A Longitudinal Study on the Dynamics of Salmonella enterica Prevalence and Serovar Composition in Beef Cattle Feces and Lymph Nodes and Potential Contributing Sources from the Feedlot Environment

Abstract

Salmonella can persist in the feedlot pen environment, acting as a source of transmission among beef cattle. Concurrently, cattle that are colonized with Salmonella can perpetuate contamination of the pen environment through fecal shedding. To study these cyclical dynamics, pen environment and bovine samples were collected for a 7-month longitudinal comparison of Salmonella prevalence, serovar, and antimicrobial resistance profiles. These samples included composite environment, water, and feed from the feedlot pens (n = 30) and cattle (n = 282) feces and subiliac lymph nodes. Salmonella prevalence across all sample types was 57.7%, with the highest prevalence in the pen environment (76.0%) and feces (70.9%). Salmonella was identified in 42.3% of the subiliac lymph nodes. Based on a multilevel mixed-effects logistic regression model, Salmonella prevalence varied significantly (P < 0.05) by collection month for most sample types. Eight Salmonella serovars were identified, and most isolates were pansusceptible, except for a point mutation in the parC gene, associated with fluoroquinolone resistance. There was a proportional difference in serovars Montevideo, Anatum, and Lubbock comparing the environment (37.2, 15.9, and 11.0%, respectively), fecal (27.5, 22.2, and 14.6%, respectively), and lymph node (15.6, 30.2, and 17.7%, respectively) samples. This suggests that the ability of Salmonella to migrate from the pen environment to the cattle host-or vice versa-is serovar specific. The presence of certain serovars also varied by season. Our results provide evidence that Salmonella serovar dynamics differ when comparing environment and host; therefore, developing serovar-specific preharvest environmental Salmonella mitigation strategies should be considered. IMPORTANCE Salmonella contamination of beef products, specifically from the incorporation of bovine lymph nodes into ground beef, remains a food safety concern. Current postharvest Salmonella mitigation techniques do not address Salmonella bacteria that are harbored in the lymph nodes, nor is it well understood how Salmonella invades the lymph nodes. Alternatively, preharvest mitigation techniques that can be applied to the feedlot environment, such as moisture applications, probiotics, or bacteriophage, may reduce Salmonella before dissemination into cattle lymph nodes. However, previous research conducted in cattle feedlots includes study designs that are cross-sectional, are limited to point-in-time sampling, or are limited to sampling of the cattle host, making it difficult to assess the Salmonella interactions between environment and hosts. This longitudinal analysis of the cattle feedlot explores the Salmonella dynamics between the feedlot environment and beef cattle over time to determine the applicability of preharvest environmental treatments.

Keywords: Salmonella enterica; feedlot cattle; feedlot pen environment; subiliac lymph nodes.

FIG 1

(A) Salmonella prevalence (percentage, number of positive samples) by sample type (Env, pen environment-manure pack; Env-D, pen environment-dry; LN, lymph node; All, all sample types) with the total number of each sample type provided in the parentheses below each bar. The bar farthest to the right provides the overall Salmonella prevalence for all samples. (B) Salmonella prevalence by collection month with the number of total samples per month provided in parentheses below each bar.

FIG 2

Predictive margin plot from a multilevel mixed-effects logistic regression model of Salmonella prevalence by collection month and sample type, indicated by color as shown in the figure legend.

FIG 3

(A) Salmonella serovar composition (percentage, number of isolates) by sample type (Env, environment-manure pack; LN, lymph nodes; All, all sample types) with the furthest bar on the right representing the overall serovar proportions across all sample types and collection months. Serovar is represented by color as shown in the legend. (B) Salmonella serovar composition by collection month.

FIG 4

(A) Multiple correspondence analysis (MCA) plot depicting the relationships of serovar and both sample type and collection month; these can be interpreted based on distances and angles from the plot origin. (B) Correspondence analysis plot investigating pen clustering relationships between Salmonella serovars and feedlot pen number.

FIG 5

Phylogenetic tree of Salmonella isolates from lymph nodes and terminal fecal samples paired by cattle identifier. If available, pen environment-manure pack isolates (OP, original pen; TP, terminal pen) were matched to the isolates from cattle samples, signified by corresponding colors in the middle column. There could be more than one animal associated with a pen sample, as seen in TP 10 and OP 28, which had two cattle (2678 and 2849) associated with them. The bootstrap values are limited to values 0.8 to 1.0.

FIG 6

Phylogenetic tree of all sequenced Salmonella isolates arising from the longitudinal study (n = 524). The metadata included in the four rings surrounding the tree from the inside out are as follows: Salmonella serovar, sample type, collection month, and pen number. The pen colors are shaded by the 12 feedlot blocks. Bootstrap values are represented by the gray circles including values 0.8 to 1.0.

FIG 7

Phylogenetic tree of all Salmonella Kentucky (n = 162) isolates in this study. Metadata from inner ring to outer ring are as follows: original sample type, collection month, and feedlot pen, which is shaded by the 12 feedlot pen blocks. The bootstrap values are limited to values 0.8 to 1.0. The setup is the same as in Fig. 6.

FIG 8

The geographical layout (3 rows [i.e., front, back, and side], 12 blocks, and 60 pens) of the West Texas A&M University Research Feedlot. Dietary treatments (yellow, WDGS; red, SB; green, COMBO; blue, CON; white, unused) were provided to cattle at the pen level. Dietary treatments and geographic locations were evenly distributed across the 30 pens selected for the longitudinal study (black stars). NCBA, National Cattlemen’s Beef Association.

FIG 9

Sample collection timeline of the longitudinal study which consisted of three main types of events. Sample collections are signified by blue markers, cattle slaughter dates by red markers, and movements of cattle between pens by yellow markers.