Abstract
Cotton (Gossypium hirsutum) is the most important non-food plant in the world. Studies concerning the fiber quality and plant fitness of cotton at molecular level depend on high sensitive and reproducible gene-expression assays. However, only a few reports have described genes suitable for normalizing gene expression data. In this study, we report for the first time that microRNAs (miRNAs) are reliable reference genes (RGs) for cotton gene expression normalization in quantitative real-time reverse transcription (RT)-PCR. The stability of cotton miRNAs was assayed in root, stem, leaf and flower samples from three different cultivars [FiberMax (FM966), Delta Opal (DO) and Cedro] and under conditions of biotic stress caused by infection with Cotton leafroll dwarf virus (CLRDV). The stability of mRNAs already described as reference genes in cotton was also assessed. The geNorm, NormFinder, BestKeeper and ΔCt algorithms were used to select the best reference genes. In 8 of the 12 sets tested, miRNAs (miR172, 168 and 390) were found to be the best RGs. To validate the best selected RGs, miR159, miR164, miR2118, miR2910, miR3476, GhDCL2 and GhDCL4 expression levels were evaluated under biotic stress conditions, and miR164 and a putative myo-inositol oxigenase gene (GhMIOX) were assessed in leaves and flowers. The RGs selected in this work proved to be excellent reference genes in the two cases studied. Our results support the use of miRNAs as reference genes for miRNA and protein-coding genes.