Analysis of Tear Protein Profiles in Dry-Eye Patients by Mass Spectrometry

Abstract

Purpose: : The composition of proteins and peptides in tears plays an important role in ocular surface diseases. In previous studies we could demonstrate changes in the tear protein patterns of dry-eye patients compared to controls.The aim of this study was to analyze if there are changes in the tear protein patterns between different subgroups of dry-eye patients. For this purpose we applied the SELDI-TOF MS ProteinChip® Array technology for an automated analysis of proteins and peptides of tear fluid. Additionally we used a MALDI-TOF MS approach for the identification of detected biomarkers.

Methods: : 39 patients, subdivided into healthy subjects (n=9), aqueous deficient dry-eye patients (n=10), dry-eye patients with changes in the lipid phase (n=10), and patients with a combination of both (n=10) were examined. Tears were collected on Schirmer strips, eluted, and incubated on SELDI-TOF-MS ProteinChip Arrays using a robotic laboratory automation station for all washing and binding steps. Protein fractionation was carried out using three different types of arrays. Samples were analyzed on a ProteinChip Reader. For workup and analysis raw data was transferred to the CiphergenExpress 2.1 database software. Detected peaks were clustered and statistical analysis incl. biomarker detection was performed using various algorithms like multivariate statistics and artificial neural networks.

Results: : In all subjects complex patterns of tear proteins and peptides were detected. More than 300 protein clusters were identified across three different chip surfaces, fractions, and laser energy settings. Analysis of discriminance demonstrated highly significant differences in the protein profiles between dry-eye patients and healthy subjects (P<0.001, AUC of ROC-curve r>0.92) and moreover between the different dry-eye groups (P<0.001). Some of the most important tear protein biomarkers could be identified as calgranulin, proline-rich protein 4 (PRP4) and alpha anti-trypsin.

Conclusions: : We detected several highly significant differences in the protein patterns of dry-eye patients versus healthy subjects and could differentiate between subgroups of dry-eye based on these biomarker patterns. These biomarkers such as calgranulin and PRP4 are thought to be inflammatory markers or mediate protective functions in the eye. The successful discrimination between the dry-eye subgroups and healthy subjects by proteomic markers could lead to both new innovative drug-targets and a diagnostic tool for dry-eye, which will lead to optimized treatments.

Keywords: proteomics • lacrimal gland