Salmonella spp. represent a food safety risk in the production chain because of their potential for biofilm development. This study examined the biofilm formation of eight Salmonella serotypes from diverse foodborne outbreaks on three food-contact surfaces, stainless steel, silicone, and nylon, at 10 °C and 37 °C. The effect of temperature was observed in slower biofilm formation at 10 °C with about 5-log (cfu/cm²) after 24 h, contrasting with 7-log (cfu/cm²) at 37 °C. The material also influenced biofilm formation, with the strongest biofilms on stainless steel at 10 °C and silicone at 37 °C. The serotypes producing the strongest biofilms were S. Enteritidis, S. Saint Paul, and S. Montevideo. The weakest serotypes were S. Senftenberg, S. Anatum, and the avirulent S. Typhimurium. The production of extra-polymeric substances was evident with S. Enteritidis. The biofilm index showed the highest value for low temperature, nylon, and silicone, and for S. Montevideo, S. Enteritidis, and S. Saint Paul. The whole-genome sequencing of each serovar suggested that single nucleotide polymorphisms in the curli (csg) genes may have contributed to the strong biofilm-forming ability of S. Montevideo and S. Saint Paul and the weaker ability of S. Senftenberg. These results can help with the correct development of sanitizing interventions based on the Salmonella strain of concern.

Foodborne pathogens are a significant cause of illness, and infection with Shiga toxin-producing Escherichia coli (STEC) may lead to life-threatening complications. The current methods to identify STEC in meat involve culture-based, molecular, and proteomic assays and take at least four days to complete. This time could be reduced by using long-read whole-genome sequencing to identify foodborne pathogens. Therefore, the goal of this project was to evaluate the use of long-read sequencing to detect STEC in ground beef. The objectives of the project included establishing optimal sequencing parameters, determining the limit of detection of all STEC virulence genes of interest in pure cultures and spiked ground beef, and evaluating selective sequencing to enhance STEC detection in ground beef. Sequencing libraries were run on the Oxford Nanopore Technologies’ MinION sequencer. Optimal sequencing output was obtained using the default parameters in MinKNOW, except for setting the minimum read length to 1 kb. All genes of interest (eae, stx1, stx2, fliC, wzx, wzy, and rrsC) were detected in DNA extracted from STEC pure cultures within 1 h of sequencing, and 30× coverage was obtained within 2 h. All virulence genes were confidently detected in STEC DNA quantities as low as 12.5 ng. In STEC-inoculated ground beef, software-controlled selective sequencing improved virulence gene detection; however, several virulence genes were not detected due to high bovine DNA concentrations in the samples. The growth enrichment of inoculated meat samples in mTSB resulted in a 100-fold increase in virulence gene detection as compared to the unenriched samples. The results of this project suggest that further development of long-read sequencing protocols may result in a faster, less labor-intensive method to detect STEC in ground beef.

A requirement of any foodborne pathogen testing method is that it only detects live bacteria. Ethidium monoazide (EMA) and propidium monoazide (PMA) are dyes that penetrate the membranes of dead cells and form cross-linkages in the DNA, which prevents its amplification in PCR. This study investigated whether treatment with EMA or PMA would inhibit the sequencing of DNA from dead Escherichia coli. Range finding experiments with qPCR were conducted to determine the optimal concentrations of EMA and PMA needed to inhibit the amplification of DNA from dead cells while not influencing live cells. An EMA concentration that differentiated between live and dead cells could not be established. However, a PMA concentration of 25 µM effectively prevented qPCR amplification of DNA from dead E. coli while not impacting the amplification of live E. coli DNA. Sequencing experiments were conducted with PMA-treated live, untreated live, PMA-treated dead, and untreated dead E. coli. There were no significant differences in the detection of virulence genes of interest between the PMA-treated live, untreated live, and untreated dead E. coli. However, no DNA sequencing data were obtained from the PMA-treated dead E. coli. These results suggest that PMA could be incorporated into sample preparation methods prior to sequencing to selectively detect live cells of foodborne pathogens.

The detection of foodborne pathogens is a critical aspect of ensuring food safety. Traditional methods rely on time-intensive enrichment steps and pathogen-specific assays, extending testing timelines and limiting throughput. This study evaluates an enrichment-free, multiplexed pathogen detection workflow combining the Pathotrak system for bacterial separation and the Neogen Molecular Detection System (MDS) for detection. The workflow enables simultaneous detection of Salmonella, Escherichia coli O157, Listeria monocytogenes, Listeria spp., and Campylobacter in poultry samples, significantly reducing the time taken to screen samples requiring further evaluation. The system’s performance was assessed using inoculated chicken samples over a range of bacterial concentrations (10²–10⁸ CFU/mL). The MDS system demonstrated robust detection for most pathogens, with strong correlations between theoretical inoculation levels and MDS-calculated concentrations (R² > 0.85 for all pathogens). However, detection variability was observed at lower concentrations for Salmonella and Listeria monocytogenes. The system maintained high sensitivity and specificity, achieving a Cohen’s Kappa coefficient for E. coli and Campylobacter. This study highlights the potential of enrichment-free, multiplex detection to streamline food safety testing by reducing the time to results, enhancing efficiency, and providing reliable pathogen quantification across multiple targets.

Shiga toxin-producing Escherichia coli (STEC) and Listeria monocytogenes are routinely responsible for severe foodborne illnesses in the United States. Current identification methods utilized by the U.S. Food Safety Inspection Service require at least four days to identify STEC and six days for L. monocytogenes. Adoption of long-read, whole genome sequencing for food safety testing could significantly reduce the time needed for identification, but method development costs are high. Therefore, the goal of this project was to use NanoSim-H software to simulate Oxford Nanopore sequencing reads to assess the feasibility of sequencing-based foodborne pathogen detection and guide experimental design. Sequencing reads were simulated for STEC, L. monocytogenes, and a 1:1 combination of STEC and Bos taurus genomes using NanoSim-H. At least 2500 simulated reads were needed to identify the seven genes of interest targeted in STEC, and at least 500 reads were needed to detect the gene targeted in L. monocytogenes. Genome coverage of 30x was estimated at 21,521, and 11,802 reads for STEC and L. monocytogenes, respectively. Approximately 5–6% of reads simulated from both bacteria did not align with their respective reference genomes due to the introduction of errors. For the STEC and B. taurus 1:1 genome mixture, all genes of interest were detected with 1,000,000 reads, but less than 1x coverage was obtained. The results suggested sample enrichment would be necessary to detect foodborne pathogens with long-read sequencing, but this would still decrease the time needed from current methods. Additionally, simulation data will be useful for reducing the time and expense associated with laboratory experimentation.

Six hundred million cases of disease and roughly 420,000 deaths occur globally each year due to foodborne pathogens. Current methods to screen and identify pathogens in swine, poultry, and cattle products include immuno-based techniques (e.g., immunoassay integrated biosensors), molecular methods (e.g., DNA hybridization and PCR assays), and traditional culturing. These methods are often used in tandem to screen, quantify, and characterize samples, prolonging real-time comprehensive analysis. Next-generation sequencing (NGS) is a relatively new technology that combines DNA-sequencing chemistry and bioinformatics to generate and analyze large amounts of short- or long-read DNA sequences and whole genomes. The goal of this project was to evaluate the quantitative capabilities of the real-time NGS Oxford Nanopore Technologies’ MinION sequencer through a shotgun-based sequencing approach. This investigation explored the correlation between known amounts of the analyte (lambda DNA as a pathogenic bacterial surrogate) with data output, in both the presence and absence of a background matrix (Bos taurus DNA). A positive linear correlation was observed between the concentration of analyte and the amount of data produced, number of bases sequenced, and number of reads generated in both the presence and absence of a background matrix. In the presence of bovine DNA, the sequenced data were successfully mapped to the NCBI lambda reference genome. Furthermore, the workflow from pre-extracted DNA to target identification took less than 3 h, demonstrating the potential of long-read sequencing in food safety as a rapid method for screening, identification, and quantification.