Host-pathogen transcriptomics: Trypanosoma cruzi as a model for studying RNA contamination.

Cellular infection assays constitute essential tools to understand host-pathogen interactions, particularly for intracellular microorganisms that are produced in cell lines are needed to propagate the microorganism. In this work, we demonstrate that RNA derived from Vero cells is an important contaminant to consider in order to avoid false positive results in transcriptomic experiments. We study the cross-contamination on a Trypanosoma cruzi cell infection model, the etiological agent of Chagas disease. We implemented the most frequently used trypanosome-purification protocols and, for all of them, we detected RNAs derived from Vero cells in trypomastigote extracts. For some of the protocols we also detected Vero RNAs in infected human cells. We also found this type of contamination in microarray experiments of human samples infected with T. cruzi. Concerning Illumina RNA-Seq data, we found that the contamination with Vero cells is probably introducing spurious results. Finally, we recommend a protocol to purify trypomastigotes, which showed a high percentage of trypomastigote recovery and the absence of Vero contamination in infected human samples. Avoiding this type of contamination should be an important factor to consider during experimental design, in order to minimize false positive results in transcriptomic studies as well as RNA contamination in vaccine production. SIGNIFICANCE: Transcriptomic studies are widely used to understand host-pathogen interactions. When the pathogen is an intracellular microorganism, an additional mammalian cell system can be needed to propagate it. In this work we demonstrate that pathogens purified from infected monolayers can carry RNAs from these mammalian cells, and that this ambient RNA contamination is probably producing false positive results in subsequent transcriptomic studies performed with qRT-PCR, microarrays or Next Generation Sequencing.