A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to Data Processing

Abstract

Untargeted metabolomics has rapidly become a profiling method of choice in many areas of research, including mitochondrial biology. Most commonly, untargeted metabolomics is performed with liquid chromatography/mass spectrometry because it enables measurement of a relatively wide range of physiochemically diverse molecules. Specifically, to assess energy pathways that are associated with mitochondrial metabolism, hydrophilic interaction liquid chromatography (HILIC) is often applied before analysis with a high-resolution accurate mass instrument. The workflow produces large, complex data files that are impractical to analyze manually. Here, we present a protocol to perform untargeted metabolomics on biofluids such as plasma, urine, and cerebral spinal fluid with a HILIC separation and an Orbitrap mass spectrometer. Our protocol describes each step of the analysis in detail, from preparation of solvents for chromatography to selecting parameters during data processing.

Keywords: Accurate mass; Data-dependent acquisition; HILIC; High-resolution; Liquid chromatography; Mass spectrometry; Metabolites; Metabolomics; Profiling; Quality assurance; Quality control.

Fig. 1

Steps implemented in the untargeted metabolomic workflow, from experimental design to data processing

Fig. 2

Input file table with assigned study factors. This table demonstrates how to implement a study factor within the Input File Characterization page. This example includes the user-defined study factor Phenotype where experimental samples are allocated as Lean or Fatty

Fig. 3

Defining sample groups and ratios for differential analysis. Using the ZDF rat experiment as an example, three sample groups are defined automatically by selecting Phenotype in the Study Variables check box. The comparison ratio is defined by selecting the control group to compare to

Fig. 4

Study page and analysis pane. After completing the wizard, a tab is generated consisting of the study pages (Study Definition, Input Files, Samples, and Analysis Results), the analysis pages (Grouping & Ratios, Workflows) and the Analysis pane. Within the Analysis pane, users can assign the Result File name

Fig. 5

A processing workflow tree. Nodes connected by edges indicate the selected functions for data processing. The workflow represented here is for the workflow template titled “Untargeted Metabolomics with Statistics Detect Unknowns with ID using Online Databases and mzLogic.” The workflow includes retention time alignment, unknown peak detection and ion association, gap filling, detection of background components unrelated to experimental samples, prediction of elemental composition, ChemSpider database searching, mzCloud spectral library matching, pathway mapping, and statistical analysis

Fig. 6

Results page default layout. The Main Table of the Compounds tab includes information for each of the detected compounds including annotation, elemental composition, molecular weight, peak areas, results from database searching and library matching, and univariate statistical results. Selecting a specific compound returns the associated chromatogram overlay and mass spectra

Fig. 7

Icons in the application toolbar supporting data review. When reviewing a Results File, data review icons are active. Selection of an icon brings the view to the front

Fig. 8

Recommended minimum peak intensity range for the Detect Compounds node of the Compound Discoverer software. Depending on the mass spectrometer used for data acquisition, corresponding peak intensity values should be selected for unknown peak detection