Selection and Evaluation of mRNA and miRNA Reference Genes for Expression Studies (qPCR) in Archived Formalin-Fixed and Paraffin-Embedded (FFPE) Colon Samples of DSS-Induced Colitis Mouse Model

SIMPLE SUMMARY: Quantitative real-time polymerase chain reaction is a widely used molecular technique in human and animal diagnostics and research. However, its accuracy and reliability are significantly affected by the stability of the selected reference genes. We evaluated the expression stability of reference genes in archived formalin-fixed and paraffin-embedded (FFPE) samples of C57BL/6JOlaHsd mice (males and females) from colitis (DSS-induced) experiments. Our results imply that the FFPE procedure does not change the ranking of stability of reference genes obtained in fresh tissues, which suggests that FFPE samples can be used as a source for further reference gene identification processes. Since miRNA is more stable than mRNA, archived FFPE samples offer an invaluable source for miRNA research (i.e., retrospective analysis, identification of biomarkers, and evaluation of robust reference genes) without any additional use of animals. In addition, multivariate analysis showed that the histological picture is an important factor affecting the expression levels of target genes. Thus, FFPE samples as a source for RNA analyses have an added advantage not only in terms of reduction of animal use, but also in terms of accuracy of results (expression in correlation to the histological picture). ABSTRACT: The choice of appropriate reference genes is essential for correctly interpreting qPCR data and results. However, the majority of animal studies use a single reference gene without any prior evaluation. Therefore, many qPCR results from rodent studies can be misleading, affecting not only reproducibility but also translatability. In this study, the expression stability of reference genes for mRNA and miRNA in archived FFPE samples of 117 C57BL/6JOlaHsd mice (males and females) from 9 colitis experiments (dextran sulfate sodium; DSS) were evaluated and their expression analysis was performed. In addition, we investigated whether normalization reduced/neutralized the influence of inter/intra-experimental factors which we systematically included in the study. Two statistical algorithms (NormFinder and Bestkeeper) were used to determine the stability of reference genes. Multivariate analysis was made to evaluate the influence of normalization with different reference genes on target gene expression in regard to inter/intra-experimental factors. Results show that archived FFPE samples are a reliable source of RNA and imply that the FFPE procedure does not change the ranking of stability of reference genes obtained in fresh tissues. Multivariate analysis showed that the histological picture is an important factor affecting the expression levels of target genes.