Quantitative real-time polymerase chain reaction (qPCR) is a widely used method to analyse the gene expression pattern in the reproductive tissues along with detecting gene levels in mutant backgrounds. This technique requires stable reference genes to normalise the expression level of target genes. Nonetheless, a considerable number of publications continue to present qPCR results normalised to a single reference gene and, to our knowledge, no comparative evaluation of multiple reference genes has been carried out in specific reproductive tissues of Arabidopsis thaliana. Herein, we assessed the expression stability levels of ten candidate reference genes (UBC9, ACT7, GAPC-2, RCE1, PP2AA3, TUA2, SAC52, YLS8, SAMDC and HIS3.3) in two conditional sets: one across flower development and the other using inflorescences from different genotypes. The stability analysis was performed using the RefFinder tool, which combines four statistical algorithms (geNorm, NormFinder, BestKeeper and the comparative ΔCt method). Our results showed that RCE1, SAC52 and TUA2 had the most stable expression in different flower developmental stages while YLS8, HIS3.3 and ACT7 were the top-ranking reference genes for normalisation in mutant studies. Furthermore, we validated our results by analysing the expression pattern of genes involved in reproduction and examining the expression of these genes in published mutant backgrounds. Overall, we provided a pool of appropriate reference genes for expression studies in reproductive tissues of A. thaliana, which will facilitate further gene expression studies in this context. More importantly, we presented a framework that will promote a consistent and accurate analysis of gene expression in any scientific field. Simultaneously, we highlighted the relevance of clearly defining and describing the experimental conditions associated with qPCR to improve scientific reproducibility.
Keywords: expression analysis; normalisation; qPCR; reference genes; reproductive tissues.