Abstract
Purpose:
To profile peripheral blood changes in the metabolites of patients with wet AMD using a stable isotope labeling liquid chromatography-mass spectrometry (LC-MS) method.
Methods:
76 plasma samples from 40 wet AMD patients and 36 age-matched controls were 12C2-isotope labeled and mixed 1:1 with 13C2-isotope labeled pooled reference (made up by pooling 76 plasma samples together). Two sets of LC-quadrupole time of flight (LC-QTOF) runs were performed; each set contained 76 plasma sample injections. Principal component analysis (PCA) score plot was used to show the separation of AMD and controls in a 2D plot. Metabolite differences in AMD versus controls were established with three statistical methods: volcano plot; orthogonal partial least squares discriminant analysis (OPLS-DA); receiver operating characteristic (ROC) curve. Three empirical criteria defined the top ranked metabolites varying in levels between the two groups: variable importance in projection (VIP) score > 2.0 (OPLS-DA); fold change > 1.5 with p < 0.05 (volcano plot); area under the ROC curve > 0.75. The Human Metabolome Database (HMDB) was used for putative metabolite identification.
Results:
Over 2,000 peak pairs were detected in each plasma sample. A total of 608 peak pairs were commonly exited in all of the plasma samples analyzed. There was a trend for AMD samples to cluster in the top and control samples to cluster in the bottom of the PCA plot. Concentration differences between AMD and controls were observed for 19 metabolites; these included dipeptides, tripeptides, hybrid amino acids, alkylamines and phenolamines.
Conclusions:
Stable isotope-assisted metabolomics is a method that overcomes ionization suppression and that normalizes the variation in the process of sample extraction and derivatization. We identified putative compounds of significant concentration difference in peripheral blood of wet AMD patients when compared to controls. These metabolites will be studied for future validation as markers of wet AMD and their potential to provide pathophysiological insight and ultimately identify novel therapeutic targets.
Keywords: 412 age-related macular degeneration •
592 metabolism •
609 neovascularization