Top-Down Detection of Oxidative Protein Footprinting by Collision-Induced Dissociation, Electron-Transfer Dissociation, and Electron-Capture Dissociation

Abstract

Fast photochemical oxidation of proteins (FPOP) footprinting is a structural mass spectrometry method that maps proteins by fast and irreversible chemical reactions. The position of oxidative modification reflects solvent accessibility and site reactivity and thus provides information about protein conformation, structural dynamics, and interactions. Bottom-up mass spectrometry is an established standard method to analyze FPOP samples. In the bottom-up approach, all forms of the protein are digested together by a protease of choice, which results in a mixture of peptides from various subpopulations of proteins with varying degrees of photochemical oxidation. Here, we investigate the possibility to analyze a specifically selected population of only singly oxidized proteins. This requires utilization of more specific top-down mass spectrometry approaches. The key element of any top-down experiment is the selection of a suitable method of ion isolation, excitation, and fragmentation. Here, we employ and compare collision-induced dissociation, electron-transfer dissociation, and electron-capture dissociation combined with multi-continuous accumulation of selected ions. A singly oxidized subpopulation of FPOP-labeled ubiquitin was used to optimize the method. The top-down approach in FPOP is limited to smaller proteins, but its usefulness was demonstrated by using it to visualize structural changes induced by co-factor removal from the holo/apo myoglobin system. The top-down data were compared with the literature and with the bottom-up data set obtained on the same samples. The top-down results were found to be in good agreement, which indicates that monitoring a singly oxidized FPOP ion population by the top-down approach is a functional workflow for oxidative protein footprinting.

Figure 1

Mass spectrum of non-oxidized (a) and oxidized (b) ubiquitin. The inset shows the mass spectrum zoomed on the 10+ charge state. Presence of additional peaks in spectrum (b) indicates the presence of singly, doubly, and triply oxidized ubiquitin in the sample subjected to FPOP.

Figure 2

CID MS/MS spectra of FPOP-labeled ubiquitin and calculated extent of oxidation. (a) Left panel: full MS/MS spectrum obtained by fragmenting the 10+ charge state of singly oxidized ubiquitin (m/z 858.6). The blue diamond marks the precursor selected for CID; Top right panel: MS/MS spectrum zoomed on the m/z range 250–415; Bottom right panels: spectra of y24 4+ and y58 7+ product ions (b) extent of oxidation in detected b (left) and y (center) ions. Error bars show the standard deviations of three independent FPOP-treated samples. The right panel shows the extent of oxidation based on internal fragment ions.

Figure 3

ECD MS/MS spectra of FPOP-labeled ubiquitin and calculated extent of oxidation. (a) Left panel: full MS/MS spectrum obtained by fragmenting the 10 + charge state of singly oxidized ubiquitin (m/z 858.6). The blue diamond marks the precursor selected for ECD; top right panel: MS/MS spectrum zoomed on the m/z range 200–650. (b) Extent of oxidation in detected c (left) and z (right) ions. Error bars show the standard deviations of three independent FPOP-treated samples.

Figure 4

Comparison of the extent of oxidation for y ions obtained at four charge states (9+, 10+, 11+, and 12+) of singly oxidized ubiquitin by CID fragmentation. Red bars show the oxidized residues and their extent of oxidation in percent. Error bars show the standard deviations of three independent FPOP samples.

Figure 5

Interpretation of FPOP experimental results for the structural analysis of heme removal from holomyoglobin (a,b) Extents of peptide oxidation calculated from multi-CASI/CID (a) and multi-CASI/ECD (b) experiments for apomyoglobin (gold) and holomyoglobin (blue), respectively. Precursors of all charged states from 15+ to 19+ were fragmented together. Error bars indicate the standard deviations of three independent FPOP samples. CID results are shown by plotting the extent of oxidation for ion series b a y, and ECD results are shown by plotting the extent of oxidation for ion series c and z. (c) Myoglobin sequence alongside with the indication of increased levels of oxidation in CID (b and y fragments) and ECD (c and z fragments). A full line indicates b ions for CID or c ions for ECD, and a dotted line indicates y ions for CID or z ions for ECD. The line above the sequence shows sequence segments that were more oxidized according to the CID results, whereas the line below the sequence shows sequence segments that were more oxidized according to the ECD results. Lines in gold show sequence segments more oxidized in apomyoglobin, whereas lines in blue depict those in holomyoglobin. The amino acid sites that were found impacted in the bottom-up control experiment are indicated in the sequence by using the same color coding—gold letters show increased oxidation detected in bottom-up in apomyoglobin and blue letters show increased oxidation detected by the bottom-up approach in holomyoglobin. The schematic representation of the helixes shows which domains have increased extents of oxidation in apo- or holomyoglobin. The two-dimensional helix domains are labeled A–H, which can be directly related to the three-dimensional representation in panels 5D and 5E. (d,e) Crystal structures of native holomyoglobin (1WLA). The regions most affected by heme presence or removal as detected by the FPOP top-down approach are highlighted on the main backbone. Gold indicates more oxidized in apomyoglobin, whereas blue indicates the same in holomyoglobin. The amino acids detected by the bottom-up workflow as oxidized are indicated using the same color coding as that of the side chains. Front orientation (d) and back orientation (e) are shown.