Avoiding Pitfalls of Internal Controls: Validation of Reference Genes for Analysis by qRT-PCR and Western Blot throughout Rat Retinal Development

<div><h3>Background</h3><p>Housekeeping genes have been commonly used as reference to normalize gene expression and protein content data because of its presumed constitutive expression. In this paper, we challenge the consensual idea that housekeeping genes are reliable controls for expression studies in the retina through the investigation of a panel of reference genes potentially suitable for analysis of different stages of retinal development.</p> <h3>Methodology/Principal Findings</h3><p>We applied statistical tools on combinations of retinal developmental stages to assess the most stable internal controls for quantitative RT-PCR (qRT-PCR). The stability of expression of seven putative reference genes (<em>Actb</em>, <em>B2m</em>, <em>Gapdh</em>, <em>Hprt1</em>, <em>Mapk1</em>, <em>Ppia</em> and <em>Rn18s</em>) was analyzed using geNorm, BestKeeper and Normfinder software. In addition, several housekeeping genes were tested as loading controls for Western blot in the same sample panel, using Image J. Overall, for qRT-PCR the combination of <em>Gapdh</em> and <em>Mapk1</em> showed the highest stability for most experimental sets. <em>Actb</em> was downregulated in more mature stages, while <em>Rn18s</em> and <em>Hprt1</em> showed the highest variability. We normalized the expression of cyclin D1 using various reference genes and demonstrated that spurious results may result from blind selection of internal controls. For Western blot significant variation could be seen among four putative internal controls (β-actin, cyclophilin b, α-tubulin and lamin A/C), while MAPK1 was stably expressed.</p> <h3>Conclusion</h3><p>Putative housekeeping genes exhibit significant variation in both mRNA and protein content during retinal development. Our results showed that distinct combinations of internal controls fit for each experimental set in the case of qRT-PCR and that MAPK1 is a reliable loading control for Western blot. The results indicate that biased study outcomes may follow the use of reference genes without prior validation for qRT-PCR and Western blot.</p>