Solar damage due to ultraviolet radiation (UVR) is implicated in the development of two proliferative lesions of the ocular surface: pterygium and pinguecula. Pterygium and pinguecula specimens were collected, along with adjacent healthy conjunctiva specimens. RNA was extracted and sequenced. Pairwise comparisons were made of differentially expressed genes (DEGs). Computational methods were used for analysis. Transcripts from 18,630 genes were identified. Comparison of two subgroups of pterygium specimens uncovered evidence of genomic instability associated with inflammation and the immune response; these changes were also observed in pinguecula, but to a lesser extent. Among the top DEGs were four genes encoding tumor suppressors that were downregulated in pterygium: C10orf90, RARRES1, DMBT1 and SCGB3A1; C10orf90 and RARRES1 were also downregulated in pinguecula. Ingenuity Pathway Analysis overwhelmingly linked DEGs to cancer for both lesions; however, both lesions are clearly still benign, as evidenced by the expression of other genes indicating their well-differentiated and non-invasive character. Pathways for epithelial cell proliferation were identified that distinguish the two lesions, as well as genes encoding specific pathway components. Upregulated DEGs common to both lesions, including KRT9 and TRPV3, provide a further insight into pathophysiology. Our findings suggest that pterygium and pinguecula, while benign lesions, are both on the pathological pathway towards neoplastic transformation.
Evidence is presented herein supporting the potential of the natural homeostatic glycoprotein CLU (clusterin) as a novel therapeutic for the treatment of dry eye. This idea began with the demonstration that matrix metalloproteinase MMP9 is required for damage to the ocular surface in mouse dry eye. Damage was characterized by degradation of OCLN (occludin), a known substrate of MMP9 and a key component of the paracellular barrier. Following up on this finding, a yeast two-hybrid screen was conducted using MMP9 as the bait to identify other proteins involved. CLU emerged as a strong interacting protein that inhibits the enzymatic activity of MMP9. Previously characterized as a molecular chaperone, CLU is expressed prominently by epithelia at fluid-tissue interfaces and secreted into bodily fluids, where it protects cells and tissues against damaging stress. It was demonstrated that CLU also protects the ocular surface in mouse dry eye when applied topically to replace the natural protein depleted from the dysfunctional tears. CLU is similarly depleted from tears in human dry eye. The most novel and interesting finding was that CLU binds selectively to the damaged ocular surface. In this position, CLU protects against epithelial cell death and barrier proteolysis, and dampens the autoimmune response, while the apical epithelial cell layer is renewed. When present at high enough concentration, CLU also blocks staining by vital dyes used clinically to diagnose dry eye. None of the current therapeutics have this combination of properties to “protect, seal, and heal”. Future work will be directed towards human clinical trials to investigate the therapeutic promise of CLU.
To investigate the relationship between tear concentration of the homeostatic protein clusterin (CLU) and dry eye signs and symptoms, and to characterize tear CLU protein. Two independent studies were conducted, one in Tucson (44 subjects), the other in Los Angeles (52 subjects). A cohort study design was employed to enroll patients without regard to dry eye diagnosis. Dry eye signs and symptoms were assessed using clinical tests. Tear samples were collected by Schirmer strip, and also by micropipette at slit lamp when possible. CLU from both sample types was quantified by immunoassay. The relationship between CLU concentration and clinical test scores was determined by Pearson's correlation coefficient (for individual eyes) and multiple linear regression analysis (including both eyes). CLU was also evaluated biochemically by western blotting. In the Tucson cohort, a positive correlation was observed between tear CLU concentration and results of the Schirmer strip test, a measure of tear flow (p = 0.021 includes both eyes). This result was corroborated in the Los Angeles cohort (p = 0.013). The mean tear CLU concentration was 31 ± 14 μg/mL (n = 18 subjects, 33 eyes; range = 7-48 μg/mL). CLU from clinical tear samples appeared biochemically similar to CLU from a non-clinical tear sample and from blood plasma. Results support the hypothesis that an optimal concentration of tear CLU is important for ocular surface health, and that this drops below the effective threshold in dry eye. Tear CLU measurement might identify patients that could benefit from supplementation. Information about concentration will aid development of therapeutic dosage parameters.