A Picture of Health?

A diverse group of metabolites, lipids are under tight homeostatic control and exhibit spatial and dynamic complexity at multiple levels. It is thus not surprising that altered lipid metabolism plays an important role in the pathogenesis of many common diseases (1).

My main research area is systems medicine. In particular, I explore metabolomics applications in biomedical research and related integrative bioinformatics in a range of conditions, including non-alcoholic fatty liver disease (NAFLD), metabolic co-morbidities in psychotic disorders, and immune-mediated inflammatory disorders, such as type 1 diabetes and celiac disease. We have found that lipid-related disturbances underlie many complex diseases and their co-morbidities.

For example, NAFLD is defined by accumulation of storage lipids in droplets in the liver. In studies of NAFLD, lipidomics not only uncovered biomarker candidates for diagnosing and monitoring the disease, but also revealed key underlying processes and disease heterogeneity. Specifically, lipidomics studies have shown that NAFLD, typically associated with insulin resistance and diabetes development, is also characterized by accumulation of triglycerides with low carbon number and double bond count, as well as increase of hepatic ceramides, which are known to cause insulin resistance (2). On the other hand and because of a common genetic variant in PNPLA3 gene, NAFLD associates only with storage of harmless dietary triglycerides, and not with insulin resistance. The NAFLD lipid signature found in liver can also be identified in the circulation, thus offering promise for new diagnostic tools for NAFLD (3)(4). Interestingly, studies by us and others have shown that the triglycerides associated with NAFLD are also associated with the risk of type 2 diabetes (5)(6).

Analysis in the balance

Lipids have high structural and functional diversity, so there are an enormous number of combinatorial possibilities. Specific specialized targeted methods are often required for specific classes of lipids, as these lipids may need specific types of sampling, extraction and analysis. Another level of complexity is introduced by the fact that even minor changes in lipid concentrations may have a large effect on cellular physiology. Therefore, accurate, quantitative measurement of lipids is important – which is, after all, the ultimate goal of lipidomic analysis.

Comprehensive “lipidomics” approaches usually refer to methods that cover the major lipid classes, including phospholipids (including major membrane lipids such as phosphatidylcholines and -ethanolamines), sphingolipids (sphingomyelins and ceramides) and neutral lipids (mono-, di- and triacylglycerols, cholesterol esters). Two main approaches have been adopted for these lipidomic analyses, based on direct infusion (shotgun approach) and on LC-MS (mainly reverse-phase LC, which is the most commonly applied approach for lipidomics). Each has its own advantages and disadvantages (7).

In the shotgun approach, because sample infusion to the mass spectrometer is at a constant concentration and the inevitable matrix effects are relatively constant, internal standard quantification may work well if the lipid profiles between the study groups do not vary too considerably. However, it comes at a cost: considerable and matrix-dependent ion suppression, as well as lower sensitivity compared to LC-MS. In LC-MS, lipids are separated prior to the introduction to the mass spectrometer, thus reducing the matrix effects and improving the sensitivity. However, because the internal standards do not elute at the same time as most of the lipids covered, the quantification may only work well for some of the compounds.

The aim of both approaches is to quantitate lipids as accurately as possible, which is aided by the increasing number of pure lipid standards becoming available. Nevertheless, strict quality control is essential (7). In our studies, about 20–25 percent of all samples analyzed are QC samples.

Over the years, much attention in lipid analytics has been devoted to somewhat contentious discussions about the superiority of one lipidomics technique over others. Confusing terminology such as “absolute quantification” has been introduced for some approaches, despite not being truly quantitative. As the field matures, and as we gain a better understanding of the advantages and limitations of different techniques, I hope we will see more attention paid to other important issues, such as having proper QC, lipid standards and harmonized ways of reporting lipidomic analysis.

Marching onwards

There has been a rapid increase in the volume of lipidomics publications over the last 15 years, partly reflecting better understanding of the importance of lipids in life sciences and medicine, and partly because of improvements in analytical tools for lipidomic analysis. Technological advances look set to continue in the near future, with better MS instrumentation, and new technical solutions for more detailed lipid identifications. LIPID MAPS (www.lipidmaps.org) has become an important resource for lipid analysis and has also contributed to the development of new pure lipid standards in collaboration with reagent companies. Lipid imaging, including in vivo imagine, is also developing fast.

Our group is currently engaged in studies of NAFLD, not only to develop better diagnostic and monitoring tools, such as in the LITMUS project, but also to study the role of fatty liver in the development of metabolic co-morbidities in other diseases. Specifically, we are interested in how metabolic co-morbidities develop in patients with psychotic disorders, a topic of considerable public health interest. Our findings indicated that fatty liver plays a crucial role (8), though this line of research will require further study of the lipidome across the gut–liver–brain axis in psychosis.

Matej Orešič is Visiting Senior Lecturer in the School of Medical Sciences, Örebro University, group leader in systems medicine at the Turku Centre for Biotechnology, University of Turku, Finland, and guest professor at the Oil Crops Research Institute of the Chinese Academy of Agricultural Sciences (Wuhan, Hubei, P. R. China). He also initiated the open source MZmine project, and is one of several experts working on LITMUS (Liver Investigation: Testing Marker Utility in Steatohepatitis).

MZmine– http://mzmine.github.io/

LITMUS– https://litmus-project.eu/