A New Lipid Software Workflow for Processing Orbitrap-based Global Lipidomics Data in Translational and Systems Biology Research

Importance of the topic

High-resolution lipidomics is critical for translational and systems-biology research because the lipidome is chemically complex (multiple categories, classes, subclasses and thousands of isomeric/isobaric species). Accurate identification and quantification of lipids enables phenotype characterization, biomarker discovery and mechanistic insight for diseases and metabolic perturbations. Combining Orbitrap high-resolution MS with MS/MS-driven database searching addresses the limitations of accurate-mass-only approaches and improves specificity in complex biological extracts.

Objectives and study overview

This study introduces a new, automated workflow based on Lipid Search software for processing Orbitrap LC-MS/MS lipidomics data. The workflow was demonstrated using a yeast model comparing wild-type (WT) versus a knockout (KO) strain defective in coenzyme Q (CoQ) biosynthesis. Goals included evaluating identification depth and speed, comparing MS2 database searching to accurate-mass metabolomics approaches, and reporting biologically meaningful lipid changes between phenotypes.

Methods and workflow

The analytical workflow combined sample preparation, reversed-phase LC separation, high-resolution full-scan and data-dependent MS/MS acquisition, and an automated MS2-driven database search and alignment. Key processing steps implemented in Lipid Search were:

• 1) Peak detection: reading raw Orbitrap files, extracting MS1 and MS2 features.
• 2) Identification: MS2 spectra matched against a comprehensive lipid database (order of 10^6 predicted entries) to assign candidate lipid structures and score fragmentation fits.
• 3) Alignment: per-sample identifications aligned across runs within a retention time tolerance (0.25 min used here) to create a combined dataset.
• 4) Quantification: extracted ion chromatograms (accurate-mass EICs) for identified precursors integrated to provide relative peak areas.
• 5) Statistics: pairwise t-tests and PCA were used to detect significantly changing lipids and to visualize group separation.

The authors contrasted this MS2-driven approach with component-finding plus accurate-mass (MW) searches used in metabolomics, arguing MS2-driven database searching is superior for lipids in complex extracts.

Used instrumentation

• Chromatography: Thermo Scientific Accela 1250 LC with Accela Open autosampler; column: C18, 2.1 × 100 mm, 2.7 µm; flow 260 µL/min; column temperature 55 °C; injection 10 µL. Reconstitution solvent for analysis: 65:35:5 acetonitrile:isopropanol:water containing 5 µg/mL 17:0 PG internal standard.
• Mass spectrometry: Thermo Scientific Q Exactive Orbitrap. Full-scan MS at 70,000 resolution (positive ESI), Top5 data-dependent MS/MS at 35,000 resolution, collision energy ~35. Mass tolerances used for Lipid Search: 5 ppm precursor, 10 ppm product ions.
• Software: Lipid Search (MKI) for lipid identification and alignment; Thermo SIEVE used for metabolomics-style accurate-mass searches for comparison.

Main results and discussion

• Identification depth: Lipid Search identified 380 distinct lipid species from yeast mitochondrial extracts using LC-MS/MS; this is comparable to numbers reported by infusion (shotgun) lipidomics studies for yeast.
• Biological differences: 112 lipid species showed significant differences between KO and WT (p < 0.05). Notable findings included reductions in CoQ6 (oxidized) and multiple diacylglycerol (DG), phospholipid (PE/PC/PG/PI/PS) and triacylglycerol (TG) species with varied fold-changes. PCA showed clear separation of WT and KO metabolomes/lipidomes, consistent with the CoQ biosynthesis defect affecting mitochondrial lipid composition.
• Method comparison: MS2-driven database searching outperformed accurate-mass-only MW searches for assigning lipids in complex extracts because it uses predicted fragment ions and an extensive structural database to resolve isomeric/isobaric ambiguities and to detect lipid mixtures in single spectra.
• Mixture handling: Lipid Search flagged spectra that contained contributions from multiple co-eluting lipids and reported the dominant identification while providing fragment-level evidence, improving confidence versus mass-only matches.
• Throughput: Data analysis was rapid—search and alignment completed in minutes on a modern laptop (example: < 8 min on an i7, 8 GB RAM), substantially reducing time compared with manual or less-automated workflows.

Practical benefits and applications

• Reliable identification: The MS2-centric database approach increases confidence in species assignments in complex biological samples where many isomers and isobars coexist.
• Quantitative comparability: LC-MS/MS combined with automated alignment and EIC-based quantification yields reproducible relative quantification across biological replicates and permits statistical comparisons (t-tests, PCA).
• Phenotype profiling: The workflow enables sensitive detection of phenotype-associated lipid changes (e.g., CoQ pathway disruption), making it applicable to disease biomarker discovery, phenotype screening, and systems-biology studies focused on membrane and energy metabolism.
• Operational efficiency: Automated processing reduces hands-on time and accelerates exploratory analyses and routine lipidome screening in translational labs.

Future trends and applications

• Deeper structural resolution: Integration of ion-mobility, targeted MSn strategies, and improved fragmentation models will further resolve positional and double-bond isomers in complex lipidomes.
• Quantitation improvement: Combining MS2-driven IDs with isotope-labelled internal standards and targeted parallel reaction monitoring (PRM) could support absolute quantitation for clinical biomarker validation.
• Database expansion and machine learning: Growing curated spectral libraries and ML-driven scoring will improve identification accuracy and permit confident annotation of low-abundance and novel lipid species.
• High-throughput translational screening: Scalable LC-MS/MS pipelines with automated identification will enable larger cohort studies, longitudinal monitoring, and integration with other omics layers in systems-biology workflows.

Conclusions

Lipid Search coupled to high-resolution Orbitrap LC-MS/MS provides an automated, fast and robust workflow for global lipidomics in translational and systems-biology contexts. MS2-driven database searching delivers superior specificity compared with accurate-mass-only approaches, supports detection of mixtures, and produced deep coverage (≈380 lipids) with dozens of statistically significant phenotype-associated changes in the yeast CoQ-deficient model. The approach improves throughput and reliability for lipid biomarker discovery and comparative lipidome profiling.

References

1. Peake DA, et al. Quantitative yeast lipidomics via LC-MS profiling using the Q Exactive Orbitrap mass spectrometer. Presented at LIPID MAPS Annual Meeting 2012, La Jolla, CA.
2. Fahy E, et al. LIPID MAPS comprehensive classification system for lipids. Journal of Lipid Research. 2009;50:S9–S14. doi:10.1194/jlr.R800095-JLR200.
3. Taguchi R, et al. Precise and global identification of phospholipid molecular species by an Orbitrap mass spectrometer and automated search engine Lipid Search. Journal of Chromatography A. 2010;1217:4229–4239. doi:10.1016/j.chroma.2010.04.034.
4. Yamada T, et al. Development of a lipid profiling system using reversed-phase LC coupled to high-resolution MS with rapid polarity switching and automated lipid identification. Journal of Chromatography A. 2013;1292:211–218. doi:10.1016/j.chroma.2013.01.078.
5. Ejsing CE, et al. Global analysis of the yeast lipidome by quantitative shotgun mass spectrometry. Proceedings of the National Academy of Sciences USA. 2009;102:17981–17986. doi:10.1073/pnas.0811700106.