Defining E. coli Diversity in Complex Samples: Methods for Surveillance & Transmission

Escherichia coli is a ubiquitous bacteria that is ordinarily harmlessly carried by humans. E. coli can also, however, cause serious infection and be complicated by resistance to antibiotics. The first step in causing infection is for the bacteria to colonise the gut of individuals. For this to happen they must be swallowed. Where more dangerous and/or drug-resistant E. coli are acquired most commonly is not clear, but follows a breakdown in personal hygiene often due to a lack of functional hygiene infrastructure whether in hospitals or the community. It is anticipated that well-designed interventions to interrupt E. coli transmission will lead to fewer infections and less antibiotics being used. Due to the complexity and expense of potential interventions to prevent transmission of these bacteria, an improved understanding of where and how E. coli are acquired is critical.
Demonstrating a person has acquired an E. coli from a specific source requires confirmation of indistinguishable E. coli from two epidemiologically connected samples. Distinguishing bacteria of the same species in complex samples is very challenging and there is currently no gold standard approach to describing the within-species diversity of E. coli in a sample at scale. We will develop genomic methods for accurately identifying the diversity and abundance of E. coli in complex samples at scale that can be used for transmission modelling. Subsequently, we will investigate its applicability to diverse samples from different countries to create a catalogue of E. coli diversity in different settings.

AMR pathogen surveillance methods often fail to address within-species diversity, which is essential to link transmission of AMR bacteria between different sources. Emerging metagenomic approaches have shown promise to resolve within-species variation, however, the most appropriate method is unclear. Our DECODE project (Defining E. coli Diversity in Complex Samples: Methods for Surveillance & Transmission) will determine the most accurate and cost-effective method at scale to define intraspecies diversity of Escherichia coli from complex samples with a resolution of epidemiologically informative single nucleotide variants to support One Health AMR surveillance and transmission modelling. To achieve this, we will create well- defined and proportion-controlled mixtures of E. coli mimicking within-human gut diversity as in vitro models and will spike stool and sewage to validate and refine laboratory and bioinformatic workflows. Concurrently, we will establish bioinformatic pipelines initially using synthetic data to determine the influence of different subtypes of E. coli on performance. Next, we will demonstrate the applicability of our methods in different complex samples (stool, sewage effluent and freshwater) in a One Health context and from different locations in sub-Saharan Africa and Europe and will create a catalogue of setting-specific E. coli diversity. Ultimately, knowledge of intraspecies diversity is key to understanding transmission and in turn designing novel interventions to interrupt transmission of AMR bacteria and thereby reduce the number of infections.