Williams, Alexander D.
(2022)
Antibiotic resistance in dairy farm soil: a metagenomic survey from slurry to field.
PhD thesis, University of Nottingham.
Abstract
Burgeoning antibiotic resistance (AR) threatens to undermine global human health by rendering antibiotic treatments ineffective. The rapid development of AR is widely attributed to excessive antibiotic use, and a greater proportion of global antibiotic sales are associated with livestock relative to human clinical use. In line with the One Health concept, increasing focus has been placed on studying the spread of AR in agricultural environments, including contamination of the human food chain. Livestock waste is a valuable resource for fertilising agricultural land worldwide; however, it also represents a source of unmetabolised antibiotics, assorted antimicrobials and antibiotic resistant bacteria (ARB). In the UK, cattle slurry comprises a substantial proportion of animal waste applied to fields. Consequently, the dispersal of AR on UK dairy farms and their environs warrants investigation.
The body of work presented here therefore aimed to characterise the dynamic spread of AR on a large, high-performance commercial dairy farm in the UK. More specifically, the occurrence and prevalence of AR determinants and bacterial taxa in slurry-amended field soil were evaluated over the course of a year, with sampling occurring in May, July, September and October 2017, as well as January and May 2018. On-site slurry samples were also characterised monthly, between June and October 2017. During the period of study, additional sampling was also carried out in January and May 2018, from a nearby field site with no history of livestock waste application. Environmental DNA was extracted from all samples, processed, and sequenced to produce metagenomic libraries. Bioinformatic analyses were carried out to annotate, quantify, and visualise the ARG profile (resistome) and taxonomic composition of samples.
By bringing together antibiotic usage records, soil physiochemistry, meteorological and metagenomic data, the current work produced a uniquely comprehensive temporal evaluation of AR in field soil following successive amendments of slurry at realistic rates of application.
Through analyses of metagenomic data it was shown that while both slurry and soil contained a diverse array of antibiotic resistance genes (ARGs), the ARG profiles of soil were distinct from those of slurry, irrespective of the origin of the soil in the context of the sites sampled. This is further reflected in the microbial composition of soils which also demonstrated robust correspondence with ARG profiles. Together, these data allowed the identification of slurry biomarkers (ARGs and taxa). ARG slurry biomarkers included gene groups belonging to macrolide, lincosamide and tetracycline resistance gene categories. Furthermore, the wider genetic context of slurry ARGs was characterised by metagenome assembly. Specifically, the beta-lactamase resistance gene cfxA was associated with NBU-1-like elements, while the tetracyline resistance gene tetM was embedded within Tn916-like transposable mobile elements. Similarly, certain ARGs such as aph(6) (aminoglycosides), aph(3”) (aminoglcosides) and sul2 (sulfonamides) were co-resident on contigs.
Taxonomic slurry biomarkers included members of phylum Bacteroidetes, Firmicutes, Synergistetes, Tenericutes and Sphaerocheata. In particular, both unassembled read data and metagenomes-assembled genomes (MAGs) indicated Proteiniphilum sp.(Bacteroidetes) was a biomarker of slurry application.
Temporal analyses showed that the slurry biomarkers exhibited two distinct lifecycles in soil, one of which appeared to be further modified by season. Firstly, select ARGs were consistently more abundant in soil with a long-term history of slurry application relative to soil from the 'untreated' site (e.g. tetM), indicating their prevalence related to longterm enrichment. In contrast, another set of ARGs exhibited transient increases following the first application of slurry in May 2017 (e.g. lnuB, mefB). These ARGs declined to pretreatment levels within eight weeks. In contrast, the same ARGs persisted for >12 weeks after the first application of slurry in February 2018. This suggests the timing of the first slurry application of the season can influence the survival of select slurry associated ARGs and taxa. Given current UK guidelines, the latter scenario may result in the contamination of grass-cut for silage intended for use as cattle-feed and may lead to a positive feed-back loop of AR in cattle, although this was beyond the scope of the current work. Further field-based and laboratory experiments should be carried out to confirm the broader scalability of these findings.
In summary, the present work explored relationships between taxa, ARGs and mobile genetic elements, as well as the dynamic nature of ARGs within soils on a working dairy farm. Analyses enabled identification of candidate biomarkers of slurry exposure (ARGs and taxa), which can be used to develop more targeted studies in the future.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of A_WILLIAMS_THESIS_RESUBMISSION.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "A_WILLIAMS_THESIS_RESUBMISSION.pdf")