Data-Independent Acquisition Represents a Promising Alternative for Fast Photochemical Oxidation of Proteins (FPOP) Samples Analysis

Abstract

Fast Photochemical Oxidation of Proteins (FPOP) is a protein footprinting method utilizing hydroxyl radicals to provide valuable information on the solvent-accessible surface area. The extensive number of oxidative modifications that are created by FPOP is both advantageous, leading to great spatial resolution, and challenging, increasing the complexity of data processing. The precise localization of the modification together with the appropriate reproducibility is crucial to obtain relevant structural information. In this paper, we propose a novel approach combining validated spectral libraries together with utilizing DIA data. First, the DDA data searched by FragPipe are subsequently validated using Skyline software to form a spectral library. This library is then matched against the DIA data to filter out nonrepresentative IDs. In comparison with FPOP data processing using only a search engine followed by generally applied filtration steps, the manually validated spectral library offers higher confidence in identifications and increased spatial resolution. Furthermore, the reproducibility of quantification was compared for DIA, DDA, and MS-only acquisition modes on timsTOF SCP. Comparison of coefficients of variation (CV) showed that the DIA and MS acquisition modes exhibit significantly better reproducibility in quantification (CV medians 0.1233 and 0.1494, respectively) compared to the DDA mode (CV median 0.2104).

Figure 1

Fragmentation spectrum of Tyr7 oxidation in the peptide SCAVAEYGVYVK in its doubly charged form at m/z 681.32 and RT 16.70 min. The spectrum contains 19 out of 22 possible fragments, with y5 and y6 providing confidence for the modification site, representing optimal entry for the spectral library.

Figure 2

Fragmentation spectrum of Tyr7 oxidation in a peptide TEGDGVYTLNNEKQWINK in its triply charged form at m/z 709.01 and RT 12.22 min. Most of the assigned fragment ions have low intensity. Additionally, the doubly charged y12 fragment, which should support the localization of the modification, has an unmodified counterpart with the same intensity (labeled *). Analysis of raw data revealed at least two coeluted peptide species separated by ion mobility. The ⧫ symbol denotes the precursor. Because of the uncertainty of the modification site, together with the low intensity of the assigned fragment ions, this ID was excluded from the spectral library.

Figure 3

Distribution of the dot product score for DIA data matched against the manually validated spectral library.

Figure 4

Variability of quantification. Coefficients of variation were calculated from the quantification of identified modifications (five replicates) for each acquisition mode. The line represents the median. Created with box and whisker plot maker.

Figure 5

Coefficients of variation for the lowest-intensity tertile of identifications.