Resource
Probing the Binding Interfaces of Protein Complexes
Using Gas-Phase H/D Exchange Mass Spectrometry
Graphical Abstract
Highlights
d
Gas-phase HDX-MS can probe the binding mode of
protein-ligand complexes
d
Side-chain sites involved in binding can be measured for both
receptor and ligand
d
Gas-phase HDX-MS results are in good agreement with
available crystal structures
d
This MS-based method can be useful in structural biology
and ligand design
Authors
Ulrik H. Mistarz, Jeffery M. Brown, Kim
F. Haselmann, Kasper D. Rand
Correspondence
[email protected]
In Brief
Mistarz et al. present the use of
gas-phase H/D exchange mass spectrometry
(gas-phase HDX-MS) to probe the
binding interface of protein-ligand
complexes, and measure the number of
heteroatom-bound non-amide side-chain
sites implicated in intermolecular
interactions in the complex. The novel
mass spectrometric approach is
demonstrated for both protein-glycan
and protein-peptide complexes.
Mistarz et al., 2016, Structure24, 310–318
February 2, 2016ª2016 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.str.2015.11.013
Structure
Resource
Probing the Binding
Interfaces of Protein Complexes
Using Gas-Phase H/D Exchange Mass Spectrometry
Ulrik H. Mistarz,1Jeffery M. Brown,2Kim F. Haselmann,3and Kasper D. Rand1,*
1Department of Pharmacy, University of Copenhagen, Universitetsparken 2, Copenhagen 2100, Denmark 2Waters MS Technologies Centre, Waters Corporation, Altrincham Road, Wilmslow SK9 4AX, UK 3Diabetes Protein Engineering, Novo Nordisk A/S, Novo Nordisk Park 1, Ma˚løv 2670, Denmark
*Correspondence:[email protected] http://dx.doi.org/10.1016/j.str.2015.11.013
SUMMARY
Fast
gas-phase
hydrogen/deuterium
exchange
mediated by ND
3gas and measured by mass
spec-trometry (gas-phase HDX-MS) is a largely
unhar-nessed, fast, and sensitive method for probing
primary- and higher-order polypeptide structure.
Labeling of heteroatom-bound non-amide hydrogens
in a sub-millisecond time span after electrospray
ioni-zation by ND
3gas can provide structural insights into
protein conformers present in solution. Here, we have
explored the use of gas-phase HDX-MS for probing
the higher-order structure and binding interfaces of
protein complexes originating from native solution
conditions. Lysozyme ions bound by an
oligosaccha-ride incorporated less deuterium than the unbound
ion. Similarly, trypsin ions showed reduced deuterium
uptake when bound by the peptide ligand
vaso-pressin. Our results are in good agreement with
crys-tal structures of the native protein complexes, and
illustrate that gas-phase HDX-MS can provide a
sen-sitive and simple approach to measure the number
of heteroatom-bound non-amide side-chain
hydro-gens involved in the binding interface of biologically
relevant protein complexes.
INTRODUCTION
The molecular interactions of proteins are intricately tied to their biological activity and function. Characterizing such interactions is therefore essential for understanding protein function at the molecular level and to inform the discovery and development of potent ligands. Nuclear magnetic resonance (NMR) spectros-copy (Wu¨thrich, 1986, 2001) and X-ray crystallography (Liljas and Rossmann, 1974; Shi, 2014) are key tools for structural anal-ysis of proteins and their molecular complexes. More sensitive and complementary methods, however, are needed to empower structural proteomics and enable a comprehensive character-ization of both protein-ligand complexes and multicomponent protein assemblies that are a key part of cell machinery. Electro-spray ionization (ESI) has enabled the direct and gentle transfer
of intact proteins and protein complexes from their native envi-ronment to the gas phase (Fenn et al., 1989). Furthermore, solu-tion-like conformational states of proteins can be retained in the gas phase for tens to several hundreds of milliseconds after ioni-zation (Badman et al., 2001; Breuker and McLafferty, 2008; Wyt-tenbach and Bowers, 2011). Mass spectrometry (MS) is thus emerging as an applicable tool for structural characterization of proteins and protein complexes in addition to mass assign-ment (Heck, 2008; Leurs et al., 2015; Rajabi et al., 2015). Inside the mass spectrometer, gaseous protein ions can be probed by a panel of gas-phase techniques (Benesch et al., 2007), for instance ion mobility spectrometry (Bohrer et al., 2008; Pritchard et al., 2012; Ruotolo et al., 2008; Salbo et al., 2012), that report on the collisional cross section of ions, and techniques that rely on chemical reactions in the gas phase such as proton-transfer re-actions (Gross et al., 1996; Loo et al., 1994), radical-based reac-tions (Breuker and McLafferty, 2003; Lermyte et al., 2014; Ly and Julian, 2010), and hydrogen/deuterium exchange (HDX). Gas-phase HDX has traditionally been performed by the infusion of a deuterated basic gas such as D2O, MeOD, or ND3into isolated parts of the mass spectrometer (Cheng and Fenselau, 1992; Heck et al., 1998; Hemling et al., 1994; Pan et al., 2012; Suckau et al., 1993; Winger et al., 1992). By this reaction, hydrogen atoms attached to oxygen, nitrogen, and sulfur (heteroatom-bound hydrogens) are replaced by deuterium. It has emerged that a key requisite to the use of gas-phase HDX to inform on so-lution-phase protein conformers is to complete the labeling reac-tion within a few tens of milliseconds after ionizareac-tion (Rand et al., 2009, 2012). Gas-phase HDX performed at such millisecond timescales using ND3gas shortly after ESI (referred to herein as gas-phase HDX-MS) allow the selective labeling of heteroat-om-bound non-amide hydrogens, and can detect and report on conformational differences between protein conformers perti-nent to the solution phase (Beeston et al., 2015; Mistarz et al., 2014; Rand et al., 2009, 2012). The hydrogen/deuterium ex-change rates of different sites in an unstructured peptide ion in the gas phase depend on the difference between the gas-phase basicity (GB) of the donor (ND3) and the acceptor site (Campbell et al., 1995; Cheng and Fenselau, 1992). A lowDGB causes fast exchange and as heteroatom-bound non-amide side-chain hydrogens, i.e. -OH, -NH, and -SH, have GBs relatively close to that of ND3, as opposed to amide NHs (Bouchoux, 2012; Hunter and Lias, 1998), efficient and selective labeling of these sites is achieved in the millisecond labeling times employed by
the gas-phase HDX-MS approach. In contrast to solution-phase HDX-MS, gas-phase HDX-MS at millisecond timescales ap-pears to report more directly on the surface accessibility and hydrogen bonding (and the availability of charged sites) with little contribution from the dynamics and flexibility of the structure, as gas-phase structural transitions occur at much lower rates due to the absence of solvent to lower the energy barrier for confor-mational rearrangements (Jarrold, 2000). Furthermore, by the gas-phase HDX-MS method, no labeling is lost during analysis as is the case for the classical solution-phase HDX-MS approach. The gas-phase HDX-MS approach allows analyses to be performed using only little material (low picomolar amounts) at low micromolar concentrations of sample, and re-quires minimal sample pretreatment (buffer exchange into an ESI-compatible buffer).
Here, we demonstrate the use of gas-phase HDX-MS to examine protein-ligand and peptide-ligand complexes, including lysozyme bound by either of two oligosaccharides (tetra- and hexa-N-acetylglucosamine), trypsin bound by the peptide ligand desGly9-vasopressin, as well as the peptide Leu-enkephalin (LE) bound by an 18-crown-6 ether (CE) ligand. By subsequent gas-phase dissociation of the complexes, it was possible to measure the labeling of individual partners in the complex and approximate the numbers of heteroatom-bound non-amide hydrogens participating in binding.
Differ-ences in deuterium uptake observed for unbound and com-plex-bound protein and ligand ions indicate that gas-phase HDX-MS can provide detailed insights into the binding interface of protein complexes present in solution.
RESULTS
Gas-Phase HDX-MS Setup
Gas-phase HDX-MS was performed in the ion source region of a commercially available hybrid QTOF mass spectrometer imme-diately downstream of the primary cone exit (sample cone) as described in more detail elsewhere (Mistarz et al., 2014). In short, an aqueous ND3/D2O reagent was added to an externally acces-sible reagent vial, conventionally used to store reagent for electron transfer dissociation (ETD) experiments. To control the flow of labeling gas, we set the mass spectrometer software in ‘‘ETD-mode’’ and applied a flow of N2 gas (‘‘make up’’ gas flow rate of 50 ml/min) across the aqueous ND3/D2O reagent contained in the reagent vial. The resulting ND3-saturated N2 gas was passed from the headspace of the reagent vial through a hollow needle immediately downstream of the primary skimmer orifice, thus enabling labeling of desolvated native-like protein and peptide complex ions immediately after ioniza-tion (Figure 1).
Investigating Peptide and 18-Crown-6 Ether Ion Complexes
As a first step, we sought to investigate the ability of our gas-phase HDX-MS method to probe the binding between a model polypeptide and a crown ether ligand. Crown ethers are cyclic polyethers that form stable structures with cations by binding to protonated primary amino groups, for instance in the N termi-nus of a peptide (Pedersen, 1967).
A mixture of Leu-enkephalin peptide (LE, aminoacid sequence: YGGFL, Figure 2) and 18-crown-6 ether (CE) was ionized from native solution conditions, and the resulting mass spectrum showed singly protonated LE both unbound and bound by CE (Figure 2A). In gas-phase HDX-MS experiments, unbound LE exchanged 5.9 D, while CE-bound LE [LE-CE + H]+exchanged 1.2 D (Figure 2B), showing a pronounced reduc-tion in HDX of the complex relative to the unbound peptide component. Thus while the unbound state of the singly proton-ated LE peptide exchanges all its heteroatom-bound non-amide hydrogens (N = 6) during the gas-phase HDX-MS experiment, the presence of additional intermolecular interactions limit the exchange of the same heteroatom-bound non-amide hydrogens in the complex-bound state of the peptide. No adducts of ND3or D2O were observed in the mass spectra. To investigate whether the labeling of the individual binding partners in such a non-co-valent ion complex could be dissected, we performed mild colli-sional activation (4 eV) of the LE-CE ion complex immediately after gas-phase HDX, which resulted in dissociation without backbone fragmentation. Adopting this strategy in combination with gas-phase HDX-MS, the deuterium uptake of both CE-bound and CE-dissociated Leu-Enk could be measured and compared in the same mass spectrum (Figures 2A and 2B). The dissociated LE ion contained 1.2 D, similar to the intact LE-CE complex. The methodology thus allowed gas-phase la-beling of an ion complex and subsequent dissection and
Figure 1. Overview of the Gas-Phase HDX-MS Setup
(A) Illustration of the externally accessible vial filled with ND3/D2O reagent and the flow of ND3-saturated N2gas.
(B) Zoom-in on the ion source region of the mass spectrometer, with illustration of the area where deuterium labeling is performed.
(C) Schematic of the QTOF mass spectrometer showing the regions of the instrument where ion selection and dissociation of ion complexes is performed following deuterium labeling.
determination of the deuterium uptake of the individual compo-nents. In the case of the studied peptide-crown ether complex, several heteroatom-bound non-amide hydrogens of the peptide were shielded from HDX upon complex formation with the crown ether, indicating extensive intermolecular interactions.
Probing the Binding Interface of Lysozyme-Oligosaccharide Ion Complexes
The protein lysozyme binds a variety of oligosaccharide sub-strates and catalyzes their site-specific hydrolysis. We next investigated the use of gas-phase HDX-MS to inform on the structure of such naturally occurring protein-oligosaccharide complexes. Complexes of chicken egg white lysozyme (Lys, 147 residues, 14.3 kDa) with tetra- or hexasaccharides of N-ace-tylglucosamine (NAG) (Figure 3A) were ionized from native solution conditions and subjected to gas-phase HDX-MS.
Non-covalent complexes of lysozyme with both NAG6and the cleavage products NAG1–5 (Ganem et al., 1991; Rupley and Gates, 1967) were present in the mass spectrum in charge states 7+ to 10+ (Figure 3B). Complexes of lysozyme with NAG6 [Lys-NAG6 + 8H]8+and NAG4 [Lys-NAG4 + 8H]8+ were chosen in this experiment for further examination, due to high ion abun-dance (Figure 3B) and a relatively low charge, and thus, an ex-pected high similarity to their native solution-phase counterparts due to minimal columbic repulsion in these gas-phase ions (Rand et al., 2009; Valentine et al., 1997). Gas-phase HDX-MS was performed in combination with ion selection in the
quadru-pole and gas-phase dissociation in the trap Traveling Wave (T-Wave; 20 eV), enabling analysis of both intact complexes, un-bound or dissociated singly protonated oligosaccharides of different length [NAGn* + H]+, and unbound or dissociated lyso-zyme ions [Lys* + 7H]7+(Figures 3C and 3D). The deuterium uptake of the intact complex, dissociated and unbound oligo-saccharides, and lysozyme is shown in Figure 3E. Gas-phase dissociation of the complex [Lys-NAG6 + 8H]8+ showed both dissociated and complex-bound NAG4 and NAG6 ions and, interestingly, also to a small extent NAG5(Figure 3C). Lysozyme is not likely to retain hydrolytic activity after ionization of the com-plex, as solvent is needed for hydrolysis. The detection of solution-phase cleavage product intermediates retained during ionization by intramolecular interactions indicates that our gas-phase experiments sample snapshots of native enzyme-substrate complexes occurring in solution. Non-covalent interactions keeping the complex intact are broken during the gas-phase collisional activation step after labeling, whereby the vibrational energy added dissociates the non-covalent com-plex before breaking any covalent bonds (Lee et al., 1998). In contrast, gas-phase activation of [Lys-NAG4 + 8H]8+ showed only dissociated and complex-bound NAG4 ions (Figure 3D). The absence of cleavage product intermediates in [Lys-NAG4+ 8H]8+could be due to the thousand-fold slower cleavage of lysozyme of NAG4than NAG6(Rupley and Gates, 1967).
Unbound lysozyme [Lys + 7H]7+ and the oligosaccharides [NAG6 + H]+ and [NAG4 + H]+ incorporated 21 D, 6 D, and 4 D, respectively (Figure 3E). Lysozyme has four intramolecular disulfide bonds that decrease flexibility and charge-mediated unfolding in the gas phase, thus stabilizing intramolecular bonds (Valentine et al., 1997). This could explain the relatively low deuterium uptake observed for [Lys + 7H]7+(21 D), which has 97 heteroatom-bound non-amide hydrogens. The deute-rium uptake for the singly protonated NAG6and NAG4 corre-sponds to approximately 1 D per saccharide monomer, likely at the hydroxyl group on either carbon 3 or carbon 6. As for polypeptides, the amide hydrogen of the saccharide monomer is not expected to exchange during our gas-phase HDX-MS ex-periments (Mistarz et al., 2014). The incomplete HDX of hetero-atom-bound non-amide hydrogens in the oligosaccharides NAG4 (4 out of 10 D) and NAG6(6 out of 14 D) indicates the presence of gas-phase structures of the unbound singly proton-ated NAG polymers with prevalent intramolecular interactions. The [Lys-NAG6 + 8H]8+ and [Lys-NAG4 + 8H]8+ complexes exchanged 20.5 D and 20 D, respectively (Figure 3E). Upon dissociation of [Lys-NAG6 + 8H]8+, the deuterium content of the dissociated lysozyme [Lys* + 7H]7+ and NAG6 [NAG6* + H]+ions was measured to 16.2 D and 4 D, respectively, whereas [Lys* + 7H]7+ and [NAG4* + H]+originating from [Lys-NAG4+ 8H]8+ contained 17.2 D and 2.5 D, respectively. As was the case for the model peptide-crown ether complexes (Figure 2), the gas-phase dissociation step effectively dissects the contri-bution of each binding partner to the overall HDX measured for the intact complex. Lysozyme showed a decrease in deuterium uptake upon complex formation with the oligosaccharides NAG6 and NAG4, of 4.8 D and 3.8 D, respectively. In turn, NAG6and NAG4exchanged 2 D and 1.5 D less upon complex formation with lysozyme, indicative of additional intermolecular interactions between lysozyme and NAG6in comparison with
Figure 2. Gas-Phase HDX-MS of the Leu-Enkephalin Peptide (LE) Bound by an 18-Crown-6 Ether (CE)
(A) Overlaid mass spectra of unlabeled LE either unbound or bound (black), unbound LE following labeling (blue) and bound LE following labeling and dissociation of the ion complex (red).
(B) Deuterium uptake of unbound LE [LE + H]+
, the LE-CE ion complex [LE-CE + H]+
and LE after complex dissociation [LE* + H]+
. The error bars shown represent 23 SD.
(C) Illustration of the primary structure of the LE peptide and the CE molecule coordinated to a protonated ammonium ion.
NAG4, which is consistent with crystal structures of both com-plexes (Blake et al., 1967; Von Dreele, 2005). In the crystal structure of lysozyme bound by NAG6, four residues of lyso-zyme form intermolecular interactions with NAG6through het-eroatom-bound non-amide hydrogens in their side chains (E35, S50, D52, and D101,Figure 3F). In good apparent agree-ment, we observed a decrease in HDX of lysozyme of 5 D upon binding to NAG6. The additional HDX reduction of 1 D that cannot be accounted for by directly inspecting the crystal struc-ture of the complex, could be due to the protection, upon complex formation, of additional one or two sites in lysozyme indirectly linked to the binding interface. The oligosaccharide binding interface of lysozyme can be divided into six subsites, A–F (Perkins et al., 1981). Hexasaccharides (i.e. NAG6) would occupy all six subsites and tetrasaccharides (i.e. NAG4) would occupy four subsites, primarily A–D (Rupley and Gates, 1967). E35 form intermolecular interactions with the saccharide in subunit sites E and F, whereas S50, D52, and D101 form inter-molecular interactions with subsites A–D. The additional approximately 1 D decrease in deuterium uptake for lysozyme in [Lys-NAG6+ 8H]8+relative to [Lys-NAG4+ 8H]8+could thus
Figure 3. Gas-Phase HDX-MS of Com-plexes of the Protein Lysozyme and, N-Ace-tylglucosamine Oligosaccharides NAG4and
NAG6
(A) The crystal structure of lysozyme bound to NAG6or NAG4(PDB: 1SFG, 1SF7).
(B) Mass spectrum of ion complexes of lysozyme and NAG oligosaccharides without gas-phase labeling.
(C and D) Mass spectra of quadrupole-selected and complex dissociated [Lys-NAG6+ 8H]8+(C) and [Lys-NAG4+ 8H]8+(D), respectively, with gas-phase labeling.
(E) Measured deuterium uptake of intact ion complexes as well as dissociated and unbound oligosaccharide and lysozyme ions. Solid bars indicate unbound or complex-bound ions, and pinstriped bars indicate complex dissociated ions. The error bars shown represent 23 SD. (F) Zoom-in on the binding interface of lysozyme and NAG6. Residues of lysozyme engaged in direct intermolecular interactions with NAG6 through heteroatom-bound non-amide hydrogens are highlighted.
See alsoFigure S1for comparison of gas-phase stability by collisional induced dissociation of the two lysozyme-oligosaccharide complexes.
nicely correlate with the inability of NAG4to form intermolecular interactions with E35 in subsites E and F of lysozyme. To probe the gas-phase stability and binding strength of examined com-plexes, we performed a set of experi-ments on [Lys-NAG4+ 8H]8+ and [Lys-NAG6 + 8H]8+ whereby the collisional activation was increased and the abun-dance of dissociated species was quan-tified (Figure S1). We observed that the NAG6complex with lysozyme was more stable than that of NAG4(Veros and Oldham, 2007), indicating a stronger network of intermolecular interactions, in support of the results obtained from gas-phase HDX-MS experiments.
Our results show that gas-phase HDX-MS can inform on intermolecular interactions in both small peptide- and larger pro-tein-ligand complexes. Subsequent gas-phase dissociation of non-covalent complexes enables a direct measure of the contri-bution of each binding partner to the binding interface by providing information on the number of residues with heteroat-om-bound non-amide hydrogens partaking directly, or indi-rectly, in binding. As intermolecular interactions between side chains typically provide a major enthalpic contribution to stabiliz-ing protein-protein/ligand complexes, deuterium labelstabiliz-ing of non-amide side-chain sites by gas-phase HDX-MS provides a quantitative view of the number of intermolecular interactions in a protein complex and, possibly, to a first approximation, the area of the binding interface. Additional studies are needed to investigate the latter aspect further. We note that such a sim-ple interpretation does assume that heteroatom-bound non-amide sites involved in direct intermolecular interactions are fully exchangeable in the unbound state, and furthermore that
changes in gas-phase HDX upon complex formation are not pri-marily due to conformational changes in either binding partner, resulting in new intramolecular hydrogen bond formation upon complex formation.
Probing the Binding Interface of Trypsin and Vasopressin Ion Complexes
Vasopressin (amino acid sequence: CYFQNCPRG) is a cyclic 9-mer peptide hormone, containing an intramolecular disulfide bridge between two cysteine residues. Besides antidiuretic hor-monal activity, vasopressin is also a serine protease inhibitor that can regulate the activity of proteolytic enzymes, like trypsin (Ibra-him and Pattabhi, 2005). Using gas-phase HDX-MS, we investi-gated the naturally occurring non-covalent complex of trypsin and vasopressin and compared measurements with available crystallographic data (Figure 4A). A mixture of trypsin and vasopressin was ionized from native solution conditions, and mass spectra showed both bound and unbound trypsin with 7–10 charges (Figure 4B). Intact vasopressin was not present in the mass spectrum, neither in unbound form nor in complex with trypsin. Only the trypsin-hydrolyzed species of vaso-pressin, desGly9-vasopressin (VasdG, amino acid sequence: CYFQNCPR) was observed, indicating rapid and complete
Figure 4. Gas-phase HDX-MS of the Vaso-pressin-Trypsin Complex
(A) The crystal structure of the vasopressin-trypsin complex (PDB: 1YF4).
(B) Mass spectrum of desGly9-vasopressin (Vasdg)-trypsin (Tryp) ion complexes in the absence of gas-phase labeling.
(C and D) Mass spectra of quadrupole selected and complex dissociated desGly9-vasopressin-trypsin ion complex [Vasdg-Tryp + 8H]8+in the absence (C) and presence (D) of gas-phase la-beling.
(E) Deuterium uptake of the intact ion complex as well as the unbound and dissociated trypsin and vasopressin ions. Solid bars indicate unbound or complex-bound ions, and pinstriped bars indicate complex dissociated ions. The error bars shown represent 23 SD.
(F) Zoom-in on the desGly9-vasopressin-trypsin binding interface. Residues of trypsin engaged in intermolecular interaction networks with vaso-pressin through heteroatom-bound non-amide hydrogens in their side chains are highlighted.
tryptic digestion of vasopressin in solu-tion prior to ESI. Complexes of trypsin and desGly9-vasopressin were observed with six to nine charges (Figure 4B). As was the case with the lysozyme-NAG complexes, we targeted the abundant [VasdG-Tryp + 8H]8+ ion for gas-phase HDX-MS, as this ion is of relatively low charge and is thus expected to retain its solution-phase conformation for a longer time frame in the gas phase (Clemmer and Jarrold, 1997).
[VasdG-Tryp + 8H]8+ dissociated into desGly9-vasopressin [VasdG* + H]+and trypsin [Tryp* + 7H]7+upon quadrupole ion se-lection of the complex and applying 20 eV collision energy. Mass spectra were obtained both in the absence and presence of gas-phase labeling (Figures 4C and 4D, respectively). Unbound trypsin [Tryp + 7H]7+exchanged 78 D (out of 143 heteroatom-bound non-amide hydrogens) and unbound desGly9-vasopressin [VasdG+ H]+exchanged 5.5 D out of eight heteroatom-bound non-amide hydrogens (Figure 4E). The [VasdG-Tryp + 8H]8+ complex exchanged 76 D, which in turn could be assigned to 71.5 D for [Tryp* + 7H]7+and 4.5 D for [VasdG* + H]+, by gas-phase dissoci-ation of the complex after labeling. Thus the deuterium uptake of trypsin and desGly9-vasopressin was reduced by 6.5 D and 1 D, respectively, when in complex, indicating a protection from HDX of specific heteroatom-bound non-amide hydrogens in both vasopressin and trypsin upon complex formation. A crystal struc-ture of the trypsin-vasopressin complex (Ibrahim and Pattabhi, 2005) shows that trypsin forms several intermolecular interaction networks with intact vasopressin through side-chain moieties containing heteroatom-bound non-amide hydrogens. Y2 and two out of three heteroatom-bound non-amide hydrogens in R8 of vasopressin is partaking in intermolecular interactions with trypsin, whereas the C-terminal carbonyl group and N-terminal
amino group remain unprotected upon complex formation. The C terminus, N terminus, charge-carrying proton, and a single NH group on R8 of vasopressin can account for an exchange of 5 D. The 1 D decrease in deuterium exchange of vasopressin upon binding to trypsin is thus likely due to partial protection from HDX of R8 or Y2.
Likewise, several residues of trypsin with heteroatom-bound non-amide hydrogens are partaking in interactions with vaso-pressin, namely, H57, D189, S190, S195, and Y217 (Figure 4F) (Ibrahim and Pattabhi, 2005). Protection from HDX of these res-idues in the complex would account for a difference in HDX be-tween the unbound and vasopressin-bound protein ion of 5 D, provided no charge-carrying protons are located on these resi-dues. Thus, the observed reduced HDX of 6.5 D in trypsin can be reconciled by involvement of these sites, in addition to one or two additional sites in trypsin indirectly involved in the binding interface. The crystallographic data of trypsin bound by the intact vasopressin ligand cannot be directly correlated to our gas-phase experiments, as G9 of vasopressin engages in inter-molecular hydrogen bonding in the crystal structure of the pep-tide-protein complex. No crystal structure is available for trypsin in complex with desGly9-vasopressin. Furthermore, it is also possible that some sites in trypsin do not exchange fully during gas-phase HDX-MS of the unbound protein ion due to intramo-lecular interactions. It is, however, apparent that both compo-nents show a decrease in deuterium exchange upon complex formation in good agreement with the crystal structure of the vasopressin-trypsin complex. This further underscores the po-tential for the gas-phase HDX-MS methodology to probe the binding interfaces of non-covalent complexes, here also illus-trated for a protein-peptide complex.
The intact singly protonated vasopressin, in the absence of trypsin, exchanged 0.5 D. This low exchange has been observed earlier, when it was also shown that the doubly protonated ion exchanged almost fully; six of its seven heteroatom-bound non-amide hydrogens (Mistarz et al., 2014). This could indicate the presence of intramolecular hydrogen bonding in the singly protonated vasopressin, making heteroatom-bound non-amide hydrogens inaccessible for exchange through ND3-mediated gas-phase HDX-MS on a millisecond timescale. Interestingly, the singly protonated trypsin-cleaved vasopressin (desGly9-vasopressin) showed a deuterium uptake similar to the doubly protonated intact vasopressin, if the 1 D difference was assigned to the absence of the additional charging proton. Evidently, the removal of the Gly9 residue removes intramolecular hydrogen bonds in the intact vasopressin ion responsible for suppressing gas-phase HDX-MS. In support, several gas-phase labeling studies on bradykinin, a similarly structured peptide, point to a correlation between low gas-phase HDX and the presence of a salt bridge between the guanidinium group of Arg residues and the C-terminal carboxyl group (Gimon-Kinsel et al., 1999; Schnier et al., 1996; Wyttenbach et al., 2003). The loss of Gly9 from vasopressin would make hydrogen bond formation be-tween Arg8 and the C-terminal carbonyl group energetically unfavorable due to steric hindrance. This could explain the 5 D increase in HDX of singly protonated desGly9-species relative to intact vasopressin, which in turn could be assigned to an un-shielding of Arg8 (3 D), C terminus (1 D), and the charge-carrying proton (1 D). Thus, the desGly-vasopressin versus vasopressin
comparison provides further evidence that gas-phase HDX-MS can report directly on structural differences between peptide/ protein ions (in terms of intramolecular interactions involving exchangeable side-chain hydrogens) provided that the charge is the same (Beeston et al., 2015; Mistarz et al., 2014; Rand et al., 2009, 2012).
DISCUSSION
Our results show that gas-phase HDX-MS mediated by ND3can be used to probe the binding interfaces of protein complexes by measuring the deuterium labeling of heteroatom-bound non-amide hydrogens. Fast and efficient gas-phase HDX-MS was achieved by directing ND3-saturated N2gas to a sub-ambient pressure region of a mass spectrometer, immediately down-stream of ionization, thereby enabling very rapid labeling of native-like conformations of peptide and protein ion complexes shortly after ESI. Subsequent collisional activation provided mild gas-phase dissociation of the complexes and enabled dissection of the contribution to HDX of the individual compo-nents of the ion complex. Thus, by a simple subtraction of the HDX of unbound ligand or protein from that of the bound coun-terpart, dissected through mild collisional activation, a measure of the number of heteroatom-bound non-amide side-chain hy-drogens involved in the interaction was obtained. This method was used to probe the interactions of two similar protein-glycan complexes and a protein-peptide complex. Gas-phase HDX-MS results were compared with the primary structure of peptides and proteins, as well as with available crystal structures, with a good apparent correlation between observed reductions in HDX upon complex formation and the presence of intermolec-ular interactions between residues with labile heteroatom-bound non-amide hydrogens in the crystal structures of the studied complexes. ND3-mediated gas-phase HDX-MS is thus highly dependent on ion structure as well as the number of charges in the ion, as increased protonation is likely to increase the rate of formation of exchange competent ion-ND3 complexes ac-cording to the proposed ‘‘onium exchange mechanism’’ (Camp-bell et al., 1995). Thus, an interpretation in terms of ion structure becomes less straightforward when comparing ions of different charge. Importantly, when measuring the reduction in HDX of protein and ligand ions upon complex formation, we compared the HDX of protein and ligand species of the same charge state, when in the unbound, bound, and subsequently dissociated states, thus facilitating a more direct interpretation of our data in terms of ion structure. Notably, the complex-bound lyso-zyme-NAG6/NAG4 and trypsin-vasopressin ions still incorpo-rated less deuterium than the unbound protein or ligand counterparts, indicating that the addition of charges from the oligosaccharide or peptide ligand to the protein had only a minor effect on gas-phase HDX of the complex, and that ion structure was the dominant factor determining the gas-phase HDX of the ion complex. We note that collisional dissociation of the non-covalent complexes studied here could facilitate intra-and/or intermolecular deuterium migration to other heteroat-om-bound hydrogens (Jørgensen et al., 2005). While not the focus of this study, no indications for intermolecular migration of deuterium were observed to occur in these experiments. The presented gas-phase HDX-MS measurements on two
protein-ligand complexes illustrate the potential of the method to not only detect ligand binding, but also distinguish between different binding modes of related ligands. This is evident from the data on lysozyme bound by NAG4or NAG6, where the reduc-tion in HDX of lysozyme due to NAG4binding was less than that of the larger NAG6, which correlates with the fewer number of intermolecular interactions of NAG4than NAG6with lysozyme. While detection of protein-ligand binding can be performed by higher-throughput methods, few methods can distinguish bind-ing modes based on differences in the number of sites involved in the interaction, i.e. the number of heteroatom-bound non-amide sites in the protein and/or ligand. Such detailed informa-tion can be achieved using X-ray crystallography, NMR, or solu-tion-phase HDX-MS methods. However, these methods are often labor intensive and suffer from low throughput. In earlier work, we have specifically shown that the current gas-phase HDX-MS methodology can be directly interfaced with automa-tion for automated sample introducautoma-tion to allow high-throughput analyses (Mistarz et al., 2014). Furthermore, compared with other methods in structural biology, this MS method allows ana-lyses to be performed (1) through a fast data acquisition obtain-ing information on both bindobtain-ing partners simultaneously from a single mass spectrum, (2) without loss of label during analysis (compared with solution-phase HDX-MS), (3) using only little material (low picomolar amounts at low micromolar concentra-tions of sample), and (4) requiring little sample pretreatment (buffer exchange into an ESI-compatible buffer). We envision considerable utility of the method in detecting and distinguishing the binding mode, in terms of the number of heteroatom-bound non-amide sites involved, of different ligands to a given protein target.
The dissociation step of the gas-phase HDX-MS method allows a dissection of the deuterium uptake of each binding partner in a protein-ligand complex but also provides the option for comparing relative binding energies between one or more protein-ligand complexes. This is illustrated by comparing the energy required to dissociate lysozyme-NAG4relative to lyso-zyme-NAG6ions (Figure S1). The collisional energy needed for dissociation of a non-covalent complex is proportional to the binding energy, and thus this measurement provides an orthog-onal analytical avenue of the gas-phase HDX-MS method for distinguishing protein-ligand binding modes. However, more work is needed to be able to infer absolute binding energies from collisional energy threshold values (Allison et al., 2015; Heck and van den Heuvel, 2004).
Like with other native MS-based methods, a limitation of the current technique is the need for ionization of the intact tein-ligand complex prior to gas-phase labeling. Not all pro-tein-ligand complexes should be expected to ionize to the same extent as intact species; in particular, those held together solely by hydrophobic forces may be more prone to dissociate in the absence of a polar solvent shell (Kitova et al., 2012; Robinson et al., 1996). Furthermore, proteins requiring the presence of complex sample matrices to remain stable, such as membrane proteins, are challenging to ionize, although significant progress has recently been made in this particular area (Mehmood et al., 2015). It is finally important to note that not all proteins or pro-tein-ligand/protein-protein complexes are expected to respond the same way to ESI even under gentle conditions. However,
provided a protein-ligand complex can be ionized as an intact species from native solution conditions to allow gas-phase HDX-MS to be performed, it is reasonable to expect this com-plex to have an overall near-native binding mode, as unfolding or rearrangement of either protein or ligand would likely have caused prior dissociation of the complex.
In summary, our results indicate that ND3-mediated gas-phase HDX immediately after ESI (gas-gas-phase HDX-MS) can be used to inform on biologically relevant conformers of pro-tein-ligand complexes. The approach should readily be appli-cable in structural proteomics in general, also including larger protein-protein complexes. The measurement of deuterium up-take for intact and dissociated protein complex components is inherently of high accuracy, as data is based on ions present in the same mass spectrum. We have previously shown that deuterium uptake during a gas-phase HDX-MS experiment can be assigned to individual residues when performed in com-bination with electron transfer dissociation (ETD) in the trap T-Wave ion guide of the same mass spectrometer used in this study (Mistarz et al., 2014; Rand et al., 2012). We thus envision that the contribution of individual amino acids to the binding interface of a protein complex could be obtained by us-ing ETD in combination with the method described herein. In addition, we have previously shown that the employed setup for gas-phase HDX-MS can be performed on a liquid chromatography-compatible timescale and is thus compatible with liquid chromatographic separation (Mistarz et al., 2014). Furthermore, using a commercially available nano-ESI auto-sampler, we have shown that the approach is adaptable to meet the demands of higher-throughput applications such as ligand screening or folding studies (Mistarz et al., 2014). The re-sults presented here thus strongly indicate that gas-phase HDX mediated by ND3 on a millisecond timescale combined with native MS represents a useful approach capable of providing valuable insights into the binding interface of protein-ligand complexes and a measure of the number of non-amide side-chain sites implicated in intermolecular interactions in the pro-tein complex.
EXPERIMENTAL PROCEDURES Overall Workflow
Samples of non-covalent peptide and protein complexes were ionized by nanoelectrospray under native conditions at pH 7.0. MS was performed with and without ND3-mediated gas-phase HDX in the ion source region of the mass spectrometer. Ion complexes were dissociated downstream of ND3 labeling but upstream of mass analysis, enabling measurement of the deute-rium content of individual binding partners in the complexes following gas-phase HDX.
Nanoelectrospray Mass Spectrometry of Intact Non-covalent Peptide and Protein Complexes
Samples were, unless stated otherwise, loaded in pulled metal-coated glass capillaries (Thermo/Proxeon) and ionized using a nanoflow Z-spray ESI source (Waters) mounted on a Synapt G2 HDMS mass spectrometer (Waters). The source and spray settings were tuned for optimal signal and transmission of the non-covalent complexes: capillary voltage 1.1–1.3 kV, sampling cone voltage 15–20 V, extraction cone voltage 2.5–3.5 V and source block temper-ature 90–110C. All other settings were kept standard operating conditions of the instrument and acquisitions were performed in ‘‘resolution mode.’’ Mass spectra were accumulated over 5 min with 0.5 s per scan from 100 to 4,000 m/z, and performed in positive ionization mode.
Gas-phase HDX-MS
Gas-phase HDX-MS was performed in the ion source region immediately downstream of the primary cone exit (sample cone) as described elsewhere (Mistarz et al., 2014). In short, 3.0 ml of aqueous ND3/D2O reagent was added to the standard ETD reagent vial, with the ETD reagent removed. To control reagent flow, the mass spectrometer software was set in ETD-mode with a ‘‘make-up’’ gas flow rate of 50 ml/min (N2gas), which is the maximum setting on the Synapt G2 HDMS instrument. The flow rate can be reduced for decreased labeling (Mistarz et al., 2014). N2gas was flowed through the headspace of the reagent pot and through a hollow glow discharge needle downstream of the primary cone exit. The normal discharging potential was switched off and the trap T-Wave wave-height was set to +2.0 V to prevent unwanted electron transfer reactions in case of residual ETD reagent in the tubing.
Gas-Phase Dissociation of Peptide and Protein Complexes
Non-covalent complexes were selected in the quadrupole and dissociated in the trap T-Wave of the mass spectrometer. The voltage settings of the trap T-Wave ion guide were individually optimized for each complex to produce a suitable abundance of both complex dissociated ions and remaining intact complex ions for reliable mass analysis and comparison of deuterium uptake of all species in a single spectrum, while ensuring no backbone fragmentation occurred. Trap T-Wave collision energies of 4 eV for the peptide complexes and 20 eV for the protein complexes were found to be optimal for these exper-iments. For deuterium-labeled peptide and protein complexes, the quadrupole selection window was increased and expanded to encompass broadening of the isotopic envelope.
Data Analysis
Processing of all mass spectra was carried out using MassLynx V4.1 software (Waters) with a Savitzky-Golay smoothing function (1, 5) and subsequent centering of the peaks. The deuterium content of oligosaccharide, peptide, protein, and complex ions were determined using Excel 2013 (Microsoft) by calculating the difference in the intensity weighted centroid average masses of deuterated ions to those from a non-deuterated control sample recorded in the absence ND3gas. All experiments were performed in at least three tech-nical replicates, unless stated otherwise. The error bars shown in all figures represent twice the standard deviation (23 SD) of such replicate measure-ments. We define any change in gas-phase HDX in excess of 23 SD to be a significant difference.
Materials
26% ammonia-D3 in D2O 99.5% (ND3/D2O) was purchased from Merck. Hexa-N-acetylglucosamine (NAG6/GlcNAc6/hexa-N-acetyl-D-glucosamine) was purchased from Megazyme. All other peptides, proteins, and reagents were purchased from Sigma-Aldrich and were used without further purification.
Sample Preparation
Lyophilized crown ethers, oligosaccharides, peptides, and proteins were, un-less stated otherwise, dissolved in distilled water (1 mg/ml) and further diluted to 100mM (1,000 mM for hexa-N-acetylglucosamine and CE) in 10 mM ammo-nium acetate adjusted to pH 7.0. Mixtures of the dissolved crown ethers, oligosaccharides, peptides, and proteins were made and diluted to varying concentrations with 10 mM ammonium acetate adjusted to pH 7.0 as stated here: (1) Leu-Enk (10mM) and CE (20 mM); (2) lysozyme (40 mM) and hexa-N-acetylglucosamine (400 mM); (3) trypsin (10 mM) and Arg8-vasopressin (40mM). All mixtures were equilibrated at 5C before analysis.
SUPPLEMENTAL INFORMATION
Supplemental Information includes one figure and can be found with this article online athttp://dx.doi.org/10.1016/j.str.2015.11.013.
AUTHOR CONTRIBUTIONS
U.H.M and K.D.R. conceived the study; U.H.M., K.F.H., and K.D.R. designed the research and analyzed the data; K.F.H., J.M.B., and K.D.R. provided tech-nical assistance and contributed reagents and analytic tools; U.H.M. and
K.D.R. wrote the paper; all authors commented on and edited the final version of the paper.
ACKNOWLEDGMENTS
K.D.R. acknowledges support from The Marie Curie Actions Programme of the EU (Grant No. PCIG09-GA-2011-294214) and the Danish Council for Indepen-dent Researchj Natural Sciences (Steno Grant No. 11-104058).
Received: August 3, 2015 Revised: November 16, 2015 Accepted: November 23, 2015 Published: December 31, 2015
REFERENCES
Allison, T.M., Reading, E., Liko, I., Baldwin, A.J., Laganowsky, A., and Robinson, C.V. (2015). Quantifying the stabilizing effects of proteligand in-teractions in the gas phase. Nat. Commun. 6, 8551.
Badman, E.R., Hoaglund-Hyzer, C.S., and Clemmer, D.E. (2001). Monitoring structural changes of proteins in an ion trap over10-200 ms: unfolding tran-sitions in cytochrome c ions. Anal. Chem. 73, 6000–6007.
Beeston, H.S., Ault, J.R., Pringle, S.D., Brown, J.M., and Ashcroft, A.E. (2015). Changes in protein structure monitored by use of gas-phase hydrogen/deute-rium exchange. Proteomics 15, 2842–2850.
Benesch, J.L.P., Ruotolo, B.T., Simmons, D.A., and Robinson, C.V. (2007). Protein complexes in the gas phase: technology for structural genomics and proteomics. Chem. Rev. 107, 3544.
Blake, C., Johnson, L.N., Mair, G., North, A., Phillips, D., and Sarma, V. (1967). Crystallographic studies of the activity of hen egg-white lysozyme. Proc. R. Soc. Lond. B Biol. Sci. 167, 378–388.
Bohrer, B.C., Mererbloom, S.I., Koeniger, S.L., Hilderbrand, A.E., and Clemmer, D.E. (2008). Biomolecule analysis by ion mobility spectrometry. Annu. Rev. Anal. Chem. 1, 293–297.
Bouchoux, G. (2012). Gas phase basicities of polyfunctional molecules. Part 3: amino acids. Mass Spectrom. Rev. 31, 391–435.
Breuker, K., and McLafferty, F.W. (2003). Native electron capture dissociation for the structural characterization of noncovalent interactions in native cyto-chrome C. Angew. Chem. Int. Ed. (English) 42, 4900–4904.
Breuker, K., and McLafferty, F.W. (2008). Stepwise evolution of protein native structure with electrospray into the gas phase, 10-12
to 102
s. Proc. Natl. Acad. Sci. U.S.A. 105, 18145–18152.
Campbell, S., Rodgers, M.T., Marzluff, E.M., and Beauchamp, J.L. (1995). Deuterium exchange reactions as a probe of biomolecule structure. Fundamental studies of gas phase H/D exchange reactions of protonated glycine oligomers with D2O, CD3OD, CD3CO2D, and ND3. J. Am. Chem.
Soc. 117, 12840–12854.
Cheng, X., and Fenselau, C. (1992). Hydrogen/deuterium exchange of mass-selected peptide ions with ND3in a tandem sector mass spectrometer. Int.
J. Mass Spectrom. 122, 109–119.
Clemmer, D.E., and Jarrold, M.F. (1997). Ion mobility measurements and their applications to clusters and biomolecules. J. Mass Spectrom. 32, 577–592. Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F., and Whitehouse, C.M. (1989). Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 64–71.
Ganem, B., Li, Y.T., and Henion, J.D. (1991). Observation of noncovalent enzyme-substrate and enzyme-product complexes by ion-spray mass spec-trometry. J. Am. Chem. Soc. 113, 7818–7819.
Gimon-Kinsel, M.E., Barbacci, D.C., and Russell, D.H. (1999). Conformations of protonated gas-phase bradykinin ions: evidence for intramolecular hydrogen bonding. J. Mass Spectrom. 34, 124–136.
Gross, D.S., Schnier, P.D., Rodriguez-Cruz, S.E., Fagerquist, C.K., and Williams, E.R. (1996). Conformations and folding of lysozyme ions in vacuo. Proc. Natl. Acad. Sci. U.S.A. 93, 3143–3148.
Heck, A.J.R. (2008). Native mass spectrometry: a bridge between interactom-ics and structural biology. Nat. Methods 5, 927–933.
Heck, A.J.R., and van den Heuvel, R.H.H. (2004). Investigation of intact protein complexes by mass spectrometry. Mass Spectrom. Rev. 23, 368–389. Heck, A.J., Jørgensen, T.J.D., O’Sullivan, M., von Raumer, M., and Derrick, P.J. (1998). Gas-phase noncovalent interactions between vancomycin-group antibiotics and bacterial cell-wall precursor peptides probed by hydrogen/ deuterium exchange. J. Am. Soc. Mass Spectrom. 9, 1255–1266.
Hemling, M.E., Conboy, J.J., Bean, M.F., Mentzer, M., and Carr, S.A. (1994). Gas phase hydrogen/deuterium exchange in electrospray ionization mass spectrometry as a practical tool for structure elucidation. J. Am. Soc. Mass Spectrom. 5, 434–442.
Hunter, E.P., and Lias, S.G. (1998). Evaluated gas phase basicities and proton affinities of molecules: an update. J. Phys. Chem. Ref. Data 27, 413–656. Ibrahim, B.S., and Pattabhi, V. (2005). Trypsin inhibition by a peptide hormone: crystal structure of trypsin-vasopressin complex. J. Mol. Biol. 348, 1191–1198. Jarrold, M.F. (2000). Peptides and proteins in the vapor phase. Annu. Rev. Phys. Chem. 51, 179–207.
Jørgensen, T.J.D., Gardsvoll, H., Ploug, M., and Roepstorff, P. (2005). Intramolecular migration of amide hydrogens in protonated peptides upon collisional activation. J. Am. Chem. Soc. 127, 2785–2793.
Kitova, E.N., El-Hawiet, A., Schnier, P.D., and Klassen, J.S. (2012). Reliable de-terminations of protein–ligand interactions by direct ESI-MS measurements. Are we there yet? J. Am. Soc. Mass Spectrom. 23, 431–441.
Lee, S.-W., Lee, H.-N., Kim, S.K., and Beauchamp, J.L. (1998). Selective bind-ing of crown ethers to protonated peptides can be used to probe mechanisms of H/D exchange and collision-induced dissociation reactions in the gas phase. J. Am. Chem. Soc. 120, 5800–5805.
Lermyte, F., Konijnenberg, A., Williams, J.P., Brown, J.M., Valkenborg, D., and Sobott, F. (2014). ETD allows for native surface mapping of a 150 kDa nonco-valent complex on a commercial Q-TWIMS-TOF instrument. J. Am. Soc. Mass Spectrom. 25, 343–350.
Leurs, U., Mistarz, U.H., and Rand, K.D. (2015). Getting to the core of protein pharmaceuticals—comprehensive structure analysis by mass spectrometry. Eur. J. Pharm. Biopharm. 93, 95–109.
Liljas, A., and Rossmann, M.G. (1974). X-ray studies of protein interactions. Annu. Rev. Biochem. 43, 475–507.
Loo, R.R.O., Winger, B.E., and Smith, R.D. (1994). Proton-transfer reaction studies of multiply-charged proteins in a high mass-to-charge ratio quadru-pole mass-spectrometer. J. Am. Soc. Mass Spectrom. 5, 1064–1071. Ly, T., and Julian, R.R. (2010). Elucidating the tertiary structure of protein ions in vacuo with site specific photoinitiated radical reactions. J. Am. Chem. Soc. 132, 8602–8609.
Mehmood, S., Allison, T.M., and Robinson, C.V. (2015). Mass spectrometry of protein complexes: from origins to applications. Annu. Rev. Phys. Chem. 66, 453–474.
Mistarz, U.H., Brown, J.M., Haselmann, K.F., and Rand, K.D. (2014). Simple setup for gas-phase H/D exchange mass spectrometry coupled to electron transfer dissociation and ion mobility for analysis of polypeptide structure on a liquid chromatographic timescale. Anal. Chem. 86, 11868–11876. Pan, J., Heath, B.L., Jockusch, R.A., and Konermann, L. (2012). Structural interrogation of electrosprayed peptide ions by gas-phase H/D exchange and electron capture dissociation mass spectrometry. Anal. Chem. 84, 373–378.
Pedersen, C.J. (1967). Cyclic polyethers and their complexes with metal salts. J. Am. Chem. Soc. 89, 7017–7036.
Perkins, S.J., Johnson, L., Phillips, D.C., and Dwek, R.A. (1981). The binding of monosaccharide inhibitors to hen egg-white lysozyme by proton magnetic resonance at 270 MHz and analysis by ring-current calculations. Biochem. J. 193, 553–572.
Pritchard, C., O’Connor, G., and Ashcroft, A.E. (2012). The role of ion mobility spectrometry-mass spectrometry in the analysis of protein reference stan-dards. Anal. Chem. 85, 7205–7212.
Rajabi, K., Ashcroft, A.E., and Radford, S.E. (2015). Mass spectrometric methods to analyze the structural organization of macromolecular complexes. Methods 89, 13–21.
Rand, K.D., Pringle, S.D., Murphy, J.P., III, Fadgen, K.E., Brown, J., and Engen, J.R. (2009). Gas-phase hydrogen/deuterium exchange in a traveling wave ion guide for the examination of protein conformations. Anal. Chem. 81, 10019–10028.
Rand, K.D., Pringle, S.D., Morris, M., and Brown, J.M. (2012). Site-specific analysis of gas-phase hydrogen/deuterium exchange of peptides and proteins by electron transfer dissociation. Anal. Chem. 84, 1931–1940.
Robinson, C.V., Chung, E.W., Kragelund, B.B., Knudsen, J., Aplin, R.T., Poulsen, F.M., and Dobson, C.M. (1996). Probing the nature of noncovalent in-teractions by mass spectrometry. A study of protein-CoA ligand binding and assembly. J. Am. Chem. Soc. 118, 8646–8653.
Ruotolo, B.T., Benesch, J.L.P., Sandercock, A.M., Hyung, S.J., and Robinson, C.V. (2008). Ion mobility-mass spectrometry analysis of large protein com-plexes. Nat. Protoc. 3, 1139–1152.
Rupley, J.A., and Gates, V. (1967). Studies on the enzymic activity of lysozyme, II. The hydrolysis and transfer reactions of N-acetylglucosamine oligosaccha-rides. Proc. Natl. Acad. Sci. U.S.A. 57, 496–510.
Salbo, R., Bush, M.F., Naver, H., Campuzano, I., Robinson, C.V., Pettersson, I., Jørgensen, T.J.D., and Haselmann, K.F. (2012). Traveling-wave ion mobility mass spectrometry of protein complexes: accurate calibrated collision cross-sections of human insulin oligomers. Rapid Commun. Mass Spectrom. 26, 1181–1193.
Schnier, P.D., Price, W.D., Jockusch, R.A., and Williams, E.R. (1996). Blackbody infrared radiative dissociation of bradykinin and its analogues: energetics, dynamics, and evidence for salt-bridge structures in the gas phase. J. Am. Chem. Soc. 118, 7178–7189.
Shi, Y. (2014). A glimpse of structural biology through X-ray crystallography. Cell 159, 995–1014.
Suckau, D., Shi, Y., Beu, S.C., Senko, M.W., Quinn, J.P., Wampler, F.M., and McLafferty, F.W. (1993). Coexisting stable conformations of gaseous protein ions. Proc. Natl. Acad. Sci. U.S.A. 90, 790–793.
Valentine, S.J., Anderson, J.G., Ellington, A.D., and Clemmer, D.E. (1997). Disulfide-intact and-reduced lysozyme in the gas phase: conformations and pathways of folding and unfolding. J. Phys. Chem. B 101, 3891–3900. Veros, C.T., and Oldham, N.J. (2007). Quantitative determination of lysozyme-ligand binding in the solution and gas phases by electrospray ionisation mass spectrometry. Rapid Commun. Mass Spectrom. 21, 3505–3510.
Von Dreele, R. (2005). Binding of N-acetylglucosamine oligosaccharides to hen egg-white lysozyme: a powder diffraction study. Acta Crystallogr. D-Biol. Crystallogr. 61, 22–32.
Winger, B.E., Light-Wahl, K.J., Rockwood, A.L., and Smith, R.D. (1992). Probing qualitative conformation differences of multiply protonated gas-phase proteins via H/D isotopic exchange with D2O. J. Am. Chem. Soc. 114, 5897.
Wu¨thrich, K. (1986). NMR of Proteins and Nucleic Acids, vol. 155, First Edition (Wiley).
Wu¨thrich, K. (2001). The way to NMR structures of proteins. Nat. Struct. Biol. 8, 923–925.
Wyttenbach, T., and Bowers, M.T. (2011). Structural stability from solution to the gas phase: native solution structure of ubiquitin survives analysis in a solvent-free ion mobility-mass spectrometry environment. J. Phys. Chem. B 115, 12266–12275.
Wyttenbach, T., Paizs, B., Barran, P., Breci, L., Liu, D.F., Suhai, S., Wysocki, V.H., and Bowers, M.T. (2003). The effect of the initial water of hydration on the energetics, structures, and H/D exchange mechanism of a family of pentapeptides: an experimental and theoretical study. J. Am. Chem. Soc. 125, 13768–13775.
