Ocular Surface Microbiome in Dry Eye Patients

The ocular surface is the first line of defense of the eye, it is therefore where external threats are sensed, and potential insults neutralised. Over the course of evolution, various microbes, especially bacteriae, have come to colonise the ocular surface as commensals. The commensals have a role to maintain the homeostasis of the ocular surface. 1 The innate immunity of the ocular surface is very active, and consists of active mechanisms to suppress inflammation 2. For example, there exist macrophages, dendritic cells, suppressor cells, regulatory cells, B cells, IgA, lysozyme, anti-microbial peptides and barriers against external agents. The normal commensals of the ocular surface maintain a basal level of activation of innate defence by stimulating the pattern recognition receptors on ocular surface epithelial cells. This normal composition of microbes is important since inflammation and infection will result if there is introduction of a pathogenic strain that overcomes the flora, or if a dominant strain secretes excessively immunogenic products, such as the exotoxin A of Staphylococcus which triggers marginal keratitis, a form of type IV hypersensitivity. The flora load of microbiome could also influence tear function as a higher flora load was found to be associated with increased mucin degradation 3 and reduced globet cell densitiy 4. Previous studies [I'm not sure which studies these are] at SERI/SNEC also point to the importance of microbes. For example, in dry eye patients, there is increased lysophospholipids in the tear, and this may contribute to inflammatory mediators such as arachidonic acid and other metabolites. The lysophospholipids are formed by phospholipase A2 reactions, and the latter may be microbial in origin. Since dry eye is a known inflammatory disease of the ocular surface, this is one way that microbes can contribute to the pathology.

The microbiome is the sum total of the microbe found in a body location. In the eye the microbiome can be investigated non-invasively because of the accessible location of the ocular surface. Despite the importance of the ocular surface microbiome, understanding of this topic is very superficial and largely based on culture methods5-10. This can be misleading because some of the microbes are not so amenable to routine cell cultures and their presence may be undetected or their representation under-estimated 11. Recently developed metagenomics sequencing, compared to cultures, are advantageous to get an un-biased view of microbiome11.

Three prior studies have evaluated the microbiome profiles of the ocular surface with metagenomics sequencing of 16s rRNA 4, 12 13. Differences in microbiome were observed between subjects with dry eyes or blepharitis, and healthy subjects. Moreover, a much greater diversity of ocular flora was discovered compared to results based on bacterial culture.

The school of chemical and environmental life science at Nanyang Technological University has evaluated the microbiomes using such techniques. The basis of this is that prokaryotic organisms have unique DNA signatures which can typically characterise the species of origin, and the relative prevalence of the organisms of different species can be deduced from the relative abundance of the different DNA detected. The global human microbiome project seeks to understand the microbiome in different parts of the human body. This is best understood in the intestine, but publications on the ocular surface have yet to emerge.

Recently our collaborators have been found that 3 bacteria dominated the ocular surface of the eye. These are Klebsiella pneumoniae, Pseudoalteromonas agarivorans and Lactobacillus rhamnosus. This dominance of 3 types of bacteria was unexpected, since the dominant view of ocular surface commensals is that coagulase negative Staphylococcus dominates, followed by Staphylococcus aureus and Streptococcus, with less gram negative bacteria, and some Propionibacter spp 11. Among the 3 species, Klebsiella pneumonia is a known gram negative pathogen that can cause infectious keratitis, and in some cases endophthalmitis. The roles of Pseudoalteromonas and Lactobacillus are entirely unknown. The former is a non-pathogenic free living gram negative bacteria known to be present in marine settings. Surprisingly the high levels of the 3 predominant bacteria are relatively constant among the 3 human participants, suggesting a certain amount of uniformity in the colonisation of the ocular surface by major bacteria microbes in non-diseased individuals.

The ocular surface microbiome can be studied by sterile swapping of the inferior fornix of the conjunctiva followed by homogenisation and extraction of DNA. The presence of bacteria DNA does not inform on the function of the microbes, since even dead and non-proliferating organisms have remnant DNA. The technique of collecting samples can also yield RNA, and from reverse transcription PCR, it is possible to determine the presence of specific microbial transcripts that are found in the ocular surface. Furthermore, our collaborators from Nanyang Polytechnic have pioneered MALDI TOF mass spectrometry techniques that allow the characterisation of bacterial proteomes, based on identification of unique bacterial peptide sequences. Together, these techniques of bacteria transcript and protein investigations permit evaluation of the gene expression and to some extent, functional status of the bacteria discovered. This would be extremely important in the case of microbes that have not been previously discovered on the ocular surface.

Apart from the obvious academic and scientific advance, this project also has medical implications and potential new Intellectual property for commercial applications. First, understanding of the microbiome is the first step to treat certain types of diseases such as marginal keratitis and unusual infectious keratitis. It suggests how the immune system can be manipulated for treatment. Secondly, prevention of diseases and inflammation may be achieved by probiotic administration of living microbes to colonise the ocular surface. Third, the study may lead to new techniques and assays that identify microbes and their components which can be developed into clinical diagnostic assays.

We have some industry collaborators who are interested to partner us to develop some of these ideas.

1. OBJECTIVES

We aim to conduct an exploratory study to:

1. Determine the composition of the bacterial microbiome of the human ocular surface in normal volunteers and dry eye patients

2. Determine the bacterial gene expression (bacterial transcripts) found in the human ocular surface.

3. Characterise any feature of the microbiome that may be linked to clinical characteristics such as age, status of tearfilm, Meibomian gland dysfunction, exposure to environmental stimulation such as smoking, etc.

2. EXPECTED RIKS AND BENEFITS Expected risks No potential problems are expected for this study. However, there will be some slight discomfort during the collection of microbiome from the lower conjunctival fornix. Non-preservative tetracaine will be given to the patient to minimise any discomfort. There will also be some slight discomfort during the Schimers I test.

Potential benefits With greater understanding of the presence of various microbiome in the ocular surface, we can gain further insight into the complex process taking place.

3. STUDY POPULATION 3.1. List the number and nature of subjects to be enrolled. 90 subjects will be recruited from SNEC dry eye clinic and general clinic serviced by the Principal Investigator. 45 subjects without dry eye symptoms will be recruited as normal controls and 45 subjects with symptomatic dry eye will be recruited as dry eye subjects.

3.2. Criteria for Recruitment and Recruitment Process

1. To recruit normal controls- All the 8 questions on dry eye symptoms must be clear of dry eye symptoms during the pre-screening stage. (Appendix 1) Questionnaire modified after Schein et al.,1997. To recruit dry eye subjects- At least one out of the 8 questions on dry eye symptoms must be ≥ grade 3

2. Subjects meet all the inclusion criteria listed below.

3. Clear of exclusion criteria. Permission would be sought from the attending doctors before subjects are being recruited. Eligible subjects will be counselled on the study by study coordinator. If the subject is interested, the study coordinator will then accompany the subject to SERI level 5 for the relevant assessments. Informed written consent will be obtained from all participants.

3.3. Inclusion Criteria

Subject must meet all of the inclusion criteria to participate in this study.

1. Subjects must be 21 years or older

2. Willing to perform all the eye examinations in this study

3.4. Exclusion Criteria

All subjects meeting any of the exclusion criteria at baseline will be excluded from participation.

1. Known history of thyroid disorders (diagnosed by physician).

2. Known history of Sjogren syndrome or rheumatoid arthritis (diagnosed by physician).

3. No ocular surgery within the last 3 months and LASIK within 1 year.

4. Ocular surface diseases such as pterygium, or obvious lid/orbital disease with lagophthalmos.

5. Any other specified reason as determined by clinical investigator.

4. STUDY DESIGN AND PROCEDURES/METHODOLOGY

Study design:

Observational -Cross sectional study

No. of visits:

There will only be 1 visit in this study.

Duration of Study:

1 year

Procedures:

Collection of Medical history and Dry Eye Questionnaire There will be some questionnaire which consists of medical history and a standard set of dry eye questionnaires which identifies factors such as contact lens use, smoking, occupation and etc.

Non-Invasive Tear Break-Up Time (OCULUS K5M) Non-Invasive Tear Break-Up Time is measured non-invasively and fully automatically using OCULUS K5M. The new infrared illumination is not visible to the human eye. This prevents glare during the examination. Patient will sit comfortably in front of the instrument and blink freely while fixing on a target directly ahead. Once the participant is ready, they will be instructed to blink twice and refrain from blinking. Keratograph 5M is fully automated and it will capture any break or distortion in the image and the time of the break will be noted. Three readings will be taken for each eye to get the average value.

Conjunctiva redness (OCULUS K5M) Conjunctiva redness will be assessed using OCULUS K5M. Patient will sit comfortably in front of the instrument while fixing on a target directly ahead. Imaged will be captured and an overview display of conjunctiva redness will be evaluated by "JENVIS Grading Scale". JENVIS scale range from 0 to 4.

Redness value Definition 0 No findings, eg. Babies or infants

1. Single injections, standard value for adults

2. Mild diffuse injections

3. Severe local injections

4. Severe diffuse injections

Collection of microbiome Two specimens of microbiome will be collected from both eyes. First: a drop of non-preserved tetracaine will be instilled into the conjunctival fornix. After the stinging sensation has resolved, a sterile cotton swap will be used to collect the microbes from the lower conjunctival fornix using a gentle rolling action (up to 8 strokes). The procedure is then repeated for the opposite eye.

These swaps will then be soaked in DNA/RNA Shield (Zymo) reagent and immediately homogenised for 1 minute, stop at 30 seconds, and keep on ice for 1 minute, and homogenized for another 30 seconds. Homogenised samples will be stored at 4°C. This ensures an optimal and sufficient DNA/RNA yield for our purpose. Subsequent nucleic acid extraction, PCR, sequencing, and bioinformatics procedures will be performed by our collaborators.

Storage: intended to be analysed within a year, not more than 5 years.

Corneal fluorescein staining (OCULUS K5M) A drop of normal saline will be instilled on the fluorescein strip (Fluorets) then shaken off so that no visible drop remains. The subject is asked to look up before the introduction of the fluoret to the inferior conjunctival fornix on the right then left eye. Corneal fluorescein staining will be imaged by Keratograph 5M. Scoring system will be based on CCLRU. Briefly, there will be 5 corneal zones. The staining scale is 0-4, with 0.5 unit steps in each of the zones. Lastly, a picture of the fluorescein stained cornea will be taken.

Schirmers I test This will be done with the standard strips currently used at SERI (5 mm wide with a notch for folding) (Schirmer Tear Test Strips, Clement Clarke International, UK). No prior anesthetic will be used. The strips will be positioned over the inferior temporal half of the lower lid margin in both eyes at the same time. The study participant will be asked to close their eyes. Any excessive irritation signs will be noted. The extent of the wetting in each strip will be recorded after 5 minutes of testing. The strip will be collected and stored in 1.5ml eppendorf tubes at -80˚C until further analysis.