Primary Supervisor: Dr Jan Kreft, School of Biosciences
Secondary supervisor: Richard Horniblow, Biomedical Sciences
PhD project title: Microbial and host interactions in the gut microbiome
University of Registration: University of Birmingham
Gut microbiome and health. Disturbance of the intestinal microbiome and antibiotic overuse is understood to contribute to the increased incidence of modern-day diseases, highlighting the importance of the intestinal microbiome in health and disease. However, only with the recent revolution of next generation sequencing that drove the surge of research on the gut microbiome have we begun to discover the multitude of ways in which the microbes affect or health and even behaviour. Furthermore, bacterial-centred bioactive therapies are undergoing extensive experimentation and human trials (including faecal microbial transplantation and probiotics) and the use of microbial therapies is changing clinical practice in some disease settings, such as antibiotic-refractory C. difficile infection.
Gaps in our understanding. An important issue with sequencing DNA extracted from poo is that the spatial structure in which the various microbes are self-organized in the gut community, facilitating local interactions between neighbours, is completely destroyed. Likewise, many mathematical models to date ignore spatial structure by effectively modelling the gut as a continuous-flow stirred tank reactor (CSTR). We know from our work and other studies how important spatial structure is and also that mathematical models can predict how competitive and cooperative interactions generate certain spatial structures (1). The ability to predict and model such ecologies will be highly beneficial as more and more bacteria are identified as beneficial (and hence, could be used as treatment probiotics) and as we need to understand how the balance of ‘good’ and ‘bad’ bacteria change in our intestines in disease.
Approach. There are other issues with current research that can only be overcome by integrating laboratory experiments with animal and clinical studies by using mathematical models as the glue between them. Mathematical models can take data from one system as input to predict dynamics in another system, which can then be contrasted with data from that system. Many challenges arising from the complexity of the gut, consisting of thousands of often poorly studied microbes interacting with each other and the body, which is also highly complex. Models, mathematical and laboratory, help by simplifying this system, but one needs to check the simplifications are not caricatures of the system.
Probiotics. The challenge of complexity is evident from research on probiotics. Most had little success (2) because they were isolated from different habitats and it has been ignored that they have to integrate into an existing community. What is needed is model-aided design of synthetic multispecies probiotics where different members work together to form a stable community.
eGut. We have been developing an agent-based model to simulate the gut microbial community called eGUT for electronic gut. This work has been funded by NC3Rs and has progressed to a stage where the model can be applied to understand and predict a variety of interactions between microbes and microbes and the host mucosa.
Project aim and objectives. Our long-term goal is the validation of our agent-based modelling platform with role model applications in order to build a ‘customer base’ of users of our platform. This will open up lots of opportunities for collaboration with gut microbiome researchers and probiotics companies in the future. The aim of this project is to study the competition of pathogens with commensal or probiotic bacteria in the gut environment using a combination of mathematical and laboratory models of the gut and time permitting animal or human study data from collaborators. As the mathematical model is challenged with experimental data, it will be improved and validated iteratively. Additionally, through a recent clinical trial examining the efficacy of faecal transplant in treating inflammatory bowel disease, a number of key bacteria that could act therapeutically have been identified; utilising this model to predict how we could best deliver these bacteria (within probiotic formulations) would be clinically useful and will be utilised in this project.
Methods and skills. The project enables learning a wide range of skills and cross disciplinary working, from computer programming, modelling, data analysis and statistical inference to culturing bacteria and running laboratory models of the gut. Samples can be analysed for specific metabolites or untargeted metabolomics. Quantifying the population dynamics can be done by various methods such as qPCR or sequencing or flow cytometry, depending on the system and the bacteria to be analysed.
- Hellweger FL, Clegg RJ, Clark JR, Plugge CM, Kreft JU (2016). Advancing microbial sciences by individual-based modelling. Nature Reviews Microbiology 14: 461–471.
- Kreft JU (2017). Mathematical modelling of the microbiota for manipulating its membership (Part of editorial series: The microbiome as a source of new enterprises and job creation). Microbial Biotechnology 11: 145–148
- Kreft JU, Gülez G (2019). Microbiology and Mathematics: microbiological meaning from mathematical models. Microbiologist (SfAM magazine) 20: 24–29
BBSRC Strategic Research Priority: Sustainable Agriculture and Food:Animal Health and Welfare & Renewable Resources and Clean Growth: Industrial Biotechnology & Understanding the Rules of Life: Systems Biology& Integrated Understanding of Health: Ageing
Techniques that will be undertaken during the project:
- General laboratory and microbiological methods
- In vitro models
- Individual-based modelling
Contact: Dr Jan Kreft, University of Birmingham