Normalization strategies for real-time expression data in Chlamydia trachomatis V. Borges a , R. Ferreira a , A. Nunes a , P. Nogueira b , M.J. Borrego a , J.P. Gomes a, ⁎ a Department of Infectious Diseases, National Institute of Health, Av. Padre Cruz, 1649-016-Lisbon, Portugal b Department of Epidemiology, National Institute of Health, Av. Padre Cruz, 1649-016-Lisbon, Portugal abstract article info Article history: Received 20 April 2010 Received in revised form 24 June 2010 Accepted 29 June 2010 Available online 6 July 2010 Keywords: Chlamydia trachomatis Expression Normalization Housekeeping Endogenous control Genomic DNA Chlamydia trachomatis is a widespread obligate intracellular pathogen genetically non-tractable for which transcriptomics is a fundamental tool to better understand its biology. However, the suitability of endogenous controls for normalization of transcriptomic data in this bacterium still needs validation. We aimed to assess the stability of 10 genes for their potential use as endogenous controls in real-time quantitative PCR assays at both normal and stress (D-cycloserine treatment) growth conditions throughout the developmental cycle of three C. trachomatis strains with different tissue tropism. Normalization was performed by real-time absolute quantification of the bacterial genomes. We also tested the applicability of two widely used softwares (geNorm and Normfinder) to our data. For all strains, we found that 16SrRNA was the most stably expressed gene throughout the chlamydial normal developmental cycle, which indicates its potential use as endogenous control in relative expression assays. However, it was highly unstable under D- cycloserine treatment (where oppA_2 was top-ranked), suggesting prudence when using ribosomal genes in expression experiments involving stress conditions. The geNorm and Normfinder algorithms revealed contrasting results and seem inappropriate for the selected pool of genes. Considering the multiplicity of experimental conditions, there should be an in loco validation of endogenous controls, where 16SrRNA appears to be in the front line. Alternatively, normalization of expression data against genomic DNA, which is less influenced by experimental constraints that are especially relevant for intracellular organisms, likely constitutes a good option. Moreover, the number of genomes also seems to be less subject to variation than expression of endogenous controls when working under stress conditions. The present study constitutes the first evaluation of putative endogenous controls for real-time expression assays in C. trachomatis. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Chlamydia trachomatis is a widespread obligate intracellular bacte- rium, where serovars A–C are the causative agents of trachoma, serovars D–K infect the urogenital tract and the invasive serovars L1–L3 cause lymphogranuloma venereum (LGV). It has a unique biphasic develop- mental cycle (~40–45 h) that alternates between an infectious form (the elementary body, EB) and a metabolically active form (the reticulate body, RB) which replicates within a host cell inclusion. The normal developmental cycle can be disturbed by several stress conditions such as nutrient starvation, temperature, host immune response or antibiotic treatment, which are known to affect transcrip- tion (Ouellette et al., 2006; Mäurer et al., 2007). Gene expression in Chlamydia has been mostly performed using real-time quantitative reverse transcription PCR (RT-qPCR), which presents well-known advantages over traditional mRNA quantifica- tion methods (Vandecasteele et al., 2002; Huggett et al., 2005). However, the requirement of a proper normalization strategy is probably the major problem when performing relative expression assays, where inappropriate methodologies can lead to inaccurate data and incorrect conclusions (Thellin et al., 1999; Bustin, 2002). To date, numerous strategies for RT-qPCR data normalization have been applied, such as the use of the total RNA mass (Bustin, 2002), the number of cells determined by quantitative culture (Vandecasteele et al., 2001) and the use of an external reference of known amount added to the cultures prior to sample processing (Huggett et al., 2005; Johnson et al., 2005). Besides the inherent and well-known disadvan- tages of these three approaches, additional drawbacks limit their application to obligate intracellular organisms like C. trachomatis: i) the total RNA that is extracted is composed not only by variable proportions of bacterial rRNA, tRNA, mRNA, small non-coding RNA but also by eukaryotic RNA (Johnson et al., 2005); ii) quantitative cultures have low sensitivity and are difficult to reproduce (Vandecasteele et al., 2002; Wooters et al., 2009); iii) the amount of external reference that enters or stays out the host cell cannot be controlled, hampering the determination of bacteria-specific mRNA. The most commonly employed method involves the use of mRNA of an endogenous control, usually a housekeeping gene, to normalize the target mRNA ensuring that both mRNAs are subjected to exactly the same processing procedure. Thus, differences associated with Journal of Microbiological Methods 82 (2010) 256–264 ⁎ Corresponding author. Tel.: + 315 217 519 241; fax: + 351 217 526 400. E-mail address: j.paulo.gomes@insa.min-saude.pt (J.P. Gomes). 0167-7012/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2010.06.013 Contents lists available at ScienceDirect Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth