Abstract
Relative gene expression data obtained from quantitative RT-PCR (RT-qPCR) experiments are dependent on appropriate normalization to represent true values. It is common to use constitutively expressed endogenous reference genes (RGs) for normalization, but for this strategy to be valid the RGs must be stably expressed across all the tested samples. Here, we have tested 10 common RGs for their expression stability in cerebral cortex from young rats after in utero exposure to thyroid hormone (TH) disrupting compounds. We found that all 10 RGs were stable according to the three algorithms geNorm, NormFinder and BestKeeper. The downstream target gene Pvalb was significantly downregulated in brains from young rats after in utero exposure to propylthiouracil (PTU), a medicinal drug inhibiting TH synthesis. Similar results were obtained regardless of which of the 10 RGs was used for normalization. Another potential gene affected by developmental TH disruption, Dio2, was either not affected, or significantly upregulated about 1.4-fold, depending on which RG was used for normalization. This highlights the importance of carefully selecting correct RGs for normalization and to take into account the sensitivity of the RT-qPCR method when reporting on changes to gene expression that are less than 1.5-fold. For future studies examining relative gene expression in rat cerebral cortex under toxicological conditions, we recommend using a combination of either Rps18/Rpl13a or Rps18/Ubc for normalization, but also continuously monitor any potential regulation of the RGs themselves following alterations to study protocols.