Purpose: The purpose of this study was to determine the possibility of achieving sufficient depth of coverage and substantial recovery of WG to assemble and identify foodborne pathogens like Salmonella from spice metagenomic datasets using WGS analysis tools.
Methods: White pepper (WP) samples implicated in a 2009 Salmonella outbreak were processed following a modified BAM protocol. Qiacube extracted DNA was collected from samples throughout enrichment. Shotgun metagenomic sequencing was performed on the MiSeq using the Nextera kit. Salmonella-specific reads were collected by mapping and assembled using CLC Workbench. Assemblies were also generated from WGS of Salmonella isolated from WP in our laboratory and from outbreak-associated NCBI archived reads. Core gene MLA and RAST were used to compare the metagenomic assembled genome (M-A) with those of the isolated colony (IC-A) and NCBI entries (NCBI-A). Additionally, Metaphlan2 analysis was performed. A similar approach was applied to spice samples spiked with Salmonella.
Results: Ability to assemble WG from metagenomic datasets by applying WGS analysis tools was demonstrated using a naturally contaminated spice sample. Salmonella was detected by microbiology and bioinformatics at various stages of enrichment. Metaphlan2 results provided relative abundance of Salmonella in the samples. The M-A was virtually identical to IC-A and to the NCBI-A. WG recovery from spiked samples was proportional to the abundance of Salmonella detected in the sample by Metaphlan2. Enrichment conditions and initial Salmonella load appear to affect WG recovery from metagenomics data sets using this approach.
Significance: Good quality WG assemblies directly from EMCs may provide rapid, high resolution molecular epidemiological data for detection and source tracking of pathogens in food.