ISC Session on Omics Highlights Metabolomics Methods Can Be Prone to Errors

The International Symposium on Chromatography (ISC) 2024 in Liverpool kicked off with a session on Omics chaired by Bogusław Buszewski from Nicolaus Copernicus University, Poland and John Langley from the University of Southampton, United Kingdom.

© Kadmy- stock.adobe.com

Gérard Hopfgartner from the University of Geneva, Switzerland started the session with a talk entitled “Liquid and Supercritical Chromatography Coupled to Multimodal Tandem Mass Spectrometry in Metabolomics." Hopfgartner stated that metabolomics involved studying a wide range of analytes, including polar metabolites such as acids, amino acids, nucleotides, sugars, and lipids, which differed in their chemical space in terms of molecular weight, polarity, pKa, and concentration dynamic range. The coupling of liquid chromatography with high-resolution atmospheric ionization tandem mass spectrometry (LC–MS/MS) is useful due to its orthogonal selectivity. However, a generic LC–MS setup is inadequate for addressing the extensive chemical space of the analytes and their isomers, he said. Hopfgartner highlighted that multimodal mass spectrometry involving liquid or supercritical fluid chromatography was useful for analysing metabolites and lipids and described an example involving the metabolomic analysis of plasma and urine samples using LC–MS and SFC–MS, on samples from drivers who tested positive for drug abuse. His second example described the structural characterization of lipids using a robotic offline multidimensional chromatography combining a hydrophilic interaction liquid chromatography (HILIC) stationary phase with a C18 column, followed by collision-induced dissociation (CID) and electron activation dissociation (EAD) tandem MS fragmentation detection.

Juraj Lenčo from Charles University, Czech Republic then gave a talk entitled “Mitigating Artifact Formation in High-Temperature LC-MS for Bottom-Up Proteomics and Quality Control of Protein Biopharmaceuticals.” Increasing column temperature enhances peptide separation in LC–MS analyses can also lead to harmful artifacts that compromise data quality, according to Lenčo. Common issues include pyroGlu formation from N-terminal Glu or Gln, Asp dehydration, and peptide bond hydrolysis at Asp, along with temperature dependence for Met oxidation, he said. Lenčo’s team developed a high-temperature LC–MS method that retains the benefits of elevated temperatures while reducing peptide modifications. This involved a low-temperature inline trap column positioned before the heated separation column. The trap column's dimensions and retention were optimized to reduce peptide residence time without sacrificing separation quality, using a mm specialized pre-column. Lenco said that this method improved peak capacity by 1.4 times and identified 10% more peptides in bottom-up proteomic studies, while minimizing modifications. This is critical for the quality control of protein biopharmaceuticals because artifacts can falsely elevate modification levels at key sites and the method, he devised offers a valuable approach to improve peptide separation and reduce unwanted modifications, he concluded.

Steven Wilson from the University of Oslo, then presented on “Liquid and Supercritical Chromatography Coupled to Multimodal Tandem Mass Spectrometry in Metabolomics.” Wilson highlighted that as well as the diverse array of analytes that need to be investigated in omics applications, biosamples present unique challenges, such as clogging and ion suppression, and require careful sample preparation. He shared his insights on various column formats and preparation methods for omics applications he has used, including open tubular columns, monoliths, packed nano liquid chromatography (nano LC) columns, and pillar arrays. The examples he described included sterols linked to breast cancer and non-alcoholic fatty liver diseases; proteomics related to cancer stem cells and glioblastoma; and pancreatic hormones from islet organoids and organ-on-chip systems. Wilson stressed there is a need for ongoing development of column formats to meet the evolving demands for sensitivity and throughput in omics and clinical applications.

This was followed by a talk entitled “Optimized UHPSFC-HRMS/MS Workflow for Molecular Networking to Explore the Specialized Metabolism of Trichoderma reesei” by Ilias El Ouar from IFPEN, France. Filamentous fungi, such as Trichoderma reesei, are subjected to various biotic and abiotic stresses throughout their lifecycle. Environmental factors stimulate the production of specialized metabolites crucial for survival. El Ouar described a study focusing on the T. reesei hyperproducer cellulase strain RUT-C30, using an ethyl acetate/butanol extract followed by an untargeted metabolomic analysis based on ultra-high-performance supercritical fluid chromatography (UHP-SFC) coupled with tandem high-resolution mass spectrometry (HRMS/MS). This method was highly effective for metabolomic analyses, providing orthogonal separation selectivity, high kinetic performance, low pressure drops, and reduced organic solvent consumption, according to El Ouar. The method incorporated open software MetGem to chemically profile the sample and highlighted numerous molecular families, andled to the discovery of previously undescribed compounds, such as potential coumarin derivatives.

Georgios Theodoridis from Aristotle University of Thessaloniki, Greece concluded this session with a talk entitled “Analyte Identification in Untargeted Analysis. Chromatography Matters.” Theodoris began by emphasizing that analyte identification presents a significant challenge and bottleneck in contemporary untargeted analysis. Annotation or identification in metabolomics and non-target screening is often prone to errors and requires considerable time and effort, knowledgeable researchers, and robust IT tools, according to Theodoris. These conditions are not always present, he said, and leads to critical errors, particularly when using LC–MS. Accurate identification is crucial because incorrect identities compromised any subsequent biochemical interpretation, pathway analysis, and biomarker discovery. Theodoris highlighted that concerns regarding the validity of proposed identities has increased recently, and he provided examples of two common types of errors: identities with poor biological plausibility and identities that did not comply with chromatographic data. By focusing on the latter, characteristic cases were presented. Theodoris concluded by proposing measures aimed at improving the situation and emphasized the need for heightened awareness and efficient use of chromatographic data.