The analysis of normal and induced variation in gene expression is important for understanding the molecular mechanism of disease and the effects of environmental stress. The advent of microarray technology has provided the means to study expression of large numbers of genes in parallel. In our study, we present a gene expression study, the array development methodology, implementation of custom software for quantifying expression ratios and the expression findings. Our gene expression study employs cDNA from genes of interest (called targets). These are immobilized onto a solid silicon support using an amino modification chemistry which resists strong washing and hybridization procedures. Pools of mRNA from tissues of interest (called probes) are labeled using multiple fluorochromes, hybridized to the array, and the relative amount of fluorescence at each spot is used as a measure of the difference in gene expression. Fluorescence probe labeling strategies were validated for specificity, sensitivity and reproducibility using FITC and Cy3. Multi-color labeling is advantageous for comparing the relative abundance of mRNA from two biological samples within the same array. For probe concentrations ranging from 20 nanograms to 10 picograms, we have demonstrated linear responses for up to 1000-fold differences in fluorescence intensities. Custom image processing software performs automated background subtraction, grid detection, spot detection and quantitation. Custom perl scripts are used to ratio and normalise data resulting from the image analysis. First, synthetic images were used to test the accuracy of the analysis, and then a known dilution series was prepared and arrayed onto slides to test the entire system.

These tested methods were finally applied to examine the biological significance of expression variation of DNA repair genes in different tissues. We are using cDNA microarrays containing genes involved in double strand break repair, nucleotide excision repair, base excision repair, direct reversal of damage, mismatch repair, cell cycle regulation, apoptosis and damage response. We have identified several differentially expressed genes that may be important for damage response and meiosis.