Download A ninth locus (RP18) for autosomal dominant retinitis pigmentosa

 1996 Oxford University Press
Human Molecular Genetics, 1996, Vol. 5, No. 8 1193–1197
A ninth locus (RP18) for autosomal dominant retinitis
pigmentosa maps in the pericentromeric region of
chromosome 1
Su Ying Xu1, Marianne Schwartz2, Thomas Rosenberg3 and Andreas Gal1,*
1Institut
für Humangenetik, Universitäts-Krankenhaus Eppendorf, Butenfeld 32, D-22529 Hamburg, Germany,
of Clinical Genetics, Department of Pediatrics, Rigshospitalet, Copenhagen, Denmark and 3National
Eye Clinic for the Visually Impaired, Hellerup, Denmark
2Section
Received April 16, 1996; Revised and Accepted May 24, 1996
We studied a large Danish family of seven generations
in which autosomal dominant retinitis pigmentosa
(adRP), a heterogeneous genetic form of retinal
dystrophy, was segregating. After linkage had been
excluded to all known adRP loci on chromosomes 3q,
6p, 7p, 7q, 8q, 17p, 17q and 19q, a genome screening
was performed. Positive lod scores suggestive of
linkage with values ranging between Z = 1.58–5.36 at θ
= 0.04–0.20 were obtained for eight loci on proximal 1p
and 1q. Close linkage without recombination and a
maximum lod score of 7.22 at θ = 0.00 was found
between the adRP locus (RP18) in this family and
D1S498 which is on 1q very near the centromere.
Analysis of multiply informative meioses suggests that
in this family D1S534 and D1S305 flank RP18 in interval
1p13–q23. No linkage has been found to loci from this
chromosomal region in six other medium sized adRP
families in which the disease locus has been excluded
from all known chromosomal regions harbouring an
adRP gene or locus suggesting that there is (at least)
one further adRP locus to be mapped in the future.
INTRODUCTION
Retinitis pigmentosa (RP) designates a frequent but heterogeneous
genetic form of retinal dystrophy affecting ∼1 in 3000 people and
responsible for the visual handicap of ∼1.5 million individuals
worldwide (for references and review, see 1,2). The condition can
be inherited as an autosomal dominant, autosomal recessive or
X-linked trait, whereas a considerable portion are simplex cases in
which the pattern of inheritance remains undetermined. Typical
clinical symptoms of RP include night blindness, reduced visual
acuity and progressive loss of peripheral vision leading to blindness
usually after the fourth decade. Ophthalmological examinations
show constricted visual fields, attenuated vessels and abnormal
accumulation of pigment in the retina.
*To whom correspondence should be addressed
Autosomal dominant retinitis pigmentosa (adRP) accounts for
∼20% of all RP cases and is characterized by significant allelic
and non-allelic heterogeneity (for references and review, see 1,3).
Eight loci for adRP have been mapped genetically to date on
chromosomes 3q, 6p, 7p, 7q, 8q, 17p, 17q and 19q (for review, see
4 and references therein, and 5). Of the eight corresponding adRP
genes, those on 3q and 6p encode rhodopsin (for review, see 1 and
references therein) and peripherin (for review, see 6 and
references therein), respectively. Here we report the mapping of
a further adRP locus (RP18) on the pericentric region of human
chromosome 1.
RESULTS
We studied a large Danish adRP family of seven generations (Fig.
1). Clinical diagnosis was based on a history of night blindness,
typical ocular fundus with peripheral bone spicule formation and
severe constriction of the retinal arterioles, and progressive visual
field defects beginning as mid-peripheral ring scotomas. Disease
manifestation was rather uniform, with both sexes equally
affected. Twelve subjects had one or several eye examinations
performed at the National Eye Clinic, including full field
electroretinography (ERG) and dark adaptometry a.m. Goldmann
Weekers. Many of the patients claimed that night blindness was
present ‘from birth’ on. Nevertheless, pathological dark
adaptation starting with a prolonged mesopic phase and elevation
of the fully dark-adapted threshold did not occur until the end of
the first decade. Atrophy of the retinal pigment epithelium with
typical bone spicule hyperpigmentation was also observed from
the second half of the first decennium. From the beginning of the
second decade, the ERG showed abolished rod responses and
strongly attenuated cone responses with prolonged implicit times.
ERG was extinguished from the beginning of the third decade if
measured by conventional techniques. Visual field defects
progressed during the third and fourth decades, leaving the central
area relatively spared. Accordingly, visual acuity remained above
0.3 in the majority of patients for the first three decades. Most of
the affected subjects had myopia in the range of –2 to –4 diopters,
but a few cases with slight hypermetropia or high myopia were
also encountered. Some of the patients developed a moderate
1194 Human Molecular Genetics, 1996, Vol. 5, No. 8
Figure 1. Pedigree of RP110. Genotypes detected at the four loci given are arranged according to most likely haplotypes. The disease phenotype of individuals VI.7,
VI.8 and VII.1 is unknown. Arrowheads mark the two meiotic breakpoints used to map the disease locus.
photophobia; glaucoma was diagnosed in three patients in their
sixties. Complicating cataracts were observed as fine subcapsular
opacifications which did not require lensectomy. Some of the
older patients were totally blind from the age of 80.
After linkage had been excluded to all known adRP loci (data
not shown), a genome screening was performed. Over 150 DNA
polymorphisms of the CA repeat type were analysed. Positive lod
scores suggestive of linkage were detected with two reference loci
of the Mappairs set (Research Genetics, Inc., Huntsville, AL,
USA; Screening Set/Weber Version 4a), D1S534 on 1p and
D1S1167 on 1q (Table 1 and Fig. 2). In order to localize the
disease locus (RP18) more precisely, the family was genotyped
for a total of eight additional ‘anonymous’ DNA polymorphisms
from this chromosomal region (7). This analysis provided
significant positive lod scores, Zmax ranging between 2.19 and
7.22, for five loci on 1q (Table 1). Close linkage without
recombination and a maximum lod score of 7.22 at θ = 0.00 was
found in this family between RP18 and D1S498, which is on 1q
very near the centromere (ref. 8 and Fig. 2). One or more
recombinants were seen with all other loci tested. Figure 1 shows
the genotype of all family members at D1S498 as well as at
D1S534 and D1S305 which are, respectively, the closest proximal
and distal markers to D1S498 analysed. Genotypes are arranged
according to the most likely haplotypes. Two recombinants define
the interval in which RP18 maps in this family. Individual V.10
is affected by RP and recombinant for D1S534 but non-recombinant for D1S498 and D1S305, which places RP18 centromeric to
D1S534 on 1p or on 1q. Individual VI.6, too, is affected by RP and
is recombinant for D1S305 but not for D1S498 and D1S534,
which maps RP18 centromeric to D1S305 on 1q or on 1p. All data
together suggest that in this family D1S534 and D1S305 flank
RP18 in interval 1p13–q23 (Fig. 2).
In this chromosomal region, there are two genes of particular
interest regarding RP, which encode two different G protein
α-subunits, GNAT2 (guanine nucleotide-binding protein,
α-transducing, polypeptide-2; ref. 9) and GNAI3 (guanine
nucleotide-binding protein, α-inhibiting, polypeptide-3; ref. 10).
Both genes have been mapped to 1p13 by in situ hybridization
(11) with human cDNA probes. However, no data were available
on the position of these two genes on the genetic map of
chromosome 1. Therefore, and using the genomic sequence
published earlier, we have designed oligonucleotide primers for
PCR amplification and single strand conformation polymorphism (SSCP) analysis to detect DNA polymorphisms in the 5′ end
of GNAI3 suitable for linkage analysis. Two different two-allele
restriction fragment length polymorphisms (RFLPs) have been
identified which are both due to single base substitutions at
nucleotide positions –72 and –61 (numbering as given in ref. 10)
and can be detected easily by restriction digestion analysis of PCR
products by BanI and BpuI, respectively (see Materials and
Methods).
1195
Human Acids
Molecular
Genetics,
Vol.No.
5, No.
Nucleic
Research,
1994,1996,
Vol. 22,
1 8 1195
Table 1. Two-point lod scores between the adRP locus (RP18) and 11 loci from chromosome 1
Physical locationa
Loci
Recombination faction (θ)
Zmax
θ (zmax)
Cytogenetic
Mb
0.000
0.01
0.05
0.10
0.20
0.30
0.40
D1S188
–
103.73
–∞
–8.59
–4.03
–2.33
–1.06
–0.68
–0.45
0.00
0.50
GNAI3
p13
–
–∞
–0.95
0.31
0.71
0.78
0.52
0.17
0.81
0.16
D1S534
–
112.36
–∞
5.11
5.35
5.05
4.08
2.85
1.42
5.36
0.04
D1S498
–
–
7.22
7.09
6.56
5.88
4.43
2.85
1.20
7.22
0.00
D1S305
–
139.56
–∞
2.18
3.23
3.36
2.92
2.07
0.98
3.37
0.09
D1S176
–
144.20
–∞
1.76
2.73
2.78
2.25
1.40
0.49
2.81
0.08
SPTA1
q21
146.31
–∞
0.45
1.60
1.86
1.74
1.29
0.67
1.88
0.12
CRP
q21–q23
146.58
–∞
1.05
1.54
1.57
1.27
0.82
0.36
1.58
0.08
D1S1167
–
–
–∞
2.02
3.64
3.93
3.51
2.57
1.30
3.93
0.10
D1S104
q21–q23
150.71
–∞
0.79
1.92
2.18
2.02
1.52
0.83
2.19
0.12
D1S196
–
152.23
–∞
–1.28
–0.03
0.38
0.57
0.48
0.27
0.57
0.20
1p
1cen
1q
aData
taken from references 8, 9 and 11.
We have genotyped members of the family for the two
above-mentioned RFLPs. In order to increase informativity,
individual genotypes obtained for each of the polymorphisms
were combined into haplotypes which were then used in linkage
analysis. Haplotypes are shown in Figure 1. Clearly, haplotype 2
is co-segregating with the disease phenotype in this family.
However, individual V.10, an affected male, inherited haplotype
3 from his affected mother, i.e. he is recombinant for GNAI3. This
finding formally excludes GNAI3 as a candidate gene for the
adRP segregating in this family. As V.10 is non-recombinant for
D1S498 and D1S305, these data place GNAI3 proximal to both
marker loci. Individual V.13, too, is recombinant for the GNAI3
versus RP18 linkage relationship. As this person is
non-recombinant for D1S534 and for D1S498, all these data taken
together map GNAI3 telomeric to D1S534 on 1p. Pairwise lod
scores between RP18 on one hand and D1S534 and GNAI3 on the
other hand reach their maximum at, respectively, θ = 0.04 and
θ = 0.20 (Table 1). As GNAI3 recombines with D1S534 at a
frequency of ∼0.15 (Z = 1.152 in this family; data not shown),
results of the two-point analyses are in line with the suggested
map position of GNAI3.
DISCUSSION
The data presented in this communication strongly suggest that a
gene implicated in autosomal dominant RP maps in the
pericentromeric region of chromosome 1. On a recent integrated
map including physical and genetic data (8), D1S534 and D1S305,
the two loci flanking RP18, are placed <30 Mb apart, which defines
the maximal size of the critical region and corresponds to about
one-ninth of the total length of chromosome 1.
RP18 represents the ninth locus for the autosomal dominant
form of RP assigned by linkage analysis and confirms the extreme
non-allelic genetic heterogeneity of the trait. It is not yet known
what proportion of patients with RP have mutations in the
corresponding gene on chromosome 1. Both linkage studies on
Figure 2. Schematic presentation of the order of marker loci typed and the most
likely position of the adRP locus (RP18) mapped in the pericentromeric region
of chromosome 1.
families and screening of larger collections of individual patients
suggest that mutations of rhodopsin and peripherin account for,
respectively, ∼25 and 5% of all adRP cases (for review see,
1196 Human Molecular Genetics, 1996, Vol. 5, No. 8
respectively, 1 and 6 and references therein). In contrast, the
majority of the remaining seven adRP loci have each been
assigned in a single large family or have been detected in a few
families (for references and review, see 1,4,12). We have found
no linkage to the corresponding marker loci on chromosome 1 in
six other medium sized adRP families in which the disease locus
has been excluded from all known chromosomal regions
harbouring an adRP gene or locus (unpublished results).
Therefore, it is likely that there is—at least—one further adRP
locus to be mapped in the future.
We have found two recombinants between GNAI3 and RP18
which practically excludes the possibility that a mutation in GNAI3
is the primary genetic cause of the disease in this family. As both
recombinants were in individuals affected by RP, phenotypic
misclassification due to reduced penetrance can also be rejected.
GNAT2 encodes the α-subunit of a retina-specific transducin
expressed exclusively in cone photoreceptors (9). Transducin is
a heterotrimeric (α,β,γ) protein whereas the α-subunit defines
both the specificity for interaction with the receptor and the
effector function. Stimulated receptor activates transducin/the
α-subunit by catalysing the exchange of the α-bound GDP for
GTP and the dissociation of the α-subunit from the β- and
γ-subunits (for review, see 11 and 13 and references therein). As
RP affects primarily the rod photoreceptors, we consider that the
probability is low that the disease segregating in the Danish
family is due to a mutation in GNAT2, although formal proof is
missing. Independent support for this assumption comes from the
observation that mammalian G protein α-subunit genes belong to
a multigene family thought to be generated by a series of genome
duplications during evolution. A recent model hypothesizes that
GNAT2 and GNAI3 evolved from a single ancestral gene by a
subsequent tandem duplication (11), i.e. that the two genes are
physically linked. A similar pair of G protein α-subunit genes
(GNAT1 and GNAI2) has been found on chromosome 3p21 (11).
Given the fact that in the Danish family GNAI3 maps ∼20 cM
from RP18 and under the assumption of close physical proximity
of GNAI3 and GNAT2 in 1p13 on one hand and in view of the
expression of GNAT2 in cone photoreceptor cells on the other
hand, our data suggest that GNAT2 is a very unlikely candidate for
the gene mutated and responsible for the adRP in the Danish
family described here.
Molecular genetic and biological analysis of other candidate
genes from this chromosomal region (8), e.g. KCNA3 (potassium
voltage-gated channel) or NGFB (nerve growth factor) will be
carried out to prove or disprove their role in the pathogenesis of
this form of retinal dystrophy.
MATERIALS AND METHODS
Family RP110
Medical files on affected members of the seven-generation
kindred studied here (Fig. 1) were retrieved from the Danish
Retinitis Pigmentosa Register (14). Autosomal dominant
inheritance was documented by several instances of male-to-male
transmission. Penetrance seems to be complete as no case of a
skipped generation was observed.
The whole study was carried out in accordance with the
Helsinki Declaration II and approved in advance both by the
Danish National Register Authority (Registertilsynet) and the
Scientific Ethical Committee (KF 01–226/93). Informed, written
consent was obtained from all participants.
Molecular genetic analysis
Blood samples for linkage analyses were obtained from 20
patients, eight unaffected relatives older than 21 and therefore
considered as not affected, three unaffected relatives below the
age of 20 and thus classified with disease phenotype unknown,
and six spouses. Experimental procedures for genotyping were
carried out following standard protocols. Detection of allelic
fragments was done by silver staining essentially as described
elsewhere (15). Linkage analysis was performed with the Liped
program (16).
For PCR amplification of a 293 bp DNA segment from the 5′ end
of the GNAI3 gene, oligonucleotide primers GNAI3-5′F
(5′-GCCA-CCGCCCAGCAATAGAC-3′)
and
GNAI3-5′R
(5′-CCCGACCAGACCCCGTTCAG-3′) were designed based on
published sequence data (10). SSCP analysis and direct sequencing
of PCR products of aberrant electrophoretic mobility were carried
out as described elsewhere (15). Single-base substitutions were
found at nucleotide positions –72 (T→C) and –61 (C→T). The T/C
and C/T changes result in gain of a recognition site for BanI and
BpuI and allelic cleavage products of 293/273 + 20 and 293/260 +
33 bp, respectively. Allele frequencies for linkage analysis were
calculated from the genotypes of eight spouses in the family.
ACKNOWLEDGEMENTS
The experimental work described here was financially supported
by the Deutsche Forschungsgemeinschaft. The Danish Retinitis
Pigmentosa Register received grants from the Danish Association
of the Blind (DBS) whereas blood sampling was financially
supported by Cykelhandler P. Th. Rasmussens og Hustru Alma
Rasmussens Mindelegat.
REFERENCES
1. Gal, A., Apfelstedt-Sylla, E., Janecke, A.R. and Zrenner, E. (1996)
Rhodopsin mutations in inherited retinal dystrophies and dysfunction. In
Osborne, N.N. and Chader, G.J. (eds), Progress in Retinal and Eye Research.
Pergamon Press, in press.
2. Haim, M., Holm, N.V. and Rosenberg, T. (1992) Prevalence of retinitis
pigmentosa and allied disorders in Denmark. I. Main results. Acta Ophthalmol., 70, 178–186.
3. Haim, M. (1993) Retinitis pigmentosa: problems associated with genetic
classification. Clin. Genet., 44, 62–70.
4. Dryja, T.P. and Li, T. (1995) Molecular genetics of retinitis pigmentosa.Hum.
Mol. Genet., 4, 1739–1743.
5. Bardien, S., Ebenezer, N., Greenberg, J., Inglehearn, C.F., Bartmann, L.,
Goliath, R., Beighton, P., Ramesar, R. and Bhattacharya, S.S. (1995) An
eighth locus for autosomal dominant retinitis pigmentosa is linked to
chromosome 17q. Hum. Mol. Genet., 4, 1459–1462.
6. Molday, R.S. (1994) Peripherin/rds and rom-1: molecular properties and role
in photoreceptor cell degeneration. In Osborne, N.N. and Chader, G.J. (eds),
Progress in Retinal and Eye Research. Pergamon Press, pp. 271–298.
7. Gyapay, G., Morissette, J., Vignal, A., Dib, C., Fizames, C., Millasseau, P.,
Marc, S., Bernardi, G., Lathrop, M. and Weissenbach, J. (1994) The 1993–94
Généthon human genetic linkage map. Nature Genet., 7, 246–339.
8. Forabosco, P., Collins, A. and Morton, N.E. (1995) Integration of gene maps:
updating chromosome 1. Ann. Hum. Genet., 59, 291–305.
9. Morris, T.A. and Fong, S-L. (1993) Characterization of the gene encoding
human cone transducin α-subunit (GNAT2). Genomics, 17, 442–448.
10. Itoh, H., Toyama, R., Kozase, T., Tsukamoto, T. Matsouka, M. and Kaziro, Y.
(1986) Presence of three distinct molecular species of Gi protein α subunit.
Structure of rat cDNAs and human genomic DNAs. J. Biol. Chem., 263,
6656–6664.
1197
Human Acids
Molecular
Genetics,
Vol.No.
5, No.
Nucleic
Research,
1994,1996,
Vol. 22,
1 8 1197
11. Wilkie, T.M., Gilbert, D.J., Olsen, A.S., Chen, X-N., Amatruda, T.T.,
Korenberg, J.R., Trask, B.J., de Jong, P., Reed, R.R., Simon, M.I. Jenkins,
N.A. and Copeland, N.G. (1992) Evolution of the mammalian G protein α
subunit multigene family. Nature Genet., 1, 85–91.
12. Xu, S., Nakazawa, M., Tamai, M. and Gal, A. (1995) Autosomal dominant
retinitis pigmentosa locus on chromosome 19q in a Japanese family. J. Med.
Genet., 32, 915–916.
13. Wittinghofer, A. (1994) The structure of transducin Gαt. More to view than
just Ras. Cell, 76, 201–204.
14. Haim, M., Holm, N.V. and Rosenberg, T. (1992) Population survey of retinitis
pigmentosa and allied disorders in Denmark. Completeness of registration
and quality of data. Acta Ophthalmol., 70, 165–177.
15. Bunge, S., Fuchs, S. and Gal, A. (1996) Simple and nonisotopic methods to
detect unknown gene mutations in nucleic acids. In Adolph, K.W. (ed.),
Methods in Molecular Genetics. Academic Press, Orlando, FL, Vol. 8, pp.
26–39.
16. Ott, J. (1974) Estimation of the recombination fraction in human pedigrees:
efficient computation of the likelihood for human linkage studies. Am. J.
Hum. Genet., 26, 588–597.