Identification of novel biomarkers for diabetic retinopathy in human tears

摘要

Diabetic retinopathy (DR) is a sight threatening disorder which develops in nearly all patients with diabetes mellitus (DM). DR centres on the abnormal growth and rupturing of the retinal blood vessels. Current modes of diagnostics rely on the clinical presentation of DR, at which stage, maturation of the disease and possible damage to the retinal tissue is significant. Due to the high incidence and severity of damage resulting from this disorder a diagnostic which allows early and accurate detection is essential.It was the aim of this project to identify biomarkers in human tears which would provide the basis for development of a non-invasive diagnostic test for DM and DR. This biomarker discovery involved extensive method development for optimal identification of quantitative changes to the tear peptidome and proteome. Proteomic technologies including MALDI-MS, iTRAQ, 2-D LC MS/MS, 2-D electrophoresis, MF10 fractionation and non-labelling mass spectrometry-based strategies were assessed in terms of their utility to investigate the tear proteome and peptidome. This thesis describes the first study to use MF10 technology and peak integration analysis to investigate the peptidome fraction of tears between different subject groups. The application of spin filters to partition higher molecular weight tear proteins from analysis, vastly improved the visualization of lower molecular weight proteins and peptides of human tears using MALDI. It is also the first study to mine the peptidomic population in tears for disease associated biomarkers. From this, one hundred and nine peptides in the 1-25 kDa MF10 fractions of the tear samples were found to alter between DM and DR subjects in comparison to healthy controls. Thirteen of these peptides were sequenced, identified and subsequently validated using an MRM assay developed specifically to quantitate and verify candidate tear peptide biomarkers.The work presented in this thesis is the first study to identify and profile tear peptide biomarkers which have correlated with the onset of DM and DR. It is envisioned that these biomarkers will form the intrinsic components of a non-invasive diagnostic for DM and DR. Furthermore, this work has established principles for future peptidomic or low abundance protein studies with complex bio-fluids. The work has demonstrated that fractionation is an important part of both biomarker discovery and validation, especially in the context of identifying biomarkers in a complex biofluid such as tears. Coupled with validation through targeted quantitative mass spectrometry, the workflow described is a powerful tool of clinical utility in biomarker discovery and validation.