Background & Objectives. The high incidence of Salmonella enterica-related foodborne illnesses in the U.S., along with increasing efforts from federal agencies and heightened public concern regarding salmonellosis, has made effective surveillance a priority. Carrier cattle can continuously shed Salmonella to the environment through fecal waste. In feedlots, liquid runoff carrying fecal waste is generated by rainfall and typically managed via holding ponds. Given the fecal content of this runoff, it could potentially carry various microorganisms, including potential pathogens such as Salmonella. Therefore, we aimed to characterize the diversity and composition of Salmonella populations in holding ponds and feces from nearby cattle pens, and conducted taxonomic, antimicrobial resistance (AMR), virulence factor, and functional profiling on both fecal and pond samples. Further, we compared the genetic potential of microbial populations (passive metagenome) to their metabolically active members (active metatranscriptome).
Methods. We analyzed fecal (n=48) and holding pond (n =13) samples from 5 cattle feedlots in the Texas Panhadle. Holding pond water was sampled from 6–12 inches below the surface, and composite floor fecal samples were collected from 4 nearby pens. All samples were tested for Salmonella using selective enrichment and PCR, and CRISPR-SeroSeq was used to identify all serovars present in culture-positive samples. DNA and RNA were co-isolated, and RNA was reverse transcribed into cDNA. Shotgun metagenomes (n = 61) and metatranscriptomes (n = 61) were sequenced on an Illumina NovaSeq 6000. Taxonomic classification was performed using Kraken2, AMR genes were identified using the AMR++ pipeline and MEGARes database, functional content was profiled using HUMAnN3, and virulence factors were identified by BLAST-based alignment against the VFDB (Virulence Factor Database). Differentially abundant features were identified using ANCOM-BC and MaAsLin3.
Results. Nearly half of the sampled feedlot ponds and nearby fecal samples were culture and PCR-positive. Further, an average of 1.88 serovars across positive samples were identified by CRISPR SeroSeq. Salmonella was detected in low abundance across all ponds and fecal samples based on meta-omics sequencing, with overall higher abundance in culture-positive samples. No differentially abundant OTUs classified as Salmonella were identified between fecal and pond samples. The composition of taxonomic profiles, AMR genes, virulence factors, functional gene families, and metabolic pathways differed between fecal and pond samples. Virulence factors related to biofilm formation, effector delivery systems, adherence and regulation of expression were differentially abundant between pond and fecal samples. While the taxonomic composition of fecal communities differed between the metagenome and metatranscriptome, no significant differences were observed in AMR or virulence gene pools, or functional pathway profiles between fecal and pond samples.
Significance. The similarities of Salmonella populations between holding ponds and feces lays the groundwork for exploring trends in community-wide Salmonella patterns in cattle through wastewater surveillance. Moreover, the observed differences between active and passive fecal microbial communities raises a consideration for future studies, where integrating both DNA and RNA analyses may offer a more comprehensive view of the fecal cattle microbiome.