Determination of selected reaction monitoring peptide transitions via multiplexed product‐ion scan modes

Abstract

Although in silico prediction of selected reaction monitoring (SRM) peptide transitions is the most commonly used approach in quantitative proteomics, systematically detectable peptide transitions selected from actual experimental data are desirable. Here, we demonstrated the use of two triple quadrupole mass spectrometry (QqQ-MS) operation modes to identify reliable SRM peptide transitions of target peptides selected from a shotgun proteomic linear ion-trap mass spectrometry (LIT-MS) profiling dataset. Transition ions (Q1 and Q3 ions) of target peptides were selected from the LIT MS/MS spectra. We performed multiplexed SRM blindly for the selected transition ions of target peptides using QqQ-MS and selected peptide transitions for which the chromatographically aligned and correlated ion intensities to the corresponding fragment ions appeared in the LIT MS/MS spectra. The identities of the peptides were further confirmed by MS/MS spectra acquired via SRM-triggered MS/MS on QqQ-MS. Despite the different MS platforms, we observed similar MS/MS patterns and relative ion abundance using both LIT-MS and QqQ-MS. Therefore, we were able to determine peptide transitions based on matching the chromatographic peak areas of all the selected Q3 ions of target peptides by the order of the corresponding ion intensities in the LIT MS/MS spectra. This approach demonstrated an efficient method to determine SRM peptide transitions, particularly when the target proteins are in low abundance and are therefore not easily detected by the QqQ full MS/MS scan mode. We employed this approach to determine the SRM peptide transitions of mitochondrial oxidative phosphorylation (OXPHOS) proteins involved in mitochondrial ATP synthesis. The multiplexed product-ion scan mode using QqQ-MS generates systematically detectable peptide transitions in a single liquid chromatography/MS run, in which we were able to identify SRM peptides that represent known target proteins in complex biological samples. The method presented here is easy to implement and has high-throughput capabilities as a result of the short analysis time. It is therefore well suited for the design of optimal SRM experiments.