Abstract
Abstract: :
Purpose: To develop a large–scale, high–throughput means to identify and prioritize cone–associated genes and candidate genes for macula and cone–rod dystrophies based on macula–enriched expression using Serial Analysis of Gene Expression (SAGE). Methods: Two human neural retina SAGE libraries were produced from matched 4 mm trephine punches taken from the macula or mid–peripheral retina of five (28–66 year old) donors. The resulting sequence tags from each of the 4 mm human retinal libraries were entered into an electronic database with cDNA microarray expression profiles previously obtained from a subset of the 4 mm–derived retina RNAs used for SAGE and the public SAGE eye libraries synthesized by Cepko et al.. Macula–enriched and cone–enriched expression profiles and expression profiles for genes within macular or cone–rod dystrophy loci and AMD–susceptibility loci were computed. Elevated macular expression of candidate disease genes identified by these large–scale methods of analysis was confirmed by real–time quantitative RT–PCR. Results: High quality SAGE libraries were generated yielding 101,420 total tags and 37,556 unique tags for the macula library, 4MAC, and 108,383 total tags and 39,831 unique tags for the matched mid–peripheral retina library, 4PERI. Macula–enriched candidate genes were identified with higher relative expression in the 4 mm diameter macula (4MAC) versus peripheral library (4PERI) and public 6 mm diameter retinal libraries. The 4MAC/4PERI profiles were also mapped to known macular or cone–rod dystrophy disease regions and AMD–susceptibility loci. Elevated transcript abundance in the macula versus mid peripheral retina for many of these candidates was validated using quantitative RT–PCR. Conclusions: Elevated regional retinal transcript expression profiles described by Cepko et al. are validated and strengthened by the expression profiles derived from multiple matched donors which decreases the effect of inconsistencies introduced by individual variation. Macula–enriched transcripts that were absent or represented at levels too low to be quantitative in a 6mm macula library were identified in quantifiable amounts in the 4mm macula library. Taken as a whole, this entire body of work, which integrates the large–scale methods of cDNA microarray expression profiles and ocular SAGE libraries with known mapping information, should yield a powerful analytical tool to identify and prioritize cone–associated genes and candidate genes for macula and cone–rod dystrophies.
Keywords: macula/fovea • photoreceptors • candidate gene analysis