Article Cite This: J. Phys. Chem. B 2017, 121, 11031−11036
pubs.acs.org/JPCB
Real-Time Investigation of the H/D Exchange Kinetics of Porphyrins and Oligopeptides by Means of Neutral Cluster-Induced Desorption/ Ionization Mass Spectrometry André Portz,† Christoph R. Gebhardt,‡ and Michael Dürr*,† †
Institut für Angewandte Physik, Justus-Liebig-Universität Giessen, Heinrich-Buff-Ring 16, D-35392 Giessen, Germany Bruker Daltonik GmbH, Fahrenheitstr. 4, D-28359 Bremen, Germany
‡
S Supporting Information *
ABSTRACT: The kinetics of the H/D exchange reaction in angiotensin II, hexaglycine (Gly6), Co(II)tetra(3-carboxyphenyl)porphyrin, and tetra(4-carboxyphenyl)porphyrin were followed in real time by mass spectrometry employing desorption/ionization induced by neutral SO2 clusters. The change of the isotope patterns with increasing degree of deuteration was recorded as a function of D2O exposure and the underlying H/ D exchange kinetics, i.e., the dependence of the different degrees of deuteration on time, were deduced. The results were modeled by means of Monte Carlo simulations taking into account different reaction constants for the H/D exchange reaction at different functional groups. In the case of the investigated porphyrins, the rate constants were directly assigned to the functional groups involved; in the case of the peptides, reaction at the explicit functional groups and the backbone chain of the molecules could be discriminated.
■
INTRODUCTION H/D exchange reactions are widely used to probe peptides and protein structures and their dynamics;1 a variety of techniques has been applied to probe H/D exchange, such as nuclear magnetic resonance,2 infrared,3 and Raman spectroscopy.4 Among these techniques, mass spectrometry (MS) with its high sensitivity on isotope exchange has emerged as a major tool for the investigation of H/D exchange reactions in biomolecules.5−8 Standard exchange experiments are conducted in solution starting with, e.g., fully undeuterated peptides or proteins, which are diluted in D2O for H/D exchange. After quenching the exchange reaction by a sudden change of the chemical environment, e.g., pH value,9 the degree of deuteration is determined, e.g., by means of electrospray ionization (ESI) mass spectrometry. Real-time measurements of H/D exchange can give access to the relevant time constants of the exchange process as well as their dependence on the chemical nature of the functional groups involved. Such realtime measurements are more difficult to be performed in the framework of the standard procedure described above, but have been realized, e.g., under ambient ionization conditions10−12 or using fast quench-flow instruments.13 In this work, we employ desorption/ionization induced by neutral SO2 clusters (DINeC) to investigate H/D exchange kinetics in oligopeptides and porphyrins in real time. DINeC makes use of neutral clusters consisting of 103−104 SO2 molecules for very soft desorption/ionization from bulk samples: During cluster surface impact, the cluster’s kinetic energy is redistributed on a short time scale activating the desorption process.14 With the high dipole moment of SO2 © 2017 American Chemical Society
(1.6 D), the cluster also serves as a transient matrix for the dissolution of polar analytes,15 thus lowering the effective energy barrier for their desorption. As a result, desorption takes place at low cluster energies ( 200 s. For comparison, the result of Monte Carlo simulations taking into account only one rate constant is also shown (dashed line, compare Figure S1 Supporting Information). For these simulations, a continuous decrease of the rate with time is obvious, as known from simple first-order reaction kinetics. Indeed, the experimentally deduced exchange kinetics shown in Figure 8a,b could be reproduced by means of Monte Carlo
Figure 4. Relative signal intensities of the five degrees of deuteration as a function of time for the porphyrin CoTCPP (symbols + solid line) together with the corresponding results of the Monte Carlo simulations (dashed lines).
Taking into account the exchange of hydrogen by deuterium (exchange rate Rf, forward reaction) and the reexchange (Rb, back reaction), the systems were modeled by means of Monte Carlo simulations. In the simulations, the H/D exchange and reexchange reactions were characterized by the probability pf ∝ Rf in forward direction and the probability pb ∝ Rb in backward direction, which include the reaction probability as such as well as the (local) concentration of D2O and H2O. Throughout the simulations, the values of pf and pb were held constant, equivalent to a constant D2O concentration as realized in our experiments. The probabilities pi can then directly be converted into pseudo-first-order rate constants ki. Further details of the Monte Carlo simulations are described in the Supporting Information. In Figure 4, the results of the optimized simulations (minimum of the mean-square deviation between measurement and simulation, pf/pb ≈ 10) are shown as dashed lines in comparison to the measured data (symbols). For the second porphyrin H2TCPP, the isotope pattern of the undeuterated, singly charged species occurs with its main peak at m/z = 791.3, i.e., ionization takes place by uptake of an additional proton, [M+H]+. Accordingly, we found a maximum degree of deuteration d = 7, as the six hydrogen atoms at the functional groups as well as the additional proton carrying the positive charge (attached to one of the nitrogen atoms) can be exchanged. When simulating the results of the second porphyrin H2TCPP, two pairs of exchange probabilities were necessary to represent the change of the measured spectra with dose by means of the Monte Carlo simulations (Figure 5). The ratio between p2,f and p1,f was deduced to be ≈0.35; p2 was applied to four exchangeable H atoms and p1 was applied to the remaining three exchangeable atoms. We furthermore note that, within the accuracy of the experiments, the values of pf,2(H2TCPP) and pf(CoTCPP) are the same. Angiotensin II. Figure 6 depicts the structural formula of angiotensin II with its 16 exchangeable hydrogen atoms. Being ionized by the uptake of an additional proton, [M + H]+, the
Figure 6. Structural formula of the neutral angiotensin II with exchangeable hydrogen atoms printed in green. H atoms at the backbone are highlighted by a blue background, all other exchangeable H atoms are highlighted by an orange background. 11033
DOI: 10.1021/acs.jpcb.7b06897 J. Phys. Chem. B 2017, 121, 11031−11036
Article
The Journal of Physical Chemistry B
Figure 7. Mass spectra of angiotensin II cations as obtained by means of DINeC−MS. Due to H/D exchange, the isotope pattern broadens and shifts toward higher m/z values in (b)−(d) when compared to the isotope pattern of the undeuterated species shown in (a). In (e), the average degree of deuteration d̅ is plotted as a function of time as deduced from the experiments (open dots), as well as deduced from Monte Carlo simulations taking into account one rate constant (black dashed curve) or three rate constants (red curve).
Figure 8. Relative signal intensities of selected degrees of deuteration for angiotensin II (symbols + solid lines) as a function of time together with the corresponding results of the Monte Carlo simulations (dashed lines). (a) Lower degrees of deuteration and (b) higher degrees of deuteration.
Table 1. Rate Constants ki (s−1) for the Exchange of Hydrogen by Deuterium As Deduced from Optimized Monte Carlo Simulations for the Four Different Molecules Investigated in this Studya
simulations only when three different pairs of exchange rates were applied (dashed lines in Figure 8a,b). The ratio pf,1:pf,2:pf,3 was deduced to be 11:7:1 for the optimized simulations. The respective subsets of exchangeable hydrogen atoms associated with one rate constant were all of comparable size, i.e., including five or six hydrogen atoms. The respective dependence of d̅ on time is shown in Figure 7e as a red line and agrees well with the experimental data. As a further example, we measured H/D exchange in hexaglycine. In that case, mass spectra were recorded in the negative-ion mode and the isotope pattern of [M−H]− was analyzed. The exchange kinetics are shown in Figure S3, together with the results of Monte Carlo simulations taking into account two different pairs of exchange probabilities, which differ by 1 order of magnitude. All rate constants as deduced from the Monte Carlo simulations are summarized in Table 1.
k1 (s−1) CoTCPP H2TCPP ATII Gly6
0.15 (3) 0.036 (5) 0.050 (2)
k2 (s−1) 0.050 (4) 0.052 (4) 0.023 (6)
k3 (s−1)
0.003 (6) 0.003 (5)
a
The numbers within parentheses indicate the size of the subset of functional groups this rate constant was applied to.
were attributed to specific functional groups: The rate constant k for CoTCPP and the lower rate constant k2 for H2TCPP are the same within the accuracy of the experiments (Table 1). In the case of CoTCPP, the respective rate constant k has to be associated with the carboxylic acid groups, as these are the only groups at which H/D exchange is possible (Figure 2). Assuming the same local concentration of D2O for both substances, we also attribute k2 in H2TCPP to the carboxylic acid groups, the more as these rates are applied to four exchangeable hydrogen atoms in H2TCPP. The rate constant k1, which is applied to the three remaining H atoms, thus describes the kinetics at the amine groups in the center of the porphyrin (Figure 2b). DINeC−MS thus allows to differentiate between the two functional groups.
■
DISCUSSION Using CoTPCC, H2TPCC, angiotensin II, and hexaglycine as test systems, DINeC has been shown to be a powerful tool for the investigation of H/D reactions in (bio)molecules. The fragmentation-free desorption/ionization allows to unambiguously follow the H/D exchange kinetics of these molecules in real time. Rate constants were deduced by comparison of experimental results with Monte Carlo simulations. In the case of the porphyrins CoTCPP and H2TCPP, these rate constants 11034
DOI: 10.1021/acs.jpcb.7b06897 J. Phys. Chem. B 2017, 121, 11031−11036
Article
The Journal of Physical Chemistry B
the application of MALDI in real-time monitoring of H/D exchange. In secondary ion mass spectrometry (SIMS), substantial fragmentation of biomolecules, as it occurs even in the case of soft cluster-SIMS,26−29 complicates the evaluation of H/D exchange.
For angiotensin II, the exchange of 17 hydrogen atoms was observed, in agreement with the maximum number of exchangeable hydrogen atoms (Figure 6). Again, one single rate constant was not sufficient to reproduce the experimentally observed exchange kinetics by means of Monte Carlo simulations, but at least three different rate constants were necessary. The number of hydrogen atoms in the respective subsets were approximately the same, but the rate constants varied by an order of magnitude. In particular, the slowest subset was much slower than the two faster subsets. For hexaglycine, two rate constants were applied with the larger subset of five hydrogen atoms being slow; its rate constant is comparable to k3(ATII). The rate constant of the two remaining H atoms is comparable to k1(ATII), the highest value in angiotensin II. With five hydrogen atoms located at the amide groups of hexaglycine’s backbone and further two exchangeable H atoms at the terminal amine group (the proton of the carboxylic acid group is detached in the case of the investigated [M−H]− species), the fast exchange was assigned to the amine groups and slow exchange was assigned to the amide groups. Thus, the slow exchange (k3) in angiotensin II is also assigned to the amide groups of the molecule’s backbone and the faster exchange is assigned to the explicit functional groups. This assignment is in agreement with the results for the porphyrins: the highest exchange rates observed for angiotensin II (k1 and k2) and for hexaglycine (k1) are in the same order of magnitude as the exchange rates for the amide and carboxylic acid groups in the porphyrins; minor deviations might be attributed to differences in the structure of the molecules. Furthermore, the assignment is compatible with MS/MS experiments with ATII after almost complete back reaction in H2O: the observed deuterated peptide fragments, i.e., with slow H/D exchange, do not indicate a preference for explicit functional groups but are distributed over the whole molecule (compare Supporting Information). Interestingly, Fouriertransform ion cyclotron resonance mass spectrometry (FTICR-MS) experiments on gas-phase H/D exchange also revealed three clearly discernible distributions of rate constants for angiotensin II.24 As the different exchange rates are associated with the chemical properties of the different functional groups and not so much with their position in the molecule, we do not expect larger changes of the kinetics for different structures in case of the investigated molecules. H/D exchange reactions are of general importance in biochemistry, especially with respect to the investigation of protein structure and dynamics. Making use of its high sensitivity and capability to discriminate isotopes, mass spectrometry is widely used as the method of choice for the detection of H/D exchange.7,8 Standard protocols involve deuteration and quenching of the deuteration process at a given time prior to the mass spectrometry measurements.9 No such measures have to be applied in real-time monitoring by means of DINeC−MS. As the samples can be gently dried prior to the DINeC experiments, they still contain a substantial amount of water,16 especially under the conditions of the H/D experiments. The results are thus expected to be also relevant for systems in solution; however, DINeC features the advantages of vacuum-based methods, such as a precise control of D2O partial pressure, different to ambient methods.10 Furthermore, DINeC is applicable in a wide range of solvent conditions, including high salt concentrations25 far beyond the limitations in ESI. In comparison to MALDI, no matrix has to be applied which may interact with the reactants and thus strongly limits
■
CONCLUSIONS DINeC−MS was applied for the real-time investigation of H/D exchange in porphyrins and peptides. The reaction kinetics of the exchange reaction could be resolved and rate constants were deduced; in particular, different rate constants of different functional groups within one molecule were discriminated. In other words, different reaction kinetics within one molecule could be observed. The H/D exchange experiments demonstrate in more general the potential of DINeC−MS for the investigation of complex reactions of (bio)molecules in real time.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.7b06897. Details on Monte Carlo simulations, further analysis and modeling of H/D exchange in angiotensin II, analysis and modeling of H/D exchange in hexaglycine, and MS/ MS experiments on H/D exchange in angiotensin II (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Phone: +0049 (0)641 9933490. ORCID
Michael Dürr: 0000-0002-4676-8715 Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Englander, S. W. Hydrogen Exchange and Mass Spectrometry: A Historical Perspective. J. Am. Soc. Mass Spectrom. 2006, 17, 1481− 1489. (2) Dempsey, C. E. Hydrogen Exchange in Peptides and Proteins Using NMR Spectroscopy. Prog. Nucl. Magn. Reson. Spectrosc. 2001, 39, 135−170. (3) Barth, A. Infrared Spectroscopy of Proteins. Biochim. Biophys. Acta 2007, 1767, 1073−1101. (4) Shashilov, V. A.; Lednev, I. K. Two-dimensional Correlation Raman Spectroscopy for Characterizing Protein Structure and Dynamics. J. Raman Spectrosc. 2009, 40, 1749−1758. (5) Engen, J. R. Analysis of Protein Conformation and Dynamics by Hydrogen/Deuterium Exchange MS. Anal. Chem. 2009, 81, 7870− 7875. (6) Benesch, J. L. P.; Ruotolo, B. T. Mass Spectrometry: Come of Age for Structural and Dynamical Biology. Curr. Opin. Struct. Biol. 2011, 21, 641−649. (7) Percy, A. J.; Rey, M.; Burns, K. M.; Schriemer, D. C. Probing Protein Interactions with Hydrogen/Deuterium Exchange and Mass Spectrometry - A Review. Anal. Chim. Acta 2012, 721, 7−21. (8) Kaltashov, I. A.; Bobst, C. E.; Abzalimov, R. R. Mass Spectrometry-based Methods to Study Protein Architecture and Dynamics. Protein Sci. 2013, 22, 530−544. (9) Bai, Y.; Milne, J. S.; Mayne, L.; Englander, S. W. Primary Structure Effects on Peptide Group Hydrogen Exchange. Proteins: Struct., Funct., Genet. 1993, 17, 75−86.
11035
DOI: 10.1021/acs.jpcb.7b06897 J. Phys. Chem. B 2017, 121, 11031−11036
Article
The Journal of Physical Chemistry B (10) Kostyukevich, Y.; Kononikhin, A.; Popov, I.; Nikolaev, E. Simple Atmospheric Hydrogen/Deuterium Exchange Method for Enumeration of Labile Hydrogens by Electrospray Ionization Mass Spectrometry. Anal. Chem. 2013, 85, 5330−5334. (11) Attygalle, A. B.; Gangam, R.; Pavlov, J. Real-Time Monitoring of In Situ Gas-Phase H/D Exchange Reactions of Cations by Atmospheric Pressure Helium Plasma Ionization Mass Spectrometry (HePI-MS). Anal. Chem. 2014, 86, 928−935. (12) Kostyukevich, Y.; Kononikhin, A.; Popov, I.; Spasskiy, A.; Nikolaev, E. In ESI-Source H/D Exchange under Atmospheric Pressure for Peptides and Proteins of Different Molecular Weights from 1 to 66kDa: the Role of the Temperature of the Desolvating Capillary on H/D Exchange. J. Mass Spectrom. 2015, 50, 49−55. (13) Keppel, T. R.; Weis, D. D. Analysis of Disordered Proteins Using a Simple Apparatus for Millisecond Quench-Flow H/D Exchange. Anal. Chem. 2013, 85, 5161−5168. (14) Gebhardt, C. R.; Tomsic, A.; Schröder, H.; Dürr, M.; Kompa, K.-L. Matrix-Free Formation of Gas-Phase Biomolecular Ions by Soft Cluster-Induced Desorption. Angew. Chem., Int. Ed. 2009, 48, 4162− 4165. (15) Portz, A.; Baur, M.; Gebhardt, C. R.; Frank, A. J.; Neuderth, P.; Eickhoff, M.; Dürr, M. Influence of the Cluster Constituents’ Reactivity on the Desorption/Ionization Process Induced by Neutral SO2 Clusters. J. Chem. Phys. 2017, 146, No. 134705. (16) Baur, M.; Lee, B.-J.; Gebhardt, C. R.; Dürr, M. Soft Clusterinduced Desorption and Ionization of Biomolecules - Influence of Surface Load and Morphology on Desorption Efficiency. Appl. Phys. Lett. 2011, 99, No. 234103. (17) Lee, B.-J.; Baur, M.; Gebhardt, C. R.; Dürr, M. Quantification of the Ionization Probability During Desorption/Ionization of Oligopeptides Induced by Neutral Cluster Impact. Rapid Commun. Mass Spectrom. 2013, 27, 1090−1094. (18) Baur, M.; Gebhardt, C. R.; Dürr, M. Desorption/Ionization Induced by Neutral Cluster Impact as a Soft and Efficient Ionization Source for Ion Trap Mass Spectrometry of Biomolecules. Rapid Commun. Mass Spectrom. 2014, 28, 290−296. (19) Lee, B.-J.; Gebhardt, C. R.; Schröder, H.; Kompa, K.-L.; Dürr, M. Observation of Ionic Desorption Channels in Cluster-induced Desorption of Alkali Halides - Influence of Surface Electronic Properties and Surface Configuration. Chem. Phys. Lett. 2013, 556, 77−81. (20) Kley, C. S.; et al. Atomic-Scale Observation of Multiconformational Binding and Energy Level Alignment of Ruthenium-Based Photosensitizers on TiO2 Anatase. Nano Lett. 2014, 14, 563−569. (21) Rauschenbach, S.; Rinke, G.; Gutzler, R.; Abb, S.; Albarghash, A.; Le, D.; Rahman, T. S.; Dürr, M.; Harnau, L.; Kern, K. TwoDimensional Folding of Polypeptides into Molecular Nanostructures at Surfaces. ACS Nano 2017, 11, 2420−2427. (22) Van Berkel, G. J.; McLuckey, S. A.; Glish, G. L. Electrospray Ionization of Porphyrins Using a Quadrupole Ion Trap for Mass Analysis. Anal. Chem. 1991, 63, 1098−1109. (23) Van Berkel, G. J.; McLuckey, S. A.; Glish, G. L. Electrochemical Origin of Radical Cations Observed in Electrospray Ionization Mass Spectra. Anal. Chem. 1992, 64, 1586−1593. (24) Freitas, M. A.; Hendrickson, C. L.; Emmett, M. R.; Marshall, A. G. High-Field Fourier Transform Ion Cyclotron Resonance Mass Spectrometry for Simultaneous Trapping and Gas-Phase Hydrogen/ Deuterium Exchange of Peptide Ions. J. Am. Soc. Mass Spectrom. 1998, 9, 1012−1019. (25) Portz, A.; Baur, M.; Gebhardt, C. R.; Dürr, M. Mass Spectrometry of Oligopeptides in the Presence of Large Amounts of Alkali Halides Using Desorption/Ionization Induced by Neutral Cluster Impact. Biointerphases 2016, 11, No. 02A316. (26) Cheng, J.; Wucher, A.; Winograd, N. Molecular Depth Profiling with Cluster Ion Beams. J. Phys. Chem. B 2006, 110, 8329−8336. (27) Ichiki, K.; Ninomiya, S.; Nakata, Y.; Honda, Y.; Seki, T.; Aoki, T.; Matsuo, J. High Sputtering Yields of Organic Compounds by Large Gas Cluster Ions. Appl. Surf. Sci. 2008, 255, 1148−1150.
(28) Mochiji, K.; Hashinokuchiy, M.; Moritani, K.; Toyoda, N. Matrix-free Detection of Intact Ions from Proteins in Argon-Cluster Secondary Ion Mass Spectrometry. Rapid Commun. Mass Spectrom. 2009, 23, 648−652. (29) Yokoyama, Y.; Aoyagi, S.; Fujii, M.; Matsuo, J.; Fletcher, J. S.; Lockyer, N. P.; Vickerman, J. C.; Passarelli, M. K.; Havelund, R.; Seah, M. P. Peptide Fragmentation and Surface Structural Analysis by Means of ToF-SIMS Using Large Cluster Ion Sources. Anal. Chem. 2016, 88, 3592−3597.
11036
DOI: 10.1021/acs.jpcb.7b06897 J. Phys. Chem. B 2017, 121, 11031−11036
