Abstract Title: | Clinical Lipidomic Quantitation Based on Mass Spectrometry: Case Study of Pancreatic Cancer |
Abstract Type: | Seminar |
Session Choice: | Big Data Chemometrics and Method Development(In-Silico)(KVCV) |
Presenter Name: | Michal Holcapek |
Co-authors: | Eva Cifkova Denise Wolrab Robert Jirasko Tereza Hrnciarova Miroslav Lisa Ladislav Kuchar Roman Hrstka |
Company/Organisation: | University of Pardubice |
Country: | Czech Republic |
Abstract Information :
A wide range of lipid molecules is present in all eukaryotic cells. They fulfill numerous important physiological functions, and their dysregulation is often related to serious human diseases. The widely accepted classification of lipids according to Lipid MAPS is based on 8 main lipid categories, and each category has numerous classes and subclasses. Furthermore, large number of lipid species can be detected in biological samples for each subclass due to various fatty acyl carbon lengths, double bond number and positions. This complexity results in the fact that multiple analytical methods may be needed to cover the broader range of lipidome. Mass spectrometry (MS) and its coupling with the liquid-phase separation techniques is strongly dominating in the lipidomic quantitation. Essential steps in the lipidomic quantitation include the following: the careful optimization of analytical conditions, the selection of appropriate internal standards (IS) not occurring in studied biological samples (at least one IS per each lipid subclass), the full validation of final methods using the addition of internal standards into pooled samples, the use of quality control samples (typically pooled sample) to control possible variability among batches. Other important aspects are the (semi)automation of the workflow and high-throughput, but these issues should not affect the analytical robustness. In our analytical workflow, we use mainly the following three MS based methods for the high-throughput clinical analysis: 1/ direct infusion tandem MS using characteristic neutral loss (NL) and precursor ion (PI) scans on triple quadrupole type mass spectrometers (shotgun MS), 2/ ultrahigh-performance supercritical fluid chromatography - mass spectrometry (UHPSFC/MS), and 3/ matrix-assisted laser desorption/ionization (MALDI) coupled to high-resolution Orbitrap mass analyzer. Shotgun MS and UHPSFC/MS techniques are applied mainly for glycerophospholipids, sphingolipids, and glycerolipids using positive-ion electrospray ionization (ESI), while MALDI is used in the negative-ion mode to obtain complementary information on sulfatides and other anionic lipid subclasses. All mentioned methods follow the basic rule of reliable lipidomic quantitation that IS should be coionized with analytes from the same lipid subclass. Our laboratory-made software LipidQuant is used for the semi-automated data processing, and then the absolute or relative concentrations of all quantified lipids in individual samples are statistically evaluated using multivariate data analysis (MDA) methods, such as nonsupervised principal component analysis (PCA), supervised orthogonal partial least square discriminant analysis (OPLS-DA), S-plots, box plots, hierarchical clustering, neuron networks, etc. The application of our approach to the analysis of serum of pancreatic cancer patients and healthy volunteers will be shown, which will illustrate the potential for the future early diagnosis screening based on the lipidomic analysis.