Evaluation of Candidate Reference Genes for Normalization of Quantitative RT-PCR in Switchgrass Under Various Abiotic Stress Conditions

Abstract

Quantitative real-time reverse transcriptase PCR (qRT-PCR) is a sensitive and powerful technique for measuring differential gene expression; however, changes in gene expression induced by abiotic stresses are complex and multifaceted. Therefore, a set of stably expressed reference genes for data normalization is required. Switchgrass (Panicum virgatum L.) is a prime candidate crop for bioenergy production. The expression stability of reference genes in switchgrass, especially under different experimental conditions, is largely unknown. In order to identify the most suitable reference genes for abiotic stress studies in switchgrass, we evaluated 14 candidate genes for their expression stability under drought, high salinity, cold, heat, and waterlogging treatments using the Delta Ct, geNorm, BestKeeper, and NormFinder approaches. Validation of reference genes indicated that the best reference genes should be selected based on the stress treatment. Actin 2 (ACT2), carotenoid-binding protein 20 (CBP20), and Tubulin (TUB) were found to have the highest expression stability to study drought stress. 18S ribosomal RNA1 (18S rRNA1), ACT2, and TUB were the most stably expressed genes under salt stress. Ubiquitin-conjugating enzyme (UBC), TUB, and cyclophilin2 (CYP2) were the most suitable reference genes across cold treatments. Likewise, 18S rRNA1, UBC, and TUB were good reference genes for studying heat stress, while ACT2, 18S rRNA1, and ubiquitin3 (UBQ3) were the top three reference genes under waterlogging treatment. Considering that reference gene expression may vary across switchgrass tissues, ACT2 and 18S ribosomal RNA2 (18S rRNA2) were shown to be the most stably expressed genes in switchgrass leaves and roots, respectively. The highly ranked reference genes that were identified in this study were shown to be capable of detecting subtle differences in the expression rates of other genes. These differences may have been missed if a less suitable reference gene was used.