Using the Microbiomes of Aquatic Organisms to Investigate the Effects of Pollution on Environmental AMR

Antimicrobial resistance (AMR) is one of the most serious global health threats facing society. Many sources, including wastewater treatment plant effluent and farming runoff, allow AMR bacteria and antimicrobials that select for resistance to enter the aquatic environment. Aquatic environmental reservoirs can act as sources of AMR transmissible to human pathogens, yet there is insufficient knowledge on the effect that environmental pollution, including antimicrobials, has on AMR in these environments. Environmental AMR is usually measured by analysing sediment and water samples, yet interpreting such samples can be made difficult by matrix variability over space and time. Currently, there is no standardised surveillance method for monitoring AMR in the environment in the UK; this research aimed to address this gap. The studies presented in this thesis aimed to determine the effect environmental pollution has on AMR in rivers, by studying changes in the microbiome of aquatic animals under varying pollution loads. A catchment-based study in the River Thames showed that the gut microbiome of the common roach (Rutilus rutilus) was significantly affected by the percentage effluent of the river at the sampling site and proportion of urban land and grassland upstream of the sampling site. This demonstrated that aquatic animal microbiomes are affected by pollution and can be indicative of the environment to which they are exposed. Another catchment-based study investigating the microbiome of the commonly found amphipod, Gammarus pulex, found that the prevalence of intI1 (a molecular marker for AMR) in the G. pulex microbiome was significantly affected by (in order of significance) increasing grassland upstream, increasing percentage effluent, and decreasing proximity to the nearest wastewater treatment plant (WWTP). This study also found that these environmental catchment variables could better explain the variance of intI1 in G. pulex (39.3% of variance explained), than in sediment (29.5%) and water (0%) samples from the same sampling sites. This study highlighted the potential use of G. pulex as a sensitive and promising approach for standardising AMR surveillance in rivers. The roach and G. pulex catchment-based studies for AMR surveillance could not discriminate between colonisation by AMR bacteria and selection by antimicrobials within the organism's microbiome. A laboratory mesocosm-based study was conducted to examine the lowest effect concentration of ciprofloxacin (CIP) impacting the prevalence of intI1 in, and composition of, the G. pulex microbiome (i.e., to determine the minimum selective concentration (MSC) of CIP). CIP did not select for intI1 in the G. pulex microbiome, however, a MSC of 2.5 µg/L was determined by quantification of intI1 in the mesocosm water column. From this, a predicted no effect concentration for the development of AMR (PNECR) of 0.046 µg/L and a risk quotient of 14.33 were determined. This indicates that CIP poses a significant risk for selection of AMR in the environment. This thesis focused on using aquatic organism microbiomes in multiple approaches of research on environmental AMR, such as in catchment-based studies, and as an environmentally-relevant, complex microbial community for determining MSCs, PNECs and RQs of an antibiotic. The results from this thesis evidence the potential for these microbiomes to be a predictive/monitoring tool for how pollution impacts environmental AMR. AMR national action plans require a standardised method for quantifying the impact of AMR pollution on the aquatic environment. The experiments may bring us one step closer to achieving this.