DNA microarray analysis of gene expression profiles in pancreatic carcinoma

Objective To identify new diagnostic markers and drug targets, the gene expression profiles of pancreatic cancer were compared with those of normal pancreatic tissues, utilizing cDNA microarray analysis.Methods cDNA microarray was set up by spotting PCR products of 12 800 human genes onto a specially prepared glass slides. The cDNA probes were prepared by labeling normal and cancer tissue mRNA with Cy3-dUTP and Cy5-dUTP separately through reverse transcription. The mixed probes were then hybridized to the cDNA microarray. The chips were scanned by ScanArray 3000 laser scanner at two wavelengths. The acquired image was analyzed by ImaGene 3.0 software. The intensities of each spot at the two wavelengths represented the signal intensities of Cy3-dUTP and Cy5-dUTP, respectively, the ratio of Cy3 to Cy5 was then computed. To screen differentially expressed genes according to the criterion that the absolute value of natural logarithm of the ratio of Cy5-dUTP to Cy3-dUTP was greater than 0.69.Results Among the six samples investigated, there were 54 genes, which exhibited differential expression, including 32 over-expressed genes (24 already in Genebank), and 22 under-expressed genes (15 already in Genebank) in cancerous tissues. Conclusions Technology of cDNA microarray provides a powerful approach to elucidate tumor-specific gene expression profiles in human pancreatic carcinoma.