Modelling the corneal microbiome to improve identification of pathogenic microorganisms

Lead Research Organisation: University of Liverpool
Department Name: Eye and Vision Science


Microbial keratitis (MK) is an infection of the cornea (the transparent front part of the eye) that may lead to ulcers, scarring and blindness. This in turn causes older people to lose their independence, people of working age to lose employment and can be devasting for families and communities, particularly in low and middle-income countries. Improving outcomes in MK depends on rapidly identifying the microorganism (bacteria or virus) causing the infection and targeting treatment towards this. One of the barriers to this has been the difficulty in collecting samples from the cornea. Until now, this has depended on the use of a sterile blade to scrape the infected area of the cornea. It can be very difficult to pick up enough bacteria on the blade for the laboratory to correctly identify what is causing the infection. A group, led by Professor Kaye at the University of Liverpool, recently developed a new, easy to perform method to obtain corneal samples called a 'corneal impression membrane (CIM)'. This is a specially adapted small piece of filter paper that is placed onto the infected area of cornea for 2-3 seconds.

Next-generation sequencing is a new technology that directly studies the genetic make up of infectious organisms and can be used to identify previously unknown infections. In addition, next-generation sequencing is rapid, with results in a few hours compared to 48 hours for current techniques. It can also detect genes responsible for resistance to antibiotics. These features give the potential for rapid, accurate and personalised treatment of MK.

A reason why next-generation sequencing is not currently used routinely is that it is very sensitive, meaning that other organisms that naturally inhabit the cornea without causing infection might be identified in error. Despite the cornea being the surface affected by MK, there is no data regarding these organisms, known together as the "corneal microbiome". As it is not invasive, the CIM sampling method offers an exciting opportunity to sample both the affected and unaffected eyes of patients presenting with MK as well as the eyes of patients unaffected by MK. By comparing these results we will gain a better understanding of which microorganisms cause infections.
The main aim of this fellowship is to better define the microorganisms causing MK, through a better understanding of the corneal and ocular surface microbiome in health and disease.

Firstly, I will set up an experimental model of corneal infection in the laboratory. This is needed to determine how well the sampling method (CIM) and the processing methods correctly identify eyes with and without MK. Alongside this, I will run a clinical study in which I will collect samples using the CIM from the affected and unaffected eyes of patients presenting with MK to St Paul's Eye Unit, Liverpool. Participants with no history of MK will be recruited to a control arm of the study.

In order to identify the organisms present, samples will be processed using three techniques: 1) conventional diagnostic culture, a technique used to multiply microorganisms through reproduction in predetermined culture medium; 2) polymerase chain reaction, a technique used to amplify small quantities of microorganism DNA and 3) next-generation sequencing, a technique used to detect all organisms in a sample.

My research will significantly benefit MK patients both nationally and internationally by increasing current understanding of corneal sample results and facilitating the potential future use of next-generation sequencing in routine ophthalmic clinical practice. Next-generation sequencing produces personalised results in a matter of hours compared to 3 days for conventional diagnostic culture. This allows the most appropriate antimicrobial therapy to be selected, potentially reducing MK disease duration and the risk of blindness. This is particularly important where it is not possible to monitor patients closely.

Technical Summary

Microbial keratitis (MK) is a major cause of blindness. The likely causative microorganism is only isolated in 40% of cases using conventional diagnostic culture (CDC). More sensitive methods such as microorganism targeted polymerase chain reaction (MTPCR) and next-generation sequencing (NGS) increase the detection of microorganisms but may increase the likelihood of picking up commensal microorganisms. A novel, minimally invasive corneal sampling method using a corneal impression membrane (CIM) enables sampling of both affected and unaffected eyes.

The primary aim of this fellowship is to better define the identification of pathogenic microorganisms in patients with MK through a better understanding of the corneal microbiome in health and disease.

I will develop an ex-vivo porcine corneal infection model to determine the sensitivity and specificity of CDC and MTPCR for different bacterial species at different stages of infection.

Simultaneously, I will run a clinical study at St Paul's Eye Unit, recruiting 151 MK participants and 80 participants to 4 control groups. CIMs and swabs will be obtained from the cornea, conjunctiva and eyelids of MK participants affected and unaffected eyes, in addition to nasal swabs. Control participants will have all samples obtained from one eye. All samples will be processed using CDC and a subset using MTPCR and NGS.

A comparison of identified bacteria obtained from CDC, MTPCR and NGS processing will be made. Microorganisms identified in the eyes with MK will be compared to the control fellow eye and other control groups and subtractive bioinformatic methodology applied to identify the likely pathogenic organism. Comparisons will be made between the microorganisms isolated from the cornea and contiguous sources to identify any possible endogenous sources of infection. This will facilitate potential future use of NGS in routine ophthalmic clinical practice, leading the way for personalised diagnosis and treatment of MK.

Planned Impact

Microbial keratitis (MK) is an ophthalmological emergency that can lead to blindness. Outcomes depend on the rapid identification of the causative microorganism and targeted treatment. This project aims to better define the pathogenic microorganisms in patients with MK through a better understanding of the corneal microbiome in health and disease. Sampling will be carried out using a novel non-invasive corneal sampling technique allowing comparison of affected and unaffected eyes, and processed using conventional diagnostic culture (CDC), microorganism-targeted PCR (MTPCR) and next-generation sequencing (NGS) in order to fully characterise the microbiological spectrum.

Both the diagnostic approach in this study and the potential to improve treatment will impact positively on patients locally and internationally. The sampling technique is much less invasive, causes minimal patient discomfort, requires much less patient corporation and is much simpler to perform than conventional scraping techniques. This allows the opportunity for sampling to be performed by allied health care professionals, which could deliver efficiency improvements to clinical services, in addition to being an option where access to specialised ophthalmic equipment is limited. From a treatment perspective, NGS processing produces personalised sample results in a matter of hours compared to 3 days for CDC. If eventually used in routine practice, this will lead to more prompt selection of the most appropriate antimicrobial therapy, with the potential to reduce protracted clinical courses and reduce the risk of sight threatening complications. This is of particular relevance in settings where close treatment response monitoring is unfeasible.

Data generated from this project will inform the treating clinician of the likely truly pathogenic organism amongst those isolated from corneal samples. This information will able to be used to build an international database that will incorporate causative microorganisms with disease outcomes for those treating patients with ophthalmic infections. This will enable treating clinicians to make more informed antimicrobial therapy choices. Furthermore, the reduced NGS processing times may offer the opportunity in future for antimicrobial therapy choices to be made on the same day as presentation, potentially reducing healthcare resources utilised for follow up.

My research will have a number of significant benefits for optometrists in the UK who are often the main point of access for contact lens wearers, a high-risk group for MK. The sampling methodology is simple to carry out and does not require specialist skills or equipment and therefore lends itself to being developed as a point of care test that can be carried out by optometrists and other allied health care professionals and not just in specialised ophthalmic services. In addition, this research will enable us to identify potential colonising reservoirs and "high risk" ocular microbiomes for MK. This could be used to prevent future recurrent episodes through decolonisation of such reservoirs and identifying those most at risk of corneal infections. High risk patients could potentially be identified in the community and prevention strategies implemented. This has the potential to significantly reduce disease burden and have substantial cost benefits for the NHS.

My study findings are also expected to impact the care of patients in low income settings where the prevalence of MK and its burden of disability are much higher. Current diagnostic techniques are too slow and dependent on specialist expertise and equipment to be used in these settings, therefore the contribution of this research towards a point of care test is expected to improve the diagnosis of these patients. Similarly, the results on causative organisms will inform more cost effective empirical treatment in these settings where access to specialised ophthalmic services are poor.


10 25 50
Title Corneal Impression Membrane 
Description The corneal impression membrane is a Millipore, Biopore 0.4 micrometer millicell membrane. The membrane is currently in clinical use to collect samples from patients with microbial keratitis and for impression cytology from patients with other disorders of the ocular surface. 
Type Diagnostic Tool - Non-Imaging
Current Stage Of Development Early clinical assessment
Year Development Stage Completed 2015
Development Status Under active development/distribution
Impact The corneal impression membrane sampling technique is not reliant on specialist equipment or specialist training and has been shown to have a significantly higher isolation rate with no loss of specificity compared to conventional corneal scraping methods. The corneal impression membrane therefore offers an alternative solution to existing sampling methods that are unfeasible in low resource settings such as Malawi. If the corneal impression membrane sampling technique can be feasibly implemented within a setting such as Malawi, it will considerably increase our understanding of the microbiological profile of microbial keratitis and enable targeted diagnostic and treatment strategies to be developed. Eventually, it is thought that medical, nursing and allied staff could be trained to apply this methodology to sustainably improve investigation and subsequent treatment of cases of microbial keratitis long term.