Abstract
Apple rust disease caused by Gymnosporangium yamadae is the one of the major threats to the development of the apple industry in China, but the pathogenic molecular mechanism of the disease remains unclear. It is imperative to screen out appropriate reference genes during the interaction between G. yamadae and apple leaves to analyze the gene expression patterns during the pathogenesis of G. yamadae. ACT, EF1, EF2, GAPDH, 40S, 60S, α-TUB, β-TUB and UBCE3 were selected as candidate reference genes based on the transcriptomic dataset of G. yamadae. The expression levels were tested by real-time quantitative PCR during time-course infection of apple leaves and the expression stabilities were evaluated by △Ct method as well as by three software (NormFinder, geNorm and BestKeeper) and one web-based analysis software (RefFinder). The expression stability of the candidate reference genes was further validated by using the effector candidate gene Cluster-3395.48660 as the target gene in RT-qPCR. According to the results by △Ct and BestKeeper, 40S, EF2 and EF1 were the most stable reference genes, while EF1, EF2 and GAPDH were the most stable reference genes based on the NormFinder analysis result. The geNorm recommended the most stable genes EF1, EF2 and α-TUB as reference genes. Comprehensive analysis results of the RefFinder indicated EF1, EF2 and α-TUB were the most suitable genes. Based on these results, EF1, EF2 and α-TUB were considered as reference genes for analyzing the gene expression profiles of Cluster-3395.48660 in different infection stages, and the results were consistent with the transcriptome data. All the results suggest that the combination of EF1, EF2 and α-TUB proved to be acceptable reference genes during the interaction between G. yamadae and apple leaves.