A Metabolomics-Inspired Strategy for the Identification of Protein Covalent Modifications.

João Nunes,Alexandra M M Antunes,Judit Morello,Catarina Charneira,Sofia Gouveia-Fernandes,Carolina Nunes,Jacinta Serpa

doi:10.3389/fchem.2019.00532

Abstract

Identification of protein covalent modifications (adducts) is a challenging task mainly due to the lack of data processing approaches for adductomics studies. Despite the huge technological advances in mass spectrometry (MS) instrumentation and bioinformatics tools for proteomics studies, these methodologies have very limited success on the identification of low abundant protein adducts. Herein we report a novel strategy inspired on the metabolomics workflows for the identification of covalently-modified peptides that consists on LC-MS data preprocessing followed by statistical analysis. The usefulness of this strategy was evaluated using experimental LC-MS data of histones isolated from HepG2 and THLE2 cells exposed to the chemical carcinogen glycidamide. LC-MS data was preprocessed using the open-source software MZmine and potential adducts were selected based on the m/z increments corresponding to glycidamide incorporation. Then, statistical analysis was applied to reveal the potential adducts as those ions are differently present in cells exposed and not exposed to glycidamide. The results were compared with the ones obtained upon the standard proteomics methodology, which relies on producing comprehensive MS/MS data by data dependent acquisition and analysis with proteomics data search engines. Our novel strategy was able to differentiate HepG2 and THLE2 and to identify adducts that were not detected by the standard methodology of adductomics. Thus, this metabolomics driven approach in adductomics will not only open new opportunities for the identification of protein epigenetic modifications, but also adducts formed by endogenous and exogenous exposure to chemical agents.

Highlights

Protein covalent adducts, which can either result from exposure to endogenous or exogenous chemical electrophiles or be enzymatically driven, have a key role at the onset of multiple health issues, including cancer and immune effects (Nunes et al, 2016; Gonzalez-Morena et al, 2017)
The major trend in such investigations is to adopt the mass spectrometry (MS)-based shotgun proteomics workflows that traditionally rely on the chromatographic separation of digested peptides followed by a data dependent analysis (DDA), where MS and MS/MS data of selected precursors are afforded in a single run, thereby allowing subsequent adduct identification using database search engines that compare experimental and theoretical MS/MS spectra
Metabolomics-Inspired Approach: LC-MS Data Preprocessing Followed by Statistical Analysis

Summary

Introduction

Protein covalent adducts, which can either result from exposure to endogenous or exogenous chemical electrophiles (i.e., non-enzymatic) or be enzymatically driven (i.e., post translational modifications- PTMs), have a key role at the onset of multiple health issues, including cancer and immune effects (Nunes et al, 2016; Gonzalez-Morena et al, 2017). The major trend in such investigations is to adopt the MS-based shotgun proteomics workflows that traditionally rely on the chromatographic separation of digested peptides followed by a data dependent analysis (DDA), where MS and MS/MS data of selected precursors are afforded in a single run, thereby allowing subsequent adduct identification using database search engines that compare experimental and theoretical MS/MS spectra (reviewed by Gan et al, 2016; Tailor et al, 2016; Sabbioni and Turesky, 2017) Despite this workflow has been successfully applied to adductomics studies for the identification of high-abundant covalent adducts (reviewed by Nunes et al, 2019), it is easy to understand the failure of this strategy in the identification of low-abundant adducted peptides in vivo and ex vivo. While this approach presents major advantages for proteomics studies (reviewed by Vidova and Spacil, 2017), its applicability to adductomics studies is still limited and focused mostly in targeted-peptide site-specific modifications (Bruderer et al, 2015; Porter and Bereman, 2015; Carlsson et al, 2017)

Methods

Results

Conclusion