Exploiting commensal-pathogen competition to treat mucosal infection

Recently, there has been a dramatic rise in the number of bacteria that cause disease acquiring resistance to antibiotics that are commonly used to treat them. These bacteria are known as antimicrobial resistant (AMR) and carry gene/s that mediate insensitivity to, degradation of, or expulsion of, these drugs. In the not too distant future, we are facing a situation where once trivial, easily treatable, infections could potentially prove fatal. Currently, E. coli, a bacterium that constitutes an important part of the normal gut flora, is one of the most frequently isolated AMR hospital acquired infections in Europe and the UK and is a major cause of blood poisoning, diarrhoea and recurrent urinary tract infections. Misuse and overuse of antibiotics (e.g. during chemotherapy or in intensive care units) is at the heart of the AMR problem. However, little research has been performed to date to elucidate the in vivo consequences of AMR in relevant animal models.

For the last 20 years my lab has been modelling human intestinal infections with pathogenic E. coli in mice, using the mouse pathogen Citrobacter rodentium. C. rodentium infection is an excellent bacterial colonisation model as it does not cause severe disease in mice. Recently, we found that when treatment of mice infected with C. rodentium that is resistant to the antibiotic kanamycin (Kan), with high level of Kan leads to a phenomenon we termed antibiotic induced bacterial persistence (AIBP). C. rodentium in the AIBP state persists within the gastrointestinal tract for many days and develops a new relationship with the host, which is different from a typical pathogen-host interaction. In particular, virulence genes are turned off in C. rodentium in the AIBP state and the pathogen is non-infectious. In contrast, treating infections with Kan resistant C. rodentium with low levels of Kan leads to delay clearance of the pathogen, which coincided with growth of commensal gut bacteria, which could potentially outcompete or kill C. rodentium.

In this study we will characterise C. rodentium in the AIBP state and its relationship with commensal bacteria. Although we appreciate that the mouse intestinal microbiota is not identical to that in humans, this project will reveal common transferable principles underpinning bacterial competition and adaptation in the intestine during antibiotic treatment, the consequences of using the wrong antibiotics to treat AR infections and lead to the identification of novel new natural products expressed by the microbiome that impact on pathogen-host interactions. These principles are likely to be applicable to other enteric pathogens that have to compete with the gut microbiota during colonisation.

1. We will use combinations of novel defined culture media, developed at the WTSI, to cultivate and profile the KanR microflora from naïve-antibiotic-treated and infected-antibiotic-treated mice. Moreover, we will treat mice with other classes of antibiotics, including the protein synthesis inhibitors chloramphenicol and tetracycline, the cell wall inhibitors beta-lactams and the DNA gyrase inhibitor Nalidixic acid, with or without infection with the corresponding antibiotic resistant C. rodentium (available in our lab). Although, we appreciate that not all bacteria can be cultured, we will characterise the potential out-competing commensals by culturing coupled with Bruker MALDI Biotyping and 16S rRNA gene sequencing.

2. We will use high concentrations of the antibiotics mentioned above to determine if they also induce AIBP C. rodentium and the status of virulence gene expression under these conditions.

3. We will use bacterial competition assays or co-infection to determine the ability of components of the murine intestinal microbiota, which expand while C. rodentium declines, to out compete/kill C. rodentium in vitro and in vivo. Real time bacterial colonisation will be monitored using multimodality imaging (bioluminescent imaging with integrated microCT) during bacterial competition and killing assays (prophylactic or therapeutic treatments of mice with components of the microbiota).

4. Selected bacterial taxa that are antagonistic to C. rodentium will be tagged using a bioluminescent reporter (e.g. click beetle luciferase). We will monitor colonisation of these strains following inoculation of naïve mice or mice that were previously treated with the relevant antibiotics (with and without C. rodentium infection).

5. We will identify, fractionate, enrich and test the efficacy of possible intestinal metabolites during infection in vivo using LC-MS and NMR.

By definition, this call is aimed at addressing a major global health problem linked to the misuse and overuse of antibiotics and the spread of AMR. For this reason, positive outcomes are bound to have an impact on human health as well as on farm animals.

The main aim of this project is to improve our understanding of how AMR pathogens behave in response to antibiotic treatment and the subsequent consequences on virulence and transmission. We aim to identify alternative treatments for AMR infections such as oral bacteriotherapy. Reducing infection rates and development of new therapies would benefit the global human population and will also have national economic benefits in both the public (healthcare) and private (biotech) sectors.