Maths-AIM: A mathematical and experimental approach for the rational assessment of bacterial Adhesion Inhibitor Materials in vivo

Using mathematical modelling combined with experimental data, this research will accelerate the development of novel treatments for bacterial infections. These treatments will be designed to minimise the likelihood that bacteria develop resistance to them. Since the discovery of penicillin in the early 20th century, antibiotics have been effective drugs for the treatment of bacterial infections. However, bacteria are resilient organisms that can evolve during the course of a single infection. If, during an infection, one bacterial cell produces offspring with a random genetic mutation enabling it to evade the action of antibiotics, it will survive the course of treatment while also producing genetically identical offspring. Thus, an antibiotic-resistant strain of bacteria can rapidly emerge. Consequently, strains of bacteria exist that are resistant to multiple antibiotics, including some that are resistant to all known antibiotics. This presents a huge problem for human and animal health and an economic burden upon the healthcare system, veterinary services and agricultural industries. Without intervention the situation will continue to deteriorate.

New types of treatment for infections must be explored as a matter of urgency and strategies to prevent bacteria from developing resistance to these treatments put in place. Rather than kill bacteria (as antibiotics do), our approach is to prevent bacteria from being able to cause infection. This should weaken them sufficiently to allow clearance by the immune system without providing an environment in which a resistant strain can flourish.

Bacteria employ a vast array of mechanisms to cause infection but one that is universal to all bacteria and absolutely required for infection is an ability to bind to proteins on host cells in animals, humans or plants. This means that inhibiting adhesion to host cells should prevent infection. Molecules can be designed that mimic either the proteins on the host cells (so that bacteria mistakenly bind them instead of host cells) or proteins on the bacterial cells (so that the molecules bind the host cells and prevent the bacteria from binding there). We call these Adhesion Inhibitor Materials (AIMs). If bacteria develop resistance to AIMs they should also lose the ability to bind to host cells, rendering them unable to cause an infection. Thus, unlike with antibiotics, developing resistance to AIMs should not benefit the bacteria and the emergence of drug-resistance should be prevented.

AIMs have been shown to weaken infections, but it is not yet proven that AIMs can be successful in treating infections (as well as preventing them) and whether they really would circumvent drug resistance. Treatment development is lengthy and costly and all avenues must be explored to accelerate it. We will employ mathematical modelling to investigate AIMs on a computer. Using data collected on Pseudomonas aeruginosa (an infectious bacteria present throughout nature) and a prototype AIM that has shown promise in our preliminary experimental studies, we will develop a set of equations to accurately simulate the dynamics of bacteria and drugs in an infection. This will enable us to test how successful the AIM would be in treating various types of infection and determine the likelihood of the bacteria to develop resistance to it. Importantly, numerous strategies will be devised from the model to improve the efficacy of the treatment and to prevent the emergence of bacterial resistance both to the AIM and to antibiotics (through alterations to its design and dosing regimens) and tested in the laboratory.

This combined modelling and experimental approach will facilitate the optimisation of a new treatment on a computer prior to it being developed on a grander scale for testing in trials. Ultimately this work will contribute to the development of future antibacterial treatments that will combat the rise in drug-resistance on a national and worldwide scale.

There is an urgent need for new treatments for bacterial infections that minimise the likelihood that resistant strains of bacteria will emerge. Building on our prototype Adhesion Inhibitor Material (AIM) (based on Multivalent Adhesion Molecules) and mathematical framework for predicting the population dynamics (of host target and immune cells, bacteria and drug) at an infection site, we will use a differential equation model to rationally assess the potential for AIMs to treat bacterial infections in vivo.

Models will be parameterised using combinations of point estimates (e.g. using SBToolbox for Matlab) and distributions of estimates (e.g. Markov Chain Monte Carlo simulations to describe uncertainty around parameters) from high-frequency time-series measurements of co-cultures of epithelium, Pseudomonas aeruginosa, macrophages, AIMs and antibiotics where appropriate.

Parameter and asymptotic analyses will identify alterations to the design and dosing regimens of AIMs to optimise their efficacy. Realistic alterations will be implemented to test the in silico model predictions in the laboratory, yielding extra data with which to further improve the model.

High-throughput evolutionary screening assays will identify strains of P. aeruginosa that are resistant to the AIMs with sequencing identifying the mechanisms of resistance. The mechanisms will be incorporated into the equations so that the population dynamics model will predict the likelihood of strains that are resistant to either the AIMs or antibiotics emerging during infection (i.e. by assessing the fitness costs of resistance at an infection site). Strategies to circumvent this (e.g. combined treatment with antibiotics or targeting an extra bacterial trait) will be predicted in silico and tested experimentally wherever possible. Thus AIMs and dosing regimens will be identified that maximise efficacy and minimise the emergence of drug-resistance.

Antimicrobial resistance (AMR) is an enormous worldwide concern, linking all living things and their environment. Farms are reservoirs of AMR that can be transferred to humans, and human and veterinary medicine activity contribute to AMR in wildlife and farms. Tackling one aspect of AMR will make significant contributions elsewhere. Moreover, its reach extends beyond the obvious failed treatment of infections: failure to prevent infection is equally important (e.g. in agriculture or during surgery or chemotherapy). The impact of developing novel treatments such as Adhesion Inhibitor Materials (AIMs) that will help tackle AMR cannot be underestimated.

The impact of AMR on animal health is huge: worldwide the majority of antibiotics are deployed for animals or agriculture. Antibiotic use for animal growth promotion is banned in the EU due to its links with AMR, but not elsewhere. Without alternative therapeutics, bans may be ignored or lead to misuse of antibiotics in ensuing infections arising as a result of unsanitary farming conditions, thus worsening AMR. We propose the development of a new treatment that could be used alongside an antibiotic ban to attenuate this issue.

Adhesion is a universal requirement for bacteria to cause infection so AIMs should be broad-spectrum. To parameterise and test our study we will use the ubiquitous bacterium Pseudomonas aeruginosa that infects humans, animals and plants. It is present in water, soil and many man-made environments (e.g. animal watering systems), capable of acute or chronic infections and a notorious biofilm former. To extend impact, we will consider bacteria more broadly through variations to model parameters that capture responses to treatment by other organisms.

This work should lead to the use of Adhesion Inhibitor Materials to treat bacterial infections in the medium term, providing a long-term legacy for interdisciplinary approaches in treatment development. Upon completion of the project, the accelerated background work will mean pharmaceutical companies gain from a treatment that is closer to their desired product, drastically reducing the work to be done before it is ready for trials. This will benefit the public, clinicians and veterinarians. We will hold a workshop to publicise the work (and interdisciplinary approaches to treatment development in general) to industry. Results will be disseminated to clinicians at symposia of the NIHR Surgical Reconstruction and Microbiology Research Centre. Impact will be enhanced significantly by our ongoing collaboration with UT Southwestern Medical Center facilitating in vivo tests.

The interdisciplinary approach will contribute to 3Rs research in the short and long-term with improved methods for the assessment and prediction of in vivo activity prior to extensive in vivo testing. UK and worldwide policy-makers will benefit from the ability to accelerate design of treatment strategies to minimise the emergence of AMR, promoting the UK worldwide as leading the way in using these cutting-edge techniques. In the short term and beyond, all models and techniques will be public so that researchers can adopt the framework to be specific to their pathogen, host or infection-site of interest. Our approach and results will be publicised through open access publications and presentations at relevant conferences.

All staff will enhance their team work, communication and interdisciplinary skills, thus adding to their skill set and contributing ultimately to the UK economy.

The work will be publicised in a variety of outreach events generating a better understanding of the good use of antimicrobials, contributing to the UK's health. Future generations of scientists will be exposed to true interdisciplinary work of timely and topical interest that illustrates the importance of encompassing multiple subjects to make real strides in research. Stimulating their interest here will have a long-term impact on the UK's economy.