Optimization by infusion of multiple reaction monitoring transitions for sensitive quantification of peptides by liquid chromatography/mass spectrometry. (4th April 2017)

• Record Type: Journal Article
• Authors: Alghanem, Bandar; Nikitin, Frédéric; Stricker, Thomas; Duchoslav, Eva; Luban, Jeremy; Strambio‐De‐Castillia, Caterina; Muller, Markus; Lisacek, Frédérique; Varesio, Emmanuel; Hopfgartner, Gérard
• Is Part Of: Rapid communications in mass spectrometry. Volume 31:Number 9(2017)
• Page Start: 753
• Page End: 761
• Publication Date: 2017-04-04
• Subjects: Mass spectrometry -- Periodicals; 543.65
• DOI: 10.1002/rcm.7839
• Languages: English
• ISSNs: 0951-4198

Abstract

Abstract : Rationale: In peptide quantification by liquid chromatography/mass spectrometry (LC/MS), the optimization of multiple reaction monitoring (MRM) parameters is essential for sensitive detection. We have compared different approaches to build MRM assays, based either on flow injection analysis (FIA) of isotopically labelled peptides, or on the knowledge and the prediction of the best settings for MRM transitions and collision energies (CE). In this context, we introduce MRMOptimizer, an open‐source software tool that processes spectra and assists the user in selecting transitions in the FIA workflow. Methods: MS/MS spectral libraries with CE voltages from 10 to 70 V are automatically acquired in FIA mode for isotopically labelled peptides. Then MRMOptimizer determines the optimal MRM settings for each peptide. To assess the quantitative performance of our approach, 155 peptides, representing 84 proteins, were analysed by LC/MRM‐MS and the peak areas were compared between: (A) the MRMOptimizer‐based workflow, (B1) the SRMAtlas transitions set used 'as‐is'; (B2) the same SRMAtlas set with CE parameters optimized by Skyline. Results: 51% of the three most intense transitions per peptide were shown to be common to both A and B1/B2 methods, and displayed similar sensitivity and peak area distributions. The peak areas obtained with MRMOptimizer for transitions sharing either the precursor ion charge state or the fragment ions with the SRMAtlas set at unique transitions were increased 1.8‐ to 2.3‐fold. The gain in sensitivity using MRMOptimizer for transitions with different precursor ion charge state and fragment ions (8% of the total), reaches a ~ 11‐fold increase. Conclusions: Isotopically labelled peptides can be used to optimize MRM transitions more efficiently in FIA than by searching databases. The MRMOptimizer software is MS independent and enables the post‐acquisition selection of MRM parameters. Coefficients of variation for optimal CE values are lower than those obtained with the SRMAtlas approach (B2) and one additional peptide was detected. Copyright © 2017 John Wiley & Sons, Ltd.