We are studying RPE-specific mechanisms, at both the regulatory and functional levels, and have been studying the function and regulation of RPE65, the key retinol isomerase enzyme of the visual cycle. Current work is focused on establishing the molecular mechanism of RPE65 catalysis, as well as its regulation and activity in the context of photoreceptor development and in disease. We are also studying the effects of bisretinoid byproducts of the visual cycle (e.g., A2E) on RPE lysosomal metabolism. In the past year we have made the following progress: a) We continued a project to study to investigate 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. Separate approaches are being used to validate the presence or absence of palmitoyl groups: acyl exchange methods, cysteine affinity capture, and mass spectrometry. We have established by multiple methods that palmitoylation of RPE65 occurs. However the level of palmitoylation does not suggest a structural or permanent acylation of RPE65. Rather, it is either dynamic or adventitious. Current work is focused on dissecting the functional relevance of RPE65 palmitoylation. In addition, we continue to conduct site-directed mutagenesis experiments to dissect the role of various key residues of RPE65 in aspects of its catalytic mechanism. b) We continue to study the D477G (a presumptive dominant-acting RPE65 mutation) knock-in mice that we have made by CRISPR/Cas9 technology. The human D477G phenotype has aspects similar to choroideremia, and is quite distinct from the usual recessive RPE65 phenotypes. However, the presumptive dominant acting nature of the human disease is not reproduced in the mouse model (i.e., no phenotype is evident in KI/+ mice that is different from +/+), instead a complex interaction is revealed in KI/KI animals, consistent with a recessive mode of inheritance. Similar to P25L mice, the D477G mice have close to wildtype phenotype in homozygous state when held under regular vivarium (i.e., low light) conditions. Despite very low levels of RPE65 protein, D477G KI/KI mice have close to wildtype levels of 11-cis retinal in photoreceptors, but do have higher than wildtype levels of retinyl esters. To help unravel the complex phenotype, human iPS-derived RPE cell experiments are planned. Further light-stress experiments (similar to experiments done for the P25L model) will be done to better understand the KI/KI phenotype. In addition, experiments are being designed to expose KI/KI and KI/KO mice to higher but moderate levels of light (1000 lux) to further stress the visual cycle. c) We completed a project aimed at resolving the pleiotropic effects of fenretinide on several enzymes (beta-carotene oxygenase 1 (BCO1), stearoyl-CoA desaturase 1 (SCD1), and dihydroceramide delta4-desaturase 1 (DES1) by analyzing the effects of two major physiological metabolites of fenretinide. Fenretinide (N-4-hydroxyphenylretinamide (4-HPR)) is a synthetic retinoid that forms a tight complex with plasma retinol-binding protein (RBP4) and thus interferes with RBP4-transthyretin (TTR) complex formation and retinol transport. Fenretinide also prevents insulin resistance in mice, unrelated to its effect on RBP4-TTR. As fenretinide has been proposed as a therapy for age-related macular degeneration (AMD), its pleiotropic effects are a cause for concern. The effects of two physiological metabolites of fenretinide, N-4-methoxyphenylretinamide (MPR), and 4-oxo-N-(4-hydroxyphenyl)retinamide (3-keto-HPR), and the non-retinoid RBP4 ligand A1120, on BCO1, SCD1 and DES1 activities were determined in ARPE-19 cells. The activities of the three enzymes are inhibited by the parent fenretinide. We found that 3-keto-HPR is more potent than fenretinide for all three enzymes and may mediate most of the in vivo beneficial effects of fenretinide. While MPR does not affect SCD1 and DES1 activity, it is a potent specific inhibitor of BCO1. In contrast, the non-retinoid RBP4-specific ligand A1120 did not affect any of the studied enzymes. We concluded that administration of 3-keto-HPR instead of fenretinide would be preferential if inhibition of SCD1 or DES1 activity is the goal, while MPR is a specific BCO1 modulator. On the other hand, as it does not inhibit any of the enzymes, we confirm that A1120 appears to only mediate effects on RBP4. A manuscript describing these results was published during this reporting period. d) We restarted a project investigating the post-transcriptional regulation of RPE65 expression that occurs in a variety of cell culture systems including primary RPE cell cultures and cell lines such as ARPE-19. We documented this in our original description of the RPE65 cDNA (Hamel et al, JBC, 1993) when we found that RPE65 protein expression decreased to zero in RPE primary cultures by 12 days after explantation, while levels of RPE65 mRNA remained relatively stable, and we hypothesized that it involved a post-transcriptional mechanism. In subsequent experiments (Liu and Redmond, ABB, 1998), we found that the 3' UTR of RPE65 mRNA played a role in this regulation, and contained a putative translation inhibition element (TIE) in the proximal 150 nt of the 3' UTR. More recently, our efforts to link this putative TIE to possible miRNA-mediated regulation were inconclusive. Our current efforts are directed towards elucidating whether the regulation is due to association of RPE65 mRNA with RNA-binding proteins, protecting the mRNA but sequestering it from ribosomal translation. We are using a number of approaches to address this question: protein binding to synthetic RNA, RNA pulldown, and density gradient fractionation of cellular RNA. Candidate proteins have been identified and are undergoing characterization. e) In ongoing bioinformatics studies (in collaboration with NCBI) of chordate genomes into the evolutionary origins of RPE65, we discovered a new subfamily of carotenoid cleavage oxygenases (CCOs) in metazoans, the BCO2-like (BCOL) clade. These were found to be present in cephalochordate (including lancelets), nematode, and molluscan taxa. Distinctively, BCOL proteins lack the conserved PDPC(K) motif found in all previously characterized metazoan BCO proteins (e.g., BCO1, BCO2 and RPE65 in vertebrates, and numerous paralogs in other taxa). Phylogenetic analysis of CCOs in all kingdoms of life confirmed that the BCOL enzymes are an independent clade of ancient origin. We cloned one of the predicted lancelet BCOL proteins and analyzed it for carotenoid cleavage activity in a bacterial carotenoid expression system. We found it had activity similar to lancelet BCO2 proteins, although with a preference for cis-isomers. Our docking predictions correlated well with the cis-favored activity. The extensive expansions of the new animal BCOL family in some groups (e.g., lancelets) suggests that the carotenoid cleavage oxygenase superfamily has evolved in the extremely high turnover fashion: the numerous losses and duplications of this family are likely to reflect complex regulation processes during development and interactions with the environment. These findings also serve to provide a rationale for the evolution of the important BCO-related outlier RPE65 retinol isomerase, an enzyme that does not utilize carotenoids as substrate or perform double-bond cleavage. A manuscript describing these results was submitted late in this reporting period and is currently under review.
Showing the most recent 10 out of 26 publications
