Untargeted Metabolomics Strategies—Challenges and

Authors: Alexandra C SchrimpeRutledge Simona G Codreanu Stacy D Sherrod John A McLean

Publish Date: 2016/09/13

Volume: 27, Issue: 12, Pages: 1897-1905

PDF Link

Abstract

Metabolites are building blocks of cellular function These species are involved in enzymecatalyzed chemical reactions and are essential for cellular function Upstream biological disruptions result in a series of metabolomic changes and as such the metabolome holds a wealth of information that is thought to be most predictive of phenotype Uncovering this knowledge is a work in progress The field of metabolomics is still maturing the community has leveraged proteomics experience when applicable and developed a range of sample preparation and instrument methodology along with myriad data processing and analysis approaches Research focuses have now shifted toward a fundamental understanding of the biology responsible for metabolomic changes There are several types of metabolomics experiments including both targeted and untargeted analyses While untargeted hypothesis generating workflows exhibit many valuable attributes challenges inherent to the approach remain This Critical Insight comments on these challenges focusing on the identification process of LCMSbased untargeted metabolomics studies—specifically in mammalian systems Biological interpretation of metabolomics data hinges on the ability to accurately identify metabolites The range of confidence associated with identifications that is often overlooked is reviewed and opportunities for advancing the metabolomics field are describedThe ultimate goal of metabolomics is the comprehensive study of the low molecular weight molecules within an organism Metabolites are the result of both biological and environmental factors and as such provide great potential to bridge knowledge of genotype and phenotype Metabolomics is often likened to its proteomics sibling and has leveraged proteomics experience but the field has evolved with inherently different challenges including the identification process Peptides and proteins are typically a linear polymer and can be sequenced Proteins are inferred by matching of identified experimental peptides against insilico fragmentation spectra Metabolites are more challenging to annotate These small molecules often lack a common building block although there is common use of the elements C H O N S P and potentially heteroatoms The idea that untargeted mass spectrometry MSbased metabolomics analysis will result in a large list of ‘identified’ small molecules that can be mapped to networks and pathways is often assumed yet high confidence analyte assignments/identifications may not be made owing to the fundamental challenges of the metabolomic identification processes For example features ie masstocharge ratio and retention time pairs can be assigned to a vast number of tentative or preliminary structures or there may be no candidate matches in curated databases Because metabolites lack a genetic template such as that for proteins metabolomics databases are currently considered incomplete Insilico metabolite databases can provide guidance but validation of retention times and MS/MS fragmentation data with a reference standard is nearly always required for confident metabolite identificationSince its inception the metabolomics field focus has shifted from detecting changes to understanding the biology leading to the changes 1 and thus the accuracy of metabolite assignments is extremely important In this Critical Insight we will discuss various challenges inherent to LCMSbased metabolomics and describe the ranges of confidence for small molecule annotations when performing global metabolomic analyses a concept essential for applying metabolomic data toward a better understanding of the mechanisms of human health and disease

Keywords: