UC Berkeley

While the eye’s surface is constantly exposed to microbes in the environment,healthy corneas lack a microbiome and have an unusual ability to resist colonization and infection by opportunistic pathogens Pseudomonas aeruginosa (Gram-negative) and Staphylococcus aureus (Gram-positive). However, extended contact lens (CL) wear and overt injury can result in the development of sight-threatening bacterial keratitis. Essential to understanding the pathogenesis of bacterial keratitis are the defenses that normally prevent corneal colonization and infection. The cornea has a high density of sensory nerve endings expressing the polymodal ion channels TRPV1 (Vanilloid) and TRPA1 (Ankyrin) whose main function is detecting noxious stimuli and modulating pain and itch. Some recent studies have shown that in some circumstances they or the nerves expressing them can detect bacterial ligands and modulate inflammatory responses. In the healthy cornea, our lab previously showed that TRPA1 can counter the adhesion of deliberately inoculated P. aeruginosa in large numbers. Experiments that I contributed to in that publication also showed TRPV1 can counter environmental colonization (likely commensal bacteria in fewer numbers), suggesting specificity in their protective roles in the healthy cornea. Here, I tested the hypothesis that the specificity of TRPA1 and TRPV1 in contributing to corneal defense against adhesion is influenced by bacterial status as a pathogen versus a commensal, respectively, or bacterial Gram-type. Otherwise, specificity relates to bacterial inoculum size. Using a Gram-positive pathogen of the eye, Staphylococcus aureus S33 (a human, ocular clinical isolate), I found that TRPV1, not TRPA1, prevented the adhesion of this pathogen (Chapter 2). Contrasting with the TRPA1-mediated defense against P. aeruginosa, TRPV1-mediated defense against S. aureus did not require sensory nerve-firing mechanisms, suggesting a local mechanism. I further observed different TRPV1/A1 and nerve-dependent immune cell responses. P. aeruginosa increased the number of CD45+ and CD11c+ cells, the latter involving sensory nerve-mediated expression of inflammatory cytokines and chemokines including IL-6, IL-1β, CCL7, and CXCL5 (Chapter 3). Conversely, S. aureus triggered a smaller CD45+ cell response with no increase in CD11c+ cell numbers. Further immune cell analysis revealed distinct morphologies and migration changes induced by S. aureus and P. aeruginosa inoculation. While both pathogens decreased CD11c+ cell sphericity (cells appeared more dendriform), their migration patterns differed. Following S. aureus inoculation, CD11c+ cells migrated further from the corneal surface while P. aeruginosa induced their migration toward the corneal surface. Additionally, CD45+ cells became more spherical in response to S. aureus rather than less spherical as observed for P. aeruginosa. Lyz2+ cell numbers were not impacted by either pathogen; however, cells were more spherical in response to S. aureus only and moved closer to the surface for P. aeruginosa only. Since P. aeruginosa is Gram-negative, while S. aureus is Gram-positive as are most environmental bacteria that colonize TRPV1 mutated corneas, Chapter 4 investigated the hypothesis that bacterial Gram-type, rather than pathogen versus commensal, differentiates their involvement in corneal defense. This was tested using a Gram-positive mouse eyelid commensal Macrococcus epidermidis, and Corynebacterium mastitidis, a Gram-positive conjunctival commensal. The results showed that neither TRPA1 nor TRPV1 were involved for corneal defense against C. mastitidis. However, both receptors participated in corneal defense against M. epidermidis. In this instance, TRPA1-V1-mediated defenses were not dependent on sensory nerve firing suggesting local mechanisms as observed for the Gram-positive pathogen S. aureus. Using quantitative mass spectrometry analysis of corneal surface eye washes following M. epidermidis challenge (cultured supernatant), I discovered significant differences in the abundance of tear-fluid associated proteins mediated by TRPV1, some with known antimicrobial functions. These included the anterior gradient protein 2 (AGR2) an important mucin-producing factor, Polymeric Immunoglobulin Receptor (pIgR) for SIgA, Lipocalin 11, and S100 A11 calcium-binding protein. A mouse model of CL wear developed by our lab was previously used to show that a parainflammatory response (subclinical inflammation) occurs after 24 h and 6 days of lens wear as observed during human lens wear. This parainflammatory response was found to be driven by microbes at the ocular surface during contact lens wear, with an antibiotic inhibiting the response, and restoration after the addition of various commensal bacteria including C. mastitidis and a coagulase-negative spp (later identified as M. epidermidis). Thus, there are parallels between the phenotype I have studied and contact lens-mediated immune cell responses. As part of my dissertation research, I contributed data to a subsequent publication that showed contact lens-related parainflammation is modulated by both TRPA1 and TRPV1. Further, unpublished data that I contributed to collecting showed that the TRPA1/V1 mediated CL-induced parainflammatory response is associated with increased corneal defense against the commensal M. epidermidis, suggesting it can protect against commensal bacteria adhesion. Taken together, the findings presented in this dissertation revealed differences in the healthy cornea’s ability to recognize, respond, and initiate defense against bacteria, involving TRPA1/V1-sensory nerves. These findings also advance our understanding of the local and immune cell responses that can occur within the healthy infection-resistant cornea upon bacterial inoculation and the relative roles of sensory nerves/TRP receptors in modulating these. Which aspects of those cellular responses contribute to defense against bacterial adhesion, and the molecular factors that prevent adhesion at the corneal surface, remain to be determined.