linkage to chromosome 17 in this family is important not only in relation to this disease but also as it may highlight genes which may be implicated in the pathology of age related macular degeneration. This is the major cause of blindness in our elderly population and an underlying aetiology has yet to be discovered.
We identified and examined a four generation family with CACD (Fig.1). Diagnosis was based on clinical ophthalmic examination including stereo fundus photography, fluorescein angiography and electrophysiological testing (33 ).Nineteen patients were found to be affected. Their findings, varying with disease state, agreed with Krill's previous classification of CACD (34 ). Individuals in generation IV showing signs suggestive of early disease were diagnosed as affected. All members of this generation were assigned to liability classes for linkage analysis as described below. The inheritance pattern in this family is autosomal dominant. One second generation member had also been seen in a reputable American retinal centre where an independent diagnosis of CACD had been made.
Approval for the study was obtained from the local Research Ethics Committee.
Microsatellite polymorphisms were typed by PCR amplification of genomic DNA using standard methods. Mapping primers were purchased from Research Genetics. One primer of each pair was 5' end-labelled with [gamma] [32P] ATP. The amplification products were separated by denaturing gel electrophoresis and alleles were detected by autoradiography. Alternative primers (GTGGTCCTTGAAATCCTGGAGC and ATGTGTGAATCTGGTTTGCCCC) were synthesised for typing of D17S796. Alleles were approximately 180 base pairs in length. D17S5 was typed by Southern blotting of TaqI digested genomic DNA and hybridisation with probe pYNZ22 (35 ).
The MLINK and LINKMAP programs of FASTLINK (version 3.0) were used for computerised two-point and multipoint analysis of data (36 ,37 ), assuming autosomal dominant inheritance with a disease allele frequency of 0.0001 and complete penetrance in members of generations I, II and III. Members of generation IV were assigned to one of four age-related liability classes ranging from 80-95% likelihood of the assigned disease status. This was achieved by varying the penetrance to allow for either reduced penetrance or the presence of phenocopies in young individuals.
Messenger RNA was extracted from whole blood using a Purescript kit (Gentra). Approximately 0.25 [mu]g of total RNA was primed with 0.5 [mu]g oligo(dT) in a reaction volume of 12 [mu]l. This was heated to 70oC for 10 min and chilled on ice. Reverse transcription was performed in a 20 [mu]l reaction mixture containing 50 mM Tris, pH 8.3; 75 mM KCl; 3 mM MgCl2; 10 mM DTT; 0.5 mM of each dNTP and 200 U of M-MLV reverse transcriptase and incubated at 37oC for 60 min.
The coding region of the PITPN gene was amplified from 2 [mu]l cDNA, in a 10 [mu]l reaction mixture, initially by PCR using PIT1F and PIT1R primers (94oC for 3 min; 10 cycles of 94oC for 30 s, 70oC for 1 min, 72oC for 1 min, followed by 30 cycles of 94oC for 30 s, 68oC for 1 min, and 72oC for 1 min, with a final extension of 72oC for 10 min) and PIT3F and PIT3R (96oC for 3 min; 20 cycles of 96oC for 1 min, 67oC for 1 min, 72oC for 1 min, followed by 25 cycles of 96oC for 1 min, 64oC for 1 min, 72oC for 1 min, with a final extension of 72oC for 10 min).
This was followed by PCR with nested primers PIT2F and PIT2R or PIT4F and PIT3R (96oC for 3 min; 40 cycles of 96oC for 1 min, 65oC for 1.5 min, 72oC for 1 min, with a final extension of 72oC for 10 min).
Both strands of PCR products were sequenced on an ABI Prism 377 DNA sequencer using dye-labelled terminators. All primers described below were used as sequencing primers with AmpliTaq® DNA polymerase in accordance with Perkin Elmer protocol P/N 402078.
Primers are numbered according to the sequence in GenBank accession number M73704. Primer PIT3F contains a single base mis-match nine bases from its 3' end to reduce the possibility of primer dimer formation.
This work was supported by grants from the MRC, British Council for Prevention of Blindness and the Department of Health and Social Services, Northern Ireland. We thank Dr Morna Murphy for assistance with sequencing.
Human Molecular Genetics
Pages
Introduction
Results
Discussion
Materials And Methods
Clinical description
Genetic markers
Linkage analysis
mRNA extraction and cDNA synthesis
RT-PCR and DNA sequencing
Acknowledgements
References
Marker
Lod score (Z) at recombination fraction ([theta])
Zmax[theta] *
0
0.01
0.05
0.1
0.2
0.3
D17S849
-[infinity]
0.13
1.26
1.51
1.37
0.95
1.52
0.12
D17S5
-[infinity]
-1.17
0.01
0.35
0.43
0.29
0.44
0.17
D17S1298
0.44
0.42
0.35
0.26
0.13
0.05
0.44
0.00
D17S938
3.84
3.78
3.50
3.12
2.33
1.49
3.84
0.00
D17S796
4.70
4.62
4.29
3.86
2.95
1.97
4.70
0.00
D17S786
1.76
3.69
4.04
3.88
3.15
2.16
4.04
0.05
D17S804
0.47
0.56
0.75
0.83
0.77
0.55
0.84
0.12
D17S520
-[infinity]
1.12
1.71
1.81
1.58
1.11
1.81
0.10
PIT1FCGGAGCAGAGCACGACGAAGAC94-116PIT1RCCTGTCACTTGGGAGGGTGGAG1088-1109PIT2FTGACCGTCAAGTTCAAGTGGTGGGG803-827PIT2RCAAAAAGTGCAGAGGGGAAAGGCG1032-1057PIT3FGCAGCCACCGAGCCGTGAAGCGAC193-216PIT3RCTTGCTTATGGATGAAGTTCTCCAC841-865PIT4FCGTTATTACAAATGAGTACATGAAAGAAG504-532PIT4RAGGGGTAAGCATTCCAGGCTTTC471-493
PIT1F
CGGAGCAGAGCACGACGAAGAC
94-116
PIT1R
CCTGTCACTTGGGAGGGTGGAG
1088-1109
PIT2F
TGACCGTCAAGTTCAAGTGGTGGGG
803-827
PIT2R
CAAAAAGTGCAGAGGGGAAAGGCG
1032-1057
PIT3F
GCAGCCACCGAGCCGTGAAGCGAC
193-216
PIT3R
CTTGCTTATGGATGAAGTTCTCCAC
841-865
PIT4F
CGTTATTACAAATGAGTACATGAAAGAAG
504-532
PIT4R
AGGGGTAAGCATTCCAGGCTTTC
471-493
REFERENCES
This page is maintained by OUP admin. Last updated Thu Oct 31 15:23:49 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996