Selection and validation of suitable reference genes for mRNA qRT-PCR analysis using somatic embryogenic cultures, floral and vegetative tissues in citrus

Abstract

Accuracy of the quantitative real-time reverse transcription-PCR (qRT-PCR) depends on the stability of the reference gene(s), i.e. housekeeping gene(s) used for data normalization. Recent studies have shown that the expression of common reference genes can vary considerably in certain experimental conditions. However, reference genes of qRT-PCR in fruit trees have not been well identified. In this study, stability of expression of ten potential reference genes in citrus, including CitACT7, CiteIF-1A, CiteIF4α, CitHistone H3, CitHistone H4, CitTUA3, CitTUB8, CitUBL5, CitUBQ1 and CitUBQ14 was assessed. Furthermore, this was validated by qRT-PCR in diverse sets of biological samples, including embryonic callus at seven stages, embryos maintained at three different temperatures, five distinct plant organs, four floral tissues and four stages of flower development. Three distinct statistical algorithms, geNorm, NormFinder and Bestkeeper, were used to evaluate the expression stability of the candidate reference genes. The three outputs were merged by means of a non-weighted unsupervised rank aggregation method. GeNorm was also used to determine both the optimal number and the best combination of reference genes for each experimental set. The expression of CitUBQ1 was the most stable one across the set of all samples, flower developmental stages and somatic embryogenesis process under two conditions i.e. different temperature treatment and normal temperature treatment. CitUBQ14 presented more stable expression in different plant organs and floral tissues. CitHistone H3 was the most unsuitable reference gene in many citrus sample sets. In addition, the relative gene expression profile of citrus receptor-like kinase gene CitSERK1-like was conducted to confirm the validity of the reference genes in this study. Taken together, this study identified the reference genes that are most suitable for normalizing the gene expression data in citrus. These results provide guidelines for the selection of reference gene(s) under different experimental conditions, and will benefit future research on more accurate gene expression studies in a wide variety of samples in citrus.