The devil we know? Using sequencing and metagenomics to assess the diversity of bacteria and antimicrobial resistance captured by passive surveillance

Antimicrobial resistance (AMR) of bacterial infections is a serious threat to public health, and in order to halt or control its rise, understanding where it comes from and how it spreads is essential. Resistance to antimicrobial drugs in non-typhoidal Salmonella (NTS), a bacterium that infects both animals and humans, is an increasing problem, one that can lead to more severe infections and complicates treatment. A great deal of our understanding of the patterns and sources of NTS and AMR is based on data from laboratory-based surveillance and from disease outbreak investigations. These systems collect data and samples from primarily clinically ill individuals, for which there is significant under-reporting and which represent only those bacteria that cause disease. Therefore, these data characterise a relatively small proportion of the overall bacterial population. The research outlined in this fellowship proposal builds on my previous experience in epidemiology, ecology and genomics, and will address five key questions: 1) How similar are NTS and AMR from clinically diseased and non-diseased hosts? 2) For common AMR patterns found in different subtypes of NTS, are the resistance genes the same? 3) How is AMR distributed between different host populations in the same location, and how well does the observed resistance correlate with identified AMR genes? 4) How does AMR in different host populations compare between an industrialised and an industrialising country? 5) How much of the total AMR in a sample, including in bacteria that cannot be grown in standard laboratory conditions, is represented by the AMR within a single organism, NTS, which can be grown in standard laboratory conditions? I will address these questions using NTS collections from Canada and Vietnam, where in each location NTS are available from diseased and non-diseased individuals. This research will be undertaken at the University of Cambridge, in conjunction with collaborators at the Oxford University Clinical Research Unit/Wellcome Trust Major Overseas Programme in Vietnam, the University of Ottawa, Canada, and the Wellcome Trust Sanger Institute, UK.

Whole genome sequencing provides the highest resolution available to investigate how organisms are related to each other on a genetic level, and I will utilise this technology to assess how similar isolates from diseased and non-diseased hosts are to each other, the type and diversity of AMR found in each, and the degree of mixture of the populations of isolates from differing disease status hosts. I will also generate a pilot dataset applying metagenomic sequencing to faecal samples, examining all bacteria in the samples, including those can and cannot grow in standard laboratory conditions, and compare how well the AMR in the NTS isolate represents the AMR found within the total bacterial population. In addition, I will use these isolate collections to examine AMR and NTS from animal and human populations in Vietnam; given the global nature of the food supply and the wide variability of agricultural, political and sociological settings across the world, it is important to conduct such assessments in many different countries, as there is unlikely to be a single universal condition for AMR. If we are going to have the ability to do something to halt the rise in AMR, we need to know where it is, how well we are currently measuring it, and what we may be missing - the research encompassed by this proposal will provide important insight and critical data with which to address these questions.

Antimicrobial resistance (AMR) of bacterial infections is a serious threat to public health; in order to halt or control its spread, understanding the drivers and reservoirs of AMR in diverse populations is essential. Non-typhoidal Salmonella (NTS), which causes tens of millions of gastroenteritis cases per year, is becoming increasingly drug resistant. Much of our knowledge of AMR and NTS is based on isolates from clinically diseased individuals, which represent only a relatively small proportion of the overall bacterial diversity. In this research, I will use whole genome sequencing and isolate collections from Vietnam and Canada to address critical gaps in our knowledge: 1) How similar are NTS and AMR from clinically diseased and asymptomatic hosts? 2) For common AMR patterns found in different serotypes of NTS, is the genetic basis the same? 3) How is AMR distributed between different contemporaneous and sympatric host populations, and how do the genetic determinants of resistance correlate to the phenotypic resistance? 4) How does AMR in different host populations compare between an industrialised and an industrialising country, where the agricultural, political and sociological conditions are different? 5) How much of the total AMR in a sample is represented by the AMR within a culturable pathogenic bacterium? If we are going to have the ability to do something to halt the rise in AMR, we need to know where it is, how well we are currently measuring it, and what we may be missing - the research encompassed by this proposal will provide important insight and critical data with which to address these questions.

The research I have outlined has the potential to have both societal and economic impacts. One of the primary beneficiaries will be the communities from whom the isolates and samples were collected, particularly the individuals participating in the study in Vietnam. The results obtained from this component of my research will clarify the extent of reservoirs of antimicrobial resistance (AMR) and Salmonella in animals and humans in this setting, and may help identify transmission pathways that could then be targeted for control, or lead to changed practices such as reduced antimicrobial usage. This is particularly important given the global nature of the food supply. This information would also be relevant for health officials and government officials. I will work with the public engagement team at the Wellcome Trust Major Overseas Programme in Ho Chi Minh City to ensure that my research is disseminated in the most effective way and to the relevant stakeholders. The results relating to the Salmonella isolates from swine in Canada will be relevant to the swine industry in that country and others, as Salmonella is a targeted organism for surveillance in swine in Canada and other countries.

Another beneficiary of my research is the general public. Over the past several years, the profile of AMR has risen considerably, and individuals are increasingly aware of and have an interest in learning about the dangers and sources of AMR. What they may not be as familiar with is the under-representation of the overall bacterial populations by outbreak investigations and passive surveillance, and the impact this may on our understanding of where and how humans become infected. To facilitate the dissemination of my research to this group, I will volunteer at events like the Cambridge Science Festival, interacting with schoolchildren and the general public, and will communicate with journalists to ensure that the widest possible audience is reached.