Identification of reference genes for quantitative RT-PCR analysis of microRNAs and mRNAs in castor bean (Ricinus communis L.) under drought stress

Abstract Quantitative real-time PCR (RT-qPCR) is one of the most powerful and sensitive techniques to the study of gene expression. Several factors influence RT-qPCR performance though, including the stability of the reference genes used for data normalization. While the selection of appropriate reference genes is crucial for accurate and reliable gene expression analysis, no suitable reference genes have been previously identified in castor bean under drought stress. In this study, the expression stability of eleven mRNAs, thirteen microRNAs (miRNAs) and one small nuclear RNA were analyzed in roots and leaves across different levels of water deficit. Three different algorithms were employed to analyze the RT-qPCR data, and the resulting outputs were merged using a non-weighted unsupervised rank aggregation method. Our analysis indicated that the Elongation factor 1-beta (EF1B), Protein phosphatase 2A (PP2A) and ADP-ribosylation factor (ADP) ranked as the best candidates across diverse samples submitted to different levels of drought conditions. EF1B and Glyceraldehyde 3-phosphate dehydrogenase (GAPDH), and EF1B and SKP1/ASK-interacting protein 16 (SKIP16) were found as the most suitable reference genes for expression analysis in roots and leaves, respectively. In addition, miRNAs miR168, miR160 and miR397 were selected as optimal reference genes across all tissues and treatments. miR168 and miR156 were recommended as reference for roots, while miR168 and miR160 were recommended for leaves. Together, our results constitute the first attempt to identify and validate the most suitable reference genes for accurate normalization of gene expression in castor bean under drought stress.