Search

Many methods, often depending on tandem mass spectrometry, have been developed for analysis of complex mixtures of triacylglycerols (TAGs), especially in clinical diagnostics and food authentication. Understanding the fragmentation mechanisms of cationized TAGs has proved problematic. To obtain a better understanding of viable mechanisms, detailed studies including double- and triple-stage tandem mass spectrometry were made using electrospray ionization on lithiated and sodiated tripropanoyl- and trihexanoylglycerols. Density functional theory computations, including a functional parameterized for van der Waals interactions, were used to correlate computed energies with mass spectra. Losses of both a neutral salt and a neutral acid corresponding to a glycerol side chain were observed as major product ions in MS 2 experiments. Signal intensities at low collision energies correlated well with computed energies. However, an important difference between the lithiated and sodiated ions was the appearance of the sodium cation as a major fragmentation product. Computations on the product ions resulting from the loss of a neutral acid indicated multiple structures for the lithiated ions but mainly a single structure for the sodiated ions. The lithiated product ions could be fragmented further (pseudo-MS 3 ) to give additional structural information, whereas the sodiated ions gave only m/z 23. The longer chain TAG, while giving a much less intense mass spectrum than the shorter chain TAG, gave comparable MS 2 and MS 3 product ion spectra. Taken together, the spectral and computational work described herein offer a new and detailed pathway for collision-induced fragmentation of lithiated and sodiated saturated TAGs.

Reversed-phase high performance liquid chromatography (RP-HPLC), followed by post-column addition of lithium salts and electrospray ionisation triple-stage mass spectrometry (ESI-MS(3)) of lithiated TAG adducts, is shown to provide a useful method for the positional analysis of triacylglycerols (TAGs) in fish oils containing eicosapentaenoic (EPA, 20:5) and docosahexaenoic acids (DHA, 22:6). One prominent fragmentation pathway in the ESI-MS(3) of these adduct ions involves the loss of a fatty acid from the sn-1/3 position in the first step followed by the loss of an α,β-unsaturated fatty acid from the sn-2 position in the second. Regioisomeric TAGs of the type ABA and AAB produced abundant product ions - [ABA+Li-RACOOH-R'BCHCHCOOH](+) and [AAB+Li-RACOOH-R'ACHCHCOOH](+) - the relative intensities of which were dependent on the position of acyl substituents. Standard solutions of TAGs containing different ratios of the regioisomeric pairs MME/MEM, PPE/PEP, PPD/PDP, EEP/EPE and DDP/DPD (M=14:0, P=16:0, E=20:5, D=22:6) were analysed by ESI-MS(3) with a quadrupole linear ion trap instrument. Methodology developed on the standards was applied to quantifying the relative isomeric abundances of EPA and DHA in several fish oil samples. DHA was preferentially located at the sn-2 position in both DHA-containing TAGs studied, while EPA was either observed at near equal levels in all positions, or predominantly at the sn-1 and -3 positions in some cases. The analysis protocol allows for quantification of the designated regioisomers in one simple, rapid chromatographic procedure using a single column and has the advantage of specificity over other methods for the positional analysis of TAGs, since it eliminates interferences associated with co-eluting TAGs of the same molecular weight that yield isobaric diacylglycerol-like product ions., (Crown Copyright © 2013. Published by Elsevier Ltd. All rights reserved.)

Rationale: Regioisomeric analysis of triacylglycerols is important in understanding lipid biochemistry and the involvement of lipids in disease and nutrition. The use of calibration plots employing fractional abundances provides a simple and rapid method for such analyses. These plots are believed to be linear, but evidence exists for non-linearity. The behavior of such plots needs to be understood to allow for proper interpretation of regioisomeric data., Methods: Solutions of five regioisomer pairs were prepared from pure standards and used to construct calibration plots using triple-stage tandem mass spectrometry (MS(3) ) with electrospray ionization (ESIMS(3) ) and cationization by lithium ions. The data were taken by direct infusion with an AB SCIEX QTRAP 2000 QqLIT mass spectrometer., Results: Non-linear calibration plots were observed for the four isomer pairs containing the polyunsaturated eicosapentaenoic (20:5) and docosahexaenoic (22:6) acids paired with palmitic acid (16:0) or myristic acid (14:0), while the pair including palmitic and stearic (18:0) acids provided a linear plot. A non-linear model was developed for these plots and then verified experimentally., Conclusions: The fractional abundance calibration plots used in regioisomeric analysis of triacylglycerols are intrinsically non-linear, but may appear linear if the scatter in data points obscures the curvature, if the curvature is slight, or if the response factors for the two isomers in the regioisomer pair are similar. Therefore, linearity should not be assumed for these types of measurements until confirmed experimentally., (Copyright © 2013 John Wiley & Sons, Ltd.)