Validation of Candidate Reference Genes for the Accurate Normalization of Real-Time Quantitative RT-PCR Data in Rice During Seed Development

Rice seed, a natural storage organ for starch and protein, is also an ideal bioreactor for the production of valuable proteins. Increasingly, studies focused on rice have tried to determine the functions of its genes and also to improve its yield and quality. Real-time RT-PCR is the best available choice at present for gene expression analysis due to its accuracy, sensitivity, and reproducibility. The right choice of reference genes for normalization, however, is a critical precondition for reliable results. In this study, the expression stabilities of nine commonly used housekeeping genes in rice were carefully assessed using the software geNorm. Our results showed that eIF-4a and ACT1 were the most suitable reference genes among almost all the tested samples from two rice varieties, including different temporal and spatial-specific tissues, especially in seeds at different developmental stages. In contrast, 18S and 25S rRNAs, two common reference genes, were found to have the least stable expression. Moreover, it is necessary to use multiple suitable reference genes together for normalization to get a more reliable result in temporal and spatial expression analysis during rice seed development. The validated reference genes were further relied when used to quantify the expression of several genes of interest during rice seed development.