The often low abundance and uneven distribution of bacterial foodborne pathogens in food samples makes rapid, sensitive and specific detection of these pathogens an enduring challenge. Selective concentration of a target organism is often required to bring pathogen cells to a detectable level for subsequent analyses. For example, culture enrichment is still the most widely used method in spite of its lengthy process (i.e., more than 12 hours) that is intrinsically restricted in speed by the length of cell cycle.

Alternative concentration approaches to culture enrichment have been developed. For instance, immunomagnetic separation (IMS) has been applied to certain food matrices. In addition to enriching bacterial cells, amplification of bacterial genomes, for instance, through multiple displacement amplification (MDA), has been shown to facilitate molecular detection of difficult-to-culture pathogens from clinical samples.

We have recently shown that by concatenating IMS, MDA and real-time PCR, we could detect low levels of Salmonella enterica (~10 CFU/g) in raw chicken breast without culture enrichment (Hyeon and Deng, Food Microbiology, 2017)

In this study, we aimed to develop a workflow that consists of 1) specific concentration of Salmonella cells, 2) high-efficiency amplification of bacterial genomes, and 3) molecular detection of Salmonella. The modular design of this workflow allows customized combination of different sample processing and detection methods to meet specific detection need for speed, sensitivity or specificity. In particular, we will demonstrate that the use of a novel magnetoelastic biosensor (Chai et al., Journal of Food Protection, 2012) and whole genome sequencing (WGS) can allow strain-level identification of Salmonella from raw chicken with no or shortened culture enrichment.