Validation of microarray data by quantitative real-time RT-PCR

For further verification of the transcript profiling results, selected genes were analyzed by quantitative real time RT-PCR with cDNAs prepared from the identical RNA samples used for microarray target preparations. For these analyses, eleven genes associated with gene ontology categories discussed above were selected, including genes involved in lipid metabolism (UPC2, OAF1, RSB1 and INO4), sulfur metabolism (MET3), response to drug (AZR1 and PDR3), mitochondrial functions (RTG1 and RTG3) and cell wall-related functions (RLM1 and BAG7). The results (Figure (Figure3)3) are in agreement with the microarray experiments, with ten genes showing up-regulation and one gene (MET3) showing down-regulation in response to pterostilbene treatment. A few discrepancies in fold-change values between real-time RT-PCR and microarray analyses could be attributed to technical differences between the two methods. For example, the genes RSB1, AZR1, and BAG7 showed a higher level of induction in the microarray data compared to the real-time RT-PCR data. This discrepancy could be attributed to the fact that the Affymetrix software assigned an "absent" call to these genes in the untreated sample – thus the signal values assigned were artifactually low in the untreated sample compared to the treated sample, making the fold-change values very high [18]. Other discrepancies could be attributed to the greater dynamic range of real-time RT-PCR compared to microarray analysis.