HUMAN MUTATION 1:240-247 (1992)
RESEARCH ARTICLE
Polymorphic Variation Within “Conserved” Sequences at the 3‘ End of the Human RDS Gene Which Results in Amino Acid Substitutions Siobhh A. Jordan,* G. Jane Farrar, Paul Kenna, and Peter Humphries Department of Genetics, Trinity CoIkge, Dublin 2, Ireland Communicated by David Vulk
T h e human RDS gene, previously mapped to chromosome 6p, encodes a protein found in the outer disc membrane of the photoreceptor cells of the retina. The cDNA sequence of the human gene shows 85% identity with the bovine peripherin gene and the rds (retinal degeneration s h ) genes from mouse and rat. Mutations in the RDS gene have recently been implicated in autosomal dominant retinitis pig. mentosa (adRP) in some families. Here we present evidence that the third exon of this gene is subject to polymorphic variation in humans. T h e three sequence alterations described in this paper give rise to amino acid substitutions. However, as these missense mutations also occur in the normal population they are not implicated as causing adRP. Interestingly such sequence variation is not found within other species examined including mouse and bovine. These intragenic polymorphisms will be of future potential value in studies to locate further disease causing mutations in adRP patients in the RDS gene. 0 1992 Wiley-Liss, Inc. KEY WORDS:
Human RDS gene, Retinitis pigmentosa, Polymorphic variation, Sequence conservation, Amino acid substitution
lNTRODUCTlON The RDS gene encodes a membrane associated glycoprotein which is confined to the photoreceptor outer segment discs (Travis et al., 1991a; Connell et al., 1990; Begy et al., 1990). The gene has been mapped to the proximal end of chromosome 6 (Travis et al., 1991b) and from recent studies, three exons of 580, 246, and 209 bp, respectively, have been characterised in the 2.97 kb cDNA (Kajiwara et al., 1991). The RDS cDNA sequence shows 85% identity with the bovine peripherin and mouse rds genes (Travis et al., 1991b). All three proteins contain 346 amino acids and conserved structural features include four stretches of 23 -26 uncharged residues suggesting membrane spanning domains, 12 cysteine residues, and a single glycosylation site at the C-terminal end of the gene (Travis et al., 1991b). Recently mutations in the first and second exons of the RDS gene have been implicated in some forms of autosomal dominant retinitis pigmentosa (adRP) (Farrar et al., 1991b, 1992; Kajiwara et al., 1991). Retinitis pigmentosa (RP) is a group of inherited human retinopathies which involves the 0 1992 WILEY-LISS, INC-
degeneration of photoreceptors and has been shown to be both clinically and genetically heterogeneous. The first adRP gene was mapped to chromosome 3q, implicating the candidate gene, rhodopsin (McWilliam et al., 1989; Farrar et al., 1990). Another adRP gene segregating in a large Irish pedigree has shown tight linkage to the RDS gene o n chromosome 6p (Farrar et al., 1991a; Jordan et al., 1992). A third adRP locus has also been identified in the centromeric region of chromosome 8, although a gene has yet to be characterised (Blanton et al., 1991). We have screened autosomal dominant and autosomal recessive RP patients for further sequence alterations in the RDS gene by single-strand conformation polymorphism electrophoresis (SSCPE) (Orita et al., 1989). Using this method we have detected a fragment of 247 bp, containing the ma-
Received April 10, 1992 accepted June 9, 1992. ‘To whom reprint requests/correspondence should be addressed.
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
jority of the third exon, from the 3’ end of this gene which shows very variable mobilities, both in RP patients and normal control individuals, suggesting the presence of a number of mutations. Characterisation of these mobility shifts by direct sequencing has revealed three mutations in the RDS gene. All three nucleotide changes lead to amino acid substitutions at sites which to date have been found to be conserved between species (bovine, mouse, and rat). Moreover the fact that these amino acid substitutions have been found in normal control populations suggests that these missense mutations do not result in a human retinopathy. Certainly some areas of the RDS gene may be functionally less important and can withstand sequence alterations without any observable loss of function. These intragenic polymorphisms will be useful in directing further analysis to locate sequence alterations in the RDS gene away from this area of sequence variation and in mapping retinal hereditary disorders in general. MATERIALS AND METHODS PCR Amplifications, SSCP Analysis, and Direct Sequencing Two oligonucleotide primers designed from the cDNA sequence (Travis et al., 1991b), 5’TTGGGCTGCGCTACCTACAG-3’ and 5’GGAGTGCACTATTTCTCAGT-3’, were used to PCR amplify a 247 bp fragment from the third exon and 3’ flanking region (positions 1082 to 1329) of the RDS gene. Single strand conformational polymorphism (SSCP) analysis (Orita et al., 1989) was used to identify sequence alterations in a panel of 60 unrelated patients from adRP families and 10 unaffected control individuals (unrelated parents from the CEPH panel of families). Patients and control individuals who exhibited mobility shifts o n polyacrylamide gels were subsequently reamplified using the above primers using PCR conditions described elsewhere (Farrar et al., 1991a). The amplified products were purified and directly sequenced (Yandell and Dryja, 1989) using nested sequencing primers in the amplified fragment (Table 1). Two pairs of oligonucleotides 5’-TTGGGCTACGCTACCTGCAC-3’ and 5 ‘-GGAGTC-
CACTACTTTTCAGT-3’; 5 ’-CCGGACTCCGCTACCTCCAC-3’ and 5 ‘-GGAGTCCACTAAGTGTTGAG-3’ were designed from the bovine peripherin (Connell e t al., 1990) and mouse rds (Travis et al., 1989) cDNA sequences, respectively. These primers were used to amplify the homologous 247 bp fragment in the third exon
241
TABLE 1. Oligonucleotide Primers Used in Directly
Sequencing the 247 bp Fragment From the PeripherinLRDS Gene Amplified From Human, Bovine, and Mouse DNA Samdes Human primers 1.
5’-TACAGACGTCGCTGGATGGT-3’
2.
5‘-GAGACCTGGAAGGCCTTTCT-3’
3.
5’-TATTTCTCAGTGTTCGGGAG-3’ 5’-AGAAAGGCCTTCCAGGTCTC-3’
4. Bovine primers 1. 2. Mouse primers 1. 2.
5’-TGCACACGGCGCTGGAAGGC-3‘ 5’-TACTTTTCAGTGTGCAGAGG-3’ 5’-TCCACACAGCGCTGGAGAGT-3’ 5‘-TAAGTGTTGAGGAGGGGGAG-3’
of the bovine peripherin gene and the mouse rds gene in a series of bovine and mouse DNA samples, respectively. The mouse oligomers were also used to amplify a laboratory rat DNA sample. The bovine DNA samples were from European Charolais and Friesian cows, Kenana and White Fulani cows of East and West African origin, respectively, and Sahiwal and Tharparkar cows from India. DNA from laboratory mice strains A/J, CBNJ, BALB/c, SHASHA, LAC A, and Q was isolated from whole blood and used for PCR amplification. The amplified products were directly sequenced using nested sequencing primers specific for each species (Table 1). All oligonucleotide primers were synthesised using an Applied Biosystems 391 DNA synthetiser. Allele Specific Oligonucleotides and Dot Blotting Genomic DNA from 160 unrelated normal individuals was amplified using the human primers. Of the amplified products 2 pl was dot-blotted onto Hybond N nylon membranes and hybridised to end-labelled normal and mutant ASOs (Farr et al., 1988) for each of the missense mutations (Table 2). The hybridisation temperature used was T, -5°C and was calculated for each specific ASO. After hybridisation the blots were washed in 4 X SSC. To ensure the specificity of the reaction, the washing temperature was predetermined for each AS0 using control mutant and normal amplification products. T h e blots were exposed to X-ray film at - 70°C for 12-24 hr. RESULTS Twenty-eight of the 60 adRP patients and 6 of the 10 control individuals displayed altered mobility shifts in SSCPE gels of a fragment encoding the third exon of the RDS gene (Fig. 1). Direct sequencing of these individuals revealed three single
242
JORDAN ET AL.
TABLE 2. ASOs Used in the Detection of the Polymorphisms
in the Amplified Human Samples” Codon
Tm (“C)
304 Normal (Al): 5’-GCGAGGGCTGGCTGCTGGAG-3’ 70 Mutant (A2): 5‘-GCCAGGGCTGGCTGCTGGAG-3’ 70
310 Normal (A3): Mutant (A4): 338 Normal (A5): Mutant (A6):
5’-GGCTGGCTGCTGGAGAAGAG-3’ 66 5’-GGCTGGCTGCTGGAGAGGAG-3’68 5’-CGAGGGCGCAGGCGCAGGCC-3’ 74 5’-CGAGGGCGCAGACGCAGGCC-3‘ 72
”The hybridisation temperature used for each A S 0 was Tm-5”C.
and amino acids which have been conserved across species (Fig. 5 and Table 4), bovine and mouse DNAs were amplified and directly sequenced to investigate whether these changes also occurred in these species and had not been previously observed. Twelve bovine DNA samples sequenced from European, African, and Indian cows did not display the sequence changes in the 3’ end of the peripherin gene as observed in the homologous region of the RDS gene. Similarly the mouse rds gene from six laboratory strains and from a single laboratory rat strain did not show any sequence alterations in this region of the gene.
DISCUSSION base changes within a 130 bp region of this fragment, with all three sequence alterations resulting in amino acid substitutions. T h e first change resulted in the replacement of glutamic acid by glutamine (E304Q), the second caused a lysine to arginine substitution (K3 lOR), while the third change resulted in a glycine being replaced by an aspartate (G338D) (Fig. 2). To eliminate the possibility that the sequence alterations were due to replication errors that may have occurred during the PCR reaction, independent reactions and sequencing of the product were carried out and at all times the polymorphisms were observed in both DNA strands. Moreover segregation of the three polymorphisms was observed in a three generation Irish family, members of which display symptoms of an early onset form of adRP. Both affected and unaffected members of this family were assessed using a variety of tests including Goldman perimetry, dark adaptometry, electroretinography, and fundus photography following the international clinical protocol for the examination of RP patients (Marmor et al., 1989). The segregation of the K310R mutation in this family (Fig. 3) clearly demonstrates that this amino acid substitution is not the causative agent for adRP in this family. In order to estimate the frequency of these apparently neutral mutations in the population, allele-specific oligonucleotides (ASOs) corresponding to the normal and mutant sequences were designed and used to test amplified DNA from 160 unrelated individuals by dot-blotting. The frequency and heterozygosity of each missense mutation in normal and adRP patient populations are outlined in Table 3 and Figure 4 shows the inheritance of the three mutations in a subset of the control individuals using dot blots. As these point mutations occur in nucleotides
The three sequence alterations reported here represent new mutations in the RDS gene, which has recently been implicated as a causative gene in some forms of adRP (Farrar et al., 1991b, 1992; Kajiwara et al., 1991). These nucleotide base changes result in amino acid substitutions in the third exon of the gene and cannot be correlated to a disease phenotype, since they occur within the normal population. Unlike the ApaI RFLP polymorphism in the 3’ untranslated region of this gene, which shows an overrepresentation of the minor allele in a set of unrelated dominant patients (Kajiwara et al., 1991, S. Jordan, unpublished observation), the sequence alterations reported here have similar frequencies in both control and RP patient populations (Table 3). Moreover, a person homozygous for the mutant allele of the K310R polymorphism in a three generation adRP pedigree does not display any symptoms of the disease (Fig. 3). In addition individuals who were homozygous for the E304Q mutation (individual N5 in Fig. 2a) and homozygous for the G338D mutation, who did not suffer from RP, were observed during this study. However, since sequencing of cloned alleles was not carried out during this study, it cannot be ruled out that a specific combination of these three mutations might cause RP. In view of the high frequency of these missense mutations in the normal population from the A S 0 data, this is an unlikely possibility. Moreover, the observation that the three amino acid substitutions exist in the normal population will allow these frequent sequence alterations to be eliminated as possible disease causing mutations in future studies implicating the RDS gene as causative of adRP. The RDS gene encodes a transmembrane protein which is putatively located in the rim region of the retinal photoreceptor cell discs (Connell et al., 1990). Initial evidence suggests that it may be
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
243
Controls
adRP Patients
1 2 3 4 5 6 7 8 9
10 11 12 13
FIGURE 1. Autoradiogram of a single strand conformationpolymorphismelectrophoresis (SSCPE) gel showing mobility shifts in amplified fragments from the third exon of the RDS gene. Lanes 1-9 are amplified from adRP patient DNA samples while lanes 10-13 are from normal control individuals (parents of the CEPH families). The mobility shift seen in lane 5 is different from that seen in lane 8.
involved in the maintenance of the disc rim structure possibly as an adhesion molecule (Connell et al., 1991). The 3’ end of the gene encodes the C-terminal domain which in the proposed structural model of the homologous bovine peripherin gene (Connell et al., 1991) is located at the cytoplasmic side of the outer disc membrane. Our observation of amino acid variation resulting in no observable functional change at this region of the protein may be indicative of a more generalised phenomenon. Mutant amino acid substitutions in the C-terminal domain may not seriously disrupt the existing function of this protein unlike mutations in the transmembrane domain (Farrar et al., 1991b) and in the intradiscal loop (Kajiwara et al., 1991; Farrar et al., 1992), which have been implicated as possible causes of adRP. Interestingly, mutations in the cytoplasmic domain have been found in adRP patients in another transmembrane photoreceptor specific protein, rhodopsin (Dryja et al., 1990; Horn et al., 1992), emphasising the different structural/functional relationships of these two proteins within the photoreceptor cells. Substitution of a chemically similar amino acid may have a small deleterious effect on the protein’s structure and function (Kimuru and Ohta, 1974). The K310R substitution observed in the RDS gene results in no overall change in the amino acid charge and both amino acids have a similar size.
However, the G338D replacement with a negatively charged and slightly larger amino acid replacing a neutral one may be more deleterious to the protein. Similiarly the substitution of negatively charged amino acid by a neutral one (E304Q) with no overall change in size may have an effect on the protein’s structure and function. The deleterious effect at one site may possibly be negated by a subsequent change in another amino acid. Exact three-dimensional models of the RDS protein elucidated from X-ray crystallography are required before accurate predictions can be made about the apparent lack of effect these mutations have o n the function of the protein. In addition reproducing these missense mutations by site-directed mutagenesis and their subsequent expression in transgenic mice will also aid in understanding the lack of effect of these mutations. From the RDS cDNA sequence the three amino acid substitutions observed in this study are in close proximity to one another, being located in the third exon of the gene within a 130 bp region. From a cross-species comparison of the cDNA sequences encoding the peripheridRDS proteins, it appears that this region is highly conserved between human, cow, mouse, and rat at both the amino acid and nucleotide level (Fig. 5). Of the amino acids encoded in the 247 bp-amplified fragment 69.6% are conserved between bovine,
244
JORDAN ET AL.
FIGURE2. Nucleotide sequence of polymorphic codons in the third exon of the RDS gene. (a) Individual N-5 has a G-to-C tranversion in the first base of codon 304 changing the specificity of the codon from glutamic acid to glutamine (E304Q). (b)Individual 1-3has an A-to-G transition a t codon 310 changing the specificity of this codon from lysine to arginine (K310R). (c) Individual N-5 has a G-to-A transition at codon 338 changing the specificity of this codon from glycine to aspartate (G338D).
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
245
Allele Frequencies and Percent Heterozygosity for Each of the Polymorphisms in the Human RDS Gene as Estimated From 160 Unrelated Control Individuals and 60 Unrelated adRP Patientsa TABLE 3.
O - - P 11
Codon position 304
0 12
6 12
of the K310R mutation in a three generation Irish family exhibiting symptoms of early onset adRP (Al, normal allele; A2, mutant allele). Affected and unaffected individuals are indicated by filled and open symbols, respectively. FIGURE 3. Segregation
310 338
Allele frequency Normals A1:0.49 A20.51 A1:0.47 A2:0.53 A1:0.41 A20.59
Heterozygosity
Patients 0.42 0.58 0.55 0.45 0.30 0.70
(%I 49.98 49.82 48.38
"Al, normal allele; A2, mutant allele.
FIGURE 4. Dot-blots showing the inheritance of the three missense mutations in a subset of the control individuals used in this study. A1 and A2 refer to the normal and mutant ASOs for the E304Q mutation, respectively. A3 and A4 refer to the normal and mutant ASOs for the K310R mutation while A5 and A6 refer to the normal and mutant ASOs for the G338D mutation.
mouse, rat, and human cDNA sequences with 82.74% conservation at the nucleotide level. E304 was previously completely conserved among all species sequenced to date (human, mouse, rat, and bovine). K310 is conserved between human, mouse, and cow, but not rat. Interestingly, the residue in cow, mouse, and rat cDNA sequences corresponding to G338 in humans is an aspartate.
Therefore the G338D mutation observed here increases the conservation between species. Little previous interspecies variation has been observed at the nucleotides which we have found to be polymorphic in humans. Moreover we have sequenced 12 bovine samples from six different breeds, mice from six different laboratory strains, and a single rat DNA, all of which showed no
246
JORDAN ET AL. TABLE 4.
Summaw of Base Changes Which Occurred in the Third Exon of the Human RDS Gene" Other species
Position
Change
Mouse
Base
Codon
Base
Amino acid
1150 1169 1253
304
GAGjCAG AAGjAGG GGCjGAC
E304Q K310R G338D
310 338
"The amino acids at the same positions in
cow,
MALLKVKFDaKKRVKLAQGLWLMNWFSVLAGIIIFSLGLFLK1ELRKRSDWI"SESHFV ...... KFDQKKRVKLAQGLWLMNWFSVLAGIIIFGLGLFLKIELRKRSDVMNNSESHFV MALLKVKFDQKKRVKLAaGLh"WLSVLAGIVLFSLGLFLK1ELRKRSEWI"SESHFV MliLLKVKFDQKKRVKLAQGLWLMNWLSVLAGIVLFSLGLFLKIELRKRSD~DNSESH~
HUMRDS BOVPER MUSRDS RATRDS
PNSLIGMGVLSCVFNSLAGKICYDALDPAKYARWKPWLKPYLAICVLFNIILFLV~CCF PNSLIGVGVLSCVFNSLAGKICYDRLDPAKYAKWKPWLKP~KPYLAVCVLF~FLV~CCF PNSLIGVGVLSCVFNSLAGKICYDALDPAKYAKWKP~KPYLAVCIFFNVILFLVALCCF PNSLIGVGVLSCVFNSLAGKICYD~DPAKYAKWKPWLKLYLAVCVFF~ILFLVALCCF
HUMRDS BOVPER MUSRDS RATRDS
LLRGSLENTLGQGLKNGMKYYRDTDTPGRCFMKKTIDMLPFRDWFEIQWI
**ff*tf*t*ttt*ttttt,**ft**__*~****.*****t**,*.(**.*ttt
....................
LLRGSLESTLAHGLKNGMKFYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI LLRGSLESTLAYGLKNGMKYYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI LLRGSLESTLAYGLKNGMKYYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI * * I * * * * , * * , ................................................
HUMRDS BOVPER MUSRDS RATRDS
HUMRDS BOVPER MUSRDS RATRDS
EELHLklI.RCCRAATI.NYY p?l-W.;X~.FITl !.IWLFB.'SITAi!PF! HTALE^VSlll'E!)I E
HUMRDS BOVPER MUSRDS RATRDS
CESEGWLLENSVSETWKAFLESFKKLGKSNQVEAEAADAGQAPEAG
...... ........ ... ... . .... ..... .. ... . . . ... .
*.*fffff,11,ff*.*.fff
Rat
GAGlE AAGK GACID
GAGlE AAGIK GAUD
GAGIE AATN GACID
mouse, and rat are also included.
HUMRDS BOVPER MUSRDS RATRDS
**ft*f,*t*******t***************,f*ff**
Bovine (base/amino acid)
1....,......,,,.*~..~..
FIGURE5. Amino acid sequence alignment of the human RDS, bovine peripherin, mouse and rat rds cDNA sequences using the CLUSTAL program (Higgins and Sharp, 1988). The position of the amino acid substitutions in the 3' end of the RDS gene is doubly underlined.
polymorphism. One could speculate that this initial observation might suggest that the human RDS protein is more subject to change in this region than in other species. However, it is noteworthy that the bovine samples and the mice strains used in this analysis are quite inbred, hence reducing the probability of observing polymorphism. Therefore further sequencing of human, bovine, and mouse samples from a range of ethnic populations would be required to lend weight to this speculation. Technical advancements which now enable rapid amplification and sequencing of genes from multiple individuals may reveal further polymorphisms resulting in amino acid substitutions in the normal population. The amino acid coding sequence and functional relationship of proteins may be more flexible than first envisaged (Lau and Dill, 1990). Missense mutations have also been observed in other proteins
which do not result in a pathological phenotype. Evidence from a study of 95 independent missense mutations in the Factor IX gene correlated to a hemophilia B phenotype suggests that mammalian sequence homology is not sufficient to define essential conserved residues (Bottema et al., 1991). It is proposed that in the Factor IX gene approximately 60% of the residues act as spacers to maintain side chain interaction and virtually no missense mutation in these residues will cause disease. Indeed a missense mutation has been observed in Factor IX (H257Y) which does not cause hemophilia B in a male (Montandon et al., 1990). Moreover from a survey of orthinine 6-aminotransferase (OAT) mutations in gyrate atrophy patients, one mutation L437F, at the extreme C terminus of the protein, had no disernible effect on OAT activity (Brody e t al., 1992). The three new mutations which result in neutral polymorphisms observed in the RDS gene will aid in future mapping of retinal hereditary disorders. They allow us to speculate that the degree of conservation of amino acids within humans may be less than first envisaged, particularly as the three amino acids appear to be well conserved between mouse, rat, and bovine species. Finally they also allow a brief glimpse of the flexibility of the RDS protein's amino acid sequence in tolerating alterations without a resulting loss of function. This may aid in an overall understanding of the role of the RDS protein in the maintenance of the disc rim structure of the rod photoreceptor cell.
ACKNOWLEDGMENTS The authors thank the United States RP Foundation, the George Gund Foundation, the Wellcome Trust, RP Ireland Fighting Blindness, the British RP Association, the Health Research Board of Ireland, Bioresearch Ireland, and the Concerted Action and Science Programmes of the Commision of the European Communities for their continuing support. They also thank Professor A.S. Whitehead and Ronan Loftus for mouse and bovine DNA samples.
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
REFERENCES Begy C , Bridges CD (1990) Nucleotide and predicted protein sequence of rat retinal degeneration slow (rds). Nucleic Acids Res 18:3058. Blanton SH, Heckenlively JR, Cottinghain AW, Freidman J , Sadler LA, Wagner M, Freidman LH, Daiger SP (1991) Linkage mapping of autosomal dominant retinitis pigmentosa (RPl) to the pericentric region of human chromosome 8. Genomics 113357-869. Bottema CDK, Ketterling RP, Ii S, Yoon H-S, Phillips Ill JA, Sommer SS (1991) Missense mutations and evolutionary conservation of amino acids: Evidence that many of the amino acids in Factor IX function as “spacer” elements. Am J Hum Genet 49:820-838. Brody LC, Mitchell GA, Obie C , Michard J, Steel G, Fontaine G, Robert M-F, Sipila I, Kaiser-Kupfer M, Valle D (1992) Orthinine 6-aminotransferase mutations in gyrate atrophy. J Biol Chem 267:3302-3307. Connell G, Molday RS (1990) Molecular cloning, primary structure and orientation of the vertebrate photoreceptor cell protein peripherin in the rod outer segment disc membrane. Biochemistry 29:4691-4698. Connell G, Bascom R, Molday L, Reid D, Mclnnes RR, Molday RS (1991) Photoreceptor peripherin is the normal product of the gene responsible for retinal degeneration in the rds mouse. Proc Natl Acad Sci USA 88:723-726. Dryja TP, McGee TL, Hahn LB, Cowley GS, Olsson JE, Reichel E, Sandberg MA, Bearson EL (1990) Mutations within the rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. N Engl J Med 323:1302-1307. Farr CJ, Saiki RK, Erlich HA, McCormick F, Marshall CJ (1988) Analysis of RAS gene mutations in acute myeloid leukaemia by polymerase chain reaction and oligonucleotide probes. Proc Natl Acad Sci USA 85:1629-1633. Farrar GJ, McWilliam P, Bradley DG, Kenna P, Lawler M, Sharp EM, Humphries MM, Eiberg H , Conneally PM, Trotfatter JA, Humphries P (1990) Autosomai dominant retinitis pigmentosa: Linkage to rhodopsin and evidence for genetic heterogeneity. Genomics 8:35-40. Farrar GJ, Jordan SA, Kenna P, Humphries MM, Kumar-Singh R, McWilliam P, Allamand V, Sharp E, Humphries P (1991a) Autosomal dominant retinitis pigmentosa: Localisation of a disease gene (RP6) to the short arm of chromosome 6. Genomics 11:870-874. Farrar GJ, Kenna P, Jordan SA, Kumar-Singh R, Humphries MM, Sharp EM, Sheils DM, Humphries P (1991b) A three-base-pair deletion in the peripherin-RDS gene in one form of retinitis pigmentosa. Nature (London) 354:478-480. Farrar GI, Kenna P, Jordan SA, Kumar-Singh R, Humphries MM, Sharp EM, Sheils D, Humphries P (1992) Autosomal dominant retinitis pigmentosa: A novel mutation at the peripherid RDS locus in the original 6p linked pedigree. Genomics (in press).
247
Horn M, Marchese C , Kunisch M, Fusi L, Apfelstedt-Sylla E, Zrenner E, Gal A, Humphries P, Farrar GJ (1992) Deletions in exon 5 of the human rhodopsin gene causing shift in the reading frame and autosomal dominant retinitis pigmentosa. Hum Genet (in press). Higgins DG, Sharp PM (1988) CLUSTAL: A package for performing multiple sequence alignments on a microcomputer. Gene 73:237-244. Jordan SA, Farrar GI, Kenna P, Humphries MM, Kumar-Singh R, Allamand V, Sharp EM, Humphries P (1992) Autosomal dominant retinitis pigmentosa (RP6): Co-segregation of RP6 and the peripherin-RDS locus in a late-onset family of Irish origin. Am J Hum Genet 50634-639. Kajiwara K, Hahn LB, Mukai S, Travis G H , Berson EL, Dryja T P (1991) Mutations in the human retinal degeneration slow gene in autosomal dominant retinitis pigmentosa. Nature (London) 354:480-483. Kimuru M, Ohta T (1974) On some principles governing molecular evolution. Proc Natl Acad Sci USA 71:2848-2852. Lau KF, Dill KA (1990) Theory for protein mutability and biogenesis. Proc Natl Acad Sci USA 87:638-642. Marmor MF, Arden GB, Nilsson SE, Zrenner E (1989) Standard for clinical electroretinography. Arch Ophthalmol 107:816819. McWilliam P, Farrar GI, Kenna P, Bradley DG,Humphries MM, Sharp EM, McConnell DJ, Lawler M, Sheils DM, Ryan C, Stevens K, Daiger SP, Humphries P (1989) Autosomal dominant retinitis pigmentosa (ADRP): Localization of an ADRP gene to the long arm of chromosome 3. Genomics 5:619-622. Montandon A], Green PM, Bentley DR, Ljung R, Nilsson IM, Giannelli F (1990) Two factor IX mutations in the family of an isolated hemophilia B patient: Direct carrier diagnosis by amplification mismatch detection. (AMD). Hum Genet 85:200204. Orita M, Suzuki Y, Sekiya T, Hayashi K (1989) Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5574-879. Travis GH, Brennan MB, Danielson PE, Kozak C A , Sutcliffe JG (1989) Identification of a photoreceptor-specific mRNA encoded by the gene responsible for retinal degeneration slow (rds). Nature (London) 338:70-73. Travis GH, Sutcliffe JC, Bok D (1991a) The retinal degeneration slow (rds) gene product is a photoreceptor disc membraneassociated glycoprotein. Neuron 6:61-70. Travis GH, Christerson L, Danielson PE, Klisak I, Sparkes RS, Hahn LB, Dryja TP, Sutcliffe J G (1991b) The human retinal degeneration slow (RDS) gene: Chromosome assignment and structure of the mRNA Genomics 10:733-739. Yandell DW, Dryja TP (1989) Detection of DNA sequence polymorphisms by enzymatic amplification and direct genomic sequencing. Am J Hum Genet 45:547-555.
RESEARCH ARTICLE
Polymorphic Variation Within “Conserved” Sequences at the 3‘ End of the Human RDS Gene Which Results in Amino Acid Substitutions Siobhh A. Jordan,* G. Jane Farrar, Paul Kenna, and Peter Humphries Department of Genetics, Trinity CoIkge, Dublin 2, Ireland Communicated by David Vulk
T h e human RDS gene, previously mapped to chromosome 6p, encodes a protein found in the outer disc membrane of the photoreceptor cells of the retina. The cDNA sequence of the human gene shows 85% identity with the bovine peripherin gene and the rds (retinal degeneration s h ) genes from mouse and rat. Mutations in the RDS gene have recently been implicated in autosomal dominant retinitis pig. mentosa (adRP) in some families. Here we present evidence that the third exon of this gene is subject to polymorphic variation in humans. T h e three sequence alterations described in this paper give rise to amino acid substitutions. However, as these missense mutations also occur in the normal population they are not implicated as causing adRP. Interestingly such sequence variation is not found within other species examined including mouse and bovine. These intragenic polymorphisms will be of future potential value in studies to locate further disease causing mutations in adRP patients in the RDS gene. 0 1992 Wiley-Liss, Inc. KEY WORDS:
Human RDS gene, Retinitis pigmentosa, Polymorphic variation, Sequence conservation, Amino acid substitution
lNTRODUCTlON The RDS gene encodes a membrane associated glycoprotein which is confined to the photoreceptor outer segment discs (Travis et al., 1991a; Connell et al., 1990; Begy et al., 1990). The gene has been mapped to the proximal end of chromosome 6 (Travis et al., 1991b) and from recent studies, three exons of 580, 246, and 209 bp, respectively, have been characterised in the 2.97 kb cDNA (Kajiwara et al., 1991). The RDS cDNA sequence shows 85% identity with the bovine peripherin and mouse rds genes (Travis et al., 1991b). All three proteins contain 346 amino acids and conserved structural features include four stretches of 23 -26 uncharged residues suggesting membrane spanning domains, 12 cysteine residues, and a single glycosylation site at the C-terminal end of the gene (Travis et al., 1991b). Recently mutations in the first and second exons of the RDS gene have been implicated in some forms of autosomal dominant retinitis pigmentosa (adRP) (Farrar et al., 1991b, 1992; Kajiwara et al., 1991). Retinitis pigmentosa (RP) is a group of inherited human retinopathies which involves the 0 1992 WILEY-LISS, INC-
degeneration of photoreceptors and has been shown to be both clinically and genetically heterogeneous. The first adRP gene was mapped to chromosome 3q, implicating the candidate gene, rhodopsin (McWilliam et al., 1989; Farrar et al., 1990). Another adRP gene segregating in a large Irish pedigree has shown tight linkage to the RDS gene o n chromosome 6p (Farrar et al., 1991a; Jordan et al., 1992). A third adRP locus has also been identified in the centromeric region of chromosome 8, although a gene has yet to be characterised (Blanton et al., 1991). We have screened autosomal dominant and autosomal recessive RP patients for further sequence alterations in the RDS gene by single-strand conformation polymorphism electrophoresis (SSCPE) (Orita et al., 1989). Using this method we have detected a fragment of 247 bp, containing the ma-
Received April 10, 1992 accepted June 9, 1992. ‘To whom reprint requests/correspondence should be addressed.
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
jority of the third exon, from the 3’ end of this gene which shows very variable mobilities, both in RP patients and normal control individuals, suggesting the presence of a number of mutations. Characterisation of these mobility shifts by direct sequencing has revealed three mutations in the RDS gene. All three nucleotide changes lead to amino acid substitutions at sites which to date have been found to be conserved between species (bovine, mouse, and rat). Moreover the fact that these amino acid substitutions have been found in normal control populations suggests that these missense mutations do not result in a human retinopathy. Certainly some areas of the RDS gene may be functionally less important and can withstand sequence alterations without any observable loss of function. These intragenic polymorphisms will be useful in directing further analysis to locate sequence alterations in the RDS gene away from this area of sequence variation and in mapping retinal hereditary disorders in general. MATERIALS AND METHODS PCR Amplifications, SSCP Analysis, and Direct Sequencing Two oligonucleotide primers designed from the cDNA sequence (Travis et al., 1991b), 5’TTGGGCTGCGCTACCTACAG-3’ and 5’GGAGTGCACTATTTCTCAGT-3’, were used to PCR amplify a 247 bp fragment from the third exon and 3’ flanking region (positions 1082 to 1329) of the RDS gene. Single strand conformational polymorphism (SSCP) analysis (Orita et al., 1989) was used to identify sequence alterations in a panel of 60 unrelated patients from adRP families and 10 unaffected control individuals (unrelated parents from the CEPH panel of families). Patients and control individuals who exhibited mobility shifts o n polyacrylamide gels were subsequently reamplified using the above primers using PCR conditions described elsewhere (Farrar et al., 1991a). The amplified products were purified and directly sequenced (Yandell and Dryja, 1989) using nested sequencing primers in the amplified fragment (Table 1). Two pairs of oligonucleotides 5’-TTGGGCTACGCTACCTGCAC-3’ and 5 ‘-GGAGTC-
CACTACTTTTCAGT-3’; 5 ’-CCGGACTCCGCTACCTCCAC-3’ and 5 ‘-GGAGTCCACTAAGTGTTGAG-3’ were designed from the bovine peripherin (Connell e t al., 1990) and mouse rds (Travis et al., 1989) cDNA sequences, respectively. These primers were used to amplify the homologous 247 bp fragment in the third exon
241
TABLE 1. Oligonucleotide Primers Used in Directly
Sequencing the 247 bp Fragment From the PeripherinLRDS Gene Amplified From Human, Bovine, and Mouse DNA Samdes Human primers 1.
5’-TACAGACGTCGCTGGATGGT-3’
2.
5‘-GAGACCTGGAAGGCCTTTCT-3’
3.
5’-TATTTCTCAGTGTTCGGGAG-3’ 5’-AGAAAGGCCTTCCAGGTCTC-3’
4. Bovine primers 1. 2. Mouse primers 1. 2.
5’-TGCACACGGCGCTGGAAGGC-3‘ 5’-TACTTTTCAGTGTGCAGAGG-3’ 5’-TCCACACAGCGCTGGAGAGT-3’ 5‘-TAAGTGTTGAGGAGGGGGAG-3’
of the bovine peripherin gene and the mouse rds gene in a series of bovine and mouse DNA samples, respectively. The mouse oligomers were also used to amplify a laboratory rat DNA sample. The bovine DNA samples were from European Charolais and Friesian cows, Kenana and White Fulani cows of East and West African origin, respectively, and Sahiwal and Tharparkar cows from India. DNA from laboratory mice strains A/J, CBNJ, BALB/c, SHASHA, LAC A, and Q was isolated from whole blood and used for PCR amplification. The amplified products were directly sequenced using nested sequencing primers specific for each species (Table 1). All oligonucleotide primers were synthesised using an Applied Biosystems 391 DNA synthetiser. Allele Specific Oligonucleotides and Dot Blotting Genomic DNA from 160 unrelated normal individuals was amplified using the human primers. Of the amplified products 2 pl was dot-blotted onto Hybond N nylon membranes and hybridised to end-labelled normal and mutant ASOs (Farr et al., 1988) for each of the missense mutations (Table 2). The hybridisation temperature used was T, -5°C and was calculated for each specific ASO. After hybridisation the blots were washed in 4 X SSC. To ensure the specificity of the reaction, the washing temperature was predetermined for each AS0 using control mutant and normal amplification products. T h e blots were exposed to X-ray film at - 70°C for 12-24 hr. RESULTS Twenty-eight of the 60 adRP patients and 6 of the 10 control individuals displayed altered mobility shifts in SSCPE gels of a fragment encoding the third exon of the RDS gene (Fig. 1). Direct sequencing of these individuals revealed three single
242
JORDAN ET AL.
TABLE 2. ASOs Used in the Detection of the Polymorphisms
in the Amplified Human Samples” Codon
Tm (“C)
304 Normal (Al): 5’-GCGAGGGCTGGCTGCTGGAG-3’ 70 Mutant (A2): 5‘-GCCAGGGCTGGCTGCTGGAG-3’ 70
310 Normal (A3): Mutant (A4): 338 Normal (A5): Mutant (A6):
5’-GGCTGGCTGCTGGAGAAGAG-3’ 66 5’-GGCTGGCTGCTGGAGAGGAG-3’68 5’-CGAGGGCGCAGGCGCAGGCC-3’ 74 5’-CGAGGGCGCAGACGCAGGCC-3‘ 72
”The hybridisation temperature used for each A S 0 was Tm-5”C.
and amino acids which have been conserved across species (Fig. 5 and Table 4), bovine and mouse DNAs were amplified and directly sequenced to investigate whether these changes also occurred in these species and had not been previously observed. Twelve bovine DNA samples sequenced from European, African, and Indian cows did not display the sequence changes in the 3’ end of the peripherin gene as observed in the homologous region of the RDS gene. Similarly the mouse rds gene from six laboratory strains and from a single laboratory rat strain did not show any sequence alterations in this region of the gene.
DISCUSSION base changes within a 130 bp region of this fragment, with all three sequence alterations resulting in amino acid substitutions. T h e first change resulted in the replacement of glutamic acid by glutamine (E304Q), the second caused a lysine to arginine substitution (K3 lOR), while the third change resulted in a glycine being replaced by an aspartate (G338D) (Fig. 2). To eliminate the possibility that the sequence alterations were due to replication errors that may have occurred during the PCR reaction, independent reactions and sequencing of the product were carried out and at all times the polymorphisms were observed in both DNA strands. Moreover segregation of the three polymorphisms was observed in a three generation Irish family, members of which display symptoms of an early onset form of adRP. Both affected and unaffected members of this family were assessed using a variety of tests including Goldman perimetry, dark adaptometry, electroretinography, and fundus photography following the international clinical protocol for the examination of RP patients (Marmor et al., 1989). The segregation of the K310R mutation in this family (Fig. 3) clearly demonstrates that this amino acid substitution is not the causative agent for adRP in this family. In order to estimate the frequency of these apparently neutral mutations in the population, allele-specific oligonucleotides (ASOs) corresponding to the normal and mutant sequences were designed and used to test amplified DNA from 160 unrelated individuals by dot-blotting. The frequency and heterozygosity of each missense mutation in normal and adRP patient populations are outlined in Table 3 and Figure 4 shows the inheritance of the three mutations in a subset of the control individuals using dot blots. As these point mutations occur in nucleotides
The three sequence alterations reported here represent new mutations in the RDS gene, which has recently been implicated as a causative gene in some forms of adRP (Farrar et al., 1991b, 1992; Kajiwara et al., 1991). These nucleotide base changes result in amino acid substitutions in the third exon of the gene and cannot be correlated to a disease phenotype, since they occur within the normal population. Unlike the ApaI RFLP polymorphism in the 3’ untranslated region of this gene, which shows an overrepresentation of the minor allele in a set of unrelated dominant patients (Kajiwara et al., 1991, S. Jordan, unpublished observation), the sequence alterations reported here have similar frequencies in both control and RP patient populations (Table 3). Moreover, a person homozygous for the mutant allele of the K310R polymorphism in a three generation adRP pedigree does not display any symptoms of the disease (Fig. 3). In addition individuals who were homozygous for the E304Q mutation (individual N5 in Fig. 2a) and homozygous for the G338D mutation, who did not suffer from RP, were observed during this study. However, since sequencing of cloned alleles was not carried out during this study, it cannot be ruled out that a specific combination of these three mutations might cause RP. In view of the high frequency of these missense mutations in the normal population from the A S 0 data, this is an unlikely possibility. Moreover, the observation that the three amino acid substitutions exist in the normal population will allow these frequent sequence alterations to be eliminated as possible disease causing mutations in future studies implicating the RDS gene as causative of adRP. The RDS gene encodes a transmembrane protein which is putatively located in the rim region of the retinal photoreceptor cell discs (Connell et al., 1990). Initial evidence suggests that it may be
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
243
Controls
adRP Patients
1 2 3 4 5 6 7 8 9
10 11 12 13
FIGURE 1. Autoradiogram of a single strand conformationpolymorphismelectrophoresis (SSCPE) gel showing mobility shifts in amplified fragments from the third exon of the RDS gene. Lanes 1-9 are amplified from adRP patient DNA samples while lanes 10-13 are from normal control individuals (parents of the CEPH families). The mobility shift seen in lane 5 is different from that seen in lane 8.
involved in the maintenance of the disc rim structure possibly as an adhesion molecule (Connell et al., 1991). The 3’ end of the gene encodes the C-terminal domain which in the proposed structural model of the homologous bovine peripherin gene (Connell et al., 1991) is located at the cytoplasmic side of the outer disc membrane. Our observation of amino acid variation resulting in no observable functional change at this region of the protein may be indicative of a more generalised phenomenon. Mutant amino acid substitutions in the C-terminal domain may not seriously disrupt the existing function of this protein unlike mutations in the transmembrane domain (Farrar et al., 1991b) and in the intradiscal loop (Kajiwara et al., 1991; Farrar et al., 1992), which have been implicated as possible causes of adRP. Interestingly, mutations in the cytoplasmic domain have been found in adRP patients in another transmembrane photoreceptor specific protein, rhodopsin (Dryja et al., 1990; Horn et al., 1992), emphasising the different structural/functional relationships of these two proteins within the photoreceptor cells. Substitution of a chemically similar amino acid may have a small deleterious effect on the protein’s structure and function (Kimuru and Ohta, 1974). The K310R substitution observed in the RDS gene results in no overall change in the amino acid charge and both amino acids have a similar size.
However, the G338D replacement with a negatively charged and slightly larger amino acid replacing a neutral one may be more deleterious to the protein. Similiarly the substitution of negatively charged amino acid by a neutral one (E304Q) with no overall change in size may have an effect on the protein’s structure and function. The deleterious effect at one site may possibly be negated by a subsequent change in another amino acid. Exact three-dimensional models of the RDS protein elucidated from X-ray crystallography are required before accurate predictions can be made about the apparent lack of effect these mutations have o n the function of the protein. In addition reproducing these missense mutations by site-directed mutagenesis and their subsequent expression in transgenic mice will also aid in understanding the lack of effect of these mutations. From the RDS cDNA sequence the three amino acid substitutions observed in this study are in close proximity to one another, being located in the third exon of the gene within a 130 bp region. From a cross-species comparison of the cDNA sequences encoding the peripheridRDS proteins, it appears that this region is highly conserved between human, cow, mouse, and rat at both the amino acid and nucleotide level (Fig. 5). Of the amino acids encoded in the 247 bp-amplified fragment 69.6% are conserved between bovine,
244
JORDAN ET AL.
FIGURE2. Nucleotide sequence of polymorphic codons in the third exon of the RDS gene. (a) Individual N-5 has a G-to-C tranversion in the first base of codon 304 changing the specificity of the codon from glutamic acid to glutamine (E304Q). (b)Individual 1-3has an A-to-G transition a t codon 310 changing the specificity of this codon from lysine to arginine (K310R). (c) Individual N-5 has a G-to-A transition at codon 338 changing the specificity of this codon from glycine to aspartate (G338D).
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
245
Allele Frequencies and Percent Heterozygosity for Each of the Polymorphisms in the Human RDS Gene as Estimated From 160 Unrelated Control Individuals and 60 Unrelated adRP Patientsa TABLE 3.
O - - P 11
Codon position 304
0 12
6 12
of the K310R mutation in a three generation Irish family exhibiting symptoms of early onset adRP (Al, normal allele; A2, mutant allele). Affected and unaffected individuals are indicated by filled and open symbols, respectively. FIGURE 3. Segregation
310 338
Allele frequency Normals A1:0.49 A20.51 A1:0.47 A2:0.53 A1:0.41 A20.59
Heterozygosity
Patients 0.42 0.58 0.55 0.45 0.30 0.70
(%I 49.98 49.82 48.38
"Al, normal allele; A2, mutant allele.
FIGURE 4. Dot-blots showing the inheritance of the three missense mutations in a subset of the control individuals used in this study. A1 and A2 refer to the normal and mutant ASOs for the E304Q mutation, respectively. A3 and A4 refer to the normal and mutant ASOs for the K310R mutation while A5 and A6 refer to the normal and mutant ASOs for the G338D mutation.
mouse, rat, and human cDNA sequences with 82.74% conservation at the nucleotide level. E304 was previously completely conserved among all species sequenced to date (human, mouse, rat, and bovine). K310 is conserved between human, mouse, and cow, but not rat. Interestingly, the residue in cow, mouse, and rat cDNA sequences corresponding to G338 in humans is an aspartate.
Therefore the G338D mutation observed here increases the conservation between species. Little previous interspecies variation has been observed at the nucleotides which we have found to be polymorphic in humans. Moreover we have sequenced 12 bovine samples from six different breeds, mice from six different laboratory strains, and a single rat DNA, all of which showed no
246
JORDAN ET AL. TABLE 4.
Summaw of Base Changes Which Occurred in the Third Exon of the Human RDS Gene" Other species
Position
Change
Mouse
Base
Codon
Base
Amino acid
1150 1169 1253
304
GAGjCAG AAGjAGG GGCjGAC
E304Q K310R G338D
310 338
"The amino acids at the same positions in
cow,
MALLKVKFDaKKRVKLAQGLWLMNWFSVLAGIIIFSLGLFLK1ELRKRSDWI"SESHFV ...... KFDQKKRVKLAQGLWLMNWFSVLAGIIIFGLGLFLKIELRKRSDVMNNSESHFV MALLKVKFDQKKRVKLAaGLh"WLSVLAGIVLFSLGLFLK1ELRKRSEWI"SESHFV MliLLKVKFDQKKRVKLAQGLWLMNWLSVLAGIVLFSLGLFLKIELRKRSD~DNSESH~
HUMRDS BOVPER MUSRDS RATRDS
PNSLIGMGVLSCVFNSLAGKICYDALDPAKYARWKPWLKPYLAICVLFNIILFLV~CCF PNSLIGVGVLSCVFNSLAGKICYDRLDPAKYAKWKPWLKP~KPYLAVCVLF~FLV~CCF PNSLIGVGVLSCVFNSLAGKICYDALDPAKYAKWKP~KPYLAVCIFFNVILFLVALCCF PNSLIGVGVLSCVFNSLAGKICYD~DPAKYAKWKPWLKLYLAVCVFF~ILFLVALCCF
HUMRDS BOVPER MUSRDS RATRDS
LLRGSLENTLGQGLKNGMKYYRDTDTPGRCFMKKTIDMLPFRDWFEIQWI
**ff*tf*t*ttt*ttttt,**ft**__*~****.*****t**,*.(**.*ttt
....................
LLRGSLESTLAHGLKNGMKFYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI LLRGSLESTLAYGLKNGMKYYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI LLRGSLESTLAYGLKNGMKYYRDTDTPGRCFMKKTIDMLQIEFKCCG~GFRDWFEIQWI * * I * * * * , * * , ................................................
HUMRDS BOVPER MUSRDS RATRDS
HUMRDS BOVPER MUSRDS RATRDS
EELHLklI.RCCRAATI.NYY p?l-W.;X~.FITl !.IWLFB.'SITAi!PF! HTALE^VSlll'E!)I E
HUMRDS BOVPER MUSRDS RATRDS
CESEGWLLENSVSETWKAFLESFKKLGKSNQVEAEAADAGQAPEAG
...... ........ ... ... . .... ..... .. ... . . . ... .
*.*fffff,11,ff*.*.fff
Rat
GAGlE AAGK GACID
GAGlE AAGIK GAUD
GAGIE AATN GACID
mouse, and rat are also included.
HUMRDS BOVPER MUSRDS RATRDS
**ft*f,*t*******t***************,f*ff**
Bovine (base/amino acid)
1....,......,,,.*~..~..
FIGURE5. Amino acid sequence alignment of the human RDS, bovine peripherin, mouse and rat rds cDNA sequences using the CLUSTAL program (Higgins and Sharp, 1988). The position of the amino acid substitutions in the 3' end of the RDS gene is doubly underlined.
polymorphism. One could speculate that this initial observation might suggest that the human RDS protein is more subject to change in this region than in other species. However, it is noteworthy that the bovine samples and the mice strains used in this analysis are quite inbred, hence reducing the probability of observing polymorphism. Therefore further sequencing of human, bovine, and mouse samples from a range of ethnic populations would be required to lend weight to this speculation. Technical advancements which now enable rapid amplification and sequencing of genes from multiple individuals may reveal further polymorphisms resulting in amino acid substitutions in the normal population. The amino acid coding sequence and functional relationship of proteins may be more flexible than first envisaged (Lau and Dill, 1990). Missense mutations have also been observed in other proteins
which do not result in a pathological phenotype. Evidence from a study of 95 independent missense mutations in the Factor IX gene correlated to a hemophilia B phenotype suggests that mammalian sequence homology is not sufficient to define essential conserved residues (Bottema et al., 1991). It is proposed that in the Factor IX gene approximately 60% of the residues act as spacers to maintain side chain interaction and virtually no missense mutation in these residues will cause disease. Indeed a missense mutation has been observed in Factor IX (H257Y) which does not cause hemophilia B in a male (Montandon et al., 1990). Moreover from a survey of orthinine 6-aminotransferase (OAT) mutations in gyrate atrophy patients, one mutation L437F, at the extreme C terminus of the protein, had no disernible effect on OAT activity (Brody e t al., 1992). The three new mutations which result in neutral polymorphisms observed in the RDS gene will aid in future mapping of retinal hereditary disorders. They allow us to speculate that the degree of conservation of amino acids within humans may be less than first envisaged, particularly as the three amino acids appear to be well conserved between mouse, rat, and bovine species. Finally they also allow a brief glimpse of the flexibility of the RDS protein's amino acid sequence in tolerating alterations without a resulting loss of function. This may aid in an overall understanding of the role of the RDS protein in the maintenance of the disc rim structure of the rod photoreceptor cell.
ACKNOWLEDGMENTS The authors thank the United States RP Foundation, the George Gund Foundation, the Wellcome Trust, RP Ireland Fighting Blindness, the British RP Association, the Health Research Board of Ireland, Bioresearch Ireland, and the Concerted Action and Science Programmes of the Commision of the European Communities for their continuing support. They also thank Professor A.S. Whitehead and Ronan Loftus for mouse and bovine DNA samples.
POLYMORPHIC VARIATION IN THE HUMAN RDS GENE
REFERENCES Begy C , Bridges CD (1990) Nucleotide and predicted protein sequence of rat retinal degeneration slow (rds). Nucleic Acids Res 18:3058. Blanton SH, Heckenlively JR, Cottinghain AW, Freidman J , Sadler LA, Wagner M, Freidman LH, Daiger SP (1991) Linkage mapping of autosomal dominant retinitis pigmentosa (RPl) to the pericentric region of human chromosome 8. Genomics 113357-869. Bottema CDK, Ketterling RP, Ii S, Yoon H-S, Phillips Ill JA, Sommer SS (1991) Missense mutations and evolutionary conservation of amino acids: Evidence that many of the amino acids in Factor IX function as “spacer” elements. Am J Hum Genet 49:820-838. Brody LC, Mitchell GA, Obie C , Michard J, Steel G, Fontaine G, Robert M-F, Sipila I, Kaiser-Kupfer M, Valle D (1992) Orthinine 6-aminotransferase mutations in gyrate atrophy. J Biol Chem 267:3302-3307. Connell G, Molday RS (1990) Molecular cloning, primary structure and orientation of the vertebrate photoreceptor cell protein peripherin in the rod outer segment disc membrane. Biochemistry 29:4691-4698. Connell G, Bascom R, Molday L, Reid D, Mclnnes RR, Molday RS (1991) Photoreceptor peripherin is the normal product of the gene responsible for retinal degeneration in the rds mouse. Proc Natl Acad Sci USA 88:723-726. Dryja TP, McGee TL, Hahn LB, Cowley GS, Olsson JE, Reichel E, Sandberg MA, Bearson EL (1990) Mutations within the rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. N Engl J Med 323:1302-1307. Farr CJ, Saiki RK, Erlich HA, McCormick F, Marshall CJ (1988) Analysis of RAS gene mutations in acute myeloid leukaemia by polymerase chain reaction and oligonucleotide probes. Proc Natl Acad Sci USA 85:1629-1633. Farrar GJ, McWilliam P, Bradley DG, Kenna P, Lawler M, Sharp EM, Humphries MM, Eiberg H , Conneally PM, Trotfatter JA, Humphries P (1990) Autosomai dominant retinitis pigmentosa: Linkage to rhodopsin and evidence for genetic heterogeneity. Genomics 8:35-40. Farrar GJ, Jordan SA, Kenna P, Humphries MM, Kumar-Singh R, McWilliam P, Allamand V, Sharp E, Humphries P (1991a) Autosomal dominant retinitis pigmentosa: Localisation of a disease gene (RP6) to the short arm of chromosome 6. Genomics 11:870-874. Farrar GJ, Kenna P, Jordan SA, Kumar-Singh R, Humphries MM, Sharp EM, Sheils DM, Humphries P (1991b) A three-base-pair deletion in the peripherin-RDS gene in one form of retinitis pigmentosa. Nature (London) 354:478-480. Farrar GI, Kenna P, Jordan SA, Kumar-Singh R, Humphries MM, Sharp EM, Sheils D, Humphries P (1992) Autosomal dominant retinitis pigmentosa: A novel mutation at the peripherid RDS locus in the original 6p linked pedigree. Genomics (in press).
247
Horn M, Marchese C , Kunisch M, Fusi L, Apfelstedt-Sylla E, Zrenner E, Gal A, Humphries P, Farrar GJ (1992) Deletions in exon 5 of the human rhodopsin gene causing shift in the reading frame and autosomal dominant retinitis pigmentosa. Hum Genet (in press). Higgins DG, Sharp PM (1988) CLUSTAL: A package for performing multiple sequence alignments on a microcomputer. Gene 73:237-244. Jordan SA, Farrar GI, Kenna P, Humphries MM, Kumar-Singh R, Allamand V, Sharp EM, Humphries P (1992) Autosomal dominant retinitis pigmentosa (RP6): Co-segregation of RP6 and the peripherin-RDS locus in a late-onset family of Irish origin. Am J Hum Genet 50634-639. Kajiwara K, Hahn LB, Mukai S, Travis G H , Berson EL, Dryja T P (1991) Mutations in the human retinal degeneration slow gene in autosomal dominant retinitis pigmentosa. Nature (London) 354:480-483. Kimuru M, Ohta T (1974) On some principles governing molecular evolution. Proc Natl Acad Sci USA 71:2848-2852. Lau KF, Dill KA (1990) Theory for protein mutability and biogenesis. Proc Natl Acad Sci USA 87:638-642. Marmor MF, Arden GB, Nilsson SE, Zrenner E (1989) Standard for clinical electroretinography. Arch Ophthalmol 107:816819. McWilliam P, Farrar GI, Kenna P, Bradley DG,Humphries MM, Sharp EM, McConnell DJ, Lawler M, Sheils DM, Ryan C, Stevens K, Daiger SP, Humphries P (1989) Autosomal dominant retinitis pigmentosa (ADRP): Localization of an ADRP gene to the long arm of chromosome 3. Genomics 5:619-622. Montandon A], Green PM, Bentley DR, Ljung R, Nilsson IM, Giannelli F (1990) Two factor IX mutations in the family of an isolated hemophilia B patient: Direct carrier diagnosis by amplification mismatch detection. (AMD). Hum Genet 85:200204. Orita M, Suzuki Y, Sekiya T, Hayashi K (1989) Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5574-879. Travis GH, Brennan MB, Danielson PE, Kozak C A , Sutcliffe JG (1989) Identification of a photoreceptor-specific mRNA encoded by the gene responsible for retinal degeneration slow (rds). Nature (London) 338:70-73. Travis GH, Sutcliffe JC, Bok D (1991a) The retinal degeneration slow (rds) gene product is a photoreceptor disc membraneassociated glycoprotein. Neuron 6:61-70. Travis GH, Christerson L, Danielson PE, Klisak I, Sparkes RS, Hahn LB, Dryja TP, Sutcliffe J G (1991b) The human retinal degeneration slow (RDS) gene: Chromosome assignment and structure of the mRNA Genomics 10:733-739. Yandell DW, Dryja TP (1989) Detection of DNA sequence polymorphisms by enzymatic amplification and direct genomic sequencing. Am J Hum Genet 45:547-555.
