Detection of tear protein biomarkers associated with contact lens use

Abstract

Purpose : To apply high-sensitivity proteomic analysis to human tears and identify differentially abundant proteins in contact lens wearers.

Methods : Tears were collected from both eyes of 48 volunteers using a 5-min Schirmer’s strip collection then stored individually at -80^oC until protein extraction. Volunteers also completed questionnaires about dry eye and ocular pain symptoms. Samples from individuals with features of interest (contact lens use; aqueous production as Schirmer's strip wetting length; ocular symptoms; topical anesthetic) were selected for quantitative proteomic analysis. Proteins were extracted from Schirmer's strips using S-Trap protocol, digested into peptides, and 16-plex Tandem Mass Tag (TMT) analyses were conducted using an Orbitrap Fusion mass spectrometer. Identified peptides were verified with open source databases and results were sorted with quality filters to identify candidates for differential abundance between groups.

Results : For contact lens use analysis, we examined samples from 14 eyes of 14 subjects (5 males; 9 females; age 23-29 years) with no history of ocular surgeries, no current eye infections, and no topical ophthalmic medication use. 5 subjects were contact lens wearers and 9 were not; half were anesthetized with topical 0.5% proparacaine HCl. Over 3000 unique proteins were identified in the volunteers’ tear fluid based on peptide sequence matching from curated databases. Application of the Benjamini-Hochberg multiple-testing correction to limit the false discovery rate and rigorous data quality filtering revealed 4 proteins with differential abundance between contact lens wearers and non-wearers, with cystatin D (CST5) demonstrating the greatest (8-fold higher) differential abundance in contact lens wearers. In addition, we found differential abundance of selected proteins previously reported in the literature, including lipocalin-1, lacritin, S100-A2, cystatins M and C, prolactin-inducible protein and zinc-alpha-2-glycoprotein. Subject age, sex, Schirmer's strip wetting length, dry eye and ocular pain scores, and topical anesthetic use were not different between the groups.

Conclusions : Unbiased TMT proteomics analyses demonstrated highly sensitive and reproducible analysis of human tear protein composition, thus showing promise as a technique in biomarker discovery. Contact lens wear significantly altered the tear proteome, while topical anesthetic use, subject sex and age did not.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.