Improved methodologies for molecular detection and quantification of viruses in food

Abstract: Foodborne viruses such as norovirus, hepatitis A virus and hepatitis E virus cause a high burden of disease worldwide. Reverse transcription (RT) quantitative real-time PCR (qPCR) is the current standard method for monitoring viral contamination in the food chain. However, quantitative detection of viruses in food is challenging and RT-qPCR has several limitations, for example in terms of biased quantification and high variability of results. Therefore, there is a high need for further developments to provide reliable data for official controls, risk assessments and surveillance studies.The aim of this thesis was to develop, improve and validate methods for molecular detection and quantification of viruses in food. Particular emphasis was placed on evaluating the usefulness of a new technique, RT droplet digital PCR (ddPCR), for food virological applications. In addition, many foodborne viruses have high sequence variability, which makes assay development for PCR-based methods time-consuming and error-prone. Another important focus was therefore to simplify and improve the design process for such assays.Five articles form the basis for this work. In Paper I, we validate and evaluate RT-ddPCR for quantitative detection of noroviruses in oysters. In Paper II we show that (RT)-ddPCR can provide less biased quantification of viruses with high sequence variability compared to (RT)-qPCR. In Paper III we develop and validate a new improved assay for hepatitis A virus in food. In Paper IV we present a new tool for the design of (RT)-PCR assays for viruses with high sequence variability. In the last study, Paper V, we optimise and validate a method for quantitative detection of hepatitis E virus in pork sausages.In addition, through a combined analysis of the validation data from Papers I, III and V, we show that RT-qPCR performs somewhat better in qualitative detection, but that RT-ddPCR is superior to RT-qPCR in quantitative detection. Furthermore, we demonstrate through comparisons with data from Poisson distributions that we achieve almost ideal precision in quantification with RT-ddPCR. In summary, this work presents methodological improvements for quantitative detection of the three most important foodborne viruses in high-risk foods. I hope that such methods will help us to better understand the transmission routes and epidemiology of foodborne viruses and reduce the burden of foodborne diseases.