Abstract
Abstract: :
Purpose: Expression analysis is a useful tool to identify large-scale gene expression patterns in a given tissue. Recent advances have concentrated on the use of cDNA microarrays; however, a large amount of data has been previously generated using nylon membrane macroarrays. Our aim was to compare human retinal gene expression qualitatively and quantitatively using these two methods to determine which method would be most useful for future experiments. Methods: GDA ver.1.3 nylon macroarrays (Incyte Genomics, MO) containing 18,376 duplicate-spotted, commercially available ESTs/membrane were hybridized to first strand, 33P-labeled cDNA generated from retina poly (A+) RNA from six individuals (3 old and 3 young). Fluorescently tagged cDNA was generated from total retina RNA of a 71 y/o normal (no ocular disease) individual. Using replicate experiments, the tagged cDNAs were hybridized to glass slide microarray 'chips' containing approximately 9000 human cDNAs from a commercial source (Hs-UniGEM2; Incyte Genomics, MO). Data from the chips and membranes was compared. Results: On the glass slide array, the expression of 179 genes was detected (∼20%), whereas on the nylon macroarray, expressed genes numbered 1100 (∼6%) for the older individuals and 1000-1200 (∼6%) for the younger individuals. A comparison of the absolute levels of gene expression indicates that relevant retinal genes, for example, B-crystallin (Hs.1940) and recoverin (Hs.80539), were expressed on both the chips and membranes. On the chips, additional relevant retinal transcripts such as retinal outer segment membrane protein (Hs.281564) were identified and shown to be expressed at least 2-fold over a universal reference (Universal Human Reference RNA; Stratagene, La Jolla, CA). Conclusions: We have identified a number of genes expressed in human retina. The glass chip microarrays may be more sensitive for detection of gene expression. The use of array technology will be useful in the detection of retinal genes associated with retinal disease. The differences in analytical techniques, dynamic range and idiopathic human variation suggest that more data must be accumulated from a variety of sources, and that community-accepted analytical standards be established.
Keywords: 417 gene/expression • 476 molecular biology • 335 candidate gene analysis