Candidate reference genes for quantitative real-time PCR (qPCR) assays during development of a diet-related enteropathy in Atlantic salmon (Salmo salar L.) and the potential pitfalls of uncritical use of normalization software tools

The use of reference genes as internal controls is commonly accepted as the most appropriate normalization strategy in quantitative real-time PCR (qPCR) assays. However, there is increasing evidence that expression of many putative reference genes may be regulated by the experimental conditions, and thereby may result in an inaccurate or incorrect quantification of target gene mRNA expression. The aim of this study was to quantitatively evaluate commonly used reference genes for their suitability as a normalization factor for gene expression analyses in the intestine during development of a soybean meal (SBM)-induced intestinal inflammation (enteropathy) in Atlantic salmon. The software applications geNorm, NormFinder and BestKeeper were used to rank eight reference genes according to their stability across 80 samples from a feeding trial with sequential sampling at 10 time points following initiation of SBM exposure. Additionally, we propose a novel statistical model for estimation and ranking of reference gene stability, based on the coefficient of variation (CV) and the Fisher test. ACTB, EF1A, G6PDH and RPS20 mRNA levels displayed a time-dependent induction during development of the enteropathy. In contrast, 18S, GAPDH, RNAPOLII and HPRTI were more stably expressed during the experiment. Overall, a combination of GAPDH, RNAPOLII and HPRTI is recommended as an internal normalization factor in qPCR assays of the distal intestine of Atlantic salmon with SBM-induced enteropathy. Furthermore, we demonstrate that ignoring underlying assumptions made by normalization software may result in an inaccurate or even completely incorrect conclusion on the selection of the best reference gene(s).