MSACL 2019 EU Abstract

LC- and CE-MS-based Workflow for Metabolic Read-Across of New Toxicants in Neuroinflammation

Víctor González-Ruiz (1,2), Domitille Schvartz (2,3), David Pamies (2,4), Marie-Gabrielle Zurich (2,4), Julien Boccard (1,2), Jean-Charles Sanchez (2,3), Serge Rudaz (1,2)
(1) Analytical Sciences, School of Pharmaceutical Sciences, Universities of Geneva and Lausanne, Switzerland (2) Swiss Centre for Applied Human Toxicology, Switzerland (3) Department of Physiology, University of Lausanne, Lausanne, Switzerland

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Víctor González-Ruiz (Presenter) University of Geneva

Presenter Bio: Víctor González-Ruiz got a M.Sc. in Pharmacy, a M.S. in Pharmaceutical R&D (2009) and a Ph.D. in Analytical Chemistry (2014) from the Complutense University of Madrid. During his Ph.D., he focused on the development of LC- and fluorescence-based analytical techniques to study antitumor drugs acting as interfacial inhibitors and to understand their mechanism of action.
In 2015 he moved to Switzerland where he joined the Analytical Sciences Laboratory in Geneva, participating in the development of new CE-MS interfaces. In 2018 he became a research associate. His research current interests include CE- and LC-MS for metabolomics, data treatment in large datasets, and quality assurance/quality control in metabolomic analyses. He is involved in several clinical and toxicological projects dealing with neuroinflammation, steroid homeostasis, cancer cell development, and microbial pathogenicity.

Relevant Financial Disclosures (within past 24 months) No relevant financial relationship(s) to disclose. Discussing Products? NO, presenter will NOT present educational content related to products/services of the ineligible company(ies) listed above. Conflict mitigation NOT REQUIRED.

Introduction: Risk assessment of new or existing chemicals finds a bottleneck in the evaluation of their potential as toxicants. Current approaches are too resource intensive in terms of time, money and animal use, thus limiting the number of substances which can be assayed. Chemical risk assessment using in vitro biological models such as human cell cultures allows to increase throughput while reducing cost and animal use. In such models, untargeted metabolomics can unveil triggered adverse outcome pathways (AOP) without the need for previous hypotheses.
Objectives: Our goal was to use a metabolomics strategy to highlight metabolic changes taking place in human astrocytes as a cost-effective model to study toxicant-induced neuroinflammation before the actual clinical effects become manifest. Since different substances will trigger this process through different AOPs, each one will provide a characteristic metabolic fingerprint. By comparison to the signature obtained for model toxicants, the neuroinflammatory potential of new chemical entities can be predicted without the need for animal testing.
Methods: Monolayer human astrocyte cultures derived from induced pluripotent stem cells (iPS) were exposed to different model neuroinflammatory substances at different doses. With the aim to enlarge the biochemical information recovered from the samples, a combination of three different liquid-chromatography methods and two capillary electrophoresis modes were coupled to high-resolution mass spectrometry detection. By using in-house developed software and reference databases, data pretreatment and metabolite identification were streamlined. Finally, multivariate analysis (MVA) allowed to cluster toxicants and doses according to the induced biological responses.
Results: MVA analysis successfully clustered the samples according to the indentity and concentration of the applied substances. The control group is noticeably separated from the treated ones. Interleukin 1β at low dose shows a characteristic profile, while low doses of other cytokines tend to have a comparable effect on metabolic patterns. Interestingly, all the cell samples show a similar behavior when exposed to high doses of the tested substances. Such a convergence phenomenon suggests that subtle metabolic differences in early-stage or mild neuroinflammation can progress towards a stronger and less specific metabolic shift. Pathway enrichment tools were used to evidence the nature of the metabolic alterations observed.
Conclusion: We developed a new workflow comprising a panel of analytical platforms, data treatment steps, and data interpretation allowing to characterize the metabolic changes induced by neuroinflammatory triggers on human astrocyte cultures. This approach paves the way to a mid-throughput chemical risk assessment strategy not relying on animal models and allowing to foresee the expected toxicological mechanism of a new substances depending on the metabolic pattern..