Abstract
Purpose: :
Unlike proteomic analysis where large protein data bases and sophisticated computer software programs are routinely used to compare human samples, identification and comparison of lipids remains laborious and is largely qualitative rather than quantitative. Existing bioinformatic software packages and techniques for comparison of lipids from dry eye and normal samples analyzed by MALDI and ESI-MS are explored.
Methods: :
Electrospray time-of-flight mass spectrometry (ESI-MS) analysis (Waters, Q-TOF II) and MALDI-TOF/TOF (Bruker ultrafleXtreme) was performed on meibum samples collected from each of three groups: postmenopausal women with dry eye (n = 10), postmenopausal women without dry eye (n = 10), and patients with contact lens-related dry eye (n = 10). Nonlinear Dynamics Progenesis MALDI software was used to quantify differences in lipids. Identity of the lipid showing statistical changes in abundance was confirmed by MS/MS.
Results: :
Direct infusion ESI-MS and MALDI-MS provides hundreds of lipid peaks for comparison in both positive and negative ion mode, consistent with previous reports. For Control vs. DE, 5 lipid peaks had a 1.25 or greater fold change with all lipids detected as down-regulated between normal and DE. The Control vs. CLDE showed thirteen lipids changing with a 1.25 of greater fold change with all 13 lipids down-regulated between control and CLDE. The DE vs. CLDE detected 7 lipids with a 1.25 fold change or higher, 5 up-regulated and 2 down regulated between DE and CLDE.
Conclusions: :
The major goal of dry eye disease biomarker discovery, and in this case lipid profiling, is to identify disease-specific lipids from human patients. Here we used a bioinformatic software package to interrogate pilot data from 3 sample groups, and discovered potential lipid candidates demonstrating up- or down-regulation. The lipids classified by the software are consistent with lipids seen by our group and others in human meibum samples. The Progenesis MALDI v1.2 software has promise in comparing human lipid samples.