BACKGROUND
Previous demonstrations of survival and persistence of Salmonella in pre-harvest agricultural systems (i.e., swine, poultry, dairy, produce, tree nut and marine) suggests that persistent strains may have the ability to adapt and respond to diverse conditions in these niches to thrive in the environment. While previous studies have demonstrated the survival and persistence of Salmonella in various pre-harvest agriculture systems, no studies on the survival and persistence of Salmonella naturally present in the feedlot environment have been published to date to our knowledge. Additionally, no studies have mechanistically determined how feedlot environmental factors collectively impact Salmonella survival and transmission dynamics (i.e., environment, vectors, and animals) within a feedlot system.
High levels of microbial diversity have been shown to decrease persistence of pathogens such as E. coli through competitive exclusion. Soil microbiome composition and soil properties, such as nutrient and water dynamics and weather, also likely play a role in Salmonella persistence. Routine Salmonella surveillance data trending in cattle slaughter facilities and a previous study we conducted indicate that the prevalence of Salmonella in beef lymph nodes and trim varies significantly by geographical location and season, where Salmonella prevalence peaks during the warmer months (summer and fall) and increases among a southern gradient. In addition, a few specific Salmonella serotypes predominate Salmonella isolated from lymph nodes and boneless beef (trim), suggesting that these strains have adapted to the environment and/or cattle in the feedlot system. Observations on the varied regional and seasonal prevalence of Salmonella in cattle lymph nodes and beef trim support the following hypotheses: (i) feedlot environmental factors (i.e., the microbiome, soil properties and weather) may impact the ability of Salmonella to survive and thrive in the feedlot environment, (ii) specific Salmonella serotypes (those commonly isolated from lymph nodes and trim) may have adapted to persist in the feedlot environment for extended periods of time and (iii) mitigation strategies to remediate Salmonella in soil may be effective in breaking the feedlot Salmonella environmental cycle and thus transmission to animals that enter the human food chain.
Microbiome Study: DNA was extracted from samples using the ZymoBIOMICS DNA/RNA Miniprep Kit to the manufacturer’s protocol and eluted using 50 uL of DNase/RNase free water. A total of 467 16S rRNA V4 libraries were generated, which included 447 actual samples and 20 negative controls.
Benchtop Soil Intervention Challenge Study: Autoclaved soil was weighed into conical tubes with 50 percent water holding capacity and inoculated with 100µL Salmonella cocktail at a high and low concentration. The application of Decon7, fertilizer, and L28 was added after a 1 h attachment time. To mimic seasonal differences, half of the samples were incubated at 35°C for the first 14 d and then incubated at 20°C for the remaining 14 d. The other half of the samples began at 20°C and were transferred to 35°C after 14 d. Samples were plated at 0d (1 h after Salmonella
inoculation), 1 h (after intervention application), 7 d, 14 d, 21 d (first measurement after temperature change), and 28 d.
The largest difference in environmental species diversity was driven by sample type, with lagoon and perimeter samples clustering separately from other sample locations. Overall, sample type explained 32% of microbiome variation, compared to <4% for season, feedlot, and pen. Perimeter and lagoon samples contained a higher diversity of unique phyla compared to the soil samples taken from the pens. Within pen-environment samples, the specific layer of soil sampled had the biggest impact on the microbiome composition, explaining 19% of the variability compared to 5% explained by season. Season had a stronger impact on the bacterial composition of the more superficial soil-layer samples. The diversity of the microbiome varied significantly by both sample type and season, with diversity decreasing in the spring and winter samples as compared to the summer and fall samples, across all sample types. Confirmed Salmonella status explained a statistically significant, yet generally small amount of the variation observed in the microbiome composition of each sample. The number of samples with Enterobacteriaceae sequences did not differ greatly across feedlots. There was high discordance between the Enterobacteriaceae counts and the confirmed Salmonella status of each sample, suggesting that the Enterobacteriaceae counts were not a robust proxy for Salmonella status.
All intervention × time interactions for Salmonella counts, soil pH and electrical conductivity (EC) were different in every inoculation × temperature combination. However, total lethality was not reached with any intervention, nor was any intervention different from the control at 28 d. Additionally, no intervention was different from itself from 14 to 21 d, suggesting the temperature change in either direction did not affect intervention efficacy. Following the same trend, soil pH was affected by the intervention × time interaction, and within intervention, soil pH gradually increased over time. EC was also affected by the intervention × time interaction, where L28 increased soil EC and fertilizer decreased soil EC, representing a positive and negative effect on soil quality, respectively.