Lipids are ubiquitous. Their hydrophobicity makes 3them “sticky” and difficult to analyze, which hinders our ability to characterize the totality of lipids (lipidome) in any organism. The team led by Makoto Arita in RIKEN Center for Medical Sciences characterize the lipidomes of living organisms by developing mass spectrometry (MS) techniques that differentiate lipids. Little were realized how diverse lipid structures would be. Initially, they ex-pected the number in the thousands. However, they were extremely diverse ; there exist more than 45,000 different lipids. This diversity facilitates the creation of complex biological systems, in which lipids act as key components of cellular membranes, signaling molecules, energy-storage molecules, and substrates.In lipidomics (and metabolomics) analysis, liquid chro-matography tandem mass spectrometry (MS/MS) and ion mobility spectrometry (IMS) are the gold-standard techniques; they provide thousands of molecular ions per specimen. A major challenge has been to develop a compu-tational platform that 1) does not target specific molecules but measures all relevant molecules (i.e., is untargeted) and 2) integrates the obtained raw MS data to accurately deter-mine molecular structures. Such a platform would allow us to semi-quantitatively elucidate complex biological systems from their lipidome (and metabolome) profiles.They recently established such a platform, packaged in Mass Spectrometry-Data Independent AnaLysis software version 4 (MS-DIAL 4; http://prime.psc.riken.jp/), which untangles lipid mass spectral fragmentations to create an atlas of lipids with an estimated false discovery rate of 1% to 2%. They tested it by analyzing 1,056 biological samples from various sources (human blood, mouse tissues, other mammalian cells, microbiota, algae, and plants) and de-rived a catalog of 8,051 different lipid species. They used those experimental data to create a comprehensive data-base of retention time and collision cross section based on machine learning, and MS/MS for 581,047 lipid structures, publishing the product as a “lipidome atlas”.MS-DIAL 4 appropriately represents the structures of 117 lipid subclasses–twice that of previous versions–based on fragment evidence for annotations at the levels of lipid class, molecular species, and acyl chain positional isomers. Importantly, MS-DIAL 4 complies with the lipidomics standards initiative: its classifications follow standardized definitions, structures are represented by an international shorthand notation system, and lipidomics results can be exported in a common data format. Thus, their platform enhances standardized lipidomics procedures and harmo-nizes data across laboratories, and is poised to accelerate lipidomics research.Dysregulated lipid metabolism is associated with dis-ease (e.g., obesity, atherosclerosis, diabetes). The ability to accurately distinguish lipid species will transform our understanding of lipid functions and facilitate discovery of bioactive lipids with therapeutic benefits. In the near future, visualization technologies, such as imaging MS, will revolutionize our understanding of the relationships among lipids.By integrating MS-DIAL 4 with other omics and imag-ing MS data, it is expected to uncover previously unknown lipid pathways and accelerate the development of biomark-ers, drugs, and clinical applications.Figure: An untargeted lipidomics platform to elucidate lipid diversity in organisms. We developed an untargeted lipidomics platform that analyzes and integrates mass spectrometry (MS) data from various techniques to ac-curately differentiate lipids. The platform, which builds upon standardized lipidomics procedures, harmonizes data obtained from different laboratories. The results of a recent study show a diversity of lipid structures that is far greater than previously realized.Original paper:Tsugawa H*, Ikeda K, Takahashi M, Satoh A, Mori Y, Uchino H, Okahashi N, Yamada Y, Tada I, Bonini P, Higashi Y, Okazaki Y, Zhou Z, Zhu Z-J, Koelmel J, Cajka T, Fiehn O, Saito K, Arita M & Arita M*. A lipidome atlas in MS-DIAL 4. Nature Biotechnology 38, 1159-1163 (2020)Makoto AritaA quantum leap for lipidomics research
元のページ ../index.html#9