Purpose: In the C57BL/6J mouse retina, hyperoxia-induced degeneration of photoreceptors shows strong regional variation, beginning at a locus ~0.5 mm inferior to the optic disc. To identify gene expression differences that might underlie this variability in vulnerability, we have used microarray techniques to describe regional (superior-inferior) variations in gene expression in the retina.
Methods: Young adult C57BL/6J mice raised in dim cyclic illumination (12 h at 5 lx and 12 h in darkness) were exposed to hyperoxia (75% oxygen for two weeks). Retinas were collected from hyperoxia-exposed and control animals without fixation and divided into superior and inferior halves. RNA was extracted from each sample, purified, and hybridized to Mouse Gene 1.0 ST arrays (Affymetrix). The consistency of the microarray results was assessed using quantitative PCR for selected genes. Expression data were analyzed to identify genes and ncRNAs whose differential expression between the superior and inferior retina could be associated with relative vulnerability to hyperoxia.
Results: In control retinas, only two genes showed a fold difference in expression >2 between the superior and inferior retina; another 25 showed a fold difference of 1.5-2.0. Of these 27, the functions of six genes, including ventral anterior homeobox containing gene 2 (Vax2) and T-box 5 (Tbox5), are related to parameters of anatomic development and the functions of five are related to sensory perception. Among the latter, short-wave-sensitive cone opsin (Opn1sw) was more strongly expressed in the inferior retina and medium-wave-sensitive cone opsin (Opn1mw) in the superior retina. This is consistent with known differences in S- and M-cone distribution, confirming our separation of retinal regions. The highest fold difference was reported for membrane metalloendopeptidase (Mme), a member from the metallothionein group of cytoprotective proteins. To identify genes whose regulation by hyperoxia was significantly different between the inferior and superior retina, we calculated the "fold margin" (FM, the difference between hyperoxia-induced regulation in the inferior and superior retina) for each gene, and identified genes for which abs(FM) > 0.5. Genes thus identified numbered 112, and included many immune-, cell defense-, and inflammation- related genes.
Conclusions: Gene expression analysis revealed relatively subtle differences between inferior and superior regions of control C57BL/6J retinas, with only 27 genes showing an expression difference >1.5 fold. Among these, genes related to cytoprotection and apoptosis were included, along with genes related to central projections and cone-type differences. After hyperoxia-induced photoreceptor degeneration had begun, the number of genes that showed significant expression differences between the inferior and superior retina more than quadrupled, with genes related to immune processes, defense processes, and inflammation being numerically dominant.