1
Introduction to Lipidomics

Harald C. Köfeler1, Kim Ekroos2, and Michal Holcapek3

1 Medical University Graz, Center for Medical Research, Stiftingtalstrasse 24, 8010, Graz, Austria

2 Lipidomics Consulting, Esbo, Finland

3 University of Pardubice, Faculty of Chemical Technology, Department of Analytical Chemistry, Pardubice, Czech Republic

1.1 Preface

We are entering a new era in lipidomic analysis. Technology advances in conjunction with community-wide collaboration efforts have prompted new ways to investigate the world of lipids. These developments have revoked interest in lipids, creating new opportunities to study lipids in different biological and biomedical settings in the hope of improving health and disease. Today, technologies allow us to dive deep into the lipid content and dissect the lipid makeup in detail, providing quantitative numbers of hundreds of lipid molecules. Lipid measurements no longer circle just around cholesterol in the context of LDL or HDL, but now the typical target is to determine the comprehensive lipidome of these particles. The new previously unseen lipid details spark curiosity and interest in reactivating research on cellular membranes, signaling cascades, and metabolic networks, among others, to shed new insights into the dysfunctions underlying a disease or a disorder. The objectives are clear. Can lipid details untangle disease biology, provide improved predictive or diagnostic biomarkers, and deliver new therapeutic strategies? However, opportunities extend further beyond, as a detailed lipid fingerprint can be envisioned, serving as a health status map of individuals. Our unique lipid code, which all of us possess, becomes a tool for precision health and medicine, which we are only beginning to explore.

The study of lipids using lipidomics can be rephrased as mass spectrometry (MS)-based lipid analysis. Until now, the field has been living its Wild West era where everything has been allowed. Although this has provided significant development, the downside is that it has resulted in inaccurate and irreproducible research results, preventing science from moving forward. With the establishment of the International Lipidomics Society (ILS), we have taken an active role in further maturing, harmonizing, and developing the lipidomics field to meet the current and future needs. By connecting the worldwide lipid community and focusing on transparent communication and collaboration, we aim to identify the common language for the entire discipline. Simply, the focus is to guide, educate, collaborate, and provide services to the academic and medical communities, industries, and the public in lipidomics. We have established several interest groups (see https://lipidomicssociety.org/working-groups) with different focuses to accelerate various angles of the field. A central program is briefly described here with the focus on the standardization of lipidomics, where we are preparing a new reporting checklist for any future lipidomics study. This is a true game changer that is needed to unlock the full potential of lipidomics. Now, we can meet the regulatory requirements for use in clinical research and diagnostics and enhance the comparability of data and understanding of the functional roles of specific lipid species. A new order in lipidomics has begun.

1.2 Historical Perspective

Although the determination of individual lipids by MS goes back to the 1970s (e.g. prostaglandins by GC/MS), the term lipidomics was introduced in 2003 by Xianlin Han and Richard Gross, defined as the system-level analysis of lipid species' abundance, biological activities, subcellular localization, and tissue distribution [1]. Lipidomics became possible by the introduction of new technologies in MS, particularly electrospray ionization (ESI), matrix-assisted laser desorption/ionization (MALDI), and Orbitrap instrumentation, resulting in a broader scope of analysis with increased sensitivity and selectivity. Fueled by these technical prerequisites and the concomitant increased biological usability of lipid data, a growing number of scientific groups have joined the field. In parallel, it soon became clear that the fast growing lipidomics field would need some sort of guidance for standards. In the early new millennium, LIPID MAPS was funded by NIH as a huge "glue grant" that included multiple labs in the United States. The most important achievement of the LIPID MAPS consortium was a comprehensive classification scheme of lipids into eight categories subdivided into dozens of lipid classes and subclasses [2, 3]. Based on this classification scheme, the LIPID MAPS Structure Database (LMSD) became the most important and comprehensive international lipid database containing 46?843 lipid structures as of December 2021, 24?815 of them experimentally proven and curated, and 22?028 of them generated in silico [4]. In parallel, a large-scale European grant LipidomicNET was awarded by the European Union and started to develop annotation rules for lipids detected by MS [5]. These rules culminated in the slogan: "Only annotate what is experimentally proven." According to this motto, a shorthand nomenclature for lipids was designed, where it is possible to simply infer the degree of annotation certainty by the nomenclature level used. In 2020, the shorthand notation for lipidomic data got a major overhaul, and now, e.g. also includes oxidized lipids and sphingolipids beyond ceramides and sphingomyelins [3]. The whole shorthand nomenclature project was performed according to the lipid categories developed by LIPID MAPS [2]. In the direct legacy of the shorthand nomenclature project, the Lipidomics Standards Initiative (LSI) was established in 2018 by Gerhard Liebisch and Kim Ekroos together with an informal group of lipidomics scientists who care for the development of standards in lipidomics (Figure 1.1). In 2019, the LSI led to the foundation of the ILS, in which the LSI constitutes one of the most important interest groups. Besides LSI, ILS hosts seven additional interest groups (applied bioinformatics, clinical lipidomics, global networking, instrumental and methodology development, lipid function, lipid ontology, reference materials, and biological reference ranges) and coordinates their activities. Some of the aforementioned interest groups and their activities will serve as a structure template for this chapter. Other community-wide standardization endeavors of the past decade worth mentioning are ring trials. Between 2014 and 2017, a ring trial organized by John A. Bowden at the National Institute of Standards and Technology (NIST) occurred [6]. The aim of this ring trial was limited to an interlaboratory lipidomics precision comparison on NIST Standard Reference Material (SRM)-1950, a reference plasma collected by NIST, because the true quantitative values of lipids in this biological material were unknown, and thus, it was impossible to determine the accuracy of the experimentally determined values. Furthermore, several community-wide position papers recently clearly defined the necessity and demand for standardization in lipidomics, including further steps to be taken toward achieving this goal [7-9].

Figure 1.1 The Lipidomics Standards Initiative (LSI) and its various fields of action within the lipidomics workflow, ranging from sample collection to data analysis.

1.3 Sampling and Preanalytics

"Without a community-wide consensus on best practices for the prevention of lipid degradation and transformations through sample collection and analysis, it is difficult to assess the quality of lipidomics data and hence trust results" [10]. Keeping this quote in mind, monitoring and documentation of the sampling step in the lipidomics workflow are of utmost importance because whatever is lost at sampling cannot be regained even by the most sophisticated analysis methods. Because of its importance in the workflow for lipidomics analysis, the LSI dedicates a separate chapter on this topic in its lipidomics guidelines (manuscript in preparation). Although stability is not as critical as when, e.g. handling RNA, there are nevertheless two big stability issues to be specifically considered when working with lipids: hydrolysis and oxidation [10, 11]. While hydrolysis affects esterified fatty acids, lipid peroxidation can occur at the methylene groups spacing two adjacent double bonds, e.g. C11 in linoleic acid. Both mechanisms may result in extensive fragmentation, truncation, and modification of lipids [12]. In contrast to lipid peroxidation, which is, in the context of sample stability, primarily a nonenzymatic chemical reaction, the threat of lipid hydrolysis also arises from enzymatic reactions catalyzed by lipases in the sample matrix. Thus, the most important measure to be taken against sample degradation is a short storage time and keeping the samples at as low temperatures as possible if storage of samples is needed. Sample workup immediately after collection is recommended because this would at least eliminate any enzymatic degradation, or, if this is not possible, the...