|Molecular Vision 2007;
Received 19 June 2007 | Accepted 23 September 2007 | Published 28 September 2007
Effect of Leu/Met variation at residue 450 on isomerase activity and protein expression of RPE65 and its modulation by variation at other residues
T. Michael Redmond,
Charles H. Weber,
Shirley Yu, Susan
Laboratory of Retinal Cell and Molecular Biology, National Eye Institute, NIH, Bethesda, MD
Correspondence to: T. Michael Redmond, NEI-LRCMB, NIH, Bldg 7, Rm 303, 7 Memorial Drive MSC 0706, Bethesda, MD 20892-0706; Phone: (301) 496-0439; FAX: (301) 402-0750; email: email@example.com
Purpose: RPE65 is the visual cycle retinol isomerase and missense mutations in its gene cause severe retinal dystrophies in man, due to lack of chromophore. While the rate of opsin regeneration in mouse is slower than in man, the methionine (M) variant of mouse RPE65 residue 450 (normally L) is associated with additionally lowered light sensitivity and with resistance to light damage in C57Bl/6 mice, consistent with lowered total activity. We wished to determine how this variant affects RPE65 and if it is modulated by other rodent-specific variations.
Methods: Site-directed mutagenesis was used to make variant constructs in mouse and dog RPE65, which were tested for isomerase activity by transient transfection in 293-F cells.
Results: The isomerase activity of dog RPE65 is slightly higher than mouse. Replacing L at aa450 with M reduces total activity of dog to approximately 70% and mouse to approximately 45% of respective wild type RPE65, and also reduces protein levels of both variants. Replacing K at aa446 in mouse with R, as in other species, reduces total activity in mouse RPE65, whereas the converse case, changing dog aa446 from R to K, increases activity. Exchanges of residues at aa457 and 459 had little overall effect. Human variants at two of these positions, L450R and T457N, had disparate effects, abolishing and augmenting activity, respectively.
Conclusions: Wildtype dog RPE65 is more active than wildtype mouse RPE65, perhaps partially explaining the slower regeneration rate in the mouse. The effect of Met at aa450 is more severe in mouse RPE65 than in dog. The effects of variation at residues 446 (K or R) modulate variation at aa450. The sensitivity of aa450 to change is underscored by the abolition of activity in the pathogenic human L450R mutation. These results suggest that subtle species-specific residue changes may be involved in "tuning" of RPE65 activity to required evolutionary criteria.
Regeneration of visual pigment opsin is ineluctably dependent upon supply of the chromophore 11-cis retinal by the retinoid visual cycle . In brief, 11-cis retinal bound, for example, to photoreceptor rhodopsin is photo-isomerized to all-trans retinal, activating rhodopsin. To regenerate rhodopsin, all-trans retinal is released from opsin, reduced to all-trans-retinol that is in turn transported to the retinal pigment epithelium (RPE) where it is esterified to all-trans retinyl esters. As the substrate for the retinol isomerase , the all-trans retinyl esters are enzymatically isomerized and hydrolyzed to yield 11-cis retinol which is oxidized to 11-cis retinal and returned to the photoreceptors. It has been found that the rate of pigment regeneration shows species-specific variations, with rodents showing slower kinetics of regeneration than humans, other primates and cats .
Recently, the highly preferentially expressed RPE protein RPE65 has been established as the isomerase central to this cycle [4-6]. Prior to this, the importance of RPE65 in chromophore regeneration had been well established. Mice carrying a targeted deletion in the gene for Rpe65 display a biochemical phenotype consisting of extreme chromophore starvation (no rhodopsin) in the photoreceptors concurrent with overaccumulation of all-trans retinyl esters in the RPE . As a result of this, Rpe65 knockout mice are extremely insensitive to light. This insensitivity to light protects Rpe65-/- mice from light damage and establishes rhodopsin as the mediator of light-induced retinal damage . Human mutations in RPE65 are associated with a spectrum of retinal dystrophies ranging from Leber congenital amaurosis/autosomal dominant childhood-onset severe retinal dystrophy (LCA/arCSRD) to later onset conditions described as autosomal dominant retinitis pigmentosa (arRP) [9-12]. Additionally, a mutation identified in the Briard dog gene for RPE65 also causes a severe retinal dystrophy in affected animals [13,14]. A naturally occurring mutation, rd12, in mouse Rpe65 has also been described .
In addition to these null mutants, a hypomorphic variant of mouse Rpe65 has contributed greatly to our understanding of the role of RPE65 in retinal physiology and retinal sensitivity to light-induced retinal damage. This variant was first detected by a quantitative trait locus analysis which linked light-damage resistance in C57Bl/6 albino B6(Cg)-Tyrc-2J/J substrain mice with the Chromosome 3 locus for Rpe65, identifying the M450 variant in light resistant mice, compared to L450 in light-sensitive albino Balb/c mice . Subsequently, resistance to light damage in C57Bl/6 mice was directly associated with the M450 variant and correlated with lowered expression of RPE65 protein in all mice of the C57Bl/6 strain . It was also shown that mice expressing M450 variant accumulate less A2E than L450 mice . Earlier, it was shown that the targeted disruption Rpe65-/- mice accumulated only 10% of the lipofuscin fluorophore, a surrogate for A2E, as control animals expressing the M450 variant . Together these findings confirmed the origin of A2E as a byproduct of the visual cycle . In comparing mice of 5 genotypes (L/L, L/M, M/M, L/-, M/-), it was found that the rhodopsin regeneration rate of these mice is proportional to RPE65 mole quantity, implying that M450 is as "active" as L450, just less of it . The M450 variant of RPE65 was also found to act as a modifier gene for other mouse retinal degenerations which were slowed on the M450 background .
In the present study we examine the effect of variation at aa450 on the expression, activity and predicted structure of RPE65 expressed in the in vitro minimal visual cycle cell culture system. We examine how variation at this residue affects and is affected by variation at neighboring residues in which rodent RPE65 differs from human, dog and cow RPE65s. We show that there are significant changes in predicted secondary structure and folding that are reflected in the observed activity and stability of the variant RPE65/isomerase proteins.
Transient transfection and cell culture
Cell culture methods and transient transfection protocols are as previously published . For any given experiment, 3x107 293-F cells were transfected with 30 mg of pVitro2 plasmid (containing dog RPE65 and bovine cellular retinaldehyde binding protein (CRALBP) open reading frame (ORF)) and 30 mg of pVitro3 plasmid (containing bovine LRAT and bovine RDH5 ORFs) in the presence of 40 ml of 293fectin transfection reagent (Invitrogen, Carlsbad, CA), all in a total volume of 30 ml.
Site-directed mutagenesis of RPE65
Site-directed mutagenesis of RPE65 ORF cloned in pVitro2  was done using QuikChange XL site-directed mutagenesis kits (Stratagene, La Jolla, CA). Oligonucleotide primer pairs used are listed in Table 1. Mutants were verified by sequence analysis (Northwoods DNA, Solway, MN) of DNA minipreps. Validated mutant and wildtype plasmids were purified using Qiagen purification kits (Maxi or Mega format, as appropriate; Qiagen, Valencia, CA).
Retinoid extractions and High performance liquid chromatography
Culture fractions of 19 ml volumes of transfected 293-F cells were centrifuged and cells were harvested and retinoids extracted and saponified by the methods previously described . Isomeric retinols were analyzed on 5 mm Lichrospher normal phase columns (2x250 mm) on an isocratic High performance liquid chromatography (HPLC) system (Agilent 1100 series) following the method of Landers and Olson  as modified by us . Data was analyzed on ChemStation software (Agilent, New Castle, DE).
Cell pellets (about 2x106 cells) from 1 ml culture aliquots were lysed in 200 ml CytoBuster detergent (Novagen, Madison, WI), incubated on ice for 10 min, centrifuged at 13,000 xg for 10 min, and the supernatant harvested for SDS-PAGE analysis. Denatured samples were separated on 12% BisTris NuPage (Invitrogen) gels and electrotransferred to nitrocellulose membranes. Blots were probed with antibodies by standard procedures and developed in color substrate. Primary antibodies used were: rabbit anti-bovine RPE65 antibody (1:4000) and rabbit anti-CRALBP antibody (1:20,000; gift of Dr. John Saari). Secondary antibody used was alkaline phosphatase-conjugated goat anti-rabbit IgG (1:10,000; Novagen). Densitometry of RPE65 and CRALBP bands on immunoblots was performed using Scion Image (release alpha 184.108.40.206) software. RPE65 expression was normalized to CRALBP expression level.
Secondary structure prediction and molecular modeling
Secondary structure modeling was done using the SSpro 4.0 software, based on an ensemble of 11 bidirectional recurrent neural networks  maintained on the SCRATCH server  at University of California at Irvine.
Tertiary structure modeling was done using SWISS-MODEL (version 36.0003). This is a protein structure homology-modeling server and was accessed via the ExPASy web server and/or locally from the program DeepView-Swiss-PdbViewer. The template for modeling RPE65 was the structure of apocarotenal oxygenase (ACO) from Synechocystis .
Sequence analysis of RPE65 and comparison with predicted structure of apocarotenoid oxygenase
RPE65 is a member of the carotenoid oxygenase family, one of which, Synechocystis ACO, has been crystallized. The main feature of the predicted structure of ACO is a seven-bladed propeller arrangement of anti-parallel β-sheets , which is proposed to hold for the other members of this family. Alignment of the RPE65 sequence with ACO, based on this structure, reveals moderate homology in the region (RPE65 aa412-463, ACO aa365-420) containing the presumptive β-sheets 24-27 of ACO , associated with blade 6 (Figure 1A). Independent analysis of this segment of RPE65 by the secondary structure prediction program SSpro predicts the presence of paralogous extended β-sheets in RPE65 at the expected regions and places RPE65 aa450 on presumptive β-sheet 26 in Blade 6 (Figure 1B). Blade 6 also contains three additional rodent-specific variations, K446 in β-sheet 26 and I457 and M459 in β-sheet 27 which are, respectively R446, V457, and T459 in other species (Figure 1C). These rodent variations in Blade 6 were thus chosen for further analysis to investigate their possible effect on the activity of both L450 and M450 RPE65 of mouse and dog.
Effect of mutation of residue 450 on RPE65 activity and expression
For all experiments, the activities are a combined phenotypic effect of each mutation on enzymatic activity and stability as if each were expressed as a homologous allele in vivo. We mutated residue 450 from leucine to methionine in both mouse and dog RPE65 and transiently transfected 293 cells as described. Activities and protein levels are given as precentage of wildtype in any given experiment. We noted that dog RPE65 wildtype L450 was somewhat more active than that of the mouse RPE65 wildtype L450 (1.22±0.18, n=14) for equivalent transfection. Isomerase activity was reduced in the L450M mutants of both species (Figure 2 and Table 2), although more severely in the mouse (44% of wt) than in the dog (71% of wt). Protein levels were likewise reduced to 55-60% of wildtype in both the L450M mutants. However, in the mouse L450M there may have also been change in the intrinsic activity of RPE65, as the reduction in protein level was not as great as the reduction in activity.
Molecular modeling was used to examine the effect of exchanging Met for Leu at residue 450. The leucine side-chain is not predicted to interact with any other residue (Figure 3A). However, there is predicted supernumerary hydrogen-bonding between the Met sidechain and two residues, T454 and E456 (Figure 3B). It is possible that the extra rigidity implied by this bonding affects both stability and activity of RPE65. Secondary structure analysis of strand 26 predicts that the substitution of Met for Leu would result in the conversion of the β-sheet into an α-helix (Figure 3C), supporting the destabilizing effect of this change on RPE65 structure.
Effect of Arg/Lys interchange at residue 446 on RPE65 activity and expression
Residue 446 of RPE65 is Arg in most species but is Lys in mouse and rat. To determine the effect of this conservative substitution on RPE65, we exchanged Arg for Lys in the mouse wildtype and L450M mutant and likewise, Lys for Arg in the dog wildtype and L450M mutant (Table 3). In general, constructs with K446 had enhanced isomerase activity compared with those with R446; the mouse L450M mutant was the only deviation from this pattern in that R446 activity was not changed compared with the K446 activity in this construct. In the L450M variant of the dog RPE65, the R446K substitution almost completely augments the loss due to the aa450 mutation.
Molecular modeling of mutants with the R446 and K446 substitutions predicted hydrogen bond formation of the R446 side-chain with the side-chain of Q414 at the beginning of Blade 6 and with the backbone of L447 (Figure 4A,B). In contrast, the side-chain of K446 is predicted to extend out toward the surface of RPE65 and form no hydrogen bonds with other residues (Figure 4C,D). Of additional interest is the prediction that the sidechains of Q414 and E456 (a residue interacting with M450) appear to be hydrogen-bonded, which suggests that the rigidity imparted by both R446 and M450 acts as a "brake" on RPE65 activity and stability. Variation at aa446 had little or no additional effect on predicted secondary structure (not shown).
Effect of interchanges of residues 457 and 459 on RPE65 activity and expression
Rodent RPE65 also differs from other species/taxa in the replacement of Ile for Thr at residue 457 and Met for Val at residue 459. We made the appropriate changes in the mouse and dog RPE65, both wildtype and L450M mutants, for testing in the 293 cell transient transfection system (Table 4). Although modest changes in isomerase activity were seen for both Ile/Thr interchanges and Met/Val interchanges, they were not significant and no pattern was observed. The slight variations in protein content in these mutants only reflected the activity changes (data not shown). Thus, these taxon-specific variations do not appear to impact either isomerase activity or stability of RPE65.
Molecular modelling of these residues in Blade 6 predicted that the sidechains of all these residues would project to the surface of RPE65 (data not shown). Thus, the small variations in activity seen may reflect more intermolecular interactions with other cellular components than internal structure/activity effects.
Effect of human mutations/variations at residues 450 and 457 on RPE65 activity and expression
Analysis of human RPE65 from retinal dystrophy patients reveals the presence of variants at two of the residues under consideration in this study, 450 and 457. The L450R alteration is associated with severe retinal dystrophy , whereas the T457N alteration was predicted to be a rare neutral variant . These variants were tested for changes in isomerase activity and expression in the dog construct in our transient transfection system (Figure 4). The L450R mutant severely reduced the activity and expression of RPE65, whereas the T457N mutant somewhat enhanced isomerase activity similarly to that of the T457I mutant (Figure 5A). The L450R mutant showed significantly lower level of RPE65 protein, whereas the T457N mutant was comparable in RPE65 protein content (Figure 5B).
Molecular modelling based on the ACO crystal structure was used to predict the effect of the L450R mutation on the local folding of RPE65. The plot predicts steric hindrance of the Arg sidechain with L383 on Blade 5 as well as a supernumerary hydrogen bond with Y431 (Figure 6A). This is in strong contrast to the lack of hydrogen bonding of the L450 wildtype and significantly more extensive than the bonding predicted in the M450 mutant. In contrast, T457N change is predicted to have little effect. When the effect on secondary structure is calculated, it was predicted that L450R would result in the conversion of the β-sheet 26 into an α-helix (Figure 6C). T457N was not predicted to change β-sheet 27 (Figure 6C).
The initiating rationale for this study was to examine the effect of the mouse L450M variant in RPE65, correlated with resistance to light damage , using site-directed mutagenesis of both mouse and dog RPE65 constructs. Though C57B1/6 mice harbor this hypomorphic mutation, they are clearly far from being blind/light-insensitive. The question can be posed: Is RPE65 as active as it can be? Is it instead more reasonable that RPE65 is "tuned" to match the particular regeneration requirement of a given species ? The advantage of faster regeneration has to be weighed against the disadvantage of a higher susceptibility to light damage  and greater A2E accumulation [18,19], both of which are linked to retinal degeneration [20,29].
We have shown that the M450 variant shows both lower protein expression and lower overall activity in transfections with our in vitro 293 cell culture system of the visual cycle. The effect was more pronounced in the mouse RPE65 (about 2/3 reduction) than in the dog RPE65 (about 1/3 reduction) and may also involve lower intrinsic activity. When we compared dog with mouse, the M450 protein levels decreased approximately the same amount but the isomerase activity decreased more in the mouse. Thus, our data suggest that it is both a decrease in isomerase activity and a decrease in the level of protein that is responsible for lower isomerase activity in M450 variant RPE65, especially in mouse. The M450 side-chain is predicted to hydrogen bond with other residues in Blade 6, whereas the L450 does not. Secondary structure analysis predicts disruption or loss of the extended beta strand number 26. Thus this mutation could affect secondary structure of the β-strand and/or the flexibility of the blade and, consequently, the stability and activity of the protein. On the other hand, it is not clear whether increasing rigidity is necessarily an adverse effect in β-propeller proteins (such as ACO and RPE65 are predicted to be) that are already known to have high structural rigidity .
Although the effect of the M450 variant on RPE65 is proposed to be one of reduction in stability of the RPE65 proteins, we have not directly measured this, though we plan to address this issue in future work. An alternative possibility is that the presence of the codon for methionine at aa450 affects translation of the mRNA species containing it, perhaps by some manifestation of leaky scanning with or without reinitiation .
Why is the effect of methionine more severe in the mouse than in the dog? One possibility is that the three other variations in predicted β-strand 26 and 27 between rodent RPE65 and dog and human RPE65s (at aa 446, 457 and 459) modify the effect of methionine at 450 in the mouse. The rationale for this is that β-strands 25, 26 and 27 are predicted to sequentially interact in anti-parallel hairpin conformation and subtle variation in one may affect interaction with the others. The residue with the most significant effect was aa446, Arg in most species but Lys in rat and mouse. The mutant R446K in dog increased RPE65 activity in both wildtype and L450M forms, whereas the comparable mutant K446R in mouse reduced wildtype activity but had little effect on the L450M mutant. Although Arg/Lys is a conservative change, the predicted effect on structure is that the Arg side-chain will bond with other residues in Blade 6, including Q414 at the N-terminal portion of the blade, whereas Lys is not predicted to form any side-chain bonds. The prediction that R446 will constrain Blade 6 in the dog wildtype (L450) could be the explanation for the lack of as great effect on the dog M450 mutant, since the blade would be expected to be already more constrained than in the mouse wildtype RPE65 (K446). It is possible that the R to K change in rodents evolved as a 'fine-tuning' alteration to compensate for other activity-reducing variations outside of Blade 6.
Overall, all β-strand 26 mutants studied (except the extreme mutation L450R) had moderate changes in transfected cells compared to wildtype. These changes could be attributed to changes in the secondary and/or tertiary structure of the blade, contributing to effects on stability as well as isomerase activity. Proper folding of this, and other blades, is crucial for orientation of the catalytic histidines and glutamates of RPE65.
The other two variants in Blade 6 on β-strand 27, Ile/Thr at 457 and Met/Val at 459 had no significant effect on activity of either mouse or dog RPE65s. As all their side-chains are predicted to project outward to the surface, they are more likely to be involved in interactions with other proteins/components than in the structural integrity and isomerase activity of RPE65.
Further support for this model of RPE65 is given by the results of the human mutations tested. The pathogenic L450R mutation  essentially abolishes isomerase activity in transfected cells and greatly reduces protein expression. This is a more extreme effect on a sensitive residue than changing it to methionine. The effect of L450R may be due to secondary structure changes converting a predicted β-sheet to an α-helix and/or to steric hindrance between L450R and other residues. In contrast, the mutation T457N , a predicted neutral variant found in cis with a presumed null mutation (1060delA, originally 1114delA) , does not have a negative effect and indeed gives rise to an increase in isomerase activity, compared to wildtype, in cells transfected with it. In the human, no phenotypic change is evident as any possible effect of T457N is negated by the null effect of its partner mutation in cis.
These findings may have some implications for gene replacement therapy. Is delivery of the most active form necessary, or is simply 'good enough' sufficient? Should a higher activity variant be used for rescue? In this regard, treatment of C57Bl/6 mice carrying the rd12 mutation  with an AAV construct carrying a normal human RPE65 gave restoration of visual function comparable to normal C57Bl/6 mice . As noted earlier, C57B1/6 mice harbor the hypomorphic M450 allele, making them less sensitive to light though far from being blind. The treated rd12 mice received a much more active RPE65 gene (human; L450) than what was lost by mutation (mouse; M450), perhaps explaining the excellent level of rescue . In contrast, even considering difference of species, therapy of affected Briard dogs with a canine wildtype RPE65 vector was less effective . Perhaps use of a "hypermorphic" replacement is the standard for rescue of individuals with a pair of null mutant alleles. This would have to be balanced against the risk of localized light damage and greater A2E accumulation, for example, at the locus of expression as discussed above [8,18-20,29]. The dynamic range of visual pigment regeneration, as governed by RPE65 activity, will become clearer as less severe, but still damaging late onset forms of human RPE65-related retinal dystrophy disease are described. In the case of human disease due to such hypomorphic missense mutations, it may be possible to forestall progression of disease by use of alternate strategies to enhance residual RPE65 activity. Understanding how these mutations affect RPE65 structure and stability will be invaluable in this regard.
We wish to thank Dr. John Saari for the generous gift of the rabbit anti-CRALBP antibody. This research was supported by the NEI Intramural Program.
1. Lamb TD, Pugh EN Jr. Phototransduction, dark adaptation, and rhodopsin regeneration the proctor lecture. Invest Ophthalmol Vis Sci 2006; 47:5137-52.
2. Moiseyev G, Crouch RK, Goletz P, Oatis J Jr, Redmond TM, Ma JX. Retinyl esters are the substrate for isomerohydrolase. Biochemistry 2003; 42:2229-38.
3. Lamb TD, Pugh EN Jr. Dark adaptation and the retinoid cycle of vision. Prog Retin Eye Res 2004; 23:307-80.
4. Jin M, Li S, Moghrabi WN, Sun H, Travis GH. Rpe65 is the retinoid isomerase in bovine retinal pigment epithelium. Cell 2005; 122:449-59.
5. Redmond TM, Poliakov E, Yu S, Tsai JY, Lu Z, Gentleman S. Mutation of key residues of RPE65 abolishes its enzymatic role as isomerohydrolase in the visual cycle. Proc Natl Acad Sci U S A 2005; 102:13658-63.
6. Moiseyev G, Chen Y, Takahashi Y, Wu BX, Ma JX. RPE65 is the isomerohydrolase in the retinoid visual cycle. Proc Natl Acad Sci U S A 2005; 102:12413-8.
7. Liu SY, Redmond TM. Role of the 3'-untranslated region of RPE65 mRNA in the translational regulation of the RPE65 gene: identification of a specific translation inhibitory element. Arch Biochem Biophys 1998; 357:37-44.
8. Grimm C, Wenzel A, Hafezi F, Yu S, Redmond TM, Reme CE. Protection of Rpe65-deficient mice identifies rhodopsin as a mediator of light-induced retinal degeneration. Nat Genet 2000; 25:63-6.
9. Thompson DA, Gyurus P, Fleischer LL, Bingham EL, McHenry CL, Apfelstedt-Sylla E, Zrenner E, Lorenz B, Richards JE, Jacobson SG, Sieving PA, Gal A. Genetics and phenotypes of RPE65 mutations in inherited retinal degeneration. Invest Ophthalmol Vis Sci 2000; 41:4293-9.
10. Marlhens F, Bareil C, Griffoin JM, Zrenner E, Amalric P, Eliaou C, Liu SY, Harris E, Redmond TM, Arnaud B, Claustres M, Hamel CP. Mutations in RPE65 cause Leber's congenital amaurosis. Nat Genet 1997; 17:139-41.
11. Gu SM, Thompson DA, Srikumari CR, Lorenz B, Finckh U, Nicoletti A, Murthy KR, Rathmann M, Kumaramanickavel G, Denton MJ, Gal A. Mutations in RPE65 cause autosomal recessive childhood-onset severe retinal dystrophy. Nat Genet 1997; 17:194-7.
12. Morimura H, Fishman GA, Grover SA, Fulton AB, Berson EL, Dryja TP. Mutations in the RPE65 gene in patients with autosomal recessive retinitis pigmentosa or leber congenital amaurosis. Proc Natl Acad Sci U S A 1998; 95:3088-93.
13. Veske A, Nilsson SE, Narfstrom K, Gal A. Retinal dystrophy of Swedish briard/briard-beagle dogs is due to a 4-bp deletion in RPE65. Genomics 1999; 57:57-61.
14. Acland GM, Aguirre GD, Ray J, Zhang Q, Aleman TS, Cideciyan AV, Pearce-Kelling SE, Anand V, Zeng Y, Maguire AM, Jacobson SG, Hauswirth WW, Bennett J. Gene therapy restores vision in a canine model of childhood blindness. Nat Genet 2001; 28:92-5.
15. Pang JJ, Chang B, Hawes NL, Hurd RE, Davisson MT, Li J, Noorwez SM, Malhotra R, McDowell JH, Kaushal S, Hauswirth WW, Nusinowitz S, Thompson DA, Heckenlively JR. Retinal degeneration 12 (rd12): a new, spontaneously arising mouse model for human Leber congenital amaurosis (LCA). Mol Vis 2005; 11:152-62 <http://www.molvis.org/molvis/v11/a17/>.
16. Danciger M, Matthes MT, Yasamura D, Akhmedov NB, Rickabaugh T, Gentleman S, Redmond TM, La Vail MM, Farber DB. A QTL on distal chromosome 3 that influences the severity of light-induced damage to mouse photoreceptors. Mamm Genome 2000; 11:422-7.
17. Wenzel A, Reme CE, Williams TP, Hafezi F, Grimm C. The Rpe65 Leu450Met variation increases retinal resistance against light-induced degeneration by slowing rhodopsin regeneration. J Neurosci 2001; 21:53-8.
18. Kim SR, Fishkin N, Kong J, Nakanishi K, Allikmets R, Sparrow JR. Rpe65 Leu450Met variant is associated with reduced levels of the retinal pigment epithelium lipofuscin fluorophores A2E and iso-A2E. Proc Natl Acad Sci U S A 2004; 101:11668-72.
19. Katz ML, Redmond TM. Effect of Rpe65 knockout on accumulation of lipofuscin fluorophores in the retinal pigment epithelium. Invest Ophthalmol Vis Sci 2001; 42:3023-30.
20. Sparrow JR, Fishkin N, Zhou J, Cai B, Jang YP, Krane S, Itagaki Y, Nakanishi K. A2E, a byproduct of the visual cycle. Vision Res 2003; 43:2983-90.
21. Lyubarsky AL, Savchenko AB, Morocco SB, Daniele LL, Redmond TM, Pugh EN Jr. Mole quantity of RPE65 and its productivity in the generation of 11-cis-retinal from retinyl esters in the living mouse eye. Biochemistry 2005; 44:9880-8.
22. Samardzija M, Wenzel A, Naash M, Reme CE, Grimm C. Rpe65 as a modifier gene for inherited retinal degeneration. Eur J Neurosci 2006; 23:1028-34.
23. Landers GM, Olson JA. Rapid, simultaneous determination of isomers of retinal, retinal oxime and retinol by high-performance liquid chromatography. J Chromatogr 1988; 438:383-92.
24. Pollastri G, Przybylski D, Rost B, Baldi P. Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles. Proteins 2002; 47:228-35.
25. Cheng J, Randall AZ, Sweredoski MJ, Baldi P. SCRATCH: a protein structure and structural feature prediction server. Nucleic Acids Res 2005; 33:W72-6.
26. Kloer DP, Ruch S, Al-Babili S, Beyer P, Schulz GE. The structure of a retinal-forming carotenoid oxygenase. Science 2005; 308:267-9.
27. Wada Y, Nakazawa M, Abe T, Fuse N, Tamai M. Clinical variability of patients associated with gene mutations of visual cycle protein; Arrestin, RPE65 and RDH5 genes. ARVO abstract. Invest Ophthalmol Vis Sci 2000;41:S617. Abstract number 3277.
28. Lorenz B, Gyurus P, Preising M, Bremser D, Gu S, Andrassi M, Gerth C, Gal A. Early-onset severe rod-cone dystrophy in young children with RPE65 mutations. Invest Ophthalmol Vis Sci 2000; 41:2735-42.
29. Reme CE. The dark side of light: rhodopsin and the silent death of vision the proctor lecture. Invest Ophthalmol Vis Sci 2005; 46:2671-82.
30. Fulop V, Jones DT. Beta propellers: structural rigidity and functional diversity. Curr Opin Struct Biol 1999; 9:715-21.
31. Kozak M. Pushing the limits of the scanning mechanism for initiation of translation. Gene 2002; 299:1-34.
32. Pang JJ, Chang B, Kumar A, Nusinowitz S, Noorwez SM, Li J, Rani A, Foster TC, Chiodo VA, Doyle T, Li H, Malhotra R, Teusner JT, McDowell JH, Min SH, Li Q, Kaushal S, Hauswirth WW. Gene therapy restores vision-dependent behavior as well as retinal structure and function in a mouse model of RPE65 Leber congenital amaurosis. Mol Ther 2006; 13:565-72.