The retinal pigment epithelium (RPE) plays a pivotal role in the development and function of the outer retina. We are interested in RPE-specific mechanisms, at both the regulatory and functional levels, and we have been studying the function and regulation of RPE65, a gene whose expression is restricted to the RPE, and mutations in which cause severe blindness in humans, known as Leber Congenital Amaurosis 2 (LCA2). Disruption of the RPE-based vitamin A visual cycle blocking regeneration of visual pigment chromophore is the common phenotype shared by humans with RPE65 gene defects (LCA2) and the Rpe65 knockout mouse (overaccumulation of all-trans-retinyl esters and total absence of 11-cis retinal, resulting in extreme insensitivity to light). We have established a catalytic role for RPE65 in the synthesis of 11-cis retinol, identifying it as the long-sought visual cycle isomerohydrolase. In the past year we have made the following progress: A) We continued to study inhibitors of RPE65 that could have future therapeutic benefit. We recently showed (Poliakov et al., Biochemistry 50: 6739-6741, 2011) that RPE65 is inhibited by aromatic lipophilic spin traps such as PBN. Inhibitors of BCMO1 were also studied and a paper was published in this reporting period. We are currently studying other small molecules unrelated to spin traps that we have found to be inhibitors of RPE65, including known inhibitors of other enzymes, as well as inhibitors of oxidative damage. B) We extended our understanding of how RPE65 directs trans to cis isomerization of retinol. Previously, we showed that mutating RPE65 residue F103 preferentially produces 13-cis retinol instead of 11-cis retinol, supporting a carbocation/radical cation mechanism of retinol isomerization. We modeled the substrate-binding cleft of RPE65 to identify residues lining its surface, including three Phe residues (F61, F312 and F526) forming an arch-like arrangement astride the cleft, and Tyr338. Also, F418 sits at the neck of the cleft, lending a bend to the volume enclosed by the cleft. All mutations of F61, F312 and F418 result in severely impaired or inactive enzyme. However, mutation of F526 and Y338, like F103, decreases 11-cis retinol formation, while increasing the 13-cis isomer. Significantly, two of these 3 residues, F103 and Y338, are located on putatively mobile inter-strand loops. We propose that residual densities in the binding cleft of the RPE65 structure represent a post-cleavage snapshot consistent not only with a fatty acid product, as originally modeled, but also an 11-cis retinol product. Substrate docking simulations permit 11-cis or 13-cis retinyl ester binding in this relatively closed cleft, with the latter favored in F103L, F526A and Y338A mutant structures, but prohibit binding of all-trans retinyl ester, suggesting that isomerization occurs early in the temporal sequence, with O-alkyl ester cleavage occurring later. Our analysis specifically favors a radical cation intermediate rather than a carbocation intermediate as the former alternative allows for the early loss of bond order crucial for docking of cis retinyl esters. A paper describing this was published during this reporting period. Further aspects of the complex mechanism of RPE65 are currently being addressed. C) We continued a project to study to establish (or disprove, as the case may be) palmitoylation of RPE65 cysteine(s), a controversial aspect of RPE65 biochemistry. Different groups have used mass spectrometry to definitively establish that RPE65 is palmitoylated, or that it is not. Clearly, only one of these alternatives is true. We are using a bioorthogonal method to determine if RPE65 is acylated by metabolic labeling in a physiologically relevant cell culture model. Existence of labeled cysteine(s) will be established by mass spectrometry of RPE65 peptides. This project was put on temporary hold due to the departure of tthe primary worker on the project, but was resumed at the end of the reporting period. D) We began and completed a project to establish the origin of the vertebrate visual cycle. This question has been somewhat controversial and inadequately addressed. There has been speculation whether more primitive chordates, such as tunicates and cephalochordates, anticipated this feature. We hypothesized that the origin of the vertebrate visual cycle is directly connected to an ancestral carotenoid oxygenase acquiring a new retinyl ester isomerohydrolase function. Our phylogenetic analyses of the RPE65/BCMO and N1pC/P60 (LRAT) superfamilies show that neither RPE65 nor LRAT orthologs occur in tunicates (Ciona) or cephalochordates (Branchiostoma), but occur in Petromyzon marinus (Sea Lamprey), a jawless vertebrate. The closest homologs to RPE65 in Ciona and Branchiostoma lacked predicted functionally diverged residues found in all authentic RPE65s, but lamprey RPE65 contained all of them. We cloned RPE65 and LRATb cDNAs from lamprey RPE and demonstrated appropriate enzymatic activities. We showed that Ciona β-carotene monooxygenase a (BCMOa;previously annotated as RPE65) has carotenoid oxygenase cleavage activity but not RPE65 activity. We verified the presence of RPE65 in lamprey RPE by immunofluorescence microscopy, immunoblot and mass spectrometry. On the basis of these data we concluded that the crucial transition from the typical carotenoid double bond cleavage functionality (BCMO) to the isomerohydrolase functionality (RPE65), coupled with the origin of LRAT, occurred subsequent to divergence of the more primitive chordates (tunicates, etc.) in the last common ancestor of jawless and jawed vertebrates. A manuscript describing these data is currently under revision at PLoS One. E) We previously generated a panel of hypomorphic knock-in mice in the mouse Rpe65 gene by homologous recombination. Following completion of the breeding strategies in previous reporting periods to establish the lines, phenotypic analysis of these mice was begun in earnest this reporting period. We generated a P25L knockin mouse to model the mild phenotype of a homozygous P25L LCA2 patient with well-preserved cone vision (Lorenz et al., IOVS 2008). Cone development and maintenance is highly dependent on an adequate supply of 11-cis retinal (RAL) and suffers, more than rods, when this is absent such as in RPE65 null mutations. Milder human RPE65 missense mutations have better preserved cone function. Also, preserving cone function is a key concern in managing RPE65 retinal dystrophy, and an important objective of RPE65 gene therapy. Existing mouse models of Rpe65 retinal dystrophy (including 2 null and 1 knockin), exhibit early (null) to midstage (R91W) cone loss. Our goal was to establish a knock-in mouse 1) to model milder RPE65 mutations and 2) to determine the lower limit of 11-cis RAL for long-term preservation of cone structure and function. The P25L line had RPE65 mRNA levels identical to wildtype (WT). By light microscopy the retinas of P25L homozygotes were normal at 2 months and 8 months compared to WT. As expected, RPE65 protein levels were significantly lowered on western blot and immunofluorescence in P25L mice compared to WT. Electrophysiological analysis of P25L rod function yielded a- and b-wave responses close to WT, but the kinetics of a-wave recovery were significantly delayed in P25L mice. Cone responses were also close to WT. Importantly, there was no evidence of cone opsin mislocalization in P25L retina at 7 months suggestive of extended cone viability, unlike in the Rpe65 KO where this occurs by 1 month. We previously found that the Rpe65 P25L mutant retains about 8% of WT activity in vitro, yet this was associated with mild disease in a human patient. The P25L knockin mouse appears to replicate this mild disease. Further data and timepoints are currently being collected.
Showing the most recent 10 out of 26 publications
