Enumeration of Escherichia coli in ground beef and identification with long-read sequencing

Authors: Counihan, Katrina; Tilman, Shannon

Submitted to: International Association for Food Protection
Publication Type: Abstract Only
Publication Acceptance Date: 3/25/2025
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Introduction: Long-read sequencing can be used to identify foodborne pathogens, but it cannot enumerate the number of bacteria. Purpose: The objective of this study was to determine if serial dilutions of Escherichia coli O157:H7 contaminated ground beef onto selective media and subsequent sequencing of colonies would result in enumeration and identification. Methods: Four independent experiments were performed each with the following treatments in triplicate: ground beef diluted 1:4 in mTSB with 2% sorbitol inoculated with 100 – 103 cfu g-1 E. coli. The mixtures in 3 experiments were stomached and in one experiment the mixtures were homogenized with enzymes using the GentleMACS system. The liquid was removed, filtered, centrifuged, resuspended in 1 mL PBS, serially diluted, and then plated on sorbitol MacConkey agar. The plates were incubated overnight, and E. coli colonies counted. Colonies were selected for DNA extraction and sequenced on an Oxford Nanopore Technologies GridION to determine if the genes fliC, eae, rrsC, stx1A, stx1B, stx2A, stx2B, wzx, and wzy could be detected. Results: Keeping the media cold prevented growth during sample preparation and there was no significant difference between the amount of E. coli inoculated into the sample and the amount recovered (t test, P<0.05). All of the virulence genes were detected an average of 104 times and rrsC was detected an average of 838 times from each of the colonies tested. Significance: Long-read sequencing is useful for identifying and serotyping foodborne pathogens. The addition of plating serial dilutions ensures that the pathogens are live, while enumerating the contaminant level down to 1 cfu g-1.