Identification of candidate genes for human retinal degeneration loci using differentially expressed genes from mouse photoreceptor dystrophy models

PURPOSE: Retinal degeneration (RD) is a complex mechanism that appears to involve many biologic processes including oxidative stress, apoptosis, and cellular remodeling. Currently there are 51 mapped, but not identified, RD human disease loci. METHODS: To assign possible disease genes to RD loci, we have used a comparative genomics procedure that incorporates microarray gene expression data of three independent mouse models for photoreceptor dystrophy (rd1, rd2, and constant light-damage in BALB/c mice), human ortholog data, and databases of known chromosomal locations involved in human RD. Immunohistochemistry and enzyme activity assays were used to further characterize a candidate gene product. RESULTS: Our analysis yielded candidate genes for four mapped, but unsolved, human chromosomal locations and confirmed two previously identified monogenic disease loci for human RD, thus validating our approach. PLA2G7 (phospholipase A2, group VII; PAF-AH, Lp-PLA2), a candidate for a dominant form macular dystrophy (Benign Concentric Annular Macular Dystrophy [BCMAD]), was selected for further study. The PLA2G7 enzyme is known to mediate breakdown of oxidatively damaged phospholipids, a contributor to oxidative stress in the retina. PLA2G7 protein was enriched in mouse photoreceptor inner and outer segments. In the rd1, rd2, and BALB/c mice exposed to constant light, retinal tissue activity levels, but not plasma levels, were significantly reduced at the onset of photoreceptor cell death. CONCLUSIONS: We have shown that this comparative genomics approach verified existing RD genes as well as identified novel RD candidate genes. The results on the characterization of the PLA2G7 protein, one of the novel RD genes, suggests that retinal tissue PLA2G7 levels may constitute an important risk factor for BCMAD. In summary, this reverse mapping approach, using accepted mouse models of human disease and known human RD loci, may prove useful in identifying possible novel disease candidates for RD and may be applicable to other human diseases.