Lipids play a key role in many biological processes, such as cellular structure, energy storage, and cell signaling, and are dysregulated in a variety of health conditions, including aging, cancer, diabetes, cardiovascular disorders and neurodegenerative diseases. As such, there is a strong interest in the comprehensive analysis of lipids for biomarker discovery and for gaining novel insights into the onset and progression of diseases. However, global profiling of the entire collection of lipids (known as lipidomics) is challenging because the lipidome comprises thousands of different molecular species that are highly diverse in chemical structure and composition.
Mass spectrometry (MS) is the preferred technology to profile lipids as it provides resolution, sensitivity, selectivity, and throughput. MS-based lipidomics can be performed using either untargeted or targeted approaches, each with their own set of advantages and limitations. Untargeted platforms are unbiased and have broad coverage, as they potentially detect all the lipids present in a sample. However, they are often used in a semiquantitative manner (providing relative quantification) and certain aspects of the workflow, including data normalization and lipid identification, are challenging, time-consuming and not well standardized. Untargeted lipidomics can be done by direct infusion or liquid chromatography (LC)-based approaches. LC-MS is often preferred because it offers higher sensitivity as well as more accurate lipid identification and quantification. In contrast, targeted platforms traditionally focus on the absolute quantification of a small number of pre-defined lipids using isotopically labeled internal standards. The number of targets is often limited due to the lack of commercially available standards. Targeted approaches are high-throughput because data generation and analysis is fast and straightforward, quantitative, but usually have a more limited coverage.
Recently, a novel targeted lipid analysis technology has been developed called the Lipidyzer platform (SCIEX) that can quantify over 1,100 lipid molecular species across 10 lipid classes in human plasma/serum. The Lipidyzer platform estimates concentrations using a mixture of internal standards specifically designed for this purpose. Consequently, the Lipidyzer platform may constitute a good compromise between untargeted and conventional targeted approaches, providing broad coverage, accurate quantification and fast and straightforward data processing. However, its performance has not been compared to more established untargeted lipidomics approaches and it has not been tested in a biological setting.
These questions prompted us to compare a conventional untargeted LC-MS approach with the targeted Lipidyzer platform. In this study, we first conducted a cross-platform comparison of the workflow, lipid coverage, precision and accuracy, and determined how well the two platforms correlate quantitatively with each other. We then applied both platforms to profile lipids in plasma from young and old mice, because a comprehensive description of changes in lipids during healthy aging in mice is still lacking. Altogether, our study constitutes the first cross-platform comparison of two state-of-the-art lipidomics platforms and provides the first assessment of the Lipidyzer platform in the context of aging.