Molecular Vision 2004; 10:265-71 Received 11 August 2003 | Accepted 15 March 2004 | Published 8 April 2004
©2004 Molecular Vision
A three base pair deletion encoding the amino acid (lysine-270) in the α-cone transducin gene Ana Luisa Piña,1 Uwe Baumert,2 Magali Loyer,1 Robert K. Koenekoop1 1
McGill Ocular Genetics Laboratory, Montreal Children’s Hospital Research Institute, McGill University, Montreal, Canada; 2Department of Orthodontics, Craniofacial Genetics, University Clinic of Regensburg, Germany Purpose: Cone transducin plays an important role in interacting with the cone photoreceptor visual pigments and activating the cGMP-dependent phosphodiesterase. The human gene for the α-subunit of cone transducin (GNAT2) has been cloned and characterized. Recently achromatopsia has been associated with mutations in this gene. Cone and cone-rod dystrophies are a genetically heterogeneous group of photoreceptor diseases, in which mutations of a single gene may cause a variety of phenotypes. In this study we tested the hypothesis that mutations in GNAT2 cause cone-rod degeneration (CRD). Methods: PCR-SSCP and heteroduplex analysis combined with automated sequencing was used for mutation detection in GNAT2 in 13 independent pedigrees with CRD. We used co-segregation analysis to establish or reject causation, when possible. Molecular computer modeling was utilized to examine the possible consequences of mutations onto GNAT2 protein structure. Results: We found a novel 3 base-pair deletion, predicted to cause the loss of a highly conserved lysine at position 270 (K270del) in a French-Canadian CRD pedigree. We detected this deletion in a CRD proband, but also in his unaffected son, the proband's unaffected father and the proband's unaffected brother. However, we did not find this defect in 12 other CRD pedigrees, nor in 100 normal, culturally matched chromosomes. According to literature and our molecular computer modeling, the K270 plays an important role in securing the guanine ring in the nucleotide binding cleft of the molecule and in creating a salt bridge between the helical and GTPase domains of GNAT2. However, the K270del in GNAT2 does not appear to have extensive consequences to the structure and the function of the GNAT2. Apparently, there is a compensatory effect of lysine (K-271), which forms a hydrogen bond with the N1 ring nitrogen substituting for the loss of the lysine at position 270. Conclusions: We detected a deletion of a highly conserved lysine at codon 270 in a critical functional area of the α-cone transducin molecule. The co-segregation analysis showed that the deletion is not co-inherited with the disease phenotype and therefore is not the disease causing mutation. Apparently the function of this molecule is not altered by this mutation, due to a compensatory effect of aminoacid 271. Taken together, the presence of this deletion in healthy individuals, and our protein modeling results, predict that α-cone transducin molecule is able to tolerate structurally and functionally the K270del.
the first steps of the phototransduction. GNAT1 and GNAT2, the genes which encode for the human rod and cone transducin α-subunits, respectively, have been cloned and characterized [1,2]. A missense mutation (G38R) in GNAT1 has been identified in the Nougaret form of congenital stationary night blindness [3], and a homologous phenotype has been described for the corresponding knock-out-mouse model [4]. For GNAT2, recently the finding of 7 mutations has been reported, all causing frameshifts and leading to a premature translation termination in achromatopsia patients [5,6]. The involvement of GNAT2 in achromatopsia would suggest, that mutations in this gene would affect cones alone rather than rods. However, interactions between cones and rods during normal and diseased states are not well known. In disorders in which photoreceptors are involved, these interactions should be taken into account. Such is the case of cone and cone-rod dystrophies, a group of photoreceptor diseases, characterized by clinical psychophysical and electrophysiological findings which suggest that the first abnormality is in the cones, but for some cases, progress to a generalized retinal degen-
Guanine nucleotide-binding proteins (G-proteins) play a major role in signal transduction in a wide variety of cellular communication systems ranging from visual phototransduction to cellular proliferation by coupling extracellular signals to a number of intracellular second messenger pathways. The Gprotein-linked receptors, cone and rod opsins, receive an extracellular signal, a photon of light, which is then transmitted and amplified by the intracellular α-subunits of rod and cone transducin molecules. This phenomenon activates the phototransduction cascade by stimulating the enzyme cGMPphosphodiesterase, which results in closure of the cGMP-gated channels and, subsequently, in membrane hyperpolarization. Thus, transducin, a heterotrimeric G-protein, mediates one of Correspondence to: Dr. Robert Koenekoop, Ophthalmology, Montreal Children’s Hospital, 2300 Tupper, Montreal, PQ, Canada, H3H 1P3; Phone: (514) 412-4400, ext. 22891; FAX: (514) 412-4443; email:
[email protected] Dr. Piña is now at the Department of Neurosurgery, University Clinic of Regensburg, Germany. 265
Molecular Vision 2004; 10:265-71
©2004 Molecular Vision
samples were denatured at 94 °C (30 s), annealed at proper temperatures (Table 1) for 30 s, and elongation took place at 72 °C (90 s) in a DNA thermocycler (Perkin-Elmer/Cetus) for a total of 32 cycles. PCR products were then diluted using 6 µl of the PCR reaction, 4µl of water and 10 µl of formamide loading dye solution (95% formamide, 0.1% bromophenol blue, 0.1% xylene cyanol) and subjected to SSCP analysis [14]. Two sets of conditions were used for SSCP analysis: 6% polyacrylamide with either 5% or 10% glycerol. Prior to loading, the samples were denatured for 5 min at 95 °C and cooled down on ice for 5 min. Gels were run at room temperature for 2.5 h at 40 W (5% glycerol) or at 10 mA for 16 h (10% glycerol). The gels were dried on a gel dryer and then directly exposed to x-ray film overnight for DNA fragment pattern visualization. When a positive SSCP result was found, the PCR-amplified DNA was further characterized by direct sequencing using the ABI prism 377 DNA sequencer with BigDye terminator (Entelechon, Regensburg, Germany). Each probe was at least tested twice in both directions. For the heteroduplex analysis, the preparation of the PCR was identical to the one described above. The PCR product for the heteroduplex analysis was heated at 95 °C (5 min), cooled at 75 °C (5 min), 55 °C (5 min), and then 37 °C (5 min), run on 12% PAGE gels for 16 h at 250 V. Computational analysis: To evaluate GNAT2 sequence conservation, multiple sequence alignments were constructed using the ClustalW software [15], and the following protein sequences: murine, bovine, and human rod α-transducins (SwissProt sequences P20612/GBT1_MOUSE, P04695/ GBT1_BOVINE, and P11488/GBT1_HUMAN) and cone αtransducins from the same species (SwissProt sequences P50149/GBT2_MOUSE, P04696/GBT2_BOVINE, and P19087/GBT2_HUMAN). The alignments were visualized
eration has been observed [7]. Genetically these diseases are heterogeneous, in which mutations of a single gene may cause a variety of phenotypes. As examples, mutations in CNGA3, the gene encoding for the cone α-subunit of the cGMP gated channel, can cause complete and incomplete achromatopsia and progressive cone dystrophy [8,9]; or the three base pair deletion of codon 153 or 154 of the Peripherin/RDS gene can produce three different phenotypes within the same family [10]. Moreover, cone and cone-rod dystrophies, display such a high variability, that even patients with the same apparent phenotype, characterized by standard methods, do not necessarily share the same gene defect. This indicates, that refinement of classification and diagnosis methods should be developed for this group of ophthalmic genetic diseases [11]. In this study, we tested the hypothesis that mutations in GNAT2 could also cause cone-rod degeneration (CRD). In the process we found a deletion in this gene, located in an important functional region of the molecule. Here, we describe the structural and functional impact of this deletion onto the protein molecule. METHODS Patient identification and pedigrees: French Canadian patients with cone-rod degenerations (CRD) from 13 independent pedigrees were analyzed in detail as previously reported [12]. Venous blood was taken for DNA analysis after informed consent was obtained, according to Montreal Children’s Hospital Ethical Review Board guidelines. Genotyping: DNA was extracted from venous blood according to standard procedures [13]. Intronic primer pairs for each of the eight exons of GNAT2 were synthesized by phosphoramidite technology on a Gene Assembler Plus (Pharmacia Biotech). The PCR-SSCP screening was as follows. A 25 µl amplification mixture contained 100 ng of genomic DNA, 12.5µM each of dCTP, dGTP and dTTP, 6.25µM dATP, 6.25µCi α-35S-dATP (12.5 mCi/ml), 0.6 U Taq polymerase (Gibco), and 50 ng of each intronic primer. The PCR TABLE 1. PCR PRIMERS USED FOR GENOMIC AMPLIFICATION OF THE EXONS OF HUMAN GBT2 Exon ---1 2 3 4 5 6 7 8
Primer Sequence ------------------------------f: 5'-AGGAAACCTTACGAGAGAGGATTAG r: 5'-CAGGAAAGTAGGAAGGACCCCTGCC f: 5'-TGGAAGGCTGAGGAGAACTGGTGGA r: 5'-GGCTGTACTGACTTCTTCACCCTAT f: 5'-CCCTTAGGTCTGGTTACTCAGGTCC r: 5'-TCTTATTGCTAGCCTTTCTGCTTCC f: 5'-CCTCCTGGAAAGCTATTACTCCTGT r: 5'-GACTGGCTAGAAGATTGCTTAAGCA f: 5'-TCTCTTAGCCTCGTCTGTGGAAAAT r: 5'-CCTTTTCAGGATTCCACACGTGTAT f: 5'-GGCATATAGGTCGTATGTTGGGCAT r: 5'-AGGGTCATGGGGAGAACTTGTTCTA f: 5'-GGTAGATATTCTATAAGCCAAATCT r: 5'-AATTGCCAGTCTCTACTAAAAGGCA f: 5'-GCAACTAACAAGGGTTCATTTATTC r: 5'-AATTACCCAGATTCCAAGCCTGTTT
Annealing Temp (°C) --------66 66 66 68 58 68 62 62
Location --------159-183 362-386 2521-2545 2649-2673 2846-2870 3073-3097 4239-4263 4482-4506 6592-6616 6806-6830 6929-6953 7144-7168 8924-8948 9163-9187 9484-9508 9760-9784
Figure 1. A French-Canadian pedigree with one affected individual for cone-rod degeneration. Phenotype and allelic state of the family members are shown. The proband and the only affected individual is II-1. The allelic state is denoted with +/+ (wild type) and +/- (carriers of heterozygous mutation K270del). The gel underneath displays the heteroduplexes analysis that shows the K270del mutation, and its distribution in the family.
Shown are the gene-specific sequences for the forward (f) and reverse (r) primers, their annealing temperatures, and their location within the α-cone transducin gene. The sequence numbering is from GenBank accession number Z18859. 266
Molecular Vision 2004; 10:265-71
with ESPript [16], and the legend was added with Adobe Illustrator. A BLAST similarity search with the human GNAT2 sequence (SwissProt P19087) against the PDB database was used to select the appropriate templates for homology modeling with Swiss-MODEL [17]. Several different coordinate files of the bovine GNAT1 (PDB:1tnd) were chosen [18]. The quality of the resulting models was controlled using ProCheck [19]. Three different models were generated: wildtype and mutant human GNAT2 as well as a mutated bovine GNAT1(based on SwissProt P04695), that contains the mutation described in this paper at the corresponding position (K266del). Using the template coordinate file (PDB:1tnd) the ligands GTPγS and Mg2+ were introduced with SwissPDBView in all models generated [17,20]. VMD [21] was used to render the models, and we used LigPlot [22] to produce schematic drawings of the ligand contacts to the GNAT2 protein backbone and to study the consequences of the identified lysine 270 deletion. RESULTS Clinical descriptions: The proband (II-1; Figure 1) is from a French-Canadian family, and was presented with photophobia and loss of central vision at age 28. He claimed never to have seen well, had trouble with vision in school, and color vision at age 25. On examination he had no nystagmus, 20/ 200 vision OU, high myopia (-18.50+3.00x90° OD and 17.25+2.75x90° OS), central scotomas with normal peripheral isopters (V4e) on Goldmann visual fields, a protan/deutan (red-green) axis of confusion on D15 testing, and subtle atrophic lesions of the fovea, with arteriolar narrowing and myopic degeneration on retinal examination. The cone b-wave
©2004 Molecular Vision
amplitudes were dramatically reduced and delayed, while the rod b-wave amplitudes were mildy reduced. Based on the clinical phenotype and test results, the diagnosis of CRD was made. The proband’s mother (I-2), father (I-1), two brothers (II-2, II-3), two sisters (II-4, II-5), and his 4 year old son (III-1) all had normal eye examinations, ERGs of the proband’s father, and both brothers were completely normal. Molecular characterization: The SSCP analysis revealed a variant band of the PCR product of exon 7 of GNAT2 (Figure 2). Subsequent nucleotide sequencing confirmed a heterozygous deletion of basepairs AAG at positions 9066-9068 (Figure 2). Loss of nucleotides AAG is predicted to result in a deletion of lysine at position 270 (K270del). Heteroduplex analysis of the entire pedigree confirmed the three basepair deletion in the proband, his father, one of his brothers, and the proband’s son, and excluded this deletion in the proband’s mother, two sisters, and brother II-3, (Figure 1). A detailed SSCP analysis of the other exons and subsequent nucleotide sequencing of all 8 exons of the proband was performed and no further mutation was found. Computational analysis: Figure 3 shows that the overall sequence conservation is very high between α-rod and α-cone transducins (79-82%) and also between the different species (man, mouse, cow) within these two groups (Figure 3; Group 1: 99%; Group 2: 95-96%). To get insight into the structural consequences of the K270del mutation, homology modeling was performed using the Swiss-Model service. Based on the experimentally determined structure of bovine GNAT-1, the structure of the human GNAT2 (GBT2_HUMAN) was generated (Figure 4A). The polypeptide is organized into two domains; a catalytic
Figure 2. Molecular genetic analysis of the K270del mutation in II-1 and II-4. On the left, an exon 7 SSCP is shown. Variant II-1 represents the mutant, and II-4 represents the normal state, in this case the non-carrier sister of the proband. On the right, results are presented from direct automated sequencing: the top panel represents the non-carrier or wild type individual (II-4), and the lower panel one of the carriers, in this case the proband (II-1) showing the heterozygous three base pair (AAG) deletion. The sequence numbering is from GenBank accession number Z18859. 267
Molecular Vision 2004; 10:265-71
©2004 Molecular Vision
Figure 3. Mutiple species sequence alignment of α-rod and αcone transducins. Multiple sequence alignment of α-rod (group 1) and α-cone transducins (group 2) from mouse, cow, and man. The amino acids are denoted by their single-letter codes. Gaps in the aligned sequence are indicated by dots. Highly conserved alignment regions are red. The sequence numbering shown below the alignment is in accordance with group 2. Above the alignment the secondary structure elements of GBT1_BOVUS/ PDB1TND are shown (labeling and naming according to Noel [18]; also Figure 4). The amino acids that are involved in GTP binding are labeled as in Noel [18], Figure 4C, and Figure 5. The alignment was drawn with ESPript version 2.0 [16].
268
Molecular Vision 2004; 10:265-71
©2004 Molecular Vision
domain (right) with a structure similar to Ras, and a helical domain (left). The GTP analog GTPγS is bound to the catalytic domain in a cleft between the two domains [18]. The position of the amino acid that is deleted (K270) is shown and lies in the vicinity of the substrate. Using the same technique two mutated structures were modeled: the mutated human GNAT2 (K270del; Figure 4B) and a mutated bovine GNAT1(with the corresponding K266del mutation). Figure 4B shows a projection of the wildtype protein structured onto the mutated human GNAT2 protein structure. Surprisingly, despite the deleted amino acid, the overall structure of the polypeptide remains the same. Since the deleted amino acid plays a crucial role in substrate binding [18], the amino acids involved in these interactions were further investigated. We used a Ligplot analysis of the amino acids involved (Figure 5). A comparison of both wild type structures of bovine GNAT1 (Figure 5A) and human GNAT2 (Figure 5C) showed that they used the same amino acids for substrate binding (alignment in Figure 3). The
Figure 4. Computer generated model of the human α-cone transducin. A: Overall structure composition of GBT2_HUMAN (Genbank accession number P19087). The amino acid core of the protein is shown in gray, helical regions are displayed as cylinders. The GTP analog GTPγS is shown in blue and the deleted amino acid K270 is colored red. B: In this panel, the wild type (yellow) and the mutated protein structures (red) of GBT2_HUMAN were overlaid to show the overall structure integrity. The ligand is shown in blue. C: Three-dimensional representation of the essential amino acids involved in GTP binding in accordance with Figure 5D (wildtype is yellow; mutant is red).
269
Molecular Vision 2004; 10:265-71
©2004 Molecular Vision
acid appeared conserved among several species, and to be in an important functional region of the molecule [18], we chose to study the potential impact of this deletion on GNAT2 protein structure through computer modeling. Our sequence alignment showed that transducin is a highly conserved protein, as all amino acids involved in substrate binding can be found at identical positions in all homologs. This would suggest that missing a key amino acid could result in a mutated protein. Using homology modeling, we evaluated the potential impact of the K270 mutation on GNAT2 protein structure. Based on the high degree of identity both GNAT1 and GNAT2 show nearly identical secondary structure. Especially the amino acids involved in substrate binding [18] appear to be identical and lie at homologous positions within the sequence and the structure. The deletion of the amino acid Lys270 (K270del) leads to a 3-D rearrangement of three amino acids. The guanine ring of GTPγS is held in place through hydrophobic interactions with Lys-270 and Thr-327. We showed that the hydrogen bond that is lost as a result of the K270 deletion is substituted by a hydrogen bond between Lys-271 and the N1 of the guanine ring system. The hydrophobic interaction of Thr-327 is not changed at all. We conclude that the K270 deletion represents a rare polymorphism, and the GNAT2 protein is able to tolerate this deletion structurally and functionally because the next amino acid is also a lysine. Nevertheless, further functional studies should be made in order to test this prediction.
side-chains of Lys-270 (Figure 5C), and Lys-266, respectively, (Figure 5A) make hydrophobic contacts with the guanine ring system of GTPγS. Asp-268 and Asp-272, respectively, stabilize the guanine base through hydrogen bonds from their side chains to the N1 ring nitrogen and exocyclic 2-amine (Figure 5A,C). The deletion of Lys-270 (and Lys-266) cause these contacts to be lost. However, amino acid 271 is also a lysine (K-271), and this second lysine presumably forms a hydrogen bond with the N1 ring nitrogen (Figure 4C; Figure 5B,D) and substitutes for the loss of the lysine at position 270. In this way, it seems that the deletion of Lys270 appears to have little or no consequences for the function of the molecule. DISCUSSION We tested the hypothesis that mutations in the gene for αcone transducin, GNAT2, cause cone-rod degeneration (CRD), and in the process discovered a novel 3 base pair deletion that is predicted to result in the loss of a conserved Lysine at position 270 (K270del) in a French Canadian pedigree. We excluded this deletion from 100 normal control chromosomes and other large screening studies of GNAT2 have not reported this deletion either [5,6,23,24]. Therefore, we initially thought that this mutation might have been pathogenic in this pedigree. Studying further family members, we identified the same deletion in three unaffected relatives, the proband’s father, brother and son and concluded that it represents a polymorphism, because of lack of co-segregation. Because this amino
Figure 5. Ligplot-generated overview of the ligand contacts with the amino acid sidechains. Overview of the ligand contacts with the amino acid sidechains in models of bovine α-rod transducin wildtype (A), bovine K266del mutant (B), human αcone transducin wildtype (C), and human K270del mutant (D).
270
Molecular Vision 2004; 10:265-71
ACKNOWLEDGEMENTS This study was supported by the E. Mildred Kanigsberg Estate Fund and an FRSQ award (970260-103) to RK and a Retinitis Pigmentosa Foundation fellowship to ALP. We are extremely grateful to Drs. R. Gravel and M. Preising for critical reading of this manuscript, and E.-M. Stoerr for technical assistance. REFERENCES 1. Fong SL. Characterization of the human rod transducin alphasubunit gene. Nucleic Acids Res 1992; 20:2865-70. 2. Morris TA, Fong SL. Characterization of the gene encoding human cone transducin alpha-subunit (GNAT2). Genomics 1993; 17:442-8. 3. Dryja TP, Hahn LB, Reboul T, Arnaud B. Missense mutation in the gene encoding the alpha subunit of rod transducin in the Nougaret form of congenital stationary night blindness. Nat Genet 1996; 13:358-60. 4. Calvert PD, Krasnoperova NV, Lyubarsky AL, Isayama T, Nicolo M, Kosaras B, Wong G, Gannon KS, Margolskee RF, Sidman RL, Pugh EN Jr, Makino CL, Lem J. Phototransduction in transgenic mice after targeted deletion of the rod transducin alpha-subunit. Proc Natl Acad Sci U S A 2000; 97:13913-8. Erratum in: Proc Natl Acad Sci U S A 2000; 98:10515. 5. Aligianis IA, Forshew T, Johnson S, Michaelides M, Johnson CA, Trembath RC, Hunt DM, Moore AT, Maher ER. Mapping of a novel locus for achromatopsia (ACHM4) to 1p and identification of a germline mutation in the alpha subunit of cone transducin (GNAT2). J Med Genet 2002; 39:656-60. 6. Kohl S, Baumann B, Rosenberg T, Kellner U, Lorenz B, Vadala M, Jacobson SG, Wissinger B. Mutations in the cone photoreceptor G-protein alpha-subunit gene GNAT2 in patients with achromatopsia. Am J Hum Genet 2002; 71:422-5. 7. Jimenez-Sierra JM, Ogden TE, Van Boemel GB. Inherited retinal diseases: a diagnostic guide. St. Louis; Mosby; 1989. 8. Kohl S, Baumann B, Broghammer M, Jagle H, Sieving P, Kellner U, Spegal R, Anastasi M, Zrenner E, Sharpe LT, Wissinger B. Mutations in the CNGB3 gene encoding the beta-subunit of the cone photoreceptor cGMP-gated channel are responsible for achromatopsia (ACHM3) linked to chromosome 8q21. Hum Mol Genet 2000; 9:2107-16. 9. Wissinger B, Gamer D, Jagle H, Giorda R, Marx T, Mayer S, Tippmann S, Broghammer M, Jurklies B, Rosenberg T, Jacobson SG, Sener EC, Tatlipinar S, Hoyng CB, Castellan C, Bitoun P, Andreasson S, Rudolph G, Kellner U, Lorenz B, Wolff G, Verellen-Dumoulin C, Schwartz M, Cremers FP, Apfelstedt-Sylla E, Zrenner E, Salati R, Sharpe LT, Kohl S. CNGA3 mutations in hereditary cone photoreceptor disorders. Am J Hum Genet 2001; 69:722-37.
©2004 Molecular Vision
10. Weleber RG, Carr RE, Murphey WH, Sheffield VC, Stone EM. Phenotypic variation including retinitis pigmentosa, pattern dystrophy, and fundus flavimaculatus in a single family with a deletion of codon 153 or 154 of the peripherin/RDS gene. Arch Ophthalmol 1993; 111:1531-42. 11. Scholl HP, Kremers J. Alterations of L- and M-cone driven ERGs in cone and cone-rod dystrophies. Vision Res 2003; 43:233344. 12. Koenekoop RK, Fishman GA, Iannaccone A, Ezzeldin H, Ciccarelli ML, Baldi A, Sunness JS, Lotery AJ, Jablonski MM, Pittler SJ, Maumenee I. Electroretinographic abnormalities in parents of patients with Leber congenital amaurosis who have heterozygous GUCY2D mutations. Arch Ophthalmol 2002; 120:1325-30. 13. John SW, Weitzner G, Rozen R, Scriver CR. A rapid procedure for extracting genomic DNA from leukocytes. Nucleic Acids Res 1991; 19:408. 14. Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 1989; 5:874-9. 15. Aiyar A. The use of CLUSTAL W and CLUSTAL X for multiple sequence alignment. Methods Mol Biol 2000; 132:221-41. 16. Gouet P, Courcelle E, Stuart DI, Metoz F. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics 1999; 15:305-8. 17. Guex N, Peitsch MC. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 1997; 18:2714-23. 18. Noel JP, Hamm HE, Sigler PB. The 2.2 A crystal structure of transducin-alpha complexed with GTP gamma S. Nature 1993; 366:654-63. 19. Laskowski RA, MacArthur MW, Moss DS, Thornton JM. PROCHECK: a program to check the stereochemical quality of protein structures. Journal of Applied Crystallography 1993; 26:283-91. 20. Peitsch MC, Wilkins MR, Tonella L, Sanchez JC, Appel RD, Hochstrasser DF. Large-scale protein modelling and integration with the SWISS-PROT and SWISS-2DPAGE databases: the example of Escherichia coli. Electrophoresis 1997; 18:498-501. 21. Prall M. VMD: a graphical tool for the modern chemists. J Comput Chem 2001; 22:132-4. 22. Wallace AC, Laskowski RA, Thornton JM. LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein Eng 1995; 8:127-34. 23. Magovcevic I, Weremowicz S, Morton CC, Fong SL, Berson EL, Dryja TP. Mapping of the human cone transducin alphasubunit (GNAT2) gene to 1p13 and negative mutation analysis in patients with Stargardt disease. Genomics 1995; 25:288-90. 24. Gerber S, Rozet JM, Bonneau D, Souied E, Weissenbach J, Frezal J, Munnich A, Kaplan J. Exclusion of the cone-specific alphasubunit of the transducin gene in Stargardt’s disease. Hum Genet 1995; 95:382-4.
The print version of this article was created on 8 Apr 2004. This reflects all typographical corrections and errata to the article through that date. Details of any changes may be found in the online version of the article. 271
