The quantitative real-time reverse transcriptase polymerase chain reaction (qRT-PCR) is a sensitive and very efficient technique for quantification of gene expression. However, qRT-PCR relies on accurate normalization of gene expression data, as RNA recovery and cDNA synthesis efficiency might vary from sample to sample. In the present study, six putative reference genes were validated for normalization of gene expression in three different tissues and in white blood cells from pigs experimentally infected with the common respiratory pathogen Actinobacillus pleuropneumoniae. Two dedicated validation programs (geNorm and Normfinder) were used to rank the six reference genes from best to worst. qRT-PCR data for the proinflammatory cytokine IL-6 was normalized using the proposed genes from geNorm and Normfinder as well as the commonly used reference gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). IL-6 expression was quantified in white blood cells, liver, lymph nodes and tonsils from 10 infected pigs and 5 control pigs. After normalization using either geNorm or Normfinder IL-6 was shown to be significantly up-regulated (P <0.05) in all of the tissues from infected animals compared to non-infected control animals with a good agreement of expression differences between the two programs. On the contrary, normalization of IL-6 expression data from blood using GAPDH rendered the difference between infected and non-infected groups non-significant, and resulted in significantly different values compared to geNorm (P = 0.01). Based on these results, we recommend to validate putative reference genes before normalization.
Veterinary Immunology and Immunopathology, 2007, Vol 118, Issue 1-2, p. 140-146