Adapting Data-Independent Acquisition for Mass Spectrometry-Based Protein Site-Specific N‑Glycosylation Analysis

A hallmark of protein N-glycosylation is extensive heterogeneity associated with each glycosylation site. In human cells, the constituent glycoforms differ mostly in numerous ways of extensions from an invariable trimannosyl core and terminal modifications. The efficient identification of these glycoforms at the glycopeptide level by mass spectrometry (MS) requires a precursor sampling technique that is not dictated by signal intensity or by preset targets during MS2 data acquisition. We show here that the recently developed data-independent acquisition (DIA) approach is best suited to this demanding task. It allows postacquisition extraction of glycopeptide-specific fragment-ion chromatograms to be aligned with that of precursor MS1 ion by nanoLC elution time. For any target glycoprotein, judicious selection of the most favorable MS1/MS2 transitions can first be determined from prior analysis of a purified surrogate standard that carries similar site-specific glycosylation but may differ in its exact range of glycoforms. Since the MS2 transitions to be used for extracting DIA data is common to that glycosylation site and not dictated by a specific MS1 value, our workflow applies equally well to the identification of both targeted and unexpected glycoforms. Using a case example, we show that, in targeted mode, it identified more site-specific glycoforms than the more commonly used data-dependent acquisition method when the amount of the target glycoprotein was limited in a sample of high complexity. In discovery mode, it allows detection, with supporting MS2 evidence, of under-sampled glycoforms and of those that failed to be identified by searching against a predefined glycan library owing to unanticipated modifications.