Evaluation of putative reference genes for quantitative real-time PCR normalization inLilium regaleduring development and under stress
Normalization to reference genes is the most common method to avoid bias in real-time quantitative PCR (qPCR), which has been widely used for quantification of gene expression. Despite several studies on gene expression,Lilium, and particularlyL. regale, has not been fully investigated regarding the evaluation of reference genes suitable for normalization. In this study, nine putative reference genes, namely18S rRNA,ACT,BHLH,CLA,CYP,EF1,GAPDH,SANDandTIP41, were analyzed for accurate quantitative PCR normalization at different developmental stages and under different stress conditions, including biotic (Botrytis elliptica), drought, salinity, cold and heat stress. All these genes showed a wide variation in their Cq (quantification Cycle) values, and their stabilities were calculated by geNorm, NormFinder and BestKeeper. In a combination of the results from the three algorithms,BHLHwas superior to the other candidates when all the experimental treatments were analyzed together;CLAandEF1were also recommended by two of the three algorithms. As for specific conditions,EF1under various developmental stages,SANDunder biotic stress,CYP/GAPDHunder drought stress, andTIP41under salinity stress were generally considered suitable. All the algorithms agreed on the stability ofSANDandGAPDHunder cold stress, while onlyCYPwas selected under heat stress by all of them. Additionally, the selection of optimal reference genes under biotic stress was further verified by analyzing the expression level ofLrLOXin leaves inoculated withB. elliptica. Our study would be beneficial for future studies on gene expression and molecular breeding ofLilium.