Abstract
BACKGROUND: Preterm birth is the leading cause of neonatal morbidity and mortality, but research efforts in neonatology are complicated due to the unavailability of large volume blood samples. Whole blood assays can be used to overcome this problem by performing both functional and gene expression studies using small amounts of blood. Gene expression studies using RT-qPCR estimate mRNA-levels of target genes normalized to reference genes. The goal of this study was to identify and validate stable reference genes applicable to cord blood samples obtained from developing neonates of different gestational age groups as well as to adult peripheral blood samples. Eight reference gene candidates (ACTB, B2M, GAPDH, GUSB, HPRT, PPIB, RPLP0, RPL13) were analyzed using the three published software algorithms Bestkeeper, GeNorm and NormFinder. RESULTS: A normalization factor consisting of ACTB and PPIB allows for comparative expression analyses of neonatal samples from different gestational age groups. Normalization factors consisting of GAPDH and PPIB or ACTB and GAPDH are suitable when samples from preterm and full-term neonates and adults are compared. However, all candidate reference genes except RPLP0 exhibited significant intergroup gene expression variance and a higher gene expression towards an older age which resulted in a small but statistically significant systematic bias. Systematic analysis of RNA-seq data revealed new reference gene candidates with potentially superior stability. CONCLUSIONS: The current study identified suitable normalization factors and proposed the use of the additional single gene RPLP0 to avoid systematic bias. This combination will enable comparative analyses not only between neonates of different gestational ages, but also between neonates and adults, as it facilitates more detailed investigations of developmental gene expression changes. The use of software algorithms did not prevent unintended systematic bias. This generally highlights the need for careful validation of such results to prevent false interpretation of potential age-dependent changes in gene expression. To identify the most stable reference genes in the future, RNA-seq based global approaches are recommended.