Hydrogen–Deuterium Scrambling Based on Chemical Isotope Labeling Coupled with LC–MS: Application to Amine Metabolite Identification in Untargeted Metabolomics

Intensity-dependent mass search for improving metabolite database matches in chemical isotope labeling LC-QTOF-MS-based metabolomics

Liquid chromatography mass spectrometry (LC-MS) has been increasingly used for metabolome analysis. One of the critical steps in the LC-MS metabolome analysis workflow is related to metabolite identification. Among the measured parameters, peak mass is commonly used to search against a database for potential metabolite matches. Higher accuracy mass measurement allows the use of a narrower mass tolerance window for mass search. While various types of mass analyzers can routinely measure a peak mass with an error of less than a few ppm, mass measurement accuracy is not uniform for peaks with different intensities, particularly for quadrupole time-of-flight (QTOF) MS. Herein we present a simple and convenient method to determine the relation between peak intensity and mass error in LC-QTOF-MS-based metabolome analysis, followed by intensity-dependent mass search (IDMS) of a database for metabolite matches. This method is based on running a series of sodium formate mass calibrants, as part of the standard operating procedure (SOP) in LC-MS data acquisition, and then curve-fitting the measured mass errors and peak intensities. We show that, in two different quadrupole time-of-flight (QTOF) mass analyzers, mass accuracy is generally reduced as peak intensity decreases, which is independent of m/z values in the range commonly used for metabolite detection (e.g., m/z < 1000). We demonstrate the improvement in metabolite matches using IDMS in the analyses of dansyl labeled standards and human urine samples. We have implemented the IDMS method in the freely available MCID database at www.mycompoundid.org, which is composed of 8021 known human endogenous metabolites and their predicted metabolic products (375,809 compounds from one metabolic reaction and 10,583,901 compounds from two reactions).

CIL-ExPMRM: An Ultrasensitive Chemical Isotope Labeling Assisted Pseudo-MRM Platform to Accelerate Exposomic Suspect Screening

Exposome is the future of next-generation environmental health to establish the association between environmental exposure and diseases. However, due to low concentrations of exposure chemicals, exposome has been hampered by lacking an effective analytical platform to characterize its composition. In this study, by combining the benefit of chemical isotope labeling and pseudo-multiple reaction monitoring (CIL-pseudo-MRM), we have developed one highly sensitive and high-throughput platform (CIL-ExPMRM) by isotope labeling urinary exposure biomarkers. Dansyl chloride (DnsCl), N-methylphenylethylamine (MPEA), and their isotope-labeled forms were used to derivatize polar hydroxyl and carboxyl compounds, respectively. We have programmed a series of scripts to optimize MRM transition parameters, curate the MRM database (>70,000 compounds), predict accurate retention time (RT), and automize dynamic MRMs. This was followed by an automated MRM peak assignment, peak alignment, and statistical analysis. A computational pipeline was eventually incorporated into a user-friendly website interface, named CIL-ExPMRM (http://www.exposomemrm.com/). The performance of this platform has been validated with a relatively low false positive rate (10.7%) across instrumental platforms. CIL-ExPMRM has systematically overcome key bottlenecks of exposome studies to some extent and outperforms previous methods due to its independence of MS/MS availability, accurate RT prediction, and collision energy optimization, as well as the ultrasensitivity and automated robust intensity-based quantification. Overall, CIL-ExPMRM has great potential to advance the exposomic studies based on urinary biomarkers.

Enzyme-Catalyzed Hydrogen-Deuterium Exchange between Environmental Pollutants and Enzyme-Regulated Endogenous Metabolites

Environmental pollutants can disrupt the homeostasis of endogenous metabolites in organisms, leading to metabolic disorders and syndromes. However, it remains highly challenging to efficiently screen for critical biological molecules affected by environmental pollutants. Herein, we found that enzyme could catalyze hydrogen-deuterium (H-D) exchange between a deuterium-labeled environmental pollutant [D₃₈-bis(2-ethylhexyl) phthalate (D₃₈-DEHP)] and several groups of enzyme-regulated metabolites [cardiolipins (CLs), monolysocardiolipins (MLCLs), phospholipids (PLs), and lysophospholipids (LPLs)]. A high-throughput scanning identified the D-labeled endogenous metabolites in a simple enzyme [phospholipase A₂ (PLA₂)], enzyme mixtures (liver microsomes), and living organisms (zebrafish embryos) exposed to D₃₈-DEHP. Mass fragmentation and structural analyses showed that similar positions were D-labeled in the CLs, MLCLs, PLs, and LPLs, and this labeling was not attributable to natural metabolic transformations of D₃₈-DEHP or incorporation of its D-labeled side chains. Molecular docking and competitive binding analyses revealed that DEHP competed with D-labeled lipids for binding to the active site of PLA₂, and this process mediated H-D exchange. Moreover, competitive binding of DEHP against biotransformation enzymes could interfere with catabolic or anabolic lipid metabolism and thereby affect the concentrations of endogenous metabolites. Our findings provide a tool for discovering more molecular targets that complement the known toxic endpoints of metabolic disruptors.

Oxygen Isotope Exchange Reaction for Untargeted LC-MS Analysis

LC-MS is a key technique for the identification of small molecules in complex samples. Accurate mass, retention time, and fragmentation spectra from LC-MS experiments are compared to reference values for pure chemical standards. However, this information is often unavailable or insufficient, leading to an assignment to a list of candidates instead of a single hit; therefore, additional features are desired to filter candidates. One such promising feature is the number of specific functional groups of a molecule that can be counted via derivatization or isotope-exchange techniques. Hydrogen/deuterium exchange (HDX) is the most widespread implementation of isotope exchange for mass spectrometry, while oxygen ¹⁶O/¹⁸O exchange is not applied as frequently as HDX. Nevertheless, it is known that some functional groups may be selectively exchanged in ¹⁸O enriched media. Here, we propose an implementation of ¹⁶O/¹⁸O isotope exchange to highlight various functional groups. We evaluated the possibility of using the number of exchanged oxygen atoms as a descriptor to filter database candidates in untargeted LC-MS-based workflows. It was shown that ¹⁶O/¹⁸O exchange provides 62% (median, n = 45) search space reduction for a panel of drug molecules. Additionally, it was demonstrated that studying the fragmentation spectra after ¹⁶O/¹⁸O can aid in eliminating false positives and, in some cases, help to annotate fragments formed with water traces in the collisional cell.

Hydrogen/Deuterium Exchange Aiding Metabolite Identification in Single-Cell Nanospray High-Resolution Mass Spectrometry Analysis

The identification of metabolites in single-cell or small-volume tissue samples using single-cell mass spectrometry (MS) is challenging. In this study, hydrogen/deuterium (H/D) exchange was combined with microsampling nanospray high-resolution mass spectrometry (HRMS) to improve the efficiency and confidence level of metabolite identification in a single cell using commercial software. A nanospray ion source showed an improved reaction depth of 8% for H/D exchange compared with an electrospray ion source. In total, 273 metabolites were identified in Allium cepa L. single cells by searching commercial databases. Generally, more than one candidate is given for a precursor ion by MS or tandem MS (MS²) databases such as ChemSpider, MetDNA, MassBank, and mzCloud. With the help of the H/D exchange technique, the number of candidates decreased and reduction of the search space by a factor of 8 was achieved. In addition, two enzymolysis products of isoalliin, the transient intermediate and its isomer, were tracked at the single-cell level using the proposed method.

Metabolic Profiling of Urinary Chiral Amino-Containing Biomarkers for Gastric Cancer Using a Sensitive Chiral Chlorine-Labeled Probe by HPLC-MS/MS

Screening of characteristic biomarkers from chiral amino-containing metabolites in biological samples is difficult and important for the noninvasive diagnosis of gastric cancer (GC). Here, an enantiomeric pair of chlorine-labeled probes d-BPCl and l-BPCl was synthesized to selectively label d- and l-amino-containing metabolites in biological samples, respectively. Incorrect structural annotations were excluded according to the characteristic 3:1 abundance ratio of natural chlorine isotopes (³⁵Cl and ³⁷Cl) derived from the probes. A sensitive C18 HPLC-QQQ-MS/MS method in combination with the probes was then developed and applied in metabolomic analysis of amino-containing metabolites in urine samples. A total of 161 amino-containing metabolites were rapidly separated and determined, and 28 chiral amino acids and achiral glycine were quantified with good precision and accuracy. A total of 18 differential variables were discriminated by analyzing chiral amino-containing metabolites in urine samples of the GC patient and healthy person using the probe-based HPLC-MS/MS-MRM method combined with the orthogonal partial least squares discriminant analysis and Mann-Whitney U test with false discovery rate correction for multiple hypotheses. A diagnostic regression model including d-isoleucine, d-serine, and β-(pyrazol-1-yl)-l-alanine and age was then constructed with an average prediction correctness of 88.9% in the validation set. This work established a close connection between gastric cancer and chiral amino-containing metabolites. The mass spectrometry data analyzed in the study are publicly available via Mendeley Data (DOI: 10.17632/4bd93j9yrr.1).

Quantitative analysis of the relationship of derivatization reagents and detection sensitivity of electrospray ionization-triple quadrupole tandem mass spectrometry: Hydrazines as prototypes

Chemical derivatization-assisted electrospray ionization-triple quadrupole mass spectrometry (ESI-QqQ-MS) has become an efficient tool for the quantification of low-molecular-weight molecules. Many studies found that the derivatives of the same analytes derivatized by different derivatization reagents with the same reaction group had different detection sensitivity, even under the same conditions of electrospray ionization-mass spectrometry (ESI-MS). This phenomenon was suggested to be caused by the different modifying groups in the derivatization reagents. However, there is still a lack of systematic study on how modifying groups in the derivatization reagents affect the detection sensitivity of their corresponding derivatives of analytes, especially theoretical investigations. In this study, we employed a quantitative structure-activity relationship (QSAR) modeling approach to explore the relationship between modifying group structures and the detection sensitivity of derivatization reagents and their derivatives during ESI-MS detection. A total of 110 derivatization reagents of the hydrazine family and their hexanal derivatives (substituted hydrazones) were selected as the prototypes to construct QSAR models. The established models suggested that several molecular descriptors, related to hydrophobicity, electronegativity, and molecular shape, were related to the detection sensitivity of hexanal derivatives induced by different modifying groups in the derivatization reagents. Besides, we found that the detection sensitivity of compounds detected in selected ion mode (SIM) showed a positive correlation with that obtained in multiple reaction monitoring mode (MRM), and the ionization efficiency was the key factor on the detection sensitivity in both modes.

Normalization of Samples of Limited Amounts in Quantitative Metabolomics Using Liquid Chromatography Fluorescence Detection with Dansyl Labeling of Metabolites

Quantitative metabolomics requires the analysis of the same or a very similar amount of samples in order to accurately determine the concentration differences of individual metabolites in comparative samples. Ideally, the total amount or concentration of metabolites in each sample is measured to normalize all the analyzed samples. In this work, we describe a very sensitive method to measure a subclass of metabolites as a surrogate quantifier for normalization of samples with limited amounts. This method starts with low-volume dansyl labeling of all metabolites containing a primary/secondary amine or phenol group in a sample to produce a final solution of 21 μL. The dansyl-labeled metabolites generate fluorescence signals at 520 nm with photoexcitation at 250 nm. To remove the interference of dansyl hydroxyl products (Dns-OH) formed from the labeling reagents used, a fast-gradient liquid chromatography separation is used to elute Dns-OH using aqueous solution, followed by organic solvent elution to produce a chromatographic peak of labeled metabolites, giving a measurement throughput of 6 min per sample. The integrated fluorescence signals of the peak are found to be related to the injection amount of the dansyl-labeled metabolites. A calibration curve using mixtures of dansyl-labeled amino acids is used to determine the total concentration of labeled metabolites in a sample. This concentration is used for normalization of samples in the range from 2 to 120 μM in 21 μL with only 1 μL consumed for fluorescence quantification (i.e., 2-120 pmol). We demonstrate the application of this sensitive sample normalization method in comparative metabolome analysis of human cancer cells, MCF-7 cells, treated with and without resveratrol, using a starting material of as low as 500 cells.

A novel ligand with –NH2 and –COOH-decorated Co/Fe-based oxide for an efficient overall water splitting: dual modulation roles of active sites and local electronic structure

Excellent electrochemical water splitting with remarkable durability can provide a solution to satisfy the increasing global energy demand in which the electrode materials play an important role.