Identification of stable reference genes for quantitative real-time PCR in human fibroblasts from lymph nodes and synovium

Abstract

Real-time quantitative PCR (RT-qPCR) has emerged as an accurate and widely used technique for measuring gene expression levels. However, its reliability depends on the selection of appropriate reference genes to normalize for sample input. Accordingly, the identification of reference genes characterized by stable expression in cells and conditions of interest is essential for ensuring accurate expression values. To date, no study has specifically identified suitable reference genes for primary human cultured fibroblast-like synoviocytes (FLS) and lymph node stromal cells (LNSCs) within the context of rheumatoid arthritis (RA). These stromal cells play a critical role in the pathogenesis of disease. In this study, we evaluated the suitability of 15 candidate reference genes for normalizing transcript expression in FLS and LNSCs subjected to various in vitro stimuli. We included traditional reference genes often used for transcript normalization in fibroblasts as well as candidate genes identified as suitable reference genes via GeneVestigator analysis of publicly available transcriptomic data. RefFinder algorithms were used to identify the most stable reference genes for transcript normalization across the cell types and different experimental conditions. We determined that the optimal number of reference genes for every experimental condition tested was two; RPLP0 and POLR2G exhibited the greatest stability across different experimental conditions for LNSCs. However, for FLS, we observed greater variability in the most stable reference genes across different experimental conditions. Although POLR2G and TBP emerged as the most stable reference genes under unstimulated conditions, our findings indicated that FLS require distinct reference genes for transcript normalization depending on the specific experimental conditions. Validation of the selected reference genes for normalizing the expression levels of metabolic genes in unstimulated FLS emphasized the importance of prior evaluation of potential reference genes, as arbitrary selection of reference genes could lead to data misinterpretation. This study constitutes the first systematic analysis for selecting optimal reference genes for transcript normalization in different types of human fibroblasts. Our findings emphasize the importance of proper selection of reference genes for each experimental condition separately when applying standard quantitative PCR technology for assessing gene expression levels.