Higher Resolution LC-MS and MS-MS Analysis of Lipid Extracts Using Benchtop Orbitrap-based Mass Spectrometers and LipidSearch Software

Significance of the topic

High-resolution LC-MS/MS lipidomics addresses the biological and clinical need to comprehensively profile complex lipidomes in health and disease. Lipids are diverse and abundant biomolecules whose alterations are implicated in metabolic disorders, cancer and other pathologies. Accurate detection and confident structural assignment of thousands of lipid species require both chromatographic separation and high-resolution, high‑speed mass spectrometry together with robust informatic support for automated identification and relative quantification.

Study objectives and overview

This study benchmarked a benchtop ultra-high-field Orbitrap LC‑MS/MS platform to evaluate how full‑scan (MS1) resolution and MS/MS acquisition speed affect the number and confidence of lipid identifications from total lipid extracts. The work compared a Q Exactive Plus and a Q Exactive HF operated under data‑dependent dd‑MS2 workflows at multiple MS1 resolution settings and assessed automated identification and quantification using LipidSearch software.

Methods and experimental design

Samples and sample preparation:

• Total lipid extracts (Avanti Polar Lipids) from bovine brain, heart, liver and yeast were used as biological test matrices.
• Extracts were prepared as a dilution series (up to 2.5 mg/mL stock, serial dilutions down to tens of ng/µL) in 50:50 methanol:isopropanol for LC injection.

Chromatography:

• Reversed‑phase LC on an Ascentis Express C18 column (2.1 × 100 mm, 2.7 µm), 55 °C, 260 µL/min flow, 2 µL injection.
• Binary gradient with solvent A = 60:40 acetonitrile:water and solvent B = 90:10 isopropanol:acetonitrile; both had 10 mM ammonium formate and 0.1% formic acid; a multi‑step 33 min gradient was applied to separate lipid classes and isomers.

Mass spectrometry and acquisition strategy:

• Two benchtop Orbitrap platforms were evaluated: Q Exactive Plus and Q Exactive HF.
• MS1 resolution settings compared included nominal values spanning low to ultra‑high (e.g., 30k, 60k, 120k, 240k on HF; 17.5k–140k on Plus in other runs).
• Data-dependent MS2 used TopN strategies (Top15 for Plus, Top20 for HF), MS2 isolation width 1 Da, stepped normalized collision energy (Pos: 25,30; Neg: 20,24,28), and AGC targets of 1E6 for MS1 and 1E5 for MS2.

Data processing and identification workflow:

• LipidSearch software performed automated peak detection, database searching against a large lipid database (>10^6 predicted precursor and fragment ions; the vendor documentation cites >1.5 million entries), retention time alignment across runs, accurate‑mass extracted ion chromatogram integration for relative quantification, and basic statistical tests (t‑test) for group comparisons.
• Search tolerances used ±5 ppm for precursor and product ions in LC/dd‑MS2 analyses.

Used instrumentation

• LC system: Thermo Scientific Dionex UltiMate 3000 RSLC.
• Column: Ascentis Express C18 (Supelco), 2.1 × 100 mm, 2.7 µm.
• Mass spectrometers: Thermo Scientific Q Exactive Plus and Q Exactive HF (ultra‑high field Orbitrap).
• Ionization: Heated electrospray ionization (HESI) source; typical source parameters: spray voltage ≈ 4.2 kV, sheath/aux gas, capillary temp 320 °C, heater 300 °C, S‑Lens ~50.
• Data analysis: LipidSearch software (Mitsui Knowledge Industry/Thermo Scientific).

Main results and discussion

Number of identifications and effect of resolution/speed:

• The Q Exactive HF, operated with higher nominal MS1 resolution and a faster Top20 dd‑MS2 acquisition, delivered substantially more lipid identifications than the Q Exactive Plus under otherwise comparable conditions — the HF produced at least ~20% more lipid IDs in benchmark runs.
• Single LC dd‑MS2 experiments routinely yielded several hundred confidently identified lipids per extract; aligning multiple runs and merging positive/negative ion mode results produced cumulative identification lists exceeding 1,000 lipid species.

Importance of MS1 resolution to resolve isobaric/isomeric species:

• High MS1 resolving power was required to distinguish closely spaced precursor ions that coelute chromatographically. A specific example demonstrated overlapping lyso‑phospholipids (LPE 18:1 and LPC 16:0p) near m/z ~480.3: low resolution (~10–22k) failed to separate the signals, whereas a setting equivalent to ~60k MS1 (actual resolving power >40k at m/z 200) allowed unequivocal resolution of both M+H ions and confident identification of the lower‑abundance species.
• Insufficient MS1 resolution reduces the ability to detect and assign minor species and can bias relative quantification.

Positive vs negative ion mode and merged results:

• Searching both polarities and aligning identifications across modes increased overall coverage; merged results sometimes exceeded the simple sum of separate polarity IDs because alignment and correlation across samples resolved additional isomeric assignments.

Class composition findings (summary):

• Extract composition estimated from MS1 peak areas matched expected distributions: bovine heart and yeast extracts were dominated by triacylglycerols (TG ≈ 61–65% by area), whereas brain and liver extracts showed higher proportions of phosphatidylcholine (PC) and other polar classes.

Practical benefits and applications

• Higher MS1 resolving power combined with faster MS/MS acquisition increases both the number and confidence of lipid identifications in untargeted LC‑MS lipidomics workflows.
• Automated software with extensive fragmentation databases (e.g., LipidSearch) enables large‑scale screening and relative quantification in a single LC‑ddMS2 experiment, reducing manual annotation burden.
• These advances facilitate method benchmarking, discovery lipidomics, comparative studies across tissues or conditions, and the detection of low‑abundance biomarkers that might otherwise be obscured.

Future trends and potential developments

• Further gains are expected from combining ultra‑high MS resolution with increased acquisition speed and improved ion transmission to maximize MS/MS sampling of chromatographic peaks.
• Integration of orthogonal separations (e.g., ion mobility), enhanced fragmentation libraries and machine‑learning identification algorithms will improve isomer discrimination and structural elucidation.
• Standardization of sample preparation, internal standards and data reporting will be important for translating discovery lipidomics into clinical and regulatory applications.
• Advances in targeted quantification workflows and absolute quantitation strategies will complement broad profiling to validate candidate lipid biomarkers.

Conclusions

Using a benchtop ultra‑high field Orbitrap (Q Exactive HF) with optimized LC separation and a dd‑MS2 acquisition strategy provides more comprehensive lipidome coverage compared to an earlier benchtop Orbitrap (Q Exactive Plus). Higher MS1 resolution is critical for resolving coeluting isobaric species and detecting minor components, while higher MS/MS sampling yields more identifications. Automated database‑driven software enables rapid annotation and relative quantification of hundreds of lipid species per run and supports alignment to produce large cumulative lipid inventories across samples.

References

1. Fahy E., Sud M., Cotter D., Subramaniam S. LIPID MAPS comprehensive classification system for lipids. Journal of Lipid Research 2009, 50, S9–S14. doi:10.1194/jlr.R800095‑JLR200.
2. Bird S., et al. Lipidomics profiling by high‑resolution LC‑MS and high‑energy collisional dissociation fragmentation: focus on characterization of mitochondrial cardiolipins and monolysocardiolipins. Analytical Chemistry 2011, 83, 940–949. doi:10.1021/ac102598u.