Lipidomics is a branch of metabolomics that, through an in-depth characterization, investigates the structures, functions and dynamic changes of lipids in cells, tissues, or bodily fluids. Moreover, lipidomics usually studies the correlation between the lipid profiles of biological samples and the health status of the human organism. The high level of structural diversity in lipid classes present in biological samples makes their characterization a very challenging task. The aim of this work was to develop a simple, fast, and versatile UHPLC-MS method. The method developed was suitable for both ESI and APCI MS-interfaces, for untargeted lipid profile characterization of human plasma.
The lipid fraction was extracted according to a modified Folch method.1 Briefly, 1 mL of plasma was placed in a centrifuge tube with 9 mL of chloroform:methanol (2:1 v/v), extracted for three times and centrifuged for 20 minutes at 3,000 rpm. To facilitate the separation between the phases, 500 μL of water was added to the previous mixture. The lower lipid containing organic phases were combined, dried with anhydrous Na2SO4, filtered on filter paper, and then dried using a rotary evaporator. Final extracts were dissolved in 500 μL of isopropanol:methanol (1:1 v/v) and injected into the LC/MS system.
The analyses were performed on a Shimadzu Ultra High Performance Liquid Chromatograph-Nexera system. The UHPLC system was coupled to an LC-MS-2020 quadrupole mass spectrometer equipped with both ESI and APCI interfaces and coupled to an LCMS-IT-TOF equipped with an ESI interface. The samples were simultaneously analyzed in full scan mode and under selected-ion monitoring (SIM) acquisition modes.
MS Parameters: Full-scan LC/MS chromatograms were obtained by scanning from m/z 350-1250, with a scan speed of 5000 amu/sec and an event time of 0.2 sec, in positive mode both for APCI and ESI, and from m/z 150-1250 and 160-1250 with a scan speed of 6000 amu/sec and an event time of 0.2 sec, in negative mode for APCI and ESI, respectively. ESI parameters were as follows: nebulizing gas (N2) flow rate: 2 L/min; drying gas (N2) flow: 15 L/min; detector voltage: 1.5 kV; interface voltage: 4.5 kV; desolvation line (DL) temperature: 250 °C; heat block temperature: 200 °C. APCI parameters were as follows: nebulizing gas (N2) flow rate: 3 L/min; drying gas (N2) flow: 15 L/min; detector voltage: 1.5 kV; interface voltage: 4.5 kV; interface temperature: 450 °C; DL temperature: 250 °C; heat block temperature: 200 °C.
When a large number of samples is analyzed, as in clinical cohorts, the total run time for analytical analysis plays an important role. A compromise is needed to maximize the sample throughput and the lipidome coverage, shorten the analysis time, limiting the peak capacity decreasing and maintaining the high quality MS information. The most suitable choice to achieve this compromise is the employment of a UHPLC system capable of operations at high back pressures up to 1200 bar. Using a UHPLC system, columns with sub-2 μM, like the Titan™ C18 column used here, particles can be operated at high mobile phase flow rates, thus reducing the analysis time without loss of resolution under optimized conditions.2 Until 2014, 65% of lipidomics studies developed reported analysis run time of about 30 minutes.3 Different solvents in combination with water were tested to optimize the UHPLC method. The water level in the mobile phase is particularly critical at the beginning of the gradient in reversed phase (RP) LC, affecting the chromatographic resolution of polar lipids. Higher concentration of water significantly improves not only the elution profile, but also the signal-to-noise ratio.4 The use of methanol or acetonitrile as solvent B, independently from the gradient applied, required a rather long separation time (> 30 minutes) to elute TAGs (triacylglycerols) and CEs (cholesterol esters).
Better results were achieved by using water containing 20 mM ammonium formate as solvent A and a mixture of isopropanol:acetonitrile:20 mM of ammonium formate (60:36:4 v/v/v) containing formic acid 0.1% as solvent B. The relatively low pH, achieved by using formic acid, allowed minimal tailing of FFAs (free fatty acids) due to the interaction of an ionized carboxyl function with free silanol sites on the LC column packing. The chromatographic pattern obtained partially fitted to the well-known model used for TAGs identification in RP-LC, where the retention of lipids increases proportionally to their Equivalent Carbon Number (ECN).4-6 The chromatographic LC method proposed was suitable, without any adjustments, to be used with both ESI and APCI interfaces. In such a way, two chromatographic profiles, perfectly equivalent in terms of retention times, were obtained. Complementary MS information can be extrapolated for a comprehensive and detailed characterization of each single sample.
The lipid extract from human plasma was analyzed by using the described UHPLC-ESI/APCI-qMS method. Starting from polar lipids, free fatty acid (FFA), and lysophospholipid (LPL) species, to non-polar lipids, triacylglycerols (TAG), and cholesterol esters (CE), all lipids elute within a 20 minute window. Figure 1 shows TIC (+) chromatogram by RP-UHPLC-APCI-MS of a human plasma sample. Figure 2 shows the mass spectra of PC-18:2/16:0 (FAs in phospholipids follow the nomenclature previously proposed for TAGs, not considering sn-position) from a human plasma sample using ESI and APCI interfaces in both positive and negative ionization modes.6
While ESI shows related molecular ions, both in positive and negative ion modes, APCI shows as base peak, in positive ion mode, the diacylglycerol specie due to the loss of head polar group [M-head polar group]+ (head polar group: 183 m/z), followed by different related molecular ions.
[M+H]+, [M+H-CH2]+, and [M+Na-CH2]+. In negative ion mode, a set of four parent ions [M-CH3]-, [M-60]- (loss of trimethylamine group), [M-86]- (loss of choline residue group), and an undefined ion, [M-72]-, related to a partial choline fragmentation, were observed. The simultaneous loss of choline residue (86 m/z) and a fatty acid from the backbone of glycerol allowed the characterization of fatty acids profile into PCs species.
The proposed splitless method can be a comprehensive platform for lipidomics studies. It can be applied without any modification in chromatographic conditions (mobile phase composition and/ or flow rate), coupled to both ESI and APCI interfaces. Such a chromatographic method can be easily reproduced. Although the sensitivity of APCI-MS is usually less than ESI-MS when a buffer is added to the mobile phase, the structural information to be gleaned from the fragmentation is well worth the trade off in sensitivity. However, in some cases, ESI is an indispensable complement to APCI when lipids containing oxygen functional groups are investigated, or for confirming related molecular ions that are generally less expressed by APCI. LC-qMS can be a valid, simple, and economical technique for lipidomics studies, and can provide similar information compared to more sophisticated techniques.