ion reactions in biological ... - SAGE Journals

429

S.A. McLuckey, Eur. J. Mass Spectrom. 16, 429–436 (2010) Received: 12 August 2009 n Accepted: 23 September 2009 n Publication: 25 September 2009

European Journal of Mass Spectrometry Proceedings of the 18th International Conference on Mass Spectrometry

The emerging role of ion/ion reactions in biological mass spectrometry: considerations for reagent ion selection Scott A. McLuckey 560 Oval Drive, Department of Chemistry, Purdue University, West Lafayette, IN 47907-2084, USA. E-mail: [email protected]

The advent of ionization methods that can produce multiply charged gaseous ions has enabled the development of gas-phase ion/ion reactions in analytical mass spectrometry. Ion/ion chemistry has proved to be a particularly effective means for converting ions from one type to another and allows for a decoupling of the ionization method from the nature of the ion subjected to tandem mass spectrometry. A growing array of applications has been developed based on a variety of reaction types, including electron transfer, proton transfer, charge inversion, metal transfer etc. Most ion/ion reactions take place following the formation of a stable bound orbit between the reactants. As reactants approach closely enough for chemistry to occur, they can react by small charged particle transfer (i.e. electron transfer and proton transfer) at crossing points in the interaction potential. Alternatively, the reactants can collide to form a relatively long-lived complex. A wide range of chemical reactions can result from the long-lived complex, which include multiple charged particle transfers and covalent bond formation. For a given analyte ion, the major reaction pathway is determined by the characteristics of the reagent ion. An appreciation of the factors that underlie the partitioning of ion/ion reaction products is important in the design and selection of reagent ions to effect transformations of interest. Important considerations for reagent ion selection are discussed here within the context of a generalized scheme for ion/ion reaction dynamics. Keywords: ion/ion reactions, electron transfer, proton transfer, charge inversion, bio-conjugation

Introduction Ion chemistry has long played important roles in molecular mass spectrometry. For example, unimolecular reactions, specifically those that result in bond cleavages, have been the means by which ion structures (i.e. bond connectivities) are generally determined. Dissociation reactions, for example, often underlie the use of mass spectrometry for identification purposes as well as for sequencing of biopolymers. The importance of unimolecular dissociation is reflected, in part, by the wide variety of approaches that have been devised to induce ionic fragmentation. Ion/molecule reactions have also played important roles in many processes associated with mass spectrometry, particularly in high pressure ion formation scenarios, such as chemical ionization and atmospheric ISSN: 1469-0667 doi: 10.1255/ejms.1031

pressure ionization. Ion/neutral collisions are also often used either to thermalize or to excite ions, depending upon the energies of the collisions. For example, energetic collisions are often used to induce subsequent unimolecular dissociation in the widely used collision-induced dissociation (CID) technique in tandem mass spectrometry (MS/MS).1 Thermal energy ion/molecule reactions can also be used within the context of MS/MS to improve specificity.2 While reactions of oppositely charged ions are commonplace in plasma ionization sources and are important processes in plasmas, combustion and interstellar space, ion/ion reactions historically have not been exploited in analytical mass spectrometry due to the fact that reactions of singly charged © IM Publications LLP 2009 All rights reserved

430

The Role of Ion/Ion Reactions in Biological Mass Spectrometry

ions of opposite polarity lead to mutual neutralization. The advent of ionization methods that generate multiply charged ions, principally spray ionization techniques,3 has made the study of ion/ion reactions possible because at least one of the reactants can retain a net charge after a reaction, such that the products are amenable to mass spectrometry. Ion/ ion reactions are generally carried out at near atmospheric pressure prior to the sampling of products into the mass spectrometer4,5 or in the mass spectrometer at pressures in the millitorr range, usually in an electrodynamic ion trap. Ion/ ion reactions have been carried out in stand-alone 3-D ion traps,6,7 stand-alone linear ion traps,8 as well as a variety of hybrid instruments that couple electrodynamic ion traps with other mass analyzers.9–13 A growing number of instrument platforms that can execute ion/ion reaction experiments are becoming commercially available, which suggests that ion/ion reactions will be increasingly applied in bioanalysis. The ability to form multiply charged ions has opened up a range of new capabilities for transforming gaseous ions from one form to another via ion/ion reactions. A range of reaction types has been noted that provides a high degree of flexibility in transforming ions from one type (for example, a multiply protonated molecule) to another (for example, a singlydeprotonated molecule). For a given analyte ion-type, it is the selection of the oppositely-charged reagent ion that determines the partitioning of the reaction products. That is, for a given analyte ion, the major competing mechanisms, such as proton transfer, electron transfer, covalent bond formation, etc. are determined by characteristics of the analyte ion. Hence, it is desirable to be able to use whichever ionization method and conditions lead to the greatest analyte ion yield and to have a suite of reagents that can transform the analyte ion into the form that yields the greatest degree of structural information. The major reaction types and the factors that govern the course of an ion/ion reaction, as they are currently understood, are summarized here.

Thermodynamics and kinetics In the absence of a dielectric medium, virtually all combinations of oppositely charged ions will undergo an exothermic charge neutralization process. In most cases, a charge transfer reaction will take place, although the formation of an electrostatically bound complex is not precluded. The strong unshielded electrostatic attraction associated with oppositely charged ions is accounted for in the thermodynamics associated with the charged particle transfer. For example, in the case of proton transfer from a multiply protonated molecule (M + nH)n+ to a singly charged anion A–, the reaction exothermicity, DHrxn, is given by the difference in proton affinities (PA) of A– and (M + (n – 1)H)(n – 1)+, i.e.

DHrxn = PA{[M + (n – 1)H](n – 1)+} - PA[A–]

(1)

The PAs of anions are, as a rule, much larger than those of neutral molecules and larger still than those of cations. Hence,

ion/ion proton transfer reactions are highly exothermic for any combination of ions where a proton can be readily transferred from the cation. In the analogous electron transfer reaction, reaction enthalphy is determined by the difference between the electron affinity (EA) of the neutral species formed by loss of an electron from the anion and the recombination energy (RE) of the cation, i.e.

DHET = EA[A] − RE[(M + nH)n+]

(2)

All cations have positive recombination energies that generally greatly exceed the electron affinities of neutral molecules. Hence, electron transfer is thermodynamically favorable for virtually all multiply-charged cation/anion combinations. Likewise, there are many other possible reaction channels that are also thermodynamically accessible due to the large potential energy made available from mutual neutralization. In some cases, products from competing processes are observed. However, in many cases, only one reaction channel dominates. Like most processes in mass spectrometry, the reactions are under kinetic control, which makes the understanding of reaction dynamics important in choosing reagent ions for particular applications. Ion/ion reactions can be divided into two major categories: (1) those that take place via small charged particle transfers at crossing points in the energy surface and (2) those that take place through a long-lived chemical complex. Small charged particle transfers can take place via a long-lived complex as well. However, once a long-lived complex is formed, many channels other than simple proton or electron transfer can become competitive. Hence, for a given analyte ion, it is important to consider characteristics of the reagent ion that will maximize the desired dynamics. It is useful to consider the relevant processes on the basis of their cross-sections, as given by: 2 sprocess = Pprocesspbprocess

(3)

where s represents the cross-section and where P represents the average probability that the process will occur at the classic impact parameter b. A key process in the dynamics of ion/ion reactions is the formation of a bound orbit involving the oppositely charged ions. The square of the impact parameter for the formation of the orbit, borb.2, is approximated by the classical Thomson model for a three body interaction:14 2 borb . »

4Z12Z22e4

2

4pe0 (mv 2 )

(4)

where Z1 and Z2 are the unit charges of the ions, e is the electron charge, v is the relative velocity, and µ is the reduced mass. This phenomenon is important because it appears to be the rate determining step for bio-ion/ion reactions. Due to the long-range electrostatic attraction, ions of opposite charge can be captured by their mutual attraction at distances that are significantly greater than those at which chemistry can occur. In the initial interaction between two bodies, the orbit is unbound in the absence of a mechanism for removal of

S.A. McLuckey, Eur. J. Mass Spectrom. 16, 429–436 (2010)

431

some of the initial relative translational energy. Mechanisms of the reactants. Therefore, it is important to understand for the removal of translational energy include a collision with the factors that affect the relative magnitudes of these relaa third body and so-called “tidal” effects whereby the electric tionships when selecting reagents. The following sections field of the orbiting pair can induce internal motion within the mention important applications of the major reaction types particle,15 thereby decreasing the relative translational motion. and emphasize the considerations that go into maximizing These effects can also reduce the orbit size until distances the process of interest. Figure 1 shows a depiction of pb2 for are reached where chemical reactions can occur. Relation (4) the four processes indicated here, i.e. bound orbit formation, predicts a Z 2 dependence for ion/ion reaction rates, which has electron transfer, proton transfer and intimate (hard-sphere) been observed experimentally.16,17 Hence, the rate of capture collision. into the bound orbit is rate-determining because all combinaScheme 1 shows a kinetic scheme for the various competing tions of reactants will undergo some type of charge neutrali- and consecutive processes for the generic reaction between a zation reaction once the orbit decreases to the point at which multiply protonated analyte, (M + nH)n+, and a singly charged chemistry can occur. anion, A–. The reaction rate is determined by the formation rate for the bound orbit, whereas the nature of the products is deter- Electron transfer mined at closer interactions. The two important categories The major motivation for implementing ion/ion reactions in include small charged particle transfer at a crossingapoint andpointcommercially available mass spectrometers is to transfer a crossing and ion/ion reaction throughtandem a long-lived complex. In the case of electron crossing point, for example, the square of the impact parameter electron transfer, ion/ion reaction through a long-lived complex. In the case of access structural information from peptidefor and protein cationsis given by: EQN (5) electron transfer at a crossing point, for example, the square via electron transfer dissociation (ETD),8 the ion/ion analog to where the electron transfer distance, rET, for ground state reactants and products can be estimated f of the impact parameter for electron transfer, is EQN given electron capture dissociation (ECD).19 ETD and ECD are comple(6) by: mentarycan to conventional CID and are particularly useful for the Analogous relationships be written for proton transfer, i.e.: é 2Z1Z2e2 ùú 2 2 ê EQN (7) (5) localization of post-translational modifications (PTMs), such bET » rET 1 + ê 4pe r mv 2 ú and: êë úû as phosphorylation, sulfation, glycosylation and disulfide link0 ET EQN (8) ages.20,21 Essentially, all of the issues currently under discusThe other major process is an intimate collision between the reactants that allows for short range io where the electron transfer distance, rET, for ground state hydrogen sion regarding underlying mechanisms of collisions ECD alsoare apply to to here as dipole-dipole, bonding,the dispersion interactions, etc. Such referred reactants and products can be estimated from: ETD. Unlike ECD, however, ETD competes with proton transfer, sphere” to distinguish them from small charged particle transfers at crossing points. The square of impact parameter for agenerally hard sphere collision is given by18: which does not lead to fragmentation, as well as any Z1Z2e2 rET » EQN (9) (6) processes that might go through a long-lived chemical complex. 4pe0DHET The various relationships given here are revealing in several ways. First, relation (4) is always signific This issue is discussed here because it is unique to ETD. The larger than relations (5), (7), and (9), which is consistent with the formation of the bound orbit being r Analogous relationships can be written for protonlimiting. transfer, competition between electron transfer and However, the relative magnitudes of relations (5), (7), and (9) proton vary withtransfer the nature of the reac Therefore, it is important to understand the factors that context affect theof relative magnitudes i.e.: is of particular importance in the maximizing ETDof these relati when selecting reagents. The following sections mention important applications of because proton transfer is always a potentially facile processthe formajor reactio é and emphasize the considerations that go into maximizing the process of interest. Figure 1 shows a 2Z1Z2e2 ùú 2 2 ê multiply protonated analyte ions. In fact, no ETD reagent anion bPT » rPT ê1+ the four processes indicated here, i.e., bound orbit formation, electron transfer, proton tran of πb2 for(7) 2ú êë 4pe0rPT mv úû to date hascollision. been shown to avoid at least some proton transfer. and intimate (hard-sphere) and:

rPT »

Z1Z2e2 4pe0DHPT

(8)

The other major process is an intimate collision between the reactants that allows for short range ion–dipole, dipole– dipole, hydrogen bonding, dispersion interactions, etc. Such collisions are referred to here as “hard-sphere” to distinguish them from small charged particle transfers at crossing points. The square of the impact parameter for a hard sphere collision is given by:18

é 2Z1Z2e2 ùú b2h-s » rh2-s êê1+ 2ú êë 4pe0rh-smv úû

(9)

The various relationships given here are revealing in several ways. First, relation (4) is always significantly larger than relations (5), (7), and (9), which is consistent with the formation of the bound orbit being rate-limiting. However, the relative magnitudes of relations (5), (7) and (9) vary with the nature

bPT

bh‐s

bET borbit

Figure 1. Depiction of pb2 for borbit, bET, bPT, and bh–s.

Figure 1 – Depiction of πb2 for borbit, bET, bPT, and bh-s.

432


(M+nH)(n-1)+•*+A• [(M+nH+A)](n-1)+

ET

(M+nH)n++A-

orbit

A(M+nH)n+

adduct formation h-s

[(M+nH+A)](n-1)+*

PT

(M+(n-1)H)(n-1)++AH

covalent bond formation/cleavage

Scheme 1. Kinetic scheme for an ion/ion reaction of a multiply protonated analyte and a singly charged anion.

The large majority charged ionsfor thus examined of the major considerations in considering the compeSchemeof1singly – Kinetic scheme anfar ion/ion reaction ofAafew multiply protonated analyte and a singly charged as reagents anion. for polyprotonated peptides have shown proton tition between electron transfer and proton transfer are illustransfer to be the dominant process. This observation appears trated in Figure 2, which shows portions of the long-range 22,23 to be rationalized on the basis of Landau–Zener theory, interactionion/ion potentials for electron transfer and proton transfer Electron transfer. The major motivation for implementing reactions in commercially available which has been applied to gas-phase ion/ion reactions,24,25 for a given analyte/reagent combination. including thetandem case ofmass multiply-charged cations reaction with information The figure shows theand entrance forelectron the reaction spectrometers is to in access structural from peptide proteinchannel cations via singly-charged anions.26,27 of the generic analyte ion (M + nH)n+ with a singly charged transfer dissociation (ETD)8, the ion/ion analog to electron capture dissociation (ECD)19. ETD and ECD are complementary to conventional CID and are particularly useful for the localization of post-translational (M+nH)n+ + Amodifications (PTMs), such as phosphorylation, sulfation, glycosylation, and disulfide linkages20,21. Essentially (n-1)+• discussion all of the issues currently regarding the underlying mechanisms of ECD also apply to ETD. (M+nH)(n-1)+• (M+nH)under • + A• +A Unlike ECD, however, ETD competes with proton transfer, which generally does not lead to fragmentation, as

ΔVrET

≈

(M+nH) + A is discussed here well as any processes that might go through a long-lived chemical complex. This issue n+

-

rET

ΔHand because it is unique to ETD. The competition between electron transfer proton transfer is of particular rxn,ET

≈

V(r)

(M+nH)(n-1)+• + A•

n+ + Aimportance in(M+nH) the context of maximizing ETD because proton transfer is always a potentially facile process

ΔHrxn,PT

(n-1)+ + to for multiply protonated analyte ions. In fact, no ETD reagent anion to date(M+(n-1)H) has been shown HAavoid at least

some proton transfer. The large majority of singly charged ions thus far examined as reagents for

rPT

polyprotonated peptides have shown proton transfer to be the dominant process. This observation appears to

(M+A+(n-1)H)(n-1)+ long-lived complex

r((M+nH)n+--- A-) Figure 2. Hypothetical portions of the long range interaction potentials for the reaction of a generic multiply protonated molecule with a singly charged anion.


reagent anion A–. Two exit channels are indicated that reflect electron transfer to yield (M + nH)(n – 1)+• and A• (note that the anion is indicated here as an even-electron species but can also be a radical anion, in which case the neutral product would be an even-electron molecule, A) and proton transfer to yield [M + (n – 1)H](n – 1)+ and AH. The points at which the entrance channel crosses the exit channels are those where small charged particle transfers can occur. As indicated in the hypothetical case shown in Figure 2, the crossing point for electron transfer occurs at larger distances than that for proton transfer. This is a common situation because, for most analyte ion/reagent ion combinations, DHET πbPT2. Therefore, if PET is high, there is a good possibility that the cross-section for electron transfer will be greater than that for proton transfer. Characteristics of both the cation and the anion determine the magnitude of PET, although all the factors that may play important roles have not been fully explored. In the context of Landau–Zener theory, the probability for a net electron transfer is governed by the conditions in the region of the curve crossing (see expanded region of Figure 2). Of particular importance is DVrET, which is the shortest distance between the adiabatic curves at the avoided crossing. The energy gap between the adiabatic curves, DVrET, indicates the strength of electronic coupling between adiabatic states. The other important factors are the difference in the slopes of the entrance and exit channels and the Franck–Condon factors associated with the transitions between vibrational states of the reactants and products. It has been demonstrated that important characteristics of the anion include a low electron affinity for the corresponding neutral species, which affects DVrET and the difference of the slopes, and favorable Franck–Condon factors for electron loss from the anion.27 If either of these two criteria is not met, proton transfer (or some other process) tends to dominate. Most singly charged anions that are readily formed in high abundance tend to have corresponding neutral species with relatively high electron affinities, which is a major reason why there are far fewer reagent anions for electron transfer to polypeptide cations than there are proton transfer reagents. Important factors associated with the cation that affect PET have not been explored in detail. However, it is clear that charge state, or perhaps more accurately, the magnitude of the electric field associated with the cation, affects recombination energy of the cation, which plays a role in determining DVrET, as well as the difference of the slopes of the entrance and exit channels.

Proton transfer Ion/ion single proton transfer reactions have proved to be particularly useful for manipulating the charge states of multiply-protonated and multiply-deprotonated biopolymers. A number of applications have been demonstrated that take advantage of this capability, including the simplification of electrospray mass spectra derived from mixtures, 5,28 the simplification of product ion spectra derived from multiplycharged parent ions29,30 and the concentration of signal from

433

a range of charge states into a single charge state via a technique referred to as “ion parking”.31 For all such applications, it is desirable that the reaction leads exclusively to proton transfer without contributions from fragmentation or adduct formation. The former is generally not observed in the deprotonation of multiply protonated biopolymers but has been noted in the protonation of multiply deprotonated species.32 The latter is often observed33 and provides clear evidence for contributions from intimate collisions that form longlived chemical complexes. Proton transfer can readily occur from such a complex and, for many reactant ion combinations, this process probably dominates over proton transfer at a crossing point. The latter process is probably most likely when the electric field is high, as when one or both of the reactants is highly charged, because proton transfer at a crossing point has a higher Z dependence than formation of an intimate complex. 34 An example of evidence for the formation of a long-lived complex in competition with proton transfer reactions at crossing points is shown in Figure 3, which shows the positive ion spectrum generated from the reaction of the ubiquitin (M + 11H)11+ ion, denoted in the figure as U 11+, with the ubiquitin (U – 5H) 5– anion. The formation of a dimeric product resulting from the addition of the two ubiquitin ions to give (2U)6+ is apparent (some U3+ may also contribute but data collected using different proteins as the cation and anion also clearly showed complex formation34). Also, a distribution of partially deprotonated ions (see, for example, the U7+ – U4+ series of ions) is apparent. A range of anionic charge states of ubiquitin complementary to the cationic charge states was also noted in the negative ion spectrum (data not shown). (No complex formation was noted because the cationic reactant carried more total charge.) These are believed to be formed via proton transfers that occur without going through a long-lived dimeric complex because it is highly unlikely that, once formed, the complex will dissociate with charge separation. For the most part, applications that employ single proton transfer ion/ion reactions, including those that rely on several sequential single proton transfers, are little affected by the relative contributions of the two possible proton transfer mechanisms (i.e. proton transfer at a curve crossing versus via a long-lived complex), provided no long-lived complexes survive. However, if there are relatively strong non-covalent interactions between the reactants that go through the longlived collision complex channel, adduct formation can result.33 If this is a problem insofar as adduct formation complicates the spectrum, subsequent collisional excitation generally results in loss of the adduct(s). Perfluorocarbon anions have proved to be particularly useful as single proton transfer reagents for multiply protonated analyte species because they do not interact strongly with the analyte species and show little propensity for adduct ion formation.

Long-lived complex formation While electron transfer and proton transfer can take place without formation of a long-lived complex, all ion/ion reactions

434


U11+/U5-

1.5x104

(2U)6+/U3+

Abundance (arb. units)

U11+

U6+

9x103

U5+ U7+ 3x103

500

U4+

U9+

(2U)5+ 1500

2500

m/z

3500

U2+ 4500

Figure 3. Positive product ion spectrum from the reaction of the 11+ ion of bovine ubiquitin with the 5- ion of bovine ubiquitin. Products from proton transfer at crossing points and complex formation are both noted.

Figure 3 – Positive product ion spectrum from the reaction of the 11+ ion of bovine types can take place through with a long-lived withProducts basic groups in peptides, which results ubiquitin the 5-complex. ion ofIndeed, bovine strongly ubiquitin. from proton 35transfer at in many reaction types can onlypoints proceedand through a long-lived adduct with the peptide ion, while the aldehyde crossing complex formation are formation both noted.

complex. Two prominent examples discussed here include group reacts with a primary amine to form the Schiff base. For the most part, applications that employ single proton transfer reactions, including those that rely charge inversion reactions and covalent bond formation. The presence ofion/ion a “sticky” group, like the sulfonic acid, is Ion/ion charge inversion involves the transfer of two or important because adduct formation provides sufficient time 35,36 several sequential single proton transfers, are littlespecific affected by the bond relative contributions of the two For more charges in aon single ion/ion encounter. This cannot for the covalent formation reaction to occur. occur efficiently if the charges are transferred one-at-a- example, deprotonated 4-formyl-1,3-benzenedicarboxylic acid possible proton transfer mechanisms (i.e., proton transfer at a curve crossing versus via a long-lived time in separate ion/ion encounters because the analyte reacts with primary amine containing peptide ions via proton species would be neutralized in one of the steps. Hence, transfer. Adduct formation is minimal because the carboxylic complex), provided no long-lived complexes survive. However, if there are relatively strong non-covalent when charge inversion is the goal, it is desirable to minimize acid group engages in weaker interactions with basic sites in 35 single charge transfers at crossing points maximize the peptide such thatcollision the lifetime of thechannel, complex adduct is too short interactions between theand reactants thatthe go through the long-lived complex hard-sphere cross-section. Multiply-charged ions derived for Schiff base formation. Hence, the selection of reagent If this by is primary a problem insofar adduct formation complicates the of spectrum, can result33. either from d endrimers formation that are terminated ions for as gas-phase bio-conjugation reactions, which Schiff amine groups, for positively charged reagents, or carboxylate base formation is an example, will require functionalities that subsequent generally in lossthe of lifetime the adduct(s). Perfluorocarbon anions have groups, for negatively chargedcollisional reagents,excitation have proved to beresults enhance of the complex in addition to functional useful for charge inversion. The fact that dendrimers give groups that engage in reactions with the bio-ion. proved to be particularly useful as single proton transfer reagents for multiply protonated analyte species rise to a range of charge states upon electrospray ionization and are available in a range of generations (i.e. sizes) allows because they do not interact strongly with the analyte species and show little propensity for adduct ion for flexibility in choosing the size and charge of the reagent. Furthermore, primary amines and carboxylate groups do not formation. engage in particularly strong dipole–dipole interactions with Gas-phase ion/ion chemistry is beginning to play a significant complex While electron and proton transfer can take Major place applications without most analyte ions, Long-lived which minimizes theformation. formation of adduct roletransfer in bioanalysis via mass spectrometry. ions. However, the extent of adduct formation is also depen- now rely primarily on electron transfer and proton transfer formation of a long-lived complex, allas ion/ion types canreaction take place through a long-lived complex. dent on the chemical composition of the analyte ion, has reactions reactions. New types, however, are being examined been noted for charge inversion of peptides with and without that include charge inversion, metal ion transfer,39,40 and biobasic side chains.37 conjugation reactions. For a given analyte ion, the identity of A new area in ion/ion chemistry research is the explora- the reagent ion is key to determining which reaction types are tion of specific chemistries for the gas-phase modification of likely to be observed. The considerations summarized here bio-ions. An example is Schiff base formation in polypeptide are helpful in considering the selection of the reagent ion for ions that contain a primary amine group via the reaction with specific applications. Recognition of the factors that affect ion/ deprotonated 4-formyl-1,3-benzenedisulfonic acid.38 This is ion reaction product partitioning will facilitate the developan example of the use of a bifunctional reagent to effect a ment of new reagents that significantly expand the potential transformation of interest. The sulfonic acid group interacts role for ion/ion chemistry in analytical mass spectrometry.

Summary and conclusions


Acknowledgment The research conducted in the author’s laboratory that has provided the information described here was sponsored by the Office of Basic Energy Sciences, Office of Sciences, US Department of Energy, under Award No. DE-FG02-00ER15105

References 1. A.K. Shukla and J.H. Futrell, “Tandem mass

spectrometry: dissociation of ions by collisional activation”, J. Mass Spectrom. 35, 1069 (2000). doi: 10.1002/1096-9888(200009)35:93.0.CO;2-C 2. S. Gronert, “Quadrupole ion trap studies of fundamental organic reactions”, Mass Spectrom. Rev. 24, 100 (2005). doi: 10.1002/mas.20008 3. Electrospray Ionization Mass Spectrometry, Ed by R.B. Cole. John Wiley & Sons Inc., New York, USA (1997). 4. R.R. Ogorzalek Loo, H.R. Udseth and R.D. Smith, “A new approach for the study of gas-phase ion-ion reactions using electrospray ionization”, J. Am. Soc. Mass Spectrom. 3, 695 (1992). doi: 10.1016/1044-0305(92)87082-A 5. M. Scalf, M.S. Westphall, J. Krause, S.L. Kaufman and L.M. Smith, “Controlling charge states of large ions”, Science 283, 194 (1999). doi: 10.1126/science.283.5399.194 6. W.J. Herron, D.E. Goeringer and S.A. McLuckey, “Ion–ion reactions in the gas-phase: proton transfer reactions of protonated pyridine with multiply-charged oligonucleotide anions”, J. Am. Soc. Mass Spectrom. 6, 529 (1995). doi: 10.1016/1044-0305(95)00199-N 7. R.G. Hartmer, D.A. Kaplan, C.R. Gebhardt, T. Lederheil and A. Brekenfeld, “Multiple ion/ion reactions in the 3D ion trap: selective reagent anion production for ETD and PTR from a single compound”, Int. J. Mass Spectrom. 276, 82 (2008). doi: 10.1016/j.ijms.2008.05.002 8. J.E.P. Syka, J.J. Coon, M.J. Schroeder, J. Shabanowitz and D.F. Hunt, “Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry”, Proc. Natl. Acad. Sci. USA 101, 9528 (2004). doi: 10.1073/ pnas.0402700101 9. Y. Xia, J. Wu, F.A. Londry, J.W. Hager and S.A. McLuckey, “Mutual storage mode ion/ion reactions in hybrid linear ion trap”, J. Am. Soc. Mass Spectrom. 16, 71 (2005). doi: 10.1016/j.jasms.2004.09.017 10. Y. Xia, P.A. Chrisman, D.E. Erickson, J. Liu, X. Liang, F.A. Londry, M.J. Yang and S.A. McLuckey, “Implementation of Ion/ion reactions in a quadrupole/time-of-flight tandem mass spectrometer”, Anal. Chem. 78, 4146 (2006). doi: 10.1021/ac0606296 11. R.G. Hartmer, D.A. Kaplan, C. Stoermer, M. Lubeck and M.A. Park, “Data-dependent electron transfer dissociation of large peptides and medium size proteins in a QTOF instrument on a liquid chromatography t imescale”,

435

Rapid Commun. Mass Spectrom. 23, 273 (2009). doi: 10.1002/rcm.4145 12. G.C. McAlister, W.T. Berggren, J. Griep-Raming, S. Horning, A. Makarov, D. Phanstiel, G.C. Stafford, D.L. Swaney, J.E.P. Syka, V. Zabrouska and J.J. Coon, “A proteomics grade electron transfer dissociation-enabled hybrid linear ion trap-orbitrap mass spectrometer”, J. Proteome Res. 7, 3127–3136 (2008). doi: 10.1021/ pr800264t 13. D.A. Kaplan, R. Hartmer, J.P. Speir, C. Stoermer, D. Gumerov, M.L. Easterling, A. Brekenfeld, T. Kim, F. Laukien and M.A. Park, “Electron transfer dissociation in the hexapole collision cell of a hybrid quadrupole– hexapole Fourier transform ion cyclotron resonance mass spectrometer,” Rapid Commun. Mass Spectrom. 22, 271 (2008). doi: 10.1002/rcm.3356 14. J.J. Thomson, “Recombination of gaseous ions, the chemical combination of gases, and monomolecular reactions,” Philos. Mag. 47, 337 (1924). 15. D.R. Bates and W.L. Morgan, “New recombination mechanism—tidal termolecular ionic recombination”, Phys. Rev. Lett. 64, 2258 (1990). doi: 10.1103/ PhysRevLett.64.2258 16. J.L. Stephenson, Jr and S.A. McLuckey, “Ion/ion reactions in the gas phase: proton transfer reactions involving multiply charged proteins”, J. Am. Chem. Soc. 118,7390 (1996). doi: 10.1021/ja9611755 17. J.M. Wells, P.A. Chrisman and S.A. McLuckey, “Formation and characterization of protein–protein complexes in vacuo”, J. Am. Chem. Soc. 125, 7238 (2003). doi: 10.1021/ja035051l 18. B.H. Mahan in Advances in Chemical Physics, Ed by I. Prigogine and S.A. Rice. John Wiley & Sons Inc., New York, USA, pp. 1–40 (1973). doi: 10.1002/9780470143735. ch1 19. R.A. Zubarev, N.L. Kelleher and F.W. McLafferty, “Electron capture dissociation of multiply charged protein cations. a nonergodic process”, J. Am. Chem. Soc. 120, 3265 (1998). doi: 10.1021/ja973478k 20. R.A. Zubarev, “Reactions of polypeptide ions with electrons in the gas phase”, Mass Spectrom. Rev. 22, 57 (2003). doi: 10.1002/mas.10042 21. J.J. Coon, “Collisions or electrons? Protein sequence analysis in the 21st century,” Anal. Chem. 81, 3208 (2009). doi: 10.1021/ac802330b 22. L.D. Landau, “On the theory of transfer of energy at collisions II”, Phys. Z (USSR) 2, 46 (1932). 23. C. Zener, “Non-adiabatic crossing of energy levels,” Proc. Royal Soc. Lond. A136, 696 (1932). doi: 10.1098/ rspa.1932.0165 24. R.E. Olson, F.T. Smith and E. Bauer, “Estimation of coupling matrix elements for one-electron transfer systems”, Appl. Optics 10, 1848 (1971). doi: 10.1364/AO.10.001848 25. R.E. Olson, “Absorbing sphere model for calculating ionion recombination total cross-sections,” J. Chem. Phys. 56, 2979 (1972). doi: 10.1063/1.1677632

436


26. I. Anusiewicz, J. Berdys-Kochanska and J. Simons,

34. J.M. Wells, P.A. Chrisman and S.A. McLuckey,

“Electron attachment step in electron capture dissociation (ECD) and electron transfer dissociation (ETD)”, J. Phys. Chem. A 109, 5801 (2005). doi: 10.1021/jp050218d 27. H.P. Gunawardena, M. He, P.A. Chrisman, S.J. Pitteri, J.M. Hogan, B.D.M. Hodges and S.A. McLuckey, “Electron transfer versus proton transfer in gas-phase ion-ion reactions of polyprotonated peptides”, J. Am. Chem. Soc. 127, 12627 (2005). doi: 10.1021/ja0526057 28. J.L. Stephenson, Jr and S.A. McLuckey, “Ion/ion proton transfer reactions for protein mixture analysis”, Anal. Chem. 68, 4026 (1996). doi: 10.1021/ac9605657 29. J.L. Stephenson, Jr and S.A. McLuckey, “Simplification of product ion spectra derived from multiply-charged parent ions via ion/ion chemistry”, Anal. Chem. 70, 3533 (1998). doi: 10.1021/ac9802832 30. J.J. Coon, B. Ueberheide, J.E.P. Syka, D.D. Dryhurst, J. Ausio, J. Shabanowitz and D.F. Hunt, “Protein identification using sequential ion/ion reactions and tandem mass spectrometry,” Proc. Natl. Acad. Sci. USA 102, 9463 (2005). doi: 10.1073/pnas.0503189102 31. S.A. McLuckey, G.E. Reid and J.M. Wells, “Ion parking during ion/ion reactions in electrodynamic ion traps”, Anal. Chem. 74, 336 (2002). doi: 10.1021/ac0109671 32. J.J. Bowers, B.D.M. Hodges, O.M. Saad, J.A. Leary and S.A. McLuckey, “Proton hydrates as soft ion/ion proton transfer reagents for multiply deprotonated biomolecules”, Int. J. Mass Spectrom. 276, 153 (2008). doi: 10.1016/j.ijms.2008.05.038 33. J.L. Stephenson, Jr and S.A. McLuckey, “Gaseous protein cations are amphoteric”, J. Am. Chem. Soc. 119, 1688 (1997). doi: 10.1021/ja9632973

“Formation and characterization of protein-protein complexes in vacuo”, J. Am. Chem. Soc. 125, 7238 (2003). doi: 10.1021/ja035051l 35. M. He, J.F. Emory and S.A. McLuckey, “Reagent anions for charge inversion of polypeptide/protein cations in the gas phase”, Anal. Chem. 77, 3173 (2005). doi: 10.1021/ ac0482312 36. J.F. Emory and S.A. McLuckey, “Charge inversion of polypeptide anions using protein and dendrimer cations as charge inversion reagents”, Int. J. Mass Spectrom. 276, 102 (2008). doi: 10.1016/j.ijms.2008.04.013 37 J.F. Emory and S.A. McLuckey, “The role of amino acid composition on the charge inversion of deprotonated peptides via gas-phase ion/ion reactions”, J. Am. Soc. Mass Spectrom. 20, 180 (2008). doi: 10.1016/j. jasms.2008.08.015 38. H. Han and S.A. McLuckey, “Selective covalent bond formation in polypeptide ions via gas-phase ion/ion reaction chemistry”, J. Am. Chem. Soc., in press. 39. K.A. Newton and S.A. McLuckey, “Gas-phase peptide/ protein cationizing agent switching via ion/ion reactions”, J. Am. Chem. Soc. 125, 12404 (2003). doi: 10.1021/ ja036924e 40. H.P. Gunawardena, R.A.J. O’Hair and S.A. McLuckey, “Selective disulfide bond cleavage in gold(i) cationized polypeptide ions formed via gas-phase ion/ion cation switching,” J. Proteome. Res. 5, 2087 (2006). doi: 10.1021/ pr0602794