Investigation of stochastic variations in growth rate as the mechanism of drug tolerance in Mycobacterium tuberculosis

Lead Research Organisation: University of Surrey
Department Name: Microbial Sciences


Mycobacterium tuberculosis is a major pathogen of man and animals. Drug treatment is available for human disease but it takes six months, which is impractical in developing world settings where TB is most common. Consequent non-compliance with treatment regimes leads to the emergence of drug-resistance. This is now a major world-wide problem with practically incurable 'extreme drug-resistant' strains appearing in many countries, including the UK. Treatment would be much more effective if treatment regimes could be shortened. The principle reason why treatment regimes have to be so prolonged is that that a sub-population of cells in lesions are relatively tolerant to the antibiotic, a phenomenon termed phenotypic tolerance or non-inherited antibiotic resistance. The mechanism of phenotypic tolerance is currently unknown but we have recently discovered that it is associated with slow growth in M. tuberculosis. We have recently shown, through theoretical studies, that random variations of growth rate in individual bacteria cells may the mechanisms for generating persisters. This project is to investigate this hypothesis in an experimental system. The results will provide a new understanding of the mechanistic basis of phenotypic tolerance that is likely to provide new targets and opportunities to interfere with the resistance mechanisms and thereby increase efficacy of TB treatment.

Technical Summary

Mycobacterium tuberculosis (Mtb) is a major pathogen of man and animals. Drug treatment takes six months or more because the pathogen exhibits phenotypic tolerance to the antibiotic regime in vivo. The key signature of antibiotic tolerance is the biphasic kill curve which is obtained when growing bacteria are exposed to antibiotic. The killing rate is initially rapid but then slows down. This tail of slowly killed bacteria is responsible for the phenomenon of drug tolerance. However, the mechanism by which bacteria become tolerant is completely unknown. The most popular theory involves phenotypic switching of cells between two or more states (bistability) corresponding to normal and antibiotic-tolerant cells. However, there are many problems with this theory, for instance, that it has not been possible to engineer cells in pure normal or persistent states. We have developed an alternative persistence hypothesis whose starting point is the well-established fact that antibiotic killing of bacteria is growth rate-dependent. We then consider cells in which growth rate is controlled by a single gene and demonstrate that such a system would generate biphasic kill curves if the gene is expressed at low level and thereby subject to stochastic noise. The aim of this project is to investigate this hypothesis for the TB bacillus. We will engineer cells whose growth rate is controlled by a gene with a range of promoter strengths. We will investigate population variation in gene expression, growth rate and antibiotic killing by single cell studies and differences in metabolism by 13C-metabolic flux analysis. The studies may lead to develop of new therapeutic strategies that target the ability of the pathogen to enter or maintain itself in the persistent (drug-tolerant) state.

Planned Impact

This research will investigate a problem of huge importance for tuberculosis control, drug tolerance. TB affects about 15 million people in the world today causing nearly 3 million deaths. The research will therefore benefit our understanding of this devastating disease and lead to new therapeutics. The market for antituberculous drugs is estimated to be USD 612-670 million annually. In the UK there are several pharmaceutical companies that that have an interest in development of novel antuberculous drugs, such GSK and AstraZeneka. Many biotech companies are involved in the development of new antibiotics. We plan to engage with potential industrial partners in a number of ways. Firstly, we will use our existing contacts. Professor McFadden and co-I's have considerable experience interacting with industry. For instance, Professor McFadden is currently PI of a collaborative research project with Sanofi Pasteur to develop new meningococcal vaccines. The University of Surrey degree structure also provides additional routes to engagement with potential partners as most of our students spend one year in a professional training placements, often in an industrial research laboratory. For instance, bioscience students are currently placed in GSK and the company Biocompatibles (Farnham, Surrey) involved in the development of drug delivery vehicles. The applicants make regular visits of students at these placements, providing opportunities for discussing this project with industrial scientists who may be interested in partnering the work. The research will also benefit the pharmaceutical more generally as drug-tolerance is a major problem in many other diseases such as those caused by staphylococci and streptococci. By leading to new drugs capable of controlling infectious diseases more effectively, the research will therefore benefit the UK and European public and worldwide. The project will also contribute to the training of interdisciplinary scientists capable of pursuing a career in the fast-moving field of systems biology.


10 25 50
Description Our aim was to investigate growth rate and antibiotic killing at the single cell level:
1. Generation of a promoter probe library constructed and analysed: We have constructed and analysed a set of promoters which constitutively express the gene of choice at a range of levels. The Mycobacterial PL5 promoter was mutated using 8-oxo-dGTP and dPTP-PCR-mutation (as described in Zaccalo et al.,1996) to create a library of mutated promoters. Following vector cloning upstream of GFP, we screened over 60 mutated promoters for differential expression using GFP fluorescence (485/535nm). This has also been carried out successfully in M. bovis BCG. We went on to analyse persistence levels of strains where growth rate is controlled using our developed promoters and the TBLeuD gene in Mycobacterial LeuD mutants. This demonstrated (paper in preparation) that strains whose growth was slowed had a higher level of persisters, proving our hypothesis that persisters are a product solely of slow growth.
2. Single cell measurements: We have further developed out microfluidics platform to enable us to continuously monitor a large number of individual bacteria over an extended time period. Bacterial cells are immobilised against the bottom surface of the device and the plate is sealed using a pneumatic manifold which also allows user control of media administration. We have recorded growth rate in individual cells for up to 6 generations and have developed object tracking with M. smegmatis cells during the initial growth phase. This can track the complete history of each cell despite the "jumping" motion of Mycobacterial cells upon division. After recording the growth rate in individual cells in the microfluidics device we have switched the feed to media containing appropriate antibiotics and tracked killing in single cells. Cell death has been tracked using the exclusion dye propidium iodide (535/617nm). Any surviving cells or potential persisters were allowed to recover by switching the media back to growth medium only. True persisters were noted as cells that survived antibiotic killing, regrew in the growth media and were then killed again by a second round of antibiotic administration. We have tracked such persisters in the E coli high persistence mutant hipQ and analysed the individual cell parameters. Our aim as described in the grant was to enable us to track backwards and determine the growth rate of persisting cells which we have successfully carried out in the E. coli hipQ high persistent mutant. We analysed growth, division and persisters for wild-type and mutant. We discovered that growth rate was the primary determinant of persister status but we also discovered that inheritance of phenotypic variation showed a puzzling feature: sisters cells were highly correlated for growth rate but not with their mother cell. This begs the question of where and how the sister-sister correlation has been established. We are continuing to investigate this phenomenon. We also identified by whole genome sequencing the gene for the hipQ mutation and demonstrated that it both increased persister level but also decreased the level of phenotypic inheritance. This is a novel phenotype and a paper has been submitted describing this interesting finding. Mycobacterial single cell experiments have also been carried out using this system with the antibiotics streptomycin (30/50/100 µg/ml), rifampicin (25/50/100/175 µg/ml) and pyrazinamide (30 µg/ml) with varying kill times and extended recovery times to up to 2 months. Potential "persisters" in M. smegmatis::pL5m1::GFP (i.e. GFP under control of the mutated promoter 1) were identified following 12 hours growth and 12 hours kill time with 175 µg/ml rifampicin. Propidium iodide staining allows dead cells to be seen in red and potential "persister" cells are seen in green, i.e. those expressing GFP. Our experiments with M. smegmatis have demonstrated that a similar pattern of phenotypic inheritance of growth rate variation is found in this species, as in E. coli.
3. We developed a new kinetic model of metabolism of E. coli.
To conclude, we developed new tools (e.g. microfluidic device/ object tracking/ single cell parameter analysis/ promoter library and growth rate controlled strains) necessary to characterise and analyse growth rate and antibiotic killing at the single cell level. Our analysis of persisters in both E. coli and mycobacteria demonstrated that, at both a single cell level and population level, phenotypic variation in growth rate is the primary determinant of persistence. We have identified a novel gene associated with persistence in E. coli.
As well as our Hu et al (2017) paper, we have made recent presentations of our work, for example: 1) British Society for Antimicrobial Chemotherapy Antibiotic Resistance and Mechanisms workshop, Birmingham (November 2016) and 2) for American Society for Microbiology conference on Tuberculosis, New York (April 2017).
Exploitation Route The results, particularly identification of a novel gene associated with persistence, will stimulate new approaches to antibiotic killing of bacteria.
We also constructed a new kinetic model of metabolism in E. coli that will be of use for biotechnological developments
Sectors Healthcare

Pharmaceuticals and Medical Biotechnology

Description The award led to the development of new methods for real-time analysis of bacterial growth and division. and has led to 7 published papers. The two key papers are Hu et al, 2017 which describes the development of novel methods to analyse the growth and replication of bacteria in real-time. This will be very useful to researchers studying persistence. The second key paper, Hingley-Wilson et al, 2020, published in the high impact journal, Proceedings of the National Academy of Science, describes our exciting finding that bacterial persistence can be caused by loss of phenotypic memory in bacteria. We also describe a novel gene that is involved in maintaining phenotypic memory, which is an entirely novel phenotype. Elucidating the mechanism of phenotypic memory is likely to inspire new approaches to tackling drug-resistance.
Sector Digital/Communication/Information Technologies (including Software),Healthcare
Impact Types Societal


Description BBSRC Small Grants - SurreyFBA: Interactive tool for computer simulations of genome scale metabolic networks.
Amount £142,917 (GBP)
Funding ID BB/K015974/1 
Organisation Biotechnology and Biological Sciences Research Council (BBSRC) 
Sector Public
Country United Kingdom
Start 08/2013 
End 04/2015
Description Standard Novel Strategies to Detect and Mitigate the Emergence of AMR in Zoonotic Pathogens
Amount £462,910 (GBP)
Funding ID EP/M027481/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 08/2015 
End 08/2017
Title Data from: Trajectory energy minimisation for cell growth tracking and genealogy analysis 
Description Cell growth experiments with a microfluidic device produce large-scale time-lapse image data, which contain important information on cell growth and patterns in their genealogy. To extract such information, we propose a scheme to segment and track bacterial cells automatically. In contrast with most published approaches, which often split segmentation and tracking into two independent procedures, we focus on designing an algorithm that describes cell properties evolving between consecutive frames by feeding segmentation and tracking results from one frame to the next one. The cell boundaries are extracted by minimizing the distance regularized level set evolution (DRLSE) model. Each individual cell was identified and tracked by identifying cell septum and membrane as well as developing a trajectory energy minimization function along time-lapse series. Experiments show that by applying this scheme, cell growth and division can be measured automatically. The results show the efficiency of the approach when testing on different datasets while comparing with other existing algorithms. The proposed approach demonstrates great potential for large-scale bacterial cell growth analysis. 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Title Individual sequencing read data for the E. coli HipQ mutant and parental strain from: Hingley-Wilson, S. M., et al. (2020). "Loss of phenotypic inheritance associated with ydcI mutation leads to increased frequency of small, slow persisters in Escherichia 
Description Whole genome sequence data from HipQ mutant of E. coli: Biosample, (accession nos. SAMN13648605 [E.coli HipQ mutant strain] and SAMN13648604 [E. coli parental strain]). 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
Impact the data will provide researchers with new insights into persistence mechanisms 
Description Presentation on persistence and antibiotic resistance 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact We gave a presentation on persistence and antibiotic resistance at the SUPERBUGS: THE FIGHT FOR OUR LIVES, event held at the Science Museum 2018, with over 2000 people attending.
Year(s) Of Engagement Activity 2018
Description Testing a novel model of Mycobacterium tuberculosis persistence by live single cell growth studies in a microfluidic device. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Poster presentation at Keystone Tuberculosis meeting in 2013 which was read and discussed by many participants and helped to stimulate microfluidic approaches in mycobacterial disease.

no actual impacts realised to date
Year(s) Of Engagement Activity 2013