Coupling Accelerated Microdroplet Reactions on LC-MS Systems: Toward Absolute Quantification and Characterization of Lipid Isomers

Traditionally, accelerated microdroplet reactions are used for chemical synthesis. We hypothesized that the ability to perform rapid post-column reactions during the stages of microdroplet formation of electrospray process can provide an avenue to limit ion suppression effects that beset quantitative MS. Post-column reactions that are not fast will limit the resolving power of the LC, especially when separating closely related isomers. Herein, we show that charged microdroplet generated under ambient conditions, represent a unique environment to achieve rapid (ms) quantitative reactions, which we apply for absolute quantification of lipid isomers. To improve microdroplet reactivity, we developed a plasma-microdroplet fusion system that stabilizes ESI signal in the presence of plasma. Through this co-axial spray mechanism, charged microdroplets derived from LC and containing lipid were reacted in real time, which facilitated (i) positive-ion mode detection of various lipid classes and (ii) instantaneous C=C bond epoxidation via reaction with reactive oxygen species present in the plasma. Hence, conventional MS/MS based on CID was made effective in characterizing positional isomers of various lipids. The mechanism governing the baseline separation of C=C positional isomers is ascribed to an induced hydrophobic effect, which produced extended interaction for isomers having C=C bond closer to carbonyl group in fatty acid chain. Data dependent acquisition method will be discussed, which enabled automated characterization of epoxide products . The method was applied to characterize novel triglycerides not previously detected in olive oil. The co-axial spray method also allows a single internal standard to be used for the quantification of different lipid classes. The relative response factors (RRF) for six different lipids have been determined to be in the range 0.1 < RRF <10, which yielded absolute quantification with accuracy <20% error.