Characterization of fatty acid double bonds using electron-activated dissociation (EAD)-based fragmentation
Zhuo Man1, Dandan Si1, Paul RS Baker2 and Zhimin Long1
1 SCIEX, China; 2 SCIEX, USA
In this technical note, EAD was used to identify the double bond position in the carbon-carbon chain of fatty acid standards. Fatty acids can have multiple biologically relevant isomers that vary in the position and/or stereochemistry of the double bond. For example, oleic acid (9-octadecenoic acid) and vaccenic acid (11- octadecenoic acid) differ only in the double bond position but they play different metabolic roles in vivo. Using liquid chromatography-electrospray ionization mass spectrometry (HPLC ESI-MS/MS), fatty acids are typically analyzed in the negative ion mode and some isomers cannot be adequately resolved by liquid chromatography. Using collision-induced dissociation (CID) fragmentation data, the product ions derived from the individual positional and stereoisomers are indistinguishable. To improve sensitivity and generate relevant fragmentation data to determine double bond position, the oleic and vaccenic acid isomers were derivatized using a tertiary amine that enables their detection in the positive ion mode. These products generated intense precursor and fragment ions, improving assay sensitivity. The complementary EAD fragmentation mode on the ZenoTOF 7600 system was used to generate structurally diagnostic fragment ions to address structural specificity. EAD enabled the characterization of each isomer by identifying the double bond position and stereochemical configuration, as shown in Figure 1.
Fatty acids play critical roles in mammalian biochemistry as sources of energy and metabolites, structural components of complex lipids and precursors to lipid signaling molecules. Consequently, complete structural characterization of fatty acids is essential to fully understand their metabolic import. The number of carbons within the fatty acyl chain can typically vary between 2 and 30 and can be distinguished by mass. The location of any double bonds within the fatty acid chain can vary to generate distinct lipid isomers with unique biochemical fates. Fatty acid analysis by gas chromatography (GC) is a well-established method to identify fatty acids and distinguish between double-bond isomers [1]. However, this technique requires derivatization and cannot be used to evaluate fatty acids within intact complex lipids, such as triglycerides and phospholipids, without prior hydrolysis. Fatty acid analysis by ESI-MS/MS, either as free acids or as esterified components of complex lipids, presents significant challenges regarding the localization of double bonds. These challenges arise because conventional liquid chromatography (HPLC) strategies lack sufficient resolving power and CID-based fragmentation often does not generate fragments that denote double bond positions.
Several strategies have emerged to locate double bonds in fatty acids by HPLC ESI-MS/MS. The addition of ozone during mass spectrometry analysis (termed OZ ID) [2] can generate double-bond site-specific fragments. Alternatively, localization can be achieved using the Paternò-Büchi reaction carried out in the electrospray region of ionization to form an oxetane adduct from acetone with each double bond in the fatty acyl chain [3, 4]. However, these 2 methods require instrument alteration and are not conducive to high-throughput analysis.
EAD fragmentation on the ZenoTOF 7600 system [5] is capable of cleaving carbon-carbon single and double bonds to pinpoint the location of double bonds in free and acylated fatty acids on an LC time scale [6]. This novel technology enables near-complete structural characterization of lipids. In this technical note, 4 model fatty acid standards were used to demonstrate the ability to distinguish isomers.
Sample preparation: The 9-cis-octadecenoic acid (oleic acid), 9-trans-octadecenoic acid (elaidic acid), 11-cis-octadecenoic acid (vaccenic acid) and 11-trans-octadecenoic acid (trans-vaccenic acid) isomers were obtained from Avanti Polar Lipids (Alabaster, AL) and derivatized at the carbolic acid functional group using trimethylethylenediamine. The derivatized unsaturated fatty acids were prepared in a mixed standard, diluted with 1:1, methanol/water and analyzed by HPLC ESIMS/MS (Figure 2).
Chromatography: A Shimadzu Prominence LC system was used with a BEH C18 column (100 × 3.0 mm, 1.7 µm) to resolve the partially separated fatty acid isomers. The column temperature was set to 50°C. A gradient elution was used at a flow rate of 300 µL/min, as described in Table 1. The injection volume was set to 5 µL.
Mass spectrometry: Data were acquired using SCIEX OS software on the ZenoTOF 7600 system in positive polarity. Data were collected from a single injection using a combination of data-dependent acquisition (DDA) and MRMHR experiments. Both fragmentation techniques were used in separate experiments to compare the information generated from CID with EAD. Relevant MS parameters for the EAD method are described in Table 2. MRMHR settings are presented in Table 3.
Data processing: All data were analyzed using SCIEX OS software.
Prepared fatty acid derivatives were analyzed using the ZenoTOF 7600 system and analyzed in the DDA or MRMHR scan modes. The HPLC elution profile of the 4 lipid molecular species is shown in Figure 3. While HPLC chromatographic separation of the respective stereoisomers was possible, under these conditions, it was not possible to separate the double-bond positional isomers by HPLC. Traditional CID-based fragmentation of fatty acids generates minimal fragments through the loss of water(s) and CO2. For these derivatized fatty acids, the predominant fragments are related to the scission of the derivative. None of these fragments are diagnostic for double bond positions and the CID spectra of these molecules are identical (Figure 4).
Using EAD on the ZenoTOF 7600 system to fragment the fatty acids generates significantly more structural information regarding the isomers than CID. EAD and CID fragmentation modes can be run independently or sequentially in the same experiment. CID dwell times are 10 ms, whereas EAD dwell times are approximately 25 ms. To improve the overall duty cycle of the experiment, EAD-based fragmentation should only be used when it is needed to generate a compound-specific fragment that is not generated during CID. Figure 5 shows the EAD spectra for the cis isomers of oleic (top) and vaccenic (bottom) acids. Whereas the CID spectra in Figure 4 show no appreciable fragments generated from the scission of the carbon-carbon bonds in the fatty acid carbon chain, EAD shows successive cleavage of all carbon-carbon bonds (highlighted in Figure 1). For oleic acid, the fragments at m/z 239.2120 and 253.2286 are formed by successive cleavage of the carboncarbon bonds, denoting the location of the double bond at the Δ9 position. Vaccenic acid generates EAD-derived fragments that enable the assignment of the double bond to the Δ11 position (Figure 5).
It is possible to assign the stereochemistry of the double bond from EAD spectra [7]. Generally, trans double bonds referentially undergo heterolytic scission, whereas cis double bonds are equally likely to undergo homolytic and heterolytic scission during EAD fragmentation. The EAD-based MS/MS spectrum can determine these differences. However, this determination might require optimization of the electron kinetic energy (KE) for each compound. Considering that HPLC was sufficient to resolve the stereoisomers in this experiment, EAD-based stereo configuration was not addressed.
Standard curves were generated using the isomer-specific fragments identified in Figure 5 and listed in Table 3. Quantitative analysis was performed to evaluate fatty acid isomer analysis on the ZenoTOF 7600 system (Figure 6). A calibration curve was generated with a linear dynamic range from 0.48 to 7.48 ng/mL, with r values for all 4 curves >0.990. Instrument testing has shown the linear dynamic range in positive and negative ion modes to be >5 orders of magnitude. Figure 7 shows the individual concentration points of the calibration curve for oleic acid. Oleic acid and elaidic acid appear in the chromatogram and peaks for oleic acid are highlighted in blue.
The quantitative accuracy of the assay was assessed by measuring the recovery of a single-standard sample containing 1.25 ng/mL oleic acid using the calibration curve (Figure 8). Based on the peak area of the oleic acid (duplicate injections), the calculated concentration ranged from 93.8% to 107% of the nominal value. In the bottom panel of the chromatograms, MRMHR transitions were used to monitor for vaccenic acid. No cross-contamination was observed, which confirms the specificity of EAD-derived fragments and the quantitative accuracy of the assay.