Abstract

Lipidomics is the comprehensive study of the lipids present in an organism. To achieve high analytical sensitivity and good lipid coverage in global lipidomics, sample preparation, mobile phase composition, chromatographic separation, and data processing parameters must be optimized. Solid-phase microextraction (SPME) is a non-exhaustive, in vivo sample preparation method that has not previously been applied to lipidomics studies but that provides a promising alternative to microdialysis for in vivo lipid analysis. Before SPME can be successfully applied to lipidomic studies, it is important to increase the sensitivity of liquid chromatography-mass spectrometry (LC-MS) detection as much as possible. This was accomplished by comparing the effect of four different mobile phase additives on signal intensity using negative electrospray ionization (ESI): acetic acid, ammonium acetate, ammonium hydroxide and ammonium acetate with acetic acid. Acetic acid at a concentration of 3.5 mM outperformed the other additives tested causing a 42% increase in lipid coverage along with a 2-19-fold increase in signal compared to 10 mM ammonium acetate. The main disadvantages of acetic acid were (i) wide peak shape for two lipid classes: phosphatidic acid and phosphatidylserine and (ii) reduction of signal intensity for phosphatidylcholine and ceramide lipids. These results show that although basic pH promotes in-solution ionization, surface, electrochemical and gas-phase processes provide significant contributions to overall ESI efficiency thus resulting in enhanced ionization at acidic pH. Acetic acid is good choice of additive for negative mode ESI due its ability to increase formation of deprotonated adducts, low molecular volume, high gas phase proton affinity, and capability for electrochemical reduction. Traditionally, lipids are extracted from brain tissue using the Folch method, a biphasic liquid-liquid extraction method that relies on chloroform/methanol solvent. In this study, SPME was compared to the Folch method to determine if it is a viable alternative. The main SPME parameters that affect extraction efficiency of in vivo SPME were evaluated for this application: extraction time, desorption time, and desorption solvent. Analysis of SPME extracts showed 147 and 613 features versus 1368 and 1161 detected in the Folch extract in positive mode and negative mode, respectively. Using LipidSearch 127 lipids in positive mode and 57 lipids in negative mode were identified in the SPME extracts verses 145 in positive mode and 99 in negative mode for the Folch extracts. This study represents the first step towards the development and inter-laboratory validation of in vivo SPME demonstrating its ability as a new and viable alternative to Folch extractions that can be used in lipidomics studies of the brains of alive, awake animals for the first time.