Objective 1: Developing a model system for studying long-term dynamics of social behaviour in bacteria
This objective is mainly completed. I have worked with collaborators Professor Soeren Molin and Dr Helle Krogh Johansen in the Danish Technical University to characterise social traits of interest in over 300 clinical isolates from sputa samples of patients with cystic fibrosis. These traits are of clinical interest but are also amenable to experimental evolution study in the lab. We have characterised (1) iron uptake systems both experimentally and through genome sequence analysis. This includes the pyoverdine system that relies on sharing of public goods, as outlined in my proposal but has also been expanded to include alternative iron uptake systems that rely on other non-social mechanisms. This line of research has expanded to explore the evolution of iron uptake in the lung in more detail and also to explore potential co-evolutionary processes occurring between different iron uptake systems.. (2) antibiotic resistance profiles: to date, this work has focussed on beta-lactamase production that involves the release of enzymes that destroy beta-lactam antibiotics into the environment surrounding the cell. Through our collaboration with Prof Molin and Dr Krogh Johansen, we have been able to analyse antibiotic resistance behaviour in relation to duration of infection and treatment regime experienced by patients from whom they were isolated. We have recently arranged to spend time working the laboratory of Professor Derrick Crook at the John Radcliffe Hospital in Oxford (brought forward due to the loss of our building! (www.bbc.co.uk/news/uk-england-oxfordshire-38934959) to establish a similar collect of Staphylococcus aureus isolates collected as part of a carriage study in the nasal passages of subjects that may or not be symptomatic of infection. this will allow us to broaden the scope of our project to other species of bacteria, beyond a single study system.
Objective 2: To investigate effects of social evolution in pathogenic bacteria
In brief, we have already shown that the answer to this question is yes. We were able to provide definitive evidence from sequence analysis of genes involved in pyoverdine-mediated iron uptake system to show that reduced pyoverdine production over time was due to invasion by cheats (see Andersen et al. 2015 PNAS). We are now developing this line of research towards the understanding of what happens to "iron starved' populations of bacterial cells in the lung, once pyoverdine has been lost from the population. The implications of this finding are potentially far-reaching: behaviour observed in bacterial cells is typically interpreted as a response to the host environment and assumed to be optimised for survival in the host environment. We were able to show that competition with conspecifics can drive beneficial traits (such as pyoverdine production) extinct. This is a well-known consequence of cooperator cheat dynamics, also captured in the Prisoner's Dilemma and the so-called Tragedy of the Commons.
As well as cooperative behaviour, we have also examined consequence of harming social behaviours in clinical strains. Specifically, we have produced evidence to show how bacteriocin production changes through duration of infection. We have found that cells become less toxic to one another through time in infection and are working to explain this pattern in terms of selection pressures operating on maintenance of effective but potentially costly "protection" mechanisms. We have recently expanded this line of enquiry to include alternative killing mechanisms - specifically contact dependent killing involving the Type VI secretion system.
We are currently finalising experiments designed to investigate consequences of social interaction on another trait - antibiotic resistance through beta-lactamase production. It is possible, that cooperation facilitates the persistence of "susceptible" cells in situ - cells that do not produce beta-lactamase but that benefit from the protection offered by the beta-lactamase produced by neighbours.
Objective 3: Social engineering: manipulating the genetic composition of bacterial infection by cheat invasion
Work on this objective has so far focussed on the development of a cheat that has been optimised for the ability to invade. We are using evolutionary theory to guide us as to which traits are most likely not achieve maximum invasive potential. This work is on-going but so far has focussed on comparison of different quorum sensing mutants. We have chosen these strains as they can potentially save cost of performing a wide range of social behaviours that are density dependent and under quorum sensing control.
