Public-transcriptome-database-assisted selection and validation of reliable reference genes for qRT-PCR in rice

Accurate quantitative reverse transcription PCR (qRT-PCR) requires reliable reference genes whose expression does not vary in different tissues and developmental stages. However, few reliable reference genes are available for qRT-PCR in rice (Oryza sativa). Here, we established an effective strategy for identifying novel reference genes (NRGs) for reliable normalization of qRT-PCR data in various rice organs and developmental stages. We selected candidate NRGs using the Information Commons for Rice Database and confirmed their expression in Rice Expression Profile Database (RiceXPro) data. Genes with low variation (<2.5 cycle quantification) across tissues and developmental stages, and little fluctuation in expression in heatmaps from RiceXPro data were considered stable NRGs. To validate this strategy, we selected 11 candidate NRGs and calculated their expression stability in different spatio-temporal conditions using five programs, and compared these genes with five established reference genes (ERGs). Only one of the ERGs (UBQ5) was reliable and 10 of the candidate NRGs were more stable than the four remaining ERGs. Therefore, public transcriptomic databases are useful for identifying NRGs. We selected two NRGs, UFC1 (Homolog of UFM1-Conjugating Enzyme 1) and FhaB (Homolog of Adhesin FhaB) for qRT-PCR analysis in rice; their homologs might be suitable for other monocot plants.