A photo of bottles on a shelf

Clinical PhD projects 2020

These are the PhD projects available to Clinical Research Training Fellows joining the CMBI in October 2020. The links to the supervisors' webpages allow potential applicants to explore in further detail the research covered by each PI group.

Project TitlePrincipal Supervisor

Mechanisms and mitigation of colistin treatment failure

Cystic fibrosis (CF) affects >10,000 people in the UK. Early mortality characterises the disease, largely driven by chronic lung infection and inflammatory airway wall damage. By adulthood >60% patients are chronically infected with the Gram negative organism, Pseudomonas aeruginosa (Pa). Current therapy is limited, and once infection is chronic, the best that can be achieved is ‘suppression’ of bacterial load, usually with inhaled antimicrobials. The Strategic Research Centre for Pa infection was established by Prof Jane C Davies, with a focus on improving understanding of pathogenic mechanisms and therapies. Its partnership with the Royal Brompton Hospital’s CF clinic, one of the largest in Europe, through senior clinical academics, underpins the strong translational focus of the programme. The narrow pipeline of new antibiotics under development means that work to improve efficacy of existing agents is urgently needed.

Colistin is the polymyxin antibiotic used most commonly to control chronic P. aeruginosa infection in CF. Unfortunately, whilst colistin is usually effective in suppressing infection, it is almost never able to clear P. aeruginosa from the lungs once chronic infection is established. Intravenous colistin is also used as a ‘last-resort’ agent in severe disease, and thus, the emergence of resistance to polymyxin antibiotics is a growing concern. Efforts to improve colistin efficacy have been hampered by a poor understanding of the antibiotic’s mode of action and the lack of knowledge around the impact that the host environment has on bacterial susceptibility. Recent work from the Edwards lab, based in the MRC Centre for Molecular Bacteriology and Infection, has revealed that colistin targets LPS in both the outer and cytoplasmic membranes, leading to bacterial lysis and killing (Sabnis et al., 2019). We have also shown that colistin resistance due to the mobile colistin resistance (MCR) family of LPS modifying enzymes (Liu et al., 2016) is due to modification of LPS at the cytoplasmic membrane (Sabnis et al., 2019).

We exploited this information to develop a combination therapeutic approach to enhance colistin activity. We found that the experimental antibiotic murepavadin caused the accumulation of LPS in the cytoplasmic membrane of P. aeruginosa, which sensitised the bacterium >1000-fold to colistin-mediated killing. Given the geographical heterogeneity in airway deposition of inhaled agents related to airway narrowing and mucus plugging, and the resulting variability in drug concentrations, successful approaches to enhance efficacy of lower drug concentrations could have direct clinical impact.

The crucial next step in this work is to determine how the host environment influences LPS processing and transport, which we hypothesise will have significant effects on colistin susceptibility and therefore treatment outcomes. For example, during this work, we found that LPS released by bacteria exposed to colistin can sequester the antibiotic, rendering it ineffective. We also found that the presence of human serum renders P. aeruginosa tolerant of colistin. These findings indicate that colistin efficacy is affected by the in vivo environment, but this requires further investigation.

The aim of this project is to understand how the lung environment affects colistin treatment efficacy and to develop new therapeutic strategies to improve patient outcomes.

The objectives of this project are: 1, determine how the lung environment affects LPS modification and transport in P. aeruginosa and whether this modulates bacterial susceptibility to colistin. 2, investigate whether LPS released by bacteria into the lung environment influences colistin activity and whether endotoxin levels in sputum would provide a useful biomarker for treatment efficacy. 3, optimise methods to recapitulate the in vivo environment (artificial sputum media, whole sputum, bronchoalveolar lavage) to 4, test new approaches to enhance colistin efficacy including the use of molecules to modulate LPS transport and thereby increase susceptibility of bacteria to colistin. 

This will involve a multi-disciplinary approach, including patient evaluation, recovery of relevant samples and well-established laboratory-based methods. The supervision package enables exposure to tertiary level CF clinical care and an understanding of the translational pathway of drug development through to clinical trials. As such, it reflects a truly synergistic and optimal partnership between microbiology and clinical supervisors.


Sabnis, A. et al. bioRxiv, 479618 (2019).

Liu, Y. Y. et al. Lancet Infect. Dis. 16(2), 161-168 (2016).

MacNair, C. R. et al. Nat. Commun. 9(1), 458 (2018).


Dr Andrew Edwards (CMBI) and Prof Jane C Davies (NHLI)

Dr Andrew Edwards

Understanding the mechanisms of emerging polymixin resistance in carbapenemase producing organisms by combining innovative diagnostic solutions with fundamental biology

Carbapenemase Producing Organisms (CPO) are an evolving global threat which pose a substantial clinical, operational, and financial challenge. These organisms are associated with high morbidity and mortality rates and therapeutic options are severely restricted with clinical management often relying upon “last line” antibiotics. Typically polymixins B and E are used as last-line agents yet they are associated with nephrotoxic and neurotoxic side effects, and reports of colistin resistance are increasing. Lipopolyscahharide modifications, often through the addition of phosphoethanolamine or L-amino-arabinose, increases the charge of lipid A moieties resulting in reduced affinity for positively charged polymixins. This is often encoded by chromosomal mutations or through acquired mcr genes, which are frequently located on highly transmissible plasmids.

Our research teams have identified and characterised novel chromosomal and plasmid mediated resistance mechanisms. We are using spatial-temporal modelling techniques incorporating genomic, phenotypic and clinical information to better understand the molecular epidemiology of resistant isolates, and the role of horizontal gene transfer and clonal expansion in their spread. Polymixin susceptibility testing is not routinely performed in diagnostic laboratories because the methods are laborious and are associated with sensitivity and specificity issues. To address this challenge we are developing new diagnostic assays based on: (i) MALDIxin; (ii) data analytics and machine learning to increase diagnostic turnover and, (iii) microchip-based lab-on-chip technologies for rapid detection of relevant genetic variants.

In partnership with ICHT, this project will examine the emergence and prevalence of colistin resistance amongst CPO isolates with the aim of establishing a gold standard for polymixin susceptibility testing methods using innovative lipidomic based mass spectrometry methods and nucleic acid amplification tools. The data and technologies will support the development of effective interventions and diagnostic pathways for clinical, operational and infection control management. The project aims to determine the genetic basis, molecular epidemiology, prevalence and genomic/phenotypic characteristics associated with polymixin resistance using next generation sequencing, genomic analysis and epidemiological modelling. Using this data, targeted mass spectrometry and nucleic acid based technologies will be developed for the rapid detection of colistin resistance.

Supervisory team and infrastructure – Professor Alison Holmes is a Professor of Infectious Diseases at Imperial College London, Director of Infection Control at ICHT and Director of both the NIHR Health Protection Unit in Healthcare Associated Infections/AMR and Imperial College’s Centre for Antimicrobial Optimisation. She has a strong track-record in clinical infectious diseases and leads a large multidisciplinary team working on cutting edge infectious disease research, including the development of emerging, innovative technologies to address AMR. Dr Gerald Larrouy-Maumus is a Senior Lecturer in Molecular Microbiology at the MRC-Centre Molecular Biology and Infection (CMBI), at Imperial College. Over the last decade he has developed cutting edge expertise in biochemistry, molecular biology, microbiology, lipidomics and metabolomics of pathogens. His lab is pioneering bacterial antibiotics susceptibility testing on intact bacteria using lipids for identification and read-out of AMR using MALDI ToF, the workhorse of clinical microbiology labs. He has over 42 peer-reviewed publications in the fields of AMR, mycobacteria, metabolomics, lipidomics and biochemistry, which are all relevant to the present proposal.

This project builds upon existing multidisciplinary research on CPOs and colistin resistance across Imperial College and will be supported by dedicated protocols and characterised isolate collections and databases. It will benefit from partnership with ICHT, co-supervision with ICHT Microbiology colleagues, and well established cross-disciplinary collaborations, expertise, facilitates and infrastructure, including: Imperial College’s Antimicrobial Research Collaborative (ARC), the NIHR Health Protection Research Unit (HPRU) in AMR/HCAI, the BRC capital funded AMR laboratory and the multidisciplinary CPE working group comprising academic researchers, and ICHT Microbiology and IPC Departments.


Rodriguez-Manzano et al., 2019.  Analytical Chemistry

Otter et al., 2016. Clinical Microbiology and Infection, November 2016

Gharbi et al., 2015.  International Journal of Antimicrobial Agents

Freeman et al.  Journal of Antimicrobial Chemotherapy

Mookerjee et al., 2018, Journal of Hospital Infection

Knight et al., 2018. BMC Medicine

Otter et al., 2017.  Scientific Reports

Malpartida-Cardenas et al. 2019. Biosensors and Bioelectronics.


Dr Gerald Larrouy-Maumus (Dept of Life Sciences) and Professor Alison Holmes (Dept of Infection)

Dr Gerald Larrouy-Maumus

Evaluating the role of the microbiota in inflammatory bowel disease therapeutics

The inflammatory bowel diseases (IBD), Crohn’s disease (CD) and ulcerative colitis (UC) are chronic diseases of the gastrointestinal tract; their incidence is rising (1). IBD and its complications represent a growing global health and economic burden.  We have recently shown that infection of C3H/HeNCrl (C3H) with the mouse pathogen Citrobacter rodentium provides a robust system to model active UC in humans. In particular we have shown that both UC and C. rodentium infection are characterised by significant faecal neutrophil elastase (NE) activity, not seen in control samples. In contrast, C. rodentium-infected

C57BL/6 (C57) mice do not develop colitis (Barry, Williams, Frankel, Mucosal Immunology, in revision). As C. rodentium infects mice in the context of the endogenous microbiota, it provides an ideal platform to model the role of the microbiota in IBD.  There is now compelling evidence that the microbiome plays a significant role in the efficacy and toxicity of many commonly used IBD drugs. Personalised medicine approaches in IBD may elucidate the role of the microbiome in drug metabolism and function. The potential of this approach is not just in personalised drug targeting, or in the significant area of drug discovery, but also in the optimisation of current therapies through stratified microbiome manipulation and modulation.

Commonly used therapies for IBD include 5-aminosalicylates (5-ASA / Mesalazine) and thiopurines. Disease response to these drugs varies widely between patients; some respond rapidly and maintain remission long-term, whilst others do not respond, relapse early, or experience side-effects. Many of the preparations of Mesalazine which are used in the treatment of IBD rely on the bacterial cleavage of prodrugs (2). Mesalazine had been shown to alter faecal microbial profiles in irritable bowel syndrome (3). There have been no robust studies in IBD.

Intriguing recent research in a mouse model has strongly implicated the colonic microbiota in the efficacy of thioguanine (TG). Rapid local bacterial conversion of TG correlated with decreased intestinal inflammation and immune activation (4).

Brief description of proposed work

The aim of this project is to investigate the role of the microbiome in the therapeutic effects of key IBD drugs through carefully planned interventional study in murine models, a chemostat model, and humans.

We will establish the impact of the current leading drugs on the microbiome of naïve C3H and C57BL/6 mice. We will also follow faecal inflammatory markers (e.g. Calprotectin, Lipocalin 2 and NE activity). We will then study the efficacy of the drugs in the IBD C. rodentium models and correlate protection/adversity with inflammation and dysbiosis. We will conduct the treatment regimes in mice pre-treated with selective antibiotics and perform mouse faecal transplantation experiments (GF).

As we have shown significant faecal NE activity in UC patients, we will test what effect recombinant NE (available commercially) has on the microbiome of control and UC patients in a control chemostat model (JM).

The microbiome, and novel markers of inflammation, will be studied in the stool and mucosal biopsies of IBD patients before and after Mesalazine and thiopurine therapy; changes will be mapped to clinical response using validated scoring systems (HW).

The Clinical Research Fellow

A clinical research fellow in Gastroenterology will be ideally placed to undertake experimental work in the mouse model of colitis, translating this into humans. They will be able to continue their clinical training at Imperial, with endoscopy and clinical commitments as appropriate.


Ng, S. C. et al. Lancet 390, 2769–2778 (2018).

Sousa T et al. J Pharm Sci. 2014; 103: 3171-5.

Andrews CN et al. Aliment Pharmacol Ther. 2011; 34: 374-83. 4. Oancea I, et al. Gut. 2017; 66: 59-69.


Academic supervisors: Gad Frankel and Julie McDonald (CMBI)

Clinical supervisor: Horace Williams (Department of Gastroenterology, Imperial College Healthcare NHS Trust and Department of Metabolism, Digestion & Reproduction, ICL)


Prof Gad Frankel

Investigating the mechanism for the intestinal decolonisation of multidrug resistant pathogens using faecal microbiota transplantation

Antimicrobial resistance (AMR) is a serious threat to human health, resulting in treatment failures, infection relapses, longer hospitalisations, and poor clinical outcomes.1 Treatment options are limited, frequently less effective, and involve the administration of toxic antibiotics. AMR infections are currently responsible for approximately 700,000 deaths globally each year, and estimated to rise to 10 million deaths globally each year by 2050.2 There is an urgent need to develop new approaches to prevent and treat AMR infections, in particular multidrug-resistant infections.

The intestine is the primary colonisation site for multidrug-resistant organisms (MDROs) and serves as a reservoir for MDROs that are responsible for invasive infections (e.g. bacteraemia and recurrent urinary tract infections).1 Currently no clinical guidelines, standard of care, or specific treatment exists for patients with MDRO intestinal colonisation.

Studies have demonstrated intestinal decolonisation of carbapenem-resistant Enterobacteriaceae (CRE) and extended spectrum β-lactamase-producing Enterobacteriaceae (ESBL-E) following faecal microbiota transplantation (FMT), where faeces from a healthy donor are administered to a colonised patient. However, the mechanism of FMT is unknown, and there are several drawbacks to administering FMT to MDRO colonised patients. Potential risks include transmission of infections, invasive administration routes, unknown long-term effects, and concerns treating high-risk individuals (e.g. immunocompromised patients). Moreover, patients with MDRO intestinal colonisation often receive antibiotics post-FMT, which significantly reduces the decolonisation rate compared to patients that don’t receive antibiotics.3 FMT needs to be replaced with a safer, more effective treatment.

Description of proposed work:

The goal of this project is to determine the mechanism(s) by which FMT decolonises CRE and ESBL-E from the intestine, with the aim to develop a new method for MDRO intestinal decolonisation using microbial metabolites, microbial enzymes, bacteria, or bacteriophage (“agents”). This new treatment will be rationally-designed, effective, safe, and doseable.

Intestinal MDRO colonisation will be modelled in artificial gut models called “chemostats” (JM). For each experiment FMT will be administered to the first vessel, faecal filtrate to the second vessel, and saline to the third vessel. MDRO growth will be monitored over the course of the experiments, and longitudinal samples will be analysed using several omics techniques.

Potential agents identified in the chemostat experiments will be validated using human stool samples (MT). The targeted agents will be measured in stool samples from MDRO colonised patients that receive FMT, from MDRO colonised patients that don’t receive FMT, and from the FMT donors.

Batch culture experiments will be performed to demonstrate the direct inhibitory effects of the targeted agents (JM). In these experiments pure cultures of MDROs will be incubated with several defined concentrations of the targeted agent and decreases in MDRO growth will be measured.

Mouse intervention experiments will be performed, where mice with intestinal MDRO colonisation will be administered one of several defined doses of the targeted agent or PBS (GF). MDRO growth will be measured in mouse faeces to demonstrate the effectiveness of the agent as a treatment.

Clinical Research Training Fellow (CRTF):

A clinical research fellow in Infectious Diseases or Gastroenterology will be well positioned to carry out the chemostat experiments and analysis of the human stool samples. The CRTF will be able to continue their clinical training at Imperial during their research project.


Manges AR, et al. (2016) Infect. Dis. (Lond) 48(8): 587-592.

O'Neill, J. (2016). The review on antimicrobial resistance: HM Government and the Wellcome Trust.

Bilinski J, et al. (2017) Clin. Infect. Dis. 65(3): 364-370.


Academic supervisors: Dr Julie McDonald and Prof Gad Frankel (CMBI)

Clinical supervisor: Prof Mark Thursz (Department of Metabolism, Digestion and Reproduction, Faculty of Medicine, Imperial College London)

Dr Julie McDonald

 Antibiotic tolerance in the nosocomial pathogen Enterobacter cloacae after exposure to aminoglycosides

Infections caused by the Gram-negative, motile bacterium Enterobacter cloacae, a member of the ESKAPE group, are often hospital-acquired and can result in respiratory, intestinal or blood infections. E. cloacae expresses three β-lactamases and many multidrug efflux pumps, and the type strain (ATCC 13047) is also colistin resistant; emerging clinical strains often carry plasmids that confer resistance to various classes of antibiotics, including carbapenems. Despite its prevalence, the mechanisms of resistance, virulence and host-interaction of E. cloacae remain poorly understood.

Our initial work has led to a discovery that exposure to the common frontline drug gentamicin induces tolerance in E. cloacae. A portion of drug-sensitive E. cloacae (i.e. gentamicin MIC < 2 mg/L) became drug-tolerant (MIC > 4 mg/L) when exposed to gentamicin (20 mg/L for 6-24 h). Drug tolerance was transient and lost upon passaging in antibiotic-free media. Importantly, gentamicin or amikacin exposure resulted in small colony variants (SCV) of bacteria which were markedly more inflammatory, as they induced higher TNF and IL-1β after infection of human macrophages. We hypothesise that gentamicin-induced drug tolerance in E. cloacae could exacerbate disease and inflammatory responses leading to negative treatment outcomes.

This project will test this hypothesis by exploring, (1) the mechanisms of phenotypic tolerance in E. cloacae clinical isolates; (2) the altered host-interaction of bacteria before and after exposure to antibiotics. We will use RNAseq to identify differentially expressed genes after gentamicin exposure and whole-genome sequencing to identify mutations leading to tolerance. Deletion of candidate genes will help address their role(s) in tolerance and host responses. Assessment of innate immune activation (e.g. inflammasomes & TLRs) in primary human macrophages and cell lines will uncover the inflammatory potential of tolerant bacteria. Together, these studies will provide insights on altered bacterial and host responses in the face of antibiotic treatment. Gentamicin administration in patients leads to periodic peaks in serum levels, and our studies on the consequences of transient high antibiotic exposure could have profound implications on current clinical practice.

Dr Frances Davies (@DrFrancesDavies), the clinical co-supervisor, is a clinical Microbiology Consultant at Imperial healthcare NHS Trust/Northwest London Pathology, and honorary Clinical Sr Lecturer and MRC Clinical Academic Research Partnership fellow. Frances has a research interest in HCAI and CPE, and is the NHS laboratory lead for HPRU.

Dr Avinash Shenoy (@avi_shenoy), is a non-clinical Sr Lecturer in Molecular Microbiology. He is based in the Flowers bdg. (SK campus) and has a satellite laboratory at the Francis Crick Institute. Avinash has received funding from MRC, Wellcome Trust, Royal Academy, co-leads the MRes in Bacterial Pathogenesis & Infection, directs the CMBI HTSCA facility, and is a member of the DoID EDI committee.

Dr Avinash Shenoy

Enteric infection and therapy in a preclinical coloractal cancer model

Project details: The aim of this project is to establish a preclinical, mouse, coloractal cancer model (using implantation of cancer cells by colonoscopy). It is anticipated that the first year of the project would be spent on setting the model up. The second and third years would be spent on defining the molecular signatures of the tumour and its susceptibility to infection. Finally, we will test novel oncolytic bacterial therapy.   


Berger CN, Crepin VF, Roumeliotis TI, Wright JC, Carson D, Pevsner-Fischer M, Furniss RCD, Dougan G, Dori-Bachash M, Yu L, Clements A, Collins JW, Elinav E, Larrouy-Maumus GJ, Choudhary JS, Frankel G. Citrobacter rodentium Subverts ATP Flux and Cholesterol Homeostasis in Intestinal Epithelial Cells In Vivo Cell Metab. 2017 Nov 7;26(5):738-752.e6. doi: 10.1016/j.cmet.2017.09.003. 

Scott AJ, Alexander JL, Merrifield CA, Cunningham D, Jobin C, Brown R, Alverdy J, O'Keefe SJ, Gaskins HR, Teare J, Yu J, Hughes DJ, Verstraelen H, Burton J, O'Toole PW, Rosenberg DW, Marchesi JR, Kinross JM. International Cancer Microbiome Consortium consensus statement on the role of the human microbiome in  carcinogenesis.Gut. 2019 Sep;68(9):1624-1632. doi: 10.1136/gutjnl-2019-318556. 

Chen YJ, Roumeliotis TI, Chang YH, Chen CT, Han CL, Lin MH, Chen HW, Chang GC, Chang YL, Wu CT, Lin MW, Hsieh MS, Wang YT, Chen YR, Jonassen I, Ghavidel FZ, Lin ZS, Lin KT, Chen CW, Sheu PY, Hung CT, Huang KC, Yang HC, Lin PY, Yen TC, Lin YW, Wang JH, Raghav L, Lin CY, Chen YS, Wu PS, Lai CT, Weng SH, Su KY, Chang WH, Tsai PY, Robles AI, Rodriguez H, Hsiao YJ, Chang WH, Sung TY, Chen JS, Yu SL, Choudhary JS, Chen HY, Yang PC, Chen YJ. Proteogenomics of Non-smoking Lung Cancer in East Asia Delineates Molecular Signatures of Pathogenesis and Progression Cell. 2020 Jul 9;182(1):226-244.e17. doi: 10.1016/j.cell.2020.06.012.


Academic Supervisor: Gad Frankel, CMBI

Co-supervisors: James Kinross, Department of Surgery and Cancer, St. Mary’s Hospital & Jyoti Choudhary The Institute for Cancer Research

 Prof Gad Frankel

What is the role of the respiratory tract microbiome in anti-bacterial immunity and can it be targeted to improve treatment of chronic lung disease?


Chronic obstructive pulmonary disease (COPD) is an inflammatory airway condition that affects ~8% of the UK population and is a major global cause of morbidity and mortality with limited treatment options.

The disease is characterized by susceptibility to infection by bacterial pathogens such as Streptococcus pneumoniae1. Greater understanding of how mucosal immunity is dysregulated in COPD is urgently needed to facilitate the development of more effective therapeutic approaches.

Mucosal surfaces within the human body are colonised by symbiotic commensal microorganisms, referred to as ‘microbiota’. These commensals have been extensively studied in the intestinal tract where they program critical components of the immune response to infectious challenge2. Historically, it was believed that the lower respiratory tract was sterile in health, but recent molecular microbiological studies have revealed that the lungs also harbour diverse microbiota with alterations (‘dysbiosis’) observed in COPD3-4. To date, respiratory microbiome research has consisted of human studies that provide descriptive information about differences between disease and health. These studies provide no insight into whether health-associated commensals fortify host-defences or if microbiome changes observed in chronic lung disease contribute to impaired anti-bacterial immunity. A major challenge for the field is to evolve from correlative studies to begin to address causation.

Using airway-delivered antibiotics targeted against lung commensals, we have recently developed a novel mouse model of specific respiratory microbiota depletion without affecting intestinal microbiota. This unique model provides a tractable system to provide the first ever functional understanding of the respiratory microbiome in immunity.


This project will examine the hypothesis that respiratory microbiota play a central role in regulation of pulmonary immunity to Streptococcus pneumoniae. Specific aims will be to: (i) identify components of the anti-bacterial immune response that are controlled by respiratory microbiota and define key commensals involved; (ii) understand if commensals altered in COPD affect responses to infection and the mechanisms through which this occurs; and (iii) determine whether microbiota manipulation can be used to enhance immunity to infection.

These aims will be addressed using analyses of banked samples from a unique longitudinal human COPD cohort, combined with functional studies in the aforementioned mouse model of respiratory microbiota depletion. The project will combine cutting edge techniques in immunology and microbiology with access to longitudinal human disease samples and thus provide comprehensive research training for an interested clinician who wishes to pursue an academic career. The mechanistic insight gained from the project is anticipated to lead to future development of new non-antibiotic (microbiota-focused) approaches to target dysbiosis and prevent/treat bacterial respiratory infections in chronic lung disease.


Dr Aran Singanayagam (@AranSinga) is an MRC Clinician Scientist, Group Leader at the Centre for Molecular Bacteriology and Infection (CMBI) and Honorary Consultant in Respiratory Medicine at Royal Brompton and Harefield NHS Trust. His research focusses on how the respiratory tract microbiota regulates immune homeostasis in health and how perturbations that occur in chronic lung diseases lead to immune dysregulation and impaired protection against pathogens.

Dr Thomas Clarke is a Wellcome Trust Sir Henry Dale Fellow and Group Leader at the CMBI. His research focusses on understanding how programming of innate immunity by the microbiota influences host responses to bacterial infection and vaccination, and how changes to the composition of the microbiota disrupts these responses.


(1) Beasley V et al. Int J COPD 2012; 7: 555-69, (2) Brown RL et al. Nat Commun 2017; 8: 1512. (3) Hilty M et al. PLoS One 2919; 5: e8578. (4) Singanayagam A et al. Sci Transl Med 2019; 11: 509.


Dr Aran Singanayagam and Dr Tom Clarke

A skin-based human challenge model for assessing efficiacy of TB vaccine candidate


A human challenge model for tuberculosis could be used to assess the efficacy of candidate vaccines at an early stage and prioritise those that should proceed to large scale efficacy trials. Human challenge models are already in use for vaccine development for several other infections including malaria, dengue, influenza and typhoid. However, deliberate infection of humans with M. tuberculosis would be unacceptable due to the 6-month multidrug treatment regimen and potential for latent infection. Others (Minassian et al. 2012; Harris et al. 2014) have demonstrated the potential of using the current BCG vaccine administered intradermally in humans as a surrogate for M. tuberculosis infection that can detect differences in anti-mycobacterial immunity induced by vaccination with BCG (Minhinnick et al. 2016) and a novel TB vaccine candidate MVA85A (Harris et al. 2018). However, application of the published model is limited by the need for intrusive skin biopsies to sample the BCG load at the site of injection, making serial sampling impossible.

Description of Challenge Organism:

To develop a human challenge model of tuberculosis we have a constructed an improved BCG challenge organism, expressing fluorescent reporter proteins. The fluorescent signal can be measured non-invasively through the skin using a digital camera equipped with LEDs to excite the fluorophores, providing an indicator of bacterial survival over time as the fluorescent signal decreases when bacteria die as a result of vaccine-induced immune responses.

Our current reporter system consists of BCG-Pasteur expressing Yellow Fluorescent Protein (YFP) and the red fluorophore Turbo-635, both expressed from a plasmid that is stably maintained by complementation of the pantothenate genes panCD deleted from the bacterial chromosome.

Preclinical Testing:

Our system has been tested using the preclinical BCG vaccination model in mice: vaccination with standard BCG followed 4 weeks later by intradermal challenge with the BCG fluorescent reporter strain. The site of inoculation is then imaged with significant reduction in signal observed in vaccinated compared to unvaccinated animals from 4 to 7 days after challenge. We have also challenged animals with M. tuberculosis alongside the BCG-reporter and demonstrated that loss of fluorescence signal in the skin is associated with protection again challenge in the lung.

Project Proposal:

Some further development and characterisation of the BCG reporter strain is required, alongside optimisation of the imaging platform, and setting up the clinical study.

1)    Remove the Turbo-635 reporter from the current strain and confirm current vaccination results. Data indicates that YFP gives by far the better signal, so we will remove the unnecessary second reporter.

2)    Consult with our photonics partner in Stanford on the optimisation of an imager suitable for human use on the forearm or other suitable site. Collect data on human skin fluorescence in relation to skin pigmentation.

3)    We propose to carry out a dose ranging clinical study using intradermal challenge with our BCG-fluorescent reporter system in normal healthy volunteers, to determine the sensitivity and discrimination of the fluorescent signal at intradermal challenge doses above and below the standard clinical BCG vaccine dose. This will require the development of study protocols (in collaboration with Prof McShane in Oxford), Ethical approval and consultation with the HSE on the GM aspects of the system.

4)    A follow-on study would then measure the fluorescent signal after BCG challenge in groups of normal healthy volunteers previously BCG-vaccinated compared to unvaccinated normal healthy volunteers to confirm the same vaccine-accelerated loss of fluorescent signal occurs as seen in the preclinical mouse model.

Harris SA et al. Evaluation of a Human BCG Challenge Model to Assess Antimycobacterial Immunity Induced by BCG and a Candidate Tuberculosis Vaccine, MVA85A, Alone and in Combination. J Infect Dis 2014; 209:1259–68.

Harris SA et al. Development of a non-human primate BCG infection model for the evaluation of candidate tuberculosis vaccines. Tuberculosis 2018; 108:99–105.

Lienhardt C et al. Translational Research for Tuberculosis Elimination: Priorities, Challenges, and Actions. Plos Med 2016; 13:e1001965. Minassian AM, Satti I, Poulton ID et al. A Human Challenge Model for Mycobacterium tuberculosis Using Mycobacterium bovis Bacille Calmette-Guérin. J Infect Dis 2012; 205:1035–42.

Minhinnick A et al. Optimization of a Human Bacille Calmette-Guérin Challenge Model: A Tool to Evaluate Antimycobacterial Immunity. J Infect Dis 2016; 213:824–30.

Reid MJ et al. Building a tuberculosis-free world: The Lancet Commission on tuberculosis. Lancet 2019, DOI: 10.1016/s0140-6736(19)30024-8.


Dr Brian Robertson (CMBI), Professor Graham Cooke (Department of Infectious Disease)

 Dr Brian Robertson & Prof Graham Cooke
PhD Example Projects