|Home | About | Journals | Submit | Contact Us | Français|
Purpose: This study determined whether LL-37 (cathelicidin)is expressed by conjunctival and corneal epithelia as part of ocular host defense. The antimicrobial activity of LL-37 was also assessed in vitro against Pseudomonas aeruginosa (PA), Staphylococcus aureus (SA), Staphylococcus epidermidis (SE), herpes simplex virus type 1 (HSV-1), and adenovirus (Ad). Methods: Expression of LL-37/hCAP 18 mRNA and LL-37 protein was determined by reverse transcription-polymerase chain reaction (RT-PCR) and immunoblotting, respectively, in scraped human corneal epithelium and primary cultured human corneal and conjunctival epithelial cells. The EC50values for three strains of PA and one each of SA and SE were determined for LL-37. LL-37 antiviral inhibition of HSV-1 and adenovirus was assessed by direct inactivation assays. Toxicity of LL-37 to A549 cells was evaluated by a MTT assay. Results: LL-37/hCAP18 mRNA and LL-37 peptide were expressed by human corneal and conjunctival epithelial cells. Antibacterial activity for LL-37 was demonstrated (EC50values for the three PA strains were 2.8 ± 1.3, 1.9 ± 0.3, and 3.6 ± 2.1; for SA: 1.6 ± 1.5; for SE: 1.3 ± 1.9 μg/ml). LL-37 produced a significant reduction (p < 0.001 ANOVA) in HSV-1 and Ad19 viral titers with distinctly different time-kill curves (p < 0.001). LL-37 (up to 111 μM) produced no toxicity in A549 cells. Conclusions: Corneal and conjunctival epithelia express LL-37 as part of mucosal innate immunity to protect against bacterial and viral ocular infections.
The innate immune system is an ancient, complex, rapid killing, nonspecific, redundant multifunctional system that is highly effective in protecting the host.1 Scientific interest in this field is burgeoning as reflected by the growing number of publications including several excellent reviews.2-5 Studies over the past two decades have shown that small cationic peptides with potent antimicrobial activity against Gram-positive and Gram-negative bacteria, fungi, and some viruses are an important component of the innate immune system.6-9 Initially described in leukocytes, cationic antimicrobial peptides such as defensins and cathelicidins are also secreted by various epithelial tissues.2,10 The precise mechanism of antimicrobial activity of these peptides is incompletely understood, but it is widely accepted that perturbation of the microbial cell membrane is the major means by which these peptides kill.11,12 In addition to their antimicrobial activity, additional diverse functions for these peptides include important roles in inflammation, immune activation, and wound healing.13-15
Cathelicidins consist of a highly conserved N-terminal region(the cathelin domain) with homology to the cysteine protease inhibitor cathelin and a less conserved C-terminal antimicrobialregion that varies among species, yielding multiple peptides with a remarkable variety of sizes, sequences, and structures.16 Only one cathelicidin has been described in humans, which is human cationic antimicrobial peptide of 18 kDa (hCAP18). This peptide is expressed by neutrophils and epithelial tissues lining the oral cavity and respiratory, urogenital, and gastrointestinal tracts.17-19
Previous studies have shown that hCAP18 is upregulated by inflammation and injury.20-24 hCAP18 is cleaved by the action of serine proteases to generate the 37-amino-acid antimicrobial peptide LL-37 and a 103-amino-acid cathelin domain.25 A recombinant version of the hCAP18 cathelin domain was also recently shown to have antimicrobial activity and to inhibit protease activity.26 LL-37 has been shown to have potent antimicrobial activity against a variety of bacteria including staphylococcal species and Escherichia coli and fungi such as Candida albicans.20,27,28 The antiviral activity of LL-37 has received far less attention, but some activity has been reported against herpes simplex virus (HSV)29 and vaccinia virus.30
In addition to its antimicrobial function, LL-37 is chemotactic for monocytes, T cells, neutrophils, and mast cells, stimulates mast cell histamine release, modulates dendritic cell differentiation, and stimulates IL-8 secretion.14,31-34 LL-37 has also been shown to stimulate angiogenesis,35 and lack of LL-37 activity was associated with impaired epithelial healing in skin wounds.36 These observations have led to the concept that LL-37 is a multifunctional effector molecule capable of directly killing pathogens, modulating the immune response, and promoting wound healing.14,37
The ocular surface stays remarkably free from infection despite constant exposure to a hostile microbial environment. This reality is attributable, in part, to a very robust innate immune response.38 Recent studies have shown that the antimicrobial peptides human α-defensin (hBD)-1,-2, and -3 are expressed by ocular surface epithelia and therefore are likely to contribute to the innate immune response.39-43 The goals of the current study were (1) to determine whether LL-37, like α-defensins, is secreted by corneal and conjunctival epithelia as part of the innate immune response, and (2) to assess in vitro the antimicrobial activity of LL-37 against major ocular bacterial and viral pathogens.
The antimicrobial peptide, LL-37, was purchased from American Peptide Company (Sunnyvale, CA, USA) and used in all the experiments. A scrambled peptide of LL-3735 with the same amino acids ordered randomly was purchased from Global Peptide Services (Fort Collins, CO, USA).
Human corneas were obtained from the Lions Eye Banks (Central Florida and Heartlands) and used in accordance with the tenets of the Declaration of Helsinki regarding the use of human tissue for research. The epithelium was scraped from each cornea (in the cases where a pair were available, then the tissue from both corneas was pooled) using a scalpel blade, then immediately placed in RNA lysis buffer (Qiagen, Valencia, CA, USA) for reverse transcription-polymerase chain reaction (RT-PCR) assays or snap frozen in liquid nitrogen and stored at -80°C for the immunoblot assays.
Primary cultures of human corneal epithelial cells (HCECs) were prepared from single or pairs of normal eye bank corneas based on the method described previously.42 Following incubation in Dispase II (1.2 U/ml) for 4-5 hr at 37°C, the epithelial layer was scraped free from the underlying stroma with a no. 15 scalpel blade and transferred to a tube containing Dulbecco’s Modified Eagle’s Medium (DMEM) and 10% fetal bovine serum (FBS) and centrifuged. The cell pellet was resuspended in EpiLife medium (Cascade Biologics, Portland, OR, USA), and a single cell suspension was obtained by titrating through a syringe fitted with a 22G needle. The cells were transferred to a culture flask coated with a mixture of fibronectin and collagen (FNC; AthenaES, Baltimore, MD, USA) containing 5 ml of serum-free EpiLife media with human corneal growth supplement (HCGS; Cascade Biologics). Primary cultured HCECs of passages 1 to 2 were used in the experiments.
Human conjunctival tissue from two donors was obtained from the National Disease Research Interchange (NDRI, Philadelphia, PA, USA).Primary conjunctival epithelial cells were cultured as described previously.43 Briefly, conjunctival tissue was incubated overnight at 4°C in a 1:1 (v/v) solution of EpiLife medium and dispase (20 U/ml). Using a scalpel blade, the epithelial cells were then scraped free and seeded into a FNC-coated flask with EpiLife media and HCGS. Primary-cultured conjunctival epithelial cells of passages 1 to 3 were used for the experiments.
Total RNA from all cell samples was extracted using an RNeasy kit(Qiagen). Two hundred fifty nanograms of total RNA were used per RT-PCR reaction using a Superscript II (Invitrogen, Carlsbad, CA,USA) kit. Reactions containing normal human testis RNA (Clontech Laboratories, Palo Alto, CA, USA) or RNAse free water in place of the RNA served as a positive control and negative control, respectively. Reverse transcription was performed at 50°C for 60 min. In some reactions, the reverse transcriptase was omitted (-RT control). After denaturation of the reverse transcriptase (94°C, 5 min), amplification of the cDNA was performed for 40 cycles: denaturation, 94°C for 50 s; annealing, 62°C for 30 s; extension 72°C for 1 min. The specific primers used were α-actin (forward 5′-CCTCGCC-TTTGCCGATCC-3′and reverse 5′- GGATCTTCATGAGGTAGTCAGTC-3′, 626 bp44 ) and LL-37/hCAP18 (forward 5′-ATCATTGCCCAG- GTCCTCAG-3′and reverse 5′-GTCCCCATACAC- CGCTTCAC-3′, 251 bp).45 Products generated with these primers were sequenced (Seqwright, Houston, TX, USA) to confirm their identities. RT-PCR products were visualized on an agarose gel using an Alpha Imager gel documentation system (Alpha Innotec, San Leandro, CA, USA). HyperLadder I (Midwest Scientific, St. Louis, MO, USA) was used as the base pair marker.
Epithelial samples that had been snap frozen were used in immunoblots to detect LL-37 peptide. Each sample was homogenized in 100 μl of ice-cold tris buffered saline (TBS, 150 mM NaCl, 20 mM Tris- HCl, pH 7.5). Cell lysate (25 μg of total protein) was blotted directly onto a nitrocellulose membrane using a Bio-Dot Microfiltration apparatus (Life Science, Hercules, CA, USA). One nanogram of synthetic LL-37 peptide (American Peptide Company, Sunnyvale, CA, USA) was also blotted onto the membrane as a positive control. Nonspecific binding sites were blocked by incubation in 5% blotto, then the membrane was incubated with a rabbit anti- LL37 polyclonal antibody (donated by Dr. Lehrer, UCLA) diluted 1 in 5000 in 3% blotto. After an overnight incubation, the membranes were then incubated with a horseradish peroxidase linked second antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) diluted 1 in 10,000 in 3% blotto. Immunoreactivity was visualized by enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ, USA).
Pseudomonas aeruginosa (PA; ATCC 27853 and two clinical isolates, PA 2219 and PA 2212 from corneal scrapings of subjects with bacterial keratitis) were tested in this study. The ATCC strain tested is known to invade intact cornea and produce severe ocular infection in experimentally infected animal models of bacterial keratitis.46 One single isolated PA colony was used to inoculate 5 ml of nutrient broth (NB) overnight at 37°C. Fifty microliters of this bacterial suspension were used to inoculate 50 ml of fresh NB, which was then incubated for 2.5 hr with vigorous shaking at 37°C to achieve mid-log phase growth. Twenty-five milliliters of the warm PA culture were centrifuged at 3100 g for 10 min, and the bacterial cell pellet was resuspended × in cold phosphate buffer (PB, 8.2 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4). Optical density of the suspension was adjusted to 0.2 at 620 nm (approximately 107cfu/ml) by adding an appropriate volume of PB. The antimicrobial assay procedure was adapted from that described by Tomita et al.47 Reaction mixtures (final volume 50 ml) containing 10 ml of 107cfu/ml PA and 5 ml LL-37 diluted in PB (final concentration 0.05, 0.1, 0.5, 1, 10, 25, 50, and 100 μg/μl) were incubated at 37°C for 2 hr with vigorous shaking. In each experiment, reaction mixtures containing 5 μl of 0.01% acetic acid, the vehicle for diluting LL-37, acted as a control. At the end of the incubation, serial dilutions of each reaction mixture were used to inoculate NB agar plates. Samples (10 μl) were spread evenly over the surface of the plates using sterile glass spreaders. After incubation at 37°C for 24 hr, the plates were placed on a lightboard and a digital image captured using an Alpha Imager documentation system. The number of colonies was counted using the colony count software of the Alpha Imager. The percentage of bacteria killed by LL- 37 was calculated using the following equation: Killing (%) = ([number of PA colonies]control[number of PA colonies]- × LL-37)/[number of PA colonies]control 100%. EC50values were calculated using GraphPad Prism4 software (GraphPad Software, San Diego, CA, USA).
Additional experiments were also performed to test the antimicrobial activity of LL-37 against Staphylococcus aureus (SA; ATCC 29213) and Staphylococcus epidermidis (SE; ATCC 155). Testing conditions in these experiments were identical to those described for PA with the exception of replacing NB with trypticase soy broth.
A549 cells were obtained from the American Type Culture Collection (Manassas, VA, USA) and were maintained in Ham’s F12 with 10% FBS and passed by standard methods using trypsin/EDTA. For cytotoxicity assays, the cells (p89, 96) were grown to confluency in 96-well plates. LL-37 and scrambled LL- 3735 were diluted in culture medium with or without 10% FBS to give final concentrations of 50, 5, and 0.5 μg/ml. These concentrations matched those of the peptides during the adsorption phase of the antiviral assay. The cells were then exposed to the diluted peptides for 3 hr. Controls received either culture media alone or culture medium with 10% FBS. Positive controls were incubated with 0.002% benzalkonium chloride for 15 min. Four to five replicates were performed for each condition. Cytotoxicity was assessed using an MTT assay kit (Chemicon International, Temecula, CA, USA) as per the manufacturer’s instructions.
A time-kill direct inactivation assay was used to determine the antiviral inhibitory activity of the antimicrobial peptides LL-37 and a scrambled peptide35 on two different DNA viruses: herpes simplex virus type 1 (McKrae strain) and several adenovirus serotypes (Ad3, Ad5, Ad8, Ad19) isolated from epidemic keratoconjunctivitis (EKC) patients presenting to The Charles T. Campbell Ophthalmic Microbiology Laboratory (Pittsburgh, PA, USA). In a single tube, each virus sample (~104pfu) was incubated directly with LL-37 (500 μg/ml, 111 μM), scrambled peptide (500 μg/ml, 111 μM), or PBS in a 37°C water bath. At 1 and 4 hr, 35 μl aliquots were removed from the virus-peptides and virus-PBS incubation mixtures and were diluted 1:10 in fresh ice-cold (0°C) media containing 10% fetal bovine serum to dilute the remaining virus and inactivate the peptides.48,49 The mixtures were further serially diluted and titrated the same day on A549 cells as previously described.50 In order to evaluate reproducibility, a series of 6 independent experiments comprising each virus assay were carried out at different times and the data analyzed statistically using one-way analysis of variance (ANOVA) (Minitab 12 for Windows, Minitab, State College, PA, USA).
RT-PCR was performed to study LL-37/hCAP18 mRNA expression in human corneal and conjunctival epithelial cells (Fig. 1A). LL-37/hCAP18 mRNA was constitutively expressed in six of seven scraped corneal epithelial samples and in all primary cultures of corneal (n = 8) and conjunctival (n 2) epithelial cells tested. Immunoblotting was performed = to study LL-37 peptide expression by human corneal epithelium. The results from four representative samples are shown in Figure 1B. LL-37 peptide was present in both scraped epithelial samples (n 6) and primary cultured epithelial cells (n=3)
Antibacterial assays were performed to study the activity of LL-37 against PA (ATCC 27853 and two clinical isolates). As shown in Figure 2, LL-37 inhibited growth of all three PA strains in a concentration-dependent manner. The EC50was 2.8 1.3 (n 3),= ± 1.9 ± 1.3 (n = 3), and 3.6 ± (n = 3) μg/ml for ATCC 2.1 27853 and the two clinical strains (PA 2219 and PA 2212), respectively. At 100 μg/ml, LL-37 was completely bactericidal for all strains of PA. LL-37 also showed potent bactericidal activity against SA and SE with EC50 values being 1.6 ± 1.5 (n = 3) and 1.31.9 (n 3) μg/ml, respectively.
A cell survival assay was performed to ensure that LL- 37 and scrambled LL-37 were not toxic to A549 cells at the concentrations tested in the antiviral assay (Fig. 3). Exposure to benzalkonium chloride, the positive control, caused significant A549 cell death (p ≥ 0.001, Student’s t test), whereas neither LL-37 nor the scrambled peptide showed any toxicity toward A549 cells at any of the concentrations tested in the presence of 10% FBS (n = 2). Identical results were obtained when the experiment was carried out in the absence of 10% FBS (data not shown).
As LL-37 and the scrambled peptide were shown to be nontoxic in A549 cells at the concentration tested, the antiviral assays results shown in Figures 4 and and55 could be properly ascribed to the peptides’ intrinsic inhibitory capabilities. For HSV-1 (Fig. 4), LL-37 produced a highly significant (p < 0.001 ANOVA)≥ 2 log reduction in HSV-1 titers compared to PBS and scrambled peptide at 4 hr. Although the scrambled peptide demonstrated limited inhibition of HSV-1 relative to the PBS control (p < 0.001), it was significantly less (p < 0.001) than that of LL-37.
Among the Ad serotypes tested (Ad8, Ad5, Ad3) at 4 hr, all demonstrated a trend toward showing an inhibitory effect by LL-37 (Fig. 4). However, only Ad19 demonstrated statistically significant (p < 0.001 ANOVA) reduction in titers compared to the scrambled peptide and PBS controls. Like for HSV-1, the scrambled peptide again demonstrated limited inhibition of Ad19 relative to the PBS control (p < 0.001), but was significantly less (p < 0.001) than that of the LL-37.
The time-kill curves comparing HSV-1 and Ad19 over 4 hr are summarized in Figure 5. For HSV-1, LL- 37 produced a rapid, highly significant (p < 0 .007) 2 log reduction in viral titers within 1 hr, while for Ad19,there was a much slower progressive reducion in titers to reach a maximum effect at 4 hr. These differences in the slopes of the kill curves suggest a possible different mechanism of inhibitory action of LL-37 for an enveloped virus (HSV-1) compared to a nonenveloped virus (Ad19).48
LL-37 is a multifunctional peptide with many important biological activities: antimicrobial activity, chemoattraction, dendritic cell differentiation, mast cell degranulation, cytokine secretion, angiogenesis stimulation, and wound healing.2,3,14,18,31-34,36,37 The current study reports for the first time that the epithelial cells of both the cornea and conjunctiva express this important effector of innate immunity. This discovery was initially reported in 200351 and has recently been confirmed in a report by Dua et al.52 Our discovery of LL-37 as a part of ocular mucosal defense complements previous studies that have demonstrated that α-defensins hBD-1, -2, and -3 are also secreted by conjunctival and corneal epithelium.39-43,53 Taken together, these findings suggest that the ocular surface must be added to the growing list of mucosal surfaces throughout the body where multifunctional cationic peptides operate as part of innate immunity.
LL-37 may have several activities at the ocular surface. It is likely to be involved in direct killing of invading pathogens and through its various effector activities may be involved in regulating the ocular surface immune response and wound healing. Indeed, preliminary studies from our laboratory do show that LL-37 can stimulate corneal epithelial cell migration.54 In this report, we focused our attention on the role of LL-37 as an ocular antimicrobial peptide and studied its effectiveness against common ocular pathogens.
We investigated the antimicrobial activity of synthetic LL-37 against the Gram-positive (SE, SA) and Gram-negative (PA) organisms most often responsible for bacterial keratitis and the pathogens most frequently associated with keratitis in contact lens wear.55-57 We found LL-37 to have potent antibacterial activity against both a laboratory strain and clinical isolates of PA. Our EC50values of approximately 2-4 μg/ml compare well with published studies of the effectiveness of LL-37 against other PA strains.20,27,28,58,59 LL-37 also demonstrated potent antibacterial activity against the most common Gram-positive keratitis-producing organisms: SE (mean EC501.3 μg/ml) and SA (mean EC501.6 μg/ml).
To explore a possible broader antimicrobial role of LL-37 in defense of the ocular surface, we also studied the effectiveness of the peptide against ocular viral pathogens in a direct inactivation assay. We report for the first time that LL-37 demonstrates significant antiviral inhibitory activity (>98% inhibition) against HSV-1, the leading viral cause of corneal blindness in industrialized countries.60 In a previous study, Yasin et al. reported limited anti-HSV-1 activity for LL-37 compared to two other more potent antimicrobial peptides, brevidin and indolicidin, in a MTT assay.29 The differences in results between the studies may be explained by differences in HSV-1 virus strains, host cells, incubation times, peptide concentrations, and the protocol for the antiviral assays used.
Additionally, we report for the first time that LL- 37 demonstrated statistically significant inhibitory activity in vitro against Ad19, a major cause of conjunctivitis and epidemic keratoconjunctivitis in local and global epidemics.61 Previous studies have reported that certain defensins have antiviral activity against genetically engineered adenovirus type 5 vectors used in gene therapy,62,63 but there are no earlier reports that any antimicrobial peptides (defensins or LL-37) can inhibit wild-type pathogenic adenoviruses. Our observation represents a “proof-of-principle” that an antimicrobial peptide can inhibit at least one nonenveloped virus under experimental conditions by direct inactivation. Serotype differences in the susceptibility to LL-37 may be related to the known structural differences among different Ad serotypes in the hypervariable region of the hexon capsid64 and the penton fiber.65
We evaluated specificity in our in vitro antiviral assay by comparing the inhibitory activity of the LL-37 to a scrambled peptide (same size, same amino acids in a different order) that had previously demonstrated no activity in promoting angiogenesis35 or corneal epithelial cell migration.51 We conclude that “reasonable” specificity was demonstrated in our studies as limited inhibition was achieved by the scrambled peptide against HSV-1 and Ad19. Notably for both viruses, the effect of the scrambled peptide was significantly less than for LL- 37 (Fig. 4). We explain the limited activity of the scrambled peptide on the fact that the antimicrobial function of peptides is known to be multivariate being dependent not only on size and sequence but also charge, degree of structuring (helicity), overall hydrophobicity, amphipathicity, and the angles subtended by hydrophobic and hydrophilic surfaces of a helical molecule.66 Therefore, size and amino acid sequence alone appeared to be insufficient to demonstrate complete specificity for the two different viruses tested.
The current study suggests possible differences in the mechanism of inhibition based on the slopes of the LL-37 curves in the time-kill assays. The rapid killing of HSV-1 (<1 hr) suggests that permeabilization of the external lipid membrane is the most likely mechanism. This mechanism has been previously proposed for all cationic peptides (e.g., defensins) against HSV-1 and other enveloped viruses, bacteria, and fungi.2,37,48,58,67,68 Furthermore, the antiviral activity of LL-37 against vaccinia virus30 has also been attributed to its direct disruptive action on the viral envelope. Interestingly, recent studies of defensins69,70 and other peptides71,72 against HSV and defensins against HIV73-75 indicate that prevention of viral entry is another important mechanism of peptide inhibitory action against these enveloped viruses.
Unlike enveloped viruses, all adenoviridae lack a surrounding host cell-derived lipid membrane suggesting that the direct killing through permeabilization of the viral envelope cannot be the operative mechanism. Alternative mechanisms such as disruption of the adenovirus particle (detergent-effect) and/or blockage of viral entry (attachment/endocytosis) into the cell remain possibilities requiring further clarification. Further studies are planned with LL-37 at different times in the adenovirus life cycle to further delineate extracellular and possible intracellular mechanisms of inhibition.
The evolutionary selection of diverse antimicrobial peptides based on a recently described universal multidimensional signature (i.e., a common three-dimensional structure) appears to be an essential component of innate immunity.76 During the past 2.6 billion years, most life forms (bacteria, fungi, plants, invertebrates, and vertebrates) have successfully used these peptides to defend against a hostile surrounding microbial world. The rapid doubling time of bacteria has been considered by many to be the driving force for the evolution of innate immunity, as the threat is constant and the speed of host response must be appropriate to the pathogen.68 For this reason, any antiviral activity demonstrated by antimicrobial peptides against enveloped viruses has been deemed secondary to the common mechanism of action, permeabilization of membranes. The recent observations that defensins can actually prevent entry of HSV69,70 and HIV,73-75 both enveloped viruses, and our observation here that LL-37 is effective against the nonenveloped virus Ad19 strongly support the notion that the antiviral activity of antimicrobial peptides is not coincidental but an important part of innate immunity that concurrently evolved with its antibacterial and antifungal capabilities. Although the time frame of virus replication is admittedly longer (20+ hr) than bacteria, the optimal strategy to limit these obligate intracellular parasites is to destroy them rapidly externally and/or prevent their entry into uninfected host cells. Like bacteria, this strategy also requires a rapid host response. Therefore, we surmise that evolution would likely select for rapidly mobilized effector molecules with antiviral capabilities (e.g., antimicrobial peptides) to protect host cells from invasion, colonization (HSV-1 latency), and death. A number of in vivo studies have shown that antimicrobial peptides (defensins and cathelicidins) are effective in defense against bacterial infections.69,70,77-79 Whether biologically significant antiviral activity of these peptides can be expanded beyond enveloped viruses to include nonenveloped viruses remains a distinct possibility, but further confirmation is required.
This work was supported by EY 008227 (Y.J.G.), Core Grant for Vision Research EY08098, The Eye and Ear Foundation of Pittsburgh, Research for the Prevention of Blindness, University of Houston Gear grant and EY-13175 (A.M.M.). The authors are grateful to Lions Eye Banks (Central Florida and Heartlands) for supplying the human corneas and to Dr. Robert Lehrer (UCLA) for his generous donation of LL-37 antibodies. Thanks also to Dr. Bradley Mitchell (Baylor College of Medicine) for providing the clinical PA isolates, Srihari Narayanan for culturing conjunctival epithelial cells, and Daniele Jean for assisting with the antibacterial assay.