Antimicrobial resistance in the microbiome of feedlot watering bowls and bovine respiratory disease associated pathogens.

Bovine respiratory disease (BRD) is the primary disease of concern to beef production and is driven by stress, bacteria, viruses, and environmental and management practices. Antimicrobial use (AMU) to control BRD selects for antimicrobial resistance (AMR) genes (ARGs). Treatment failure due to AMR necessitates a rapid switch to an effective antimicrobial before a devastating amount of death loss occurs. Monitoring AMR in a herd through individual animals is impractical. There is a need for a fast, pragmatic, and scalable methodology to monitor AMR within cattle to guide effective AMU. Cattle watering bowls (WB) have been previously shown to harbour BRD pathogens and ARGs but have yet to be explored in the same context as wastewater AMR monitoring is for public health. This thesis outlines a pilot project of WB sampling as a proxy method to monitor the presence of ARGs of relevance to BRD associated pathogens. The microbiomes within watering bowls of two feedlots were genotypically and phenotypically assessed through water, swab, and sediment sampling. A new feedlot was sampled (68 water, 63 swab) as pens filled (up to 9 WB) over a series of 8 weeks. The other, older, feedlot was only sampled in the 8th week (sediment, swab, water) from 20 WB. For phenotypic AMR detection, samples were inoculated into antimicrobial sensitivity test (AST) panels containing microdilutions of four antimicrobials (enrofloxacin, florfenicol (FFN), tulathromycin, oxytetracycline). Bacteria that grew on these plates were isolated (n = 78). A subset (n = 28 new feedlot, n = 3 old feedlot + H. somni from a deep nasopharyngeal swab) of these underwent further characterisation on an AST panel containing 10 different antimicrobials. The whole genome of isolates (n = 10, 6 new, 4 old) was sequenced and analysed for ARGs. These detected ARGs were compared to a custom dataset of 172 BRD genomes from NCBI. The genetic neighbourhood of floR, an ARG encoding for the efflux of FFN, within the WB isolates (n = 4) was also compared to pathogen genomes. For the culture-independent methodology, samples from one feedlot underwent amplified 16S rRNA sequencing (n = 122), while samples from the other underwent shotgun metagenomic sequencing (n = 11). The ARGs detected in WB isolates were compared to the ARGs detected within metagenomic sequencing, and previous datasets of feedlot water. Multi-drug resistant bacteria were routinely isolated from WBs. When environmental samples from the last week of sampling were inoculated into media with FFN (>32 µg/mL), 93/96 samples had bacterial growth. By constructing a genotypic AMR profile from two feedlots and by consulting previous metagenomic datasets from feedlot water, the understudied resistome of the watering bowl microbiome was shown to provide outstanding coverage of the ARGs (15/16) detected within 172 BRD genomes. The overlap in genetic neighbourhoods between copies of floR in WB and BRD genomes confirms the possibility of horizontal gene transfer via a myriad of transposable elements between both populations. Though further studies are needed to reaffirm our findings, in this thesis the understudied environmental microbiome within feedlot watering bowls was found to serve as a potential source of ARGs for BRD pathogens and thus may serve as a proxy for relevant ARGs within feedlot environments.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.