Acquisition and Selection of Antibiotic Resistance in Companion and Farmed Animals and Implications for Transmission to Humans

Without antimicrobial drugs, the risk of bacterial infection would render many common medical procedures too dangerous to contemplate because of the risk of infections caused by "opportunistic bacteria". They can live on the patient's skin, or in their intestines, and infection occurs when bacteria get into parts of the body that are normally sterile. A perfect example is urinary tract infection (UTI) caused by faecal bacteria. E. coli is particularly abundant in human faeces so is perfectly placed to cause opportunistic infections. It is one of the most common causes of healthcare pneumonia, surgical site infection, bloodstream infection and UTI in the UK. In order to prevent against and treat opportunistic infections, patients are given antimicrobials.

Almost all antimicrobials are "antibiotics", which means they are derived from natural chemicals produced by microbes found in the environment. Natural antibiotics have been present in the environment for millions of years, and so bacteria living in their presence have had time to evolve mechanisms that can resist their actions, encoded by "resistance genes". Opportunistic bacteria like E. coli can randomly acquire these pre-evolved resistance genes and in a single step, they become insusceptible to a particular antimicrobial. If that insusceptible E. coli colonises a person and then causes an opportunistic infection, the infection will not be treatable with that particular antimicrobial. We refer to this as "antimicrobial resistance" (AMR); however AMR bacteria don't just resist clinical antimicrobial therapy, they beat it.

Animals also carry an abundance of E. coli in their intestines and are frequently treated with antimicrobials. This can select for the acquisition of AMR E. coli which can then be passed on to another animals, directly, or via contamination of the environment with faeces. Theoretically, the AMR E. coli could also be passed on to people, and there is much debate about whether such "zoonotic transmission" happens to any significant degree. This is an important debate because it has led to calls from some to dramatically reduce the amount of antimicrobials that are given to animals with the view that it will reduce the level of AMR in animals, and so the possibility of zoonotic transmission to people. But the potential impact on welfare and food production means this should only be done if there is evidence that it will work.

In this project we will identify what drives acquisition of AMR in animals using E. coli as the exemplar bacterium and dairy cows and dogs as exemplar farmed and companion animals. We will test whether AMR bacteria encountered by an animal as it interacts with the environment influence the AMR profile in its faeces, and/or whether early life antimicrobial use plays a part in selection of AMR bacteria in animals. We will also test whether reducing antimicrobial use in dairy cows actually does reduce AMR in the near-farm environment that is contaminated with their faeces. We will test whether exercising in these contaminated near-farm environments influences the abundance of AMR bacteria in dogs, and whether there is any evidence of direct acquisition of AMR E. coli by dogs from near-farm environments, which might be brought into the home.

Finally, we will investigate whether AMR abundance in human UTI E. coli reduces as antimicrobial drug prescribing reduces in primary care; whether living close to a farm affects AMR abundance in UTI E. coli; whether there is direct evidence for E. coli carried by dogs or found in near-farm environments contaminated by cattle faeces also causing UTIs in humans.

These interlaced studies will provide much needed data about the management changes that might reduce AMR in animals and in humans, and are designed to address the fundamental question of whether zoonotic transmission is particularly significant as a driver of AMR in people relative to antimicrobial drug use by doctors.

INDUSTRY
This project addresses the impact that the Farming and Veterinary Medicine industries might have on human health through animal-mediated contamination of publically accessible land with AMR bacteria, and because zoonosis is a recognised health and safety risk for farmers and veterinarians. Understanding risk factors for acquisition of AMR in farmed and companion animals and possible ways to mitigate them will help these industries react to growing public concern about AMR and possible legislation in an informed way. Directly involving businesses from these sectors in our project will allow them to advocate for our findings in their own industries. For example, provision of first-hand evidence on the positive effect of reducing antibiotic use on AMR levels on the farms in our study may stimulate a peer-led process to encourage other businesses to consider how to reduce levels further. Limiting the rise of AMR will make farming in the UK more productive (via less untreatable disease) and more economically successful, and will provide a model for tackling AMR elsewhere in the world.

REGULATORY BODIES
The Veterinary Medicines Directorate (VMD) has confirmed its interest in this project, and we have had meetings with the VMD AMR Surveillance Team about our work. VMD is responsible for channelling EU regulations for reducing AMR in the veterinary controlled environment. VMD recognise, however, that there are significant knowledge gaps, and this study will address three of these: how antibiotic use selects for resistance; current AMR levels in healthy animals and in the environment; what might be done to reduce these levels. Other Defra executive agencies such as Natural England are likely to be interested in our research as it pertains to footpaths and we have included significant Public Health England involvement in the project, so our findings can feed back to provide public information about the importance of reducing exposure to AMR in the environment. Local authorities also have responsibility for establishing areas in which dog faeces must be collected and enforcing bylaws, so the may also be interested.

THIRD SECTOR
Approximately 130,000 dogs were estimated to enter rescue centres run by animal charities in 2009. Dogs mixing in these environments may be an important aspect of AMR transmission, impacting on the management of such centres. Charities also place dogs into new homes (e.g. Dogs Trust re-homed almost 15,000 dogs last year), and information on risk to human health from this project will help inform the best policy. Walking groups including those convened through the Ramblers' Association and the U3A will be interested in the findings of this study. U3A members are likely to be aware through personal experience of the rise of AMR and its role in healthcare-associated and opportunistic community infections, and may benefit from any simple risk mitigation strategies that our findings might suggest.

WIDER PUBLIC
According to Natural England, >800 million visits to the natural environment were made by people in England and Wales between March and May 2015. An estimated 10.5 million dogs are kept as companion animals in the UK, and are often regarded as members of the family. If zoonotic transmission of AMR from the environment, directly or via companion animals, is a significant health risk, then our research promises considerable benefits for the health and wellbeing of the UK population by attempting to identify risk factors for acquisition of AMR in farmed and companion animals, but also by simply highlighting such a risk. For instance, basic hand hygiene is a simple intervention that might be all that is required to have a considerable effect on decreasing the transmission of AMR pathogens.