HDX-MS
Here is a summary of the history of HDX-MS, a little bit of theory, and some insight into how experiments are typically conducted. It is neither exhaustive or entirely comprehensive overview of the field but provides a (fairly) brief introduction to the technique. A peer-reviewed review of HDX-MS in relation to drug discovery can be found here.
Hydrogen Deuterium Exchange Mass Spectrometry
Hydrogen Deuterium Exchange Mass Spectrometry (HDX-MS) is a hugely versatile and powerful technique that is experiencing a surge in application following advances in both software and instrumentation (see Figure 1).
Figure 1: The increasing popularity of HDX-MS. (A) The number of articles returned when searching for HDX-MS in Pubmed (retrieved July 2014 – data up to 2013 included). (B) The Pubmed return for HDX.
HDX-MS has been used to measure protein:protein (Shukla et al. 2014; Q. Zhang et al. 2011b), protein:ligand (Chalmers et al. 2006; H. Xiao et al. 2003; Seckler et al. 2011), and protein:membrane (Burke et al. 2008) interactions, as well as being used extensively to provide insight into protein folding and dynamics (Wales & Engen 2005).
A brief history of HDX-MS application and theory
Spurred by the discovery of a-helical (PAULING et al. 1951) and b-sheet secondary structure (PAULING & COREY 1951), Linderstrøm-Lang of the Carlsberg Laboratory wished to investigate the secondary structure of proteins. Reasoning that secondary structure would retard the solvent exchange rate of the amide group hydrogen forming these structures, Linderstrøm-Lang went about deuterating proteins with the aim of distinguishing residues taking part in secondary structure. A flurry of papers started with the publication detailing the deuteration of insulin (Hvidt & LINDERSTRØM-LANG 1954), along with three papers detailing the methods employed, and a comparison of insulin to short peptides and ‘Sanger’s A-chain’ (KRAUSE & LINDSTROM-LANG 1955; Hvidt & LINDERSTRØM-LANG 1955a; Hvidt & LINDERSTRØM-LANG 1954). Using crude techniques, consisting of completely deuterating porcine insulin in solution, subsequent freezing and lyophilisation, controlled rehydration in water, and then density measurement using density gradient columns, Linderstrøm-Lang discovered many of the tenets of HDX including: HDX kinetics are non-linear and highly complex for folded proteins, folded proteins exchange far slower than simple short peptides, and that the temperature and pH of the aqueous environment also effects exchange rates non-linearly (LINDERSTRØM-LANG 1955). HDX was applied to protein (insulin) folding, with urea treatment of proteins found to greatly increase the rate of solvent exchange (Hvidt & LINDERSTRØM-LANG 1955a; Hvidt & LINDERSTRØM-LANG 1955b).
By using poly-d/l-alanine as a model protein for HDX and varying the pH (from pH 2 to 4.9), a nadir of HDX rate was discovered at pH 3.0 (BERGER & LINDERSTRØM-LANG 1957). Furthermore the chemical mechanism of solvent exchange was also discovered at this point, using NMR with N-methylacetamide (CH3CONHCH3) as a model for a single peptide, and was found to be both acid and base catalysed (BERGER et al. 1959). This allowed a theoretical framework for the HDX of proteins to be proposed.
Equation 1‑1: The Linderstrøm-Lang equation.
Linderstrøm-Lang equation (Equation 1‑1) describes the two-step mechanism of solvent incorporation. The first step is dependent on the equilibrium of the closed (N) to open (I) conformations (rates k1 and k2). The second step is the solvent exchange, which proceeds at rate k3 (Hvidt 1964). The rate k3 in Equation 1‑1, the solvent exchange step, can be further defined in Equation 1‑2, where [cat] is the catalyst concentration, consisting of either hydroxide ions [OH–] for the base catalysed reaction, or hydronium cations [H3O+] for the acid catalysed.
Equation 1‑2
Finally the observed rate of solvent exchange (kex= 1/τ) can be defined as:
Equation 1‑3
Equation 1‑3 is applicable for the vast majority of exchange reactions at physiological pH (therefore base catalysed) and describes the incorporation of D2O that occurs in the course of local, rapid folding and unfolding events. During each event, a single residue rapidly unfolds and re-folds, with a chance of solvent incorporation in the unfolded state i.e. , and this is known as EX2 kinetics (Hvidt & Nielsen 1966). A subset of HDX reactions can exhibit EX1 kinetics. EX1 kinetics occur when . This can be envisioned as a larger scale unfolding event, for instance a large conformational change, when a group or subset of residues all exchange before re-folding again. Peptides exhibiting EX1 kinetics are readily apparent from the characteristic ‘bi-modal’ distribution (see Figure 2).
Figure 2‑: Representation of EX2 and EX1 kinetics. Peptides with EX2 kinetics gradually incorporate the deuterium over time, causing a shift in the isotopic envelope to a higher m/z. In EX1 kinetics however, there are two populations of the peptide – a non-deuterated population, and a deuterated population, whose spectra overlay with one another. This creates a characteristic bi-modal distribution.
The theoretical chemical basis of proton transfer was determined subsequently (Englander & Downer 1972; Eigen 1964), as detailed in Scheme 1‑1.
Scheme 1‑1: Primary steps during a proton transfer reaction.
The reaction is initiated with the collision of the hydrogen donor (A) with the hydrogen acceptor (B), forming a hydrogen bond. There is an equilibrium (K2) in the distribution of the hydrogen between A and B, followed by dissociation between A and B, and the possibility of a transfer event occurring. In base-catalysed amide hydrogen exchange, the rate-limiting step is the removal of the hydrogen from the amide by OH–, and in acid-catalysed exchange, the proton donation from H3O+. These reactions show a high degree of temperature dependency, with the rate tripling with each 10 °C rise of temperature (Hvidt & Nielsen 1966). There are also (less dramatic) isotopic effects on the exchange rate, which can usually be ignored in biological processes (Englander & Downer 1972).
It is important to note that process shown in Scheme 1-1 is reversible. Assuming a protein is in a 50:50 solution containing H2O:D2O, after a deuterium incorporation event occurs, there is ~ 50% chance (ignoring small isotopic effects) that this deuterium incorporation signal will be ‘lost’ and replaced with hydrogen from H2O. This phenomenon, called back-exchange, essentially means that there is a loss of signal for a deuterated sample when exposed to an aqueous, protiated environment. This loss of signal has to be taken into account and minimised in an HDX experiment. Practically this is achieved by forcing a unidirectional incorporation of deuterium by diluting a concentrated protein solution in a ~100% D2O buffer solution.
This theoretical framework has altered very little since Englander, Downer and Teitelbaum summarized it in 1972. What has altered greatly since then are the techniques for determining the rate of HDX and resolution of the exchanging residues.
The initial deuteration-lyophilisation-rehydration-density gradient experiments of Linderstrøm-Lang were time consuming, demanding and prone to experimental artefact (Englander & Downer 1972). Advances were made by the use of tritium (Ikegami & Kono 1967), and the use of tritium gel filtration (Englander 1963) to separate the tritiated proteins from the tritiated solution. This technique was used extensively until the 1980s, with whole protein labelling being used to investigate protein dynamics and conformational changes, but there was little structural information other than the identification of a conformational event occurring (Englander 2006).
With the increasing popularity of Nuclear Magnetic Resonance (NMR) spectroscopy in the 1980s and 1990s, HDX became a tool for measuring protein folding, by using denaturants and temperature changes to determine unfolding rates of proteins (Wales & Engen 2005). In order to determine the residues that undergo solvent exchange, prior isotopic labelling and assignment is required, which limits protein expression systems and the size of proteins that can be studied (Itzhaki et al. 1997). HDX NMR studies identified local, subglobal and global unfolding events, and were crucial developing protein folding theory (Bai & Englander 1996).
Coupling liquid chromatography to mass spectrometry has allowed for a renaissance in HDX. This experimental setup allows for both a highly accurate and precise method for determining the level of deuterium incorporation, and a highly reproducible method for the separation of deuterated polypeptides.
Modern HDX-MS technique
Modern HDX experiments are typically conducted using an ultra-high performance liquid chromatography (UPLC) system coupled to a mass spectrometer equipped with a soft ionisation method source, typically electrospray ionisation.
Figure 3: HDX-MS Sample preparation. A concentrated protein solution is diluted in an excess of D2O buffer for a set time, t, in a temperature-controlled manner. To quench the exchange reaction, a low pH quench buffer is added, mixed, and the sample is instantly flash frozen in liquid nitrogen.
Samples are prepared ahead of measurement (see Figure 3), typically in triplicate, and are stored at -80 °C. It is important to note that there are a number of variables in this method that need to be altered and tailored to specific proteins in order to optimise digestion, peptide separation, prevent sample overlap, and minimise back exchange (by minimising the time the sample is exposed to the protiated environment). A protein solution (approximately 10 mM) is diluted in a deuterium buffer, mixed and then incubated either at room temperature (22 °C) or on ice for a set amount of time. Times used typically range from 3 s to 3000 s. A quench solution consisting of (typically) 5 M guanidinium chloride and 8.4% formic acid is added to halt the exchange reaction, which is then immediately flash frozen in liquid nitrogen. The quench solution both reduces the pH to lower the rate of exchange, and also denatures the protein for the subsequent digestion. This process may be automated, but at the cost of some of the very fast reactions (3 s) not being possible using current generation robot technology.
To analyse the sample, it is rapidly thawed at room temperature and injected into a UPLC system kept on ice, and run in an acidic (0.1% formic acid) aqueous environment to slow back-exchange (Figure 4). This UPLC system is responsible for both the digestion of the peptides and the separation of these peptides for subsequent mass analysis. The sample is proteolysed using an immobilised non-specific acid-functional protease, such as porcine pepsin – but other proteases can be used such as aspartic protease nepenthesin 1 (Kadek et al. 2014; Rey et al. 2013). Porcine pepsin has the advantage of retaining a high level of activity when immobilised, at low temperature, and in an acidic condition but does have a slight preference for cleaving C-terminal to large hydrophobic amino-acids (Z. Zhang & Smith 1993) – and rarely cleaving at all prior to histidine, lysine, proline, and arginine residues. A random cleavage pattern is desirable, as this allows for the generation of peptides with overlapping sequences, and with that an increase in resolution of the information obtained. The digestion typically lasts for three minutes (but again can be optimised for certain proteins and digestion conditions).
Figure 4: An HDX-MS UPLC system. (A) Both valves are in the “load” position. The 0.1% formic acid solution from the isocratic pump flows over both the pepsin column and the peptide trap, eventually going to waste. The Dual gradient column pump flows onto the C18 UPLC column. The peptide sample is loaded at the injection point (labelled 7) into the loop. (B) The right hand valve is switched to inject, emptying the contents onto the peptide trap. As the protein is digested, the peptides are immobilized onto the peptide trap. Digest the protein for 3 minutes. (C) The left hand valve is now switched to inject. The peptide trap is now in line with the acetonitrile gradient, but with the flow reversed. The peptide trap is also now in line with the C18 column. With the reversed flow, the peptides are eluted onto the column, and the acetonitrile gradient elutes the peptides from the C18 column to be subsequently analysed using the mass spectrometer.
The number of peptides produced from the digestion determines the ‘resolution’ of a particular experiment – each peptide is a data point where the rate of deuterium incorporation is followed over time in various conditions. The fewer the number of peptides – the less complete our understanding of the conformational alterations that are occurring. Varying the concentration of guanidinium chloride in the quench solution may alter the peptic peptides produced, as will altering the flow rate over the pepsin column. When we have plenty of overlapping peptides, the redundancy within the dataset allows for an increased confidence in our observations and allows for further insight into local perturbations.
The peptic peptides are loaded onto a reverse-phase UPLC column and subsequently eluted using an acetonitrile gradient that elutes the peptides depending on their hydrophobicity. The retention time (sometimes referred to as rt) of a peptide on a certain acetonitrile gradient is highly reproducible and is used to identify peptides between runs. Gradients are often altered to allow for greater resolution between peptides that have similar retention times. For example, a protein that produces a very hydrophobic set of peptides may saturate the mass spectrometer towards the end of the acetonitrile gradient. Therefore, we run a shallower gradient through the higher percentages of acetonitrile to achieve a greater resolution between these hydrophobic peptides – the acetonitrile gradient profile is thus tailored to each protein to some extent.
Using a UPLC coupled mass spectrometer has a number of advantages for conducting an HDX-MS experiment. First, the deuterated sample can be quickly extracted from the deuterated buffer and digested at ~ 0°C whilst maintaining a low pH (and thus limiting back-exchange). Second, the apparatus has extremely high reproducibility, as the time between thawing, proteolysis and ionisation remains constant.
Prior to conducting any deuterium exchange rate measurements, several MS/MS experiments are conducted on non-deuterated samples to correctly identify the sequence of peptic peptides (from the fragment b and y ions), identify their retention time on the reverse-phase column, and also alter the various parameters mentioned above to optimise protein digestion. In an instrument such as the Waters Xevo G2 QtoF (see Figure 5), fragment ions are produced using collision induced dissociation (CID) – this is achieved by passing precursor (parent) ions through a sequence of ring-shaped electrodes (the T-Wave Collision Cell) filled with a high energy inert gas that interacts with the precursor ion to cause dissociation into the product (fragment) b and y ions (see Figure 6). With the Orbitrap instrument, this dissociation takes place in the HCD Collision Cell instead.
Peptide identification may be conducted by using data-dependent acquisition (DDA). In DDA, a number of precursor ions of a known m/z value are identified in an initial survey scan (typically the most abundant), are then subjected to CID and subsequent mass analysis. By knowing the mass of the precursor ion, and determining the mass of the subsequent series of multiple resultant fragment ions, it is possible to determine the sequence of the peptide by using software, such as Mascot, that searches a database of protein sequences that are assumed to be present in the sample. Mascot assesses the probability of correctly identifying the observed peptide sequences with projected peptides produced from the database based on the coverage of the fragment ions, the discrepancy between the observed and calculated masses, and the redundancy of the peptide in the protein database.
Figure 5: Schematic of the Mass Spectrometers used in this study. A) Waters Xevo G2 QtoF. The peptic peptides are ionised using a Z-spray electrospray ionisation source. At this point a lockspray is also infused into the sample cone – an internal mass calibration (Glu-1-Fibriniopeptide B (GFP)) that can be used to correct the mass of the collected data in real time. Ion mass is initially measured in the first quadrupole, before entering the T-Wave Collision Cell, where the precursor ions are fragmented into fragment b and y ions via collision induced dissociation (CID). The product ions’ m/z is then measured using the reflectron TOF. B) The Thermo LTQ Orbintrap XL. Similarly to the Waters instrument, an ESI source ionises the peptic peptides. These are then filtered through a quadrupole into a linear trap quadrupole (LTQ) Linear Ion Trap. Ions are trapped here both radially, using a two-dimensional radio-frequency field, and axially, using stopping potentials on the end electrodes. Ions are then fed to the C-trap, and can by dissociated by Higher Energy Collisional Dissociation (HCD) by entering the HCD Collision Cell. Fragment ions are then fed back into the C-trap, and are injected onto the Orbitrap mass analyser, which is functionally similar to a Fourier transform ion cyclotron resonance mass spectrometer, as it detects ions based on their axial oscillation around an electrode.
CID does have a major drawback that is the phenomenon known as ‘scrambling’ – the propensity for the amide hydrogen atoms to be completely randomised by undergoing intramolecular migration (Jørgensen et al. 2005). It would be highly desirable to produce fragment ions that retain the incorporated deuterium, as this could produce single-residue resolution HDX-MS. This is possible with the use of Electron-Capture Dissociation (ECD) or Electron-Transfer Dissociation (ETD), two alterative mechanisms of fragmentation, that cause dissociation of the precursor ion by delivering negative charge either by an electron (ECD) or by radical anions (ETD) (Syka et al. 2004). ETD and ECD produce c and z fragment ions and are frequently used in identifying post-translational modifications. There has been recent progress in using these new dissociation technologies in HDX-MS (Pan et al. 2008; Rand et al. 2011; Rand et al. 2009; Abzalimov et al. 2013).
Figure 6: Fragmentation peptides produced by CID in tandem mass spectrometry. The precursor ion is shown at the top, and the various possible product ion pairs shown below.
Due to collision-induced scrambling, all deuterium incorporation measurements are thus observed as a change in mass of the precursor ion in the MS (rather the MS/MS) mode of the instrument. The progression of a single peptide is shown in Figure 7. The centroid mass, represented in Figure 7 as the coloured arrow is a single data point. The computation of the centroid mass is done semi-autonomously by the data analysis software, but this is normally checked manually to assess the quality of the spectra, and the fit of the isotopic envelope.
Figure 7: Representative peptide detailing HDX progression. A non-deuterated peptide, residues 35-42, from wtPTEN is shown on top. Upon exposure to D2O for 3, 30 or 300 s, the mass of the peptide increases, and the isotopic envelope of the peptide increases. When PTEN binds to lipid vesicles, this rate of incorporation is slowed, and the amount of D2O in the observed peptide is decreased
One of the most powerful advances in the application of HDX-MS arises from its use to assess differences in the rates of HDX brought about by the formation of complexes or upon modification of the protein, as opposed to the absolute HDX rates. If possible, the results of the HDX-MS are mapped onto the three-dimensional structures of the proteins. This gives considerably greater insight into deciphering the meaning of the changes in HDX rates, suggesting whether these differences are likely to reflect direct changes in solvent exposure or whether they are more likely to be the consequence of conformational changes.
HDX-MS has multiple benefits when compared to other, more conventional methods of structural biology, such as X-ray crystallography or NMR. HDX-MS requires far less protein than either of these two methods, requiring as little as 100 mL of protein at a concentration of 5 mM to answer questions such as the location of intrinsic disorder within a protein, or to determine whether a protein is folded. Full-length proteins with intrinsically disordered regions, which often defy crystallographic structure determination, are still amenable to the technique. Also, determining the location of membrane binding elements of a protein is possible with HDX-MS, something impossible with X-ray crystallography. Finally, there is no theoretical size limit to the proteins being analysed, or need for isotopic labelling, both of which limit the utility of NMR. This makes HDX-MS ideally suited for the study of membrane-binding signalling complexes, such as the PI3Ks, and lipid phosphatases.
REFERENCES
Abzalimov, R.R., Bobst, C.E. & Kaltashov, I.A., 2013. A new approach to measuring protein backbone protection with high spatial resolution using H/D exchange and electron capture dissociation. Analytical Chemistry, 85(19), pp.9173–9180.
Bai, Y. & Englander, S.W., 1996. Future directions in folding: the multi-state nature of protein structure. Proteins Structure Function and Genetics, 24, pp.145–151.
BERGER, A. & LINDERSTRØM-LANG, K., 1957. Deuterium exchange of poly-DL-alanine in aqueous solution. Archives of Biochemistry and Biophysics, 69, pp.106–118.
BERGER, A., Loewenstein, A. & Meiboom, S., 1959. Nuclear Magnetic Resonance Study of the Protolysis and Ionization of N-Methylacetamide 1. Journal of the American Chemical Society, 81(1), pp.62–67.
Burke, J.E. & Williams, R.L., 2013. Dynamic steps in receptor tyrosine kinase mediated activation of class IA phosphoinositide 3-kinases (PI3K) captured by H/D exchange (HDX-MS). Advances in biological regulation, 53(1), pp.97–110.
Burke, J.E. et al., 2008. A phospholipid substrate molecule residing in the membrane surface mediates opening of the lid region in group IVA cytosolic phospholipase A2. The Journal of biological chemistry, 283(45), pp.31227–31236.
Burke, J.E. et al., 2011. Dynamics of the phosphoinositide 3-kinase p110δ interaction with p85α and membranes reveals aspects of regulation distinct from p110α. Structure, 19(8), pp.1127–1137.
Burke, J.E. et al., 2012. Oncogenic mutations mimic and enhance dynamic events in the natural activation of phosphoinositide 3-kinase p110α (PIK3CA). Proceedings of the National Academy of Sciences of the United States of America, 109(38), pp.15259–15264.
Burke, J.E. et al., 2014. Structures of PI4KIIIβ complexes show simultaneous recruitment of Rab11 and its effectors. Science, 344(6187), pp.1035–1038.
Chalmers, M.J. et al., 2006. Probing protein ligand interactions by automated hydrogen/deuterium exchange mass spectrometry. Analytical Chemistry, 78(4), pp.1005–1014.
Eigen, M., 1964. Proton Transfer, Acid-Base Catalysis, and Enzymatic Hydrolysis. Part I: ELEMENTARY PROCESSES. Angewandte Chemie International Edition, 3(1), pp.1–19.
regulators of growth and metabolism. Nature Reviews Genetics, 7(8), pp.606–619.
Englander, S.W., 1963. A Hydrogen Exchange Method Using Tritium and Sephadex: Its Application to Ribonuclease *. Biochemistry, 2(4), pp.798–807.
Englander, S.W., 2006. Hydrogen Exchange and Mass Spectrometry: A Historical Perspective. Journal of The American Society for Mass Spectrometry, 17(11), pp.1481–1489.
Englander, S.W. & Downer, N.W., 1972. Hydrogen exchange. Annual review of biochemistry, 4(1), pp.903–324.
Hon, W.-C., Berndt, A. & Williams, R.L., 2012. Regulation of lipid binding underlies the activation mechanism of class IA PI3-kinases. Oncogene, 31(32), pp.3655–3666.
Hvidt, A., 1964. A DISCUSSION OF THE PH DEPENDENCE OF THE HYDROGEN-DEUTERIUM EXCHANGE OF PROTEINS. Comptes-rendus des travaux du Laboratoire Carlsberg, 34, pp.299–317.
Hvidt, A. & LINDERSTRØM-LANG, K., 1955a. Exchange of deuterium and 18O between water and other substances. III. Deuterium exchange of short peptides, Sanger’s A-chain and insulin. Comptes rendus des travaux du Laboratoire Carlsberg. Série chimique, 29(22-23), pp.385–402.
Hvidt, A. & LINDERSTRØM-LANG, K., 1954. Exchange of hydrogen atoms in insulin with deuterium atoms in aqueous solutions. Biochimica et biophysica acta, 14(4), pp.574–575.
Hvidt, A. & LINDERSTRØM-LANG, K., 1955b. The kinetics of the deuterium exchange of insulin with D2O; an amendment. Biochimica et biophysica acta, 16(1), pp.168–169.
Hvidt, A. & Nielsen, S.O., 1966. Hydrogen exchange in proteins. Advances in protein chemistry, 21, pp.287–386.
Iakoucheva, L.M. et al., 2002. Intrinsic disorder in cell-signaling and cancer-associated proteins. Journal of Molecular Biology, 323(3), pp.573–584.
Ikegami, A. & Kono, N., 1967. Tritium-hydrogen exchange of polypeptides in aqueous solutions. Journal of Molecular Biology, 29(2), pp.251–274.
Itzhaki, L.S., Neira, J.L. & Fersht, A.R., 1997. Hydrogen exchange in chymotrypsin inhibitor 2 probed by denaturants and temperature. Journal of Molecular Biology, 270(1), pp.89–98.
Jørgensen, T.J.D. et al., 2005. Intramolecular Migration of Amide Hydrogens in Protonated Peptides upon Collisional Activation. Journal of the American Chemical Society, 127(8), pp.2785–2793.
Keppel, T.R. & Weis, D.D., 2014. Mapping Residual Structure in Intrinsically Disordered Proteins at Residue Resolution Using Millisecond Hydrogen/Deuterium Exchange and Residue Averaging. Journal of The American Society for Mass Spectrometry.
KRAUSE, I.M. & LINDSTROM-LANG, K., 1955. Exchange of deuterium and 18O between water and other substances. II. Alternative-methods. Comptes rendus des travaux du Laboratoire Carlsberg. Série chimique, 29(22-23), pp.367–384.
Krissinel, E. & Henrick, K., 2007. Inference of macromolecular assemblies from crystalline state. Journal of Molecular Biology, 372(3), pp.774–797.
LINDERSTRØM-LANG, K., 1955. The pH-dependence of the deuterium exchange of insulin. Biochimica et biophysica acta, 18(2), p.308.
Pan, J. et al., 2008. Electron capture dissociation of electrosprayed protein ions for spatially resolved hydrogen exchange measurements. Journal of the American Chemical Society, 130(35), pp.11574–11575.
Pascal, B.D. et al., 2012. HDX workbench: software for the analysis of H/D exchange MS data. Journal of The American Society for Mass Spectrometry, 23(9), pp.1512–1521.
PAULING, L. & COREY, R.B., 1951. The pleated sheet, a new layer configuration of polypeptide chains. Proceedings of the National Academy of Sciences, 37(5), pp.251–256.
PAULING, L., COREY, R.B. & BRANSON, H.R., 1951. The structure of proteins; two hydrogen-bonded helical configurations of the polypeptide chain. Proceedings of the National Academy of Sciences, 37(4), pp.205–211.
Rand, K.D. et al., 2011. ETD in a traveling wave ion guide at tuned Z-spray ion source conditions allows for site-specific hydrogen/deuterium exchange measurements. Journal of The American Society for Mass Spectrometry, 22(10), pp.1784–1793.
Rand, K.D. et al., 2009. Protein hydrogen exchange measured at single-residue resolution by electron transfer dissociation mass spectrometry. Analytical Chemistry, 81(14), pp.5577–5584.
Rey, M. et al., 2013. Nepenthesin from monkey cups for hydrogen/deuterium exchange mass spectrometry. Molecular & Cellular Proteomics, 12(2), pp.464–472.
Seckler, J.M., Barkley, M.D. & Wintrode, P.L., 2011. Allosteric suppression of HIV-1 reverse transcriptase structural dynamics upon inhibitor binding. Biophysical Journal, 100(1), pp.144–153.
Shukla, A.K. et al., 2014. Visualization of arrestin recruitment by a G-protein-coupled receptor. Nature. Available at: http://www.nature.com/nature/journal/vaop/ncurrent/full/nature13430.html#f2.
Söding, J., Biegert, A. & Lupas, A.N., 2005. The HHpred interactive server for protein homology detection and structure prediction. Nucleic acids research, 33(Web Server issue), pp.W244–8.
Syka, J.E.P. et al., 2004. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proceedings of the National Academy of Sciences, 101(26), pp.9528–9533.
Uversky, V.N. & Dunker, A.K., 2010. Understanding protein non-folding. Biochimica et Biophysica Acta (BBA) – Proteins and Proteomics, 1804(6), pp.1231–1264.
Vadas, O. et al., 2013. Molecular determinants of PI3Kγ-mediated activation downstream of G-protein-coupled receptors (GPCRs). Proceedings of the National Academy of Sciences of the United States of America, 110(47), pp.18862–18867.
Wales, T.E. & Engen, J.R., 2005. Hydrogen exchange mass spectrometry for the analysis of protein dynamics. Mass Spectrometry Reviews, 25(1), pp.158–170.
Yang, X. et al., 2011. Using Tandem Mass Spectrometry in Targeted Mode to Identify Activators of Class IA PI3K in Cancer. Cancer Research, 71(18), pp.5965–5975.
Zhang, Q. et al., 2011b. Epitope mapping of a 95 kDa antigen in complex with antibody by solution-phase amide backbone hydrogen/deuterium exchange monitored by Fourier transform ion cyclotron resonance mass spectrometry. Analytical Chemistry, 83(18), pp.7129–7136.
Zhang, Z. & Smith, D.L., 1993. Determination of amide hydrogen exchange by mass spectrometry: a new tool for protein structure elucidation. Protein science : a publication of the Protein Society, 2(4), pp.522–531.