Perturbing Peptide Cation-Radical Electronic States by Thioxoamide

Article pubs.acs.org/JPCA

Perturbing Peptide Cation-Radical Electronic States by Thioxoamide Groups: Formation, Dissociations, and Energetics of Thioxopeptide Cation-Radicals Magdalena Zimnicka,† Thomas W. Chung, Christopher L. Moss, and František Tureček* Department of Chemistry, Bagley Hall, Box 351700, University of Washington, Seattle, Washington 98195-1700, United States S Supporting Information *

ABSTRACT: Thioxodipeptides Gly-thio-Lys (GtK), Ala-thio-Lys (AtK), and Ala-thio-Arg (AtR) in which the amide group has been modified to a thioxoamide were made into dications by electrospray ionization and converted to cation-radicals, (GtK + 2H)+•, (AtK + 2H)+•, and (AtR + 2H)+•, by electron transfer dissociation (ETD) tandem mass spectrometry using fluoranthene anion-radical as an electron donor. The common and dominant dissociation of these cation-radicals was the loss of a hydrogen atom. The dissociation products were characterized by collision-induced dissociation (CID) multistage tandem mass spectrometry up to CID-MS5. The ground electronic states of several (GtK + 2H)+•, (AtK + 2H)+•, and (AtR + 2H)+• conformers were explored by extensive ab initio and density functional theory calculations of the potential energy surface. In silico electron transfer to the precursor dications, (GtK + 2H)2+, (AtK + 2H)2+, and (AtR + 2H)2+, formed zwitterionic intermediates containing thioenol anion-radical and ammonium cation groups that were local energy minima on the potential energy surface of the ground electronic state. The zwitterions underwent facile isomerization by N-terminal ammonium proton migration to the thioenol anion-radical group forming aminothioketyl intermediates. Combined potential energy mapping and RRKM calculations of dissociation rate constants identified N−Cα bond cleavages as the most favorable dissociation pathways, in a stark contrast to the experimental results. This discrepancy is interpreted as being due to the population upon electron transfer of low-lying excited electronic states that promote loss of hydrogen atoms. For (GtK + 2H)+•, these excited states were characterized by time-dependent density functional theory as A−C states that had large components of Rydberg-like 3s molecular orbitals at the N-terminal and lysine ammonium groups that are conducive to hydrogen atom loss.

■

radical dissociations,9 and dissociation dynamics in the ground and excited electronic states of model systems.10 Nevertheless, the effects on ExD of peptide structure variations are best explored by well-designed experiments.11 We have recently reported a joint experimental and computational study of an aminothioketyl radical produced by hydrogen atom addition to N-methylthioacetamide.12 The energetics and dissociations of this radical system showed some analogies with those of Nmethylacetaminoketyl radical, which is a prototypical peptide radical model.13 Here, we extend these studies to cation-radicals derived from electron−ion recombination of doubly protonated thioxopeptides Gly-thio-Lys, Ala-thio-Lys, and Ala-thio-Arg.

INTRODUCTION Electron attachment to peptide and protein ions produces reactive radical intermediates that undergo facile dissociations by the loss of small neutral fragments, disulfide bond cleavages, and backbone cleavages mainly affecting the bonds between the amide nitrogens and Cα carbons of the adjacent amino acid residues (N−Cα bond cleavage for short). When applied to multiply charged peptides or proteins, the charge-reduced intermediates and their fragments are detectable by mass spectrometry and can provide valuable information on the peptide amino acid sequence. The experimental methods utilizing this phenomenon are electron capture1 and electron transfer dissociations,2 which are respectively abbreviated as ECD and ETD or collectively referred to as ExD. The N−Cα bond cleavage in ExD has been found to be remarkably sensitive to the electronic properties of the peptide radical. The presence in the peptide ion of groups functioning as electron traps,3 hydrogen atom traps,4 complexation with transition metal cations,5 or modifications to the peptide backbone by introduction of ester linkages6 or β-amino acid residues7 have been shown to have major, mainly deleterious, effects on the backbone dissociations. Current theoretical models of ExD provide general guidelines as to the electronic states accessed by electron-ion recombination,8 energetics of peptide cation© XXXX American Chemical Society

■

EXPERIMENTAL SECTION Materials and Methods. The thioxopeptides GtK, AtK, and AtR were custom synthesized by LifeTein LLC (South Plainfield, NJ, USA) and used as received. Electron-transfer dissociation (ETD) and collision-induced dissociation (CID)

Special Issue: Peter B. Armentrout Festschrift Received: June 14, 2012 Revised: July 5, 2012

A

dx.doi.org/10.1021/jp305865q | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

within 2.9 kJ mol−1 root-mean-square deviation (rmsd) (Table 1).

mass spectra were measured on a Thermo Fisher (San Jose, CA, USA) LTQ XL linear ion trap instrument, outfitted with a chemical ionization source for the production of fluoranthene anion radials as ETD reagent. Precursor dications were isolated with a window of 2−4 m/z units to accommodate the nearest 13 C isotopologues and allowed to react with fluoranthene anions for 100−500 ms. The ion−ion reactions showed pseudofirst order kinetics, as illustrated by the time dependence of the precursor ion relative intensity for (GtK + 2H)2+ (Figure S1, Supporting Information). Selected measurements were carried out on a Thermo Fisher (San Jose, CA, USA) LTQ Orbitrap Velos hybrid mass spectrometer equipped with an ion source for ETD. The mass resolution was set to 60 000 to obtain accurate m/z ratios for ion characterization by elemental compositions. Conformational Search and ab Initio Calculations. A conformational search was performed with a home-built search engine, as described previously.14 In the first step, molecular dynamics calculations were performed using NAMD15 and the CHARMM force field.16 Molecular dynamics force-field parameters have not previously been developed for peptides containing thioxoamide groups. We obtained charge distributions and force constants with ab initio calculations to develop a set of parameters for the thioxopeptides. Other residues were parametrized using established values from the CHARMM force-field.16 Conformational search was performed for tautomers protonated in the side-chain of the basic residue (Lys or Arg) and at the N-terminal amino group. Semiempirical and ab initio calculations were performed using Gaussian09.17 Once the force-field parameters had been determined, replica exchange molecular dynamics (REMD)18 calculations were performed for 10 ns with a step size of 1 fs with 8 temperature replicas from 300 to 800 K. This created 800 000 conformers for each ion tautomer. One thousand structures from each replica were selected at regular intervals for a total of 8000 initial candidate structures. Each candidate structure was then optimized with PM6.19 The PM6 optimized structures were then analyzed for potential hydrogen bonds. Structures with the same hydrogen bond patterns were grouped together and the lowest energy structure from each group was taken to form a new list of candidate structures. The single-point energy for each candidate structure was calculated at the B3LYP/631+G(d,p) level,20 80 structures that had the lowest energies were reoptimized with B3LYP/6-31++G(2d,p), and the local minima were confirmed with frequency calculations. Scaled harmonic frequencies along with moments of inertia were used to calculate 298 K enthalpies and entropies. Tables of complete Cartesian coordinates and energies for optimized structures can be obtained from the corresponding author upon request. Single-point energies were calculated for each reoptimized structure with B3LYP/6-311++G(2d,p) and Moller−Plesset theory,21 MP2(frozen core)/6-311++G(2d,p), and averaged (B3-MP2) to cancel known errors in each method.22 Additional sets of single-point energies were obtained for dications by MP2/6-311++G(3df,2p) and coupled-cluster theory23 calculations with single, double, and perturbative triple excitations,24 CCSD(T)/6-311G(d,p) single-point calculations. The MP2 and CCSD(T) energies were combined using the standard linear formula: 25 E[CCSD(T)/6-311++G(3df,2p)] ≈ E[CCSD(T)/6-311G(d,p)] + E[MP2/6-311++G(3df,2p)] − E[MP2/6-311G(d,p)]. The B3-MP2 relative energies for the dications showed a very good agreement with the effective CCSD(T)/6-311++G(3df,2p) relative energies, which were

Table 1. Relative Energies of Thioxopeptide Dications relative energy (kJ/mol)a

ion

B3LYP/631+ +G(2d,p)

B3-MP2/6 -311+ +G(2d,p)

CCSD(T)/6311+ +G(3df,2p)

expected population [%]c

GtK12+ GtK22+ GtK32+ AtK12+ AtK22+ AtK32+ AtR12+ AtR22+ AtR32+ AtR42+

0 4 33 0 4 10 0 3 5 14

0 (0)b 2 (5)b 30 (27)b 0 (0)b 2 (6)b 9 (6)b 0 (0)b 5 (3)b 7 (3)b 17 (7)b

0 1 32 0 0.1 13

88 12 0 84 8 8 52 18 27 3

a

Relative enthalpies including B3LYP/6-31++G(2d,p) zero-point energy corrections and 298 K enthalpies. bValues in parentheses are relative free energies at 298 K. cPopulation in the gas phase at 298 K.

Geometry optimization of charged-reduced cation-radicals began with the optimized structures of doubly protonated thioxopeptides. Spin unrestricted calculations were performed for all open-shell systems. Stationary points were characterized by harmonic frequency calculations with B3LYP/6-31++G(2d,p) as the local minima or first-order saddle points. Singlepoint energies were calculated with B3LYP/6-311++G(2d,p) and spin-projected MP2/6-311++G(2d,p) and averaged to provide improved relative energies for cation-radicals. The performance of the B3-PMP2 scheme for thioxopeptide cationradicals was benchmarked on the adiabatic recombination energies of (GtK + 2H)2+ ions from effective CCSD(T)/6-31+ +G(3df,2p) calculations. The two sets of data were within 5 kJ mol−1 rmsd, attesting to the reliability of the B3-PMP2 scheme. Excited states were calculated using time-dependent density functional theory (TD-DFT)26 and the B3LYP and M06-2X functionals,27 each with the 6-311++G(2d,p) basis set. Dissociations and isomerizations of charge-reduced ions were investigated by potential energy surface (PES) mapping with UB3LYP/6-31++G(2d,p). Improved energies of transition states were obtained by single-point B3LYP and MP2 calculations with the 6-311++G(2d,p) basis set and combined according to the B3-MP2 scheme. Gradient searches of the PES along the S−H dissociation coordinates did not lead to saddle points, and thus, the transition states were treated differently. Corrected PES profiles along the S−H dissociation coordinates were obtained from single-point energy calculations with B3MP2/6-311++G(2d,p) at several points on the B3LYP PES where the S···H dissociation coordinate was stretched in 0.02 Å steps, while the other internal coordinates were fully optimized. Saddle points determined from polynomial fitting of these profiles were used for frequency and B3-MP2 single-point energy calculations. Unimolecular rate constants were calculated with the Rice− Ramsperger−Kassel−Marcus (RRKM) theory28 using the program of Zhu and Hase29 that was recompiled for MSDOS and run under Windows XP or Windows 7.30 Vibrational state densities were obtained by a direct count of quantum states in 2 kJ mol−1 steps for internal energies up to 600 kJ mol−1 above the threshold. The rotational states were treated B



Article

adiabatically and the microscopic rate constants [k(E,J,K)] were Boltzmann-averaged over the thermal distribution of rotational J and K states pertaining to the ion source temperature.

■

RESULTS Electron-Induced Dissociations of Thioxopeptide Ions. Electrospray ionization of thioxopeptides produced singly and doubly charged ions in a 10:1 ratio. The doubly charged ions were selected by mass and subjected to an ion−ion reaction with fluoranthene anions to yield electron transfer dissociation (ETD) mass spectra. Figure 1a shows the ETD

Figure 2. CID-MS3 mass spectra of mass-selected ETD fragment ions from (GtK + 2H)2+: (a) m/z 220, (b) m/z 203, and (c) m/z 185.

from charge-reduced (GtK + 2H)+• to form vibrationally excited (GtK + H)+ at m/z 220, which undergoes CID-like dissociations. The consecutive dissociations of the ETD m/z 220 ion by loss of H2S (m/z 186), NH3, H2O (m/z 185), and C3H4O (m/z 129) were monitored in the MS4 and MS5 spectra (Figure S2a−c, Supporting Information). The MS3 and MS4 data showed two conspicuous differences from the ETD mass spectrum in the formation of the m/z 100 and m/z 102 fragment ions upon ETD, which were not formed by CID of any of the primary ETD fragments. It may also be noted that the relative intensities of the fragment ions in the ETD mass spectra depended on the ion− ion interaction time. The ETD spectrum obtained after 2000 ms ion−ion interaction time (Figure S3a, Supporting Information) displayed reduced relative intensities of the m/z 203, 157, 129, and 102 fragment ions. Presumably, these changes were due to species-dependent depletion of the ETD fragments by electron transfer from the fluoranthene anion radical. The overall intensity of charge-reduced ETD fragments showed a 4-fold decrease from 300 to 2000 ms interaction time (Figure S4a, Supporting Information). The relative intensities of the m/z 129 and 102 fragment ions showed a pseudofirst order decay with ion−ion interaction time, while the other fragment ions were either less sensitive to charge reduction (m/ z 185) or showed increased relative intensities (m/z 100, 186, and 220; Figure S4b, Supporting Information). Presumably,

Figure 1. ETD mass spectra taken at 100 ms ion−ion interaction time of (a) (GtK + 2H)2+ and (b) (GK + 2H)2+.

mass spectrum of (GtK + 2H)2+ (m/z 110) obtained at a 100 ms ion−ion interaction time. The spectrum displayed chargereduced dissociation products at m/z 220 (220.1112, loss of H), m/z 203 (203.0849, loss of H and NH3), and m/z 186 (186.1238, loss of H and H2S). The identity of these fragment ions was unambiguously confirmed by accurate mass measurements. Further fragments were at m/z 185 (185.0744, combined losses of H, NH3, and H2O), m/z 169 (169.0972), 157, 129 (129.1023, C6H13N2O), 102, and 100. No chargereduced ion was detected at m/z 221, indicating that electron transfer resulted in complete dissociation on the 100 ms time scale of the measurement. The m/z 220 ion from ETD was further isolated and submitted to collision-induced dissociation in an ETD-CIDMS3 experiment. The resulting MS3 mass spectrum (Figure 2a) showed most of the fragment ions that were also present in the ETD mass spectrum of (GtK + 2H)2+ (Figure 1a) with the exception of the m/z 130 fragment ion (130.0863, C6H12NO2) which was weak in the ETD mass spectrum. Likewise, the MS3 mass spectrum of the other major ETD fragment ion, m/z 203 (Figure 2b), showed secondary fragment ions that were also represented in the ETD mass spectrum of (GtK + 2H)2+. These data can be interpreted by ET-induced loss of a hydrogen atom C



Article

(Figure 4a). The dissociations were further studied by obtaining CID-MS3 spectra of the ETD fragments (Figure

these effects are due to a combination of increased fragment formation by electron transfer to (GtK + 2H)2+ and their subsequent depletion by electron transfer. The ETD mass spectrum of (GtK + 2H)2+ is compared to that of the related dipeptide (GK + H)2+ ion (Figure 1b). The latter ion showed similar types of dissociations, e.g., loss of H (m/z 204) followed by elimination of ammonia (m/z 187), water (m/z 186), and the formation of the m/z 130 and 129 ions. The main difference between the ETD mass spectra of (GtK + 2H)2+ and (GK + 2H)2+ was the absence of the abundant y1 ion (m/z 147) from the thioxopeptide, indicating that the thioxoamide C−N bond was not susceptible to protondriven dissociation. ETD of the homologous (AtK + 2H)2+ ion (m/z 117.5, Figure 3a) showed notable differences from that of (GtK +

Figure 4. (a) ETD mass spectrum of (AtR + 2H)2+ ion at m/z 131.6. CID-MS3 spectra of mass-selected ETD fragments (b) m/z 262 and (c) m/z 245.

4b,c). CID of the m/z 262 (M + H)+ ion resulted in a major loss of H2S (m/z 228.1444), followed by loss of ammonia (m/z 211.1180) (Figure 4b). CID of the m/z 245 ion formed the m/ z 203.0841 fragment by loss of CH2N2, presumably HNC NH from the Arg side chain. Further CID of the m/z 203 ion resulted in the loss of a C3H4OS neutral molecule, forming the m/z 115 ion (115.0859). This fragmentation sequence indicates that the ammonia molecules eliminated in the dissociations originated from different parts of the ion, namely, the thioxoalanine N-terminus and the Arg guanidine group. The loss of C3H4OS further indicates a cyclization in an Nterminal deaminated fragment, presumably by nucleophilic attack by the C-terminal carbonyl group. Electrospray of AtR from CD3 OD/D2 O resulted in incorporation of nine deuterons in the ten exchangeable proton positions in the doubly charged ion, producing mainly d9-species at m/z 136.0978. The distribution of the nine deuterons from the exchange was unknown. The ETD mass spectrum of AtR-d92+ (Figure S5, Supporting Information) showed loss of D and H (m/z 270.1822 and 271.1885) in a 2.6:1 ratio, which was substantially lower than the 9:1 ratio for a statistically random loss of H or D. The subsequent loss of ammonia produces the m/z 250.1369 and 251.1432 ions in a 1.5:1 ratio. The increased relative abundance of the m/z 251 ion was consistent with convergent loss of ND3 (20 Da) from the m/z 271 ion and ND2H (19 Da) from the m/z 270 ion. Previous studies reported that electron attachment to the

Figure 3. ETD mass spectra of (AtK + 2H)2+ taken after (a) 100 ms and (b) 500 ms of ion−ion interaction time. (c) CID-MS3 mass spectrum of the mass-selected ETD fragment at m/z 217.

2H)2+. In particular, the (M + H)+ fragment ion at m/z 234 was extremely weak and underwent facile elimination of ammonia to form the major fragment at m/z 217. The latter ion further eliminated water upon CID (m/z 199) and formed the common Lys fragment ion (C6H13N2O, m/z 129, Figure 3c). Facile loss of ammonia also occurred from the doubly charged precursor ion, forming the doubly charged fragment ion at m/z 109. The relative intensity of the m/z 109 ion decreased upon longer ion−ion interaction time as shown for the ETD spectrum taken after 500 ms (Figure 3b). ETD of the (AtR + 2H)2+ ion (m/z 131.5696) formed a weak charge-reduced (M + H)+ ion at m/z 262 (262.1320), an abundant m/z 245 ion (245.1055) by combined loss of H and ammonia, and an m/z 203 fragment (203.0841, C8H15N2O2S) D



Article

Figure 5. B3LYP/6-31++G(2d,p) optimized structures of lowest-energy (GtK + 2H)2+, (AtK + 2H)2+, and (AtR + 2H)2+ ion conformers. The atom color coding is as follows: green = C, blue = N, red = O, yellow = S, and gray = H. Major hydrogen bonding and dipole−dipole interactions are denoted by blue double arrows and the distances are given in Ångstrøms.

methylthioacetamide,12 is slightly higher than the intrinsic basicities of Gly or Ala residues (886−900 kJ mol−1).31 The preferential protonation at the N-terminus in GtK, AtK, and AtR may be due to the σ-electron donating effect of the thioxoamide group, increasing the amino group basicity, and Coulomb repulsion with the Lys or Arg charge that presumably destabilizes the S-protonated tautomers because of a shorter distance between the charge sites. An interesting feature of the GtK and AtK ion structures is that the Lys ε-ammonium group is internally solvated by the carboxyl group in the most stable conformers GtK12+, GtK22+, AtK12+, and AtK22+ (Figure 5) in spite of the fact that the side-chain folding brings the charged Lys NH3 group closer to the N-terminal one, thus increasing Coulomb repulsion. The AtR ion conformers consist of structures exhibiting different modes of internal solvation of the charged guanidinium group (AtR12+−AtR32+, Figure 5). The unfolded structure (AtR42+) is slightly disfavored compared to the folded ones. (GtK + 2H)+•, (AtK + 2H)+•, and (AtR + 2H)+• CationRadicals by Vertical Electron Attachment. Electron attachment to the dications was first studied as a vertical process and then following full geometry relaxation in the charge-reduced species. The vertical ion−electron recombination energies (REv) are summarized in Table 2. The data indicate a relatively weak dependence of the REv on the ion

related (AR + 2H)2+ ion triggered loss of a guanidinium hydrogen atom from an excited electronic state of the chargereduced cation-radical.10 A major difference between the ETD of (AtR + 2H)2+ and (AR + 2H)2+ is the absence of N−Cα bond cleavages in the thioxopeptide ion, which does not lose ammonia nor does it form a z1 fragment ion. In summary, electron transfer to (GtK + 2H)2+, (AtK + 2H)2+, and (AtR + 2H)2+ induces loss of a hydrogen atom as the dominant dissociation. The fragment ions observed in the ETD spectra mostly arise by secondary dissociations of (GtK + H)+, (AtK + H)+, and (AtR + H)+. To address this feature of the thioxopeptide ETD, we undertook a detailed study of precursor dication and charge-reduced cation-radical structures, as described below. (GtK + 2H)2+, (AtK + 2H)2+, and (AtR + 2H)2+ Ion Structures and Relative Stability. To aid the interpretation of the ETD spectra, we obtained optimized structures of lowenergy conformers for each system (Figure 5). The ion relative energies are summarized in Table 1. In all three thioxopeptides, the preferred protonation sites were the Lys or Arg side-chain groups and the N-terminal amino group. Tautomers in which one proton was placed on the thioxoamide sulfur atom rearranged by spontaneous proton transfer to the N-terminus. Note that the intrinsic proton affinity of the thioxoamide group, which was calculated as PA = 909 kJ mol−1 for NE



Article

Table 2. Ion−Electron Recombination Energiesa Ion 2+

GtK1 GtK22+ GtK32+ AtK12+ AtK22+ AtK32+ AtR12+ AtR22+ AtR32+ AtR42+

REverticalb

REadiabaticb,c

6.22 6.47 6.40 6.14 6.42 6.19 6.00 5.95 5.99 5.86

7.16 7.26 7.44 7.07 7.18 7.30 6.63 6.55 6.61 6.65

giving spin densities of 0.72, 0.13, and 0.02 for the thioamide, Lys, and N-terminal NH3 groups, respectively. The excited electronic states in vertically reduced GtK1+• consist of a Lys (A) and N-terminal (B by M06-2X and C by B3LYP) ammonium radicals that are represented by the corresponding Rydberg-like 3s orbitals. The other states (B by B3LYP and C by M06-2X) have major components consisting of the π* orbital of the carboxyl group. The D state is represented by a Lys ammonium 3p Rydberg-like orbital. Regarding the excitation energies, the B3LYP excited states show somewhat greater excited state compression than do the M06-2X ones, e.g., ΔE(X → A) = 0.12 and 0.50 eV by B3LYP and M06-2X, respectively (Figure 6). This results in a more extensive mixing of the X and A as well as B and C states in the B3LYP wave functions and leads to more extensive electron delocalization. The vertical and adiabatic recombination energies of (GtK + 2H)2+ (Table 2) indicate substantial excitation in chargereduced (GtK + 2H)+•. The approximate energy balance for the electron transfer, ignoring the ion-induced dipole potential33 between (GtK + 2H)+• and neutral fluoranthene in the exit channel, is given by ΔE = RE(ion) − EA(fluoranthene)34 = 6.2 − 0.6 = 5.6 eV = 540 kJ mol−1. The ΔE is partitioned between (GtK + 2H)+• and neutral fluoranthene, ΔE = Eexc(ion) − Eexc(fluoranthene). Note that fluoranthene can be formed in the S0, S1 (Eexc = 3.45 eV), S2 (Eexc = 4.32 eV),35 or higher excited electronic states.36 The electronic energy deposited in (GtK + 2H)+• is further augmented by Franck−Condon energy (EFC) upon geometry relaxation to the potential energy minimum of the pertinent electronic state of the chargereduced ion. For example, relaxation to the ground electronic state of (GtK + 2H)+• is associated with a substantial EFC = 7.16 − 6.22 eV = 0.94 eV = 91 kJ mol−1, which contributes to vibrational excitation upon electron transfer. The low electronic excited states of (GtK + 2H)+• (A, B, and D) are all Rydberglike states (3s or 3p, Figure 6) that are known to have relatively low EFC (≤20 kJ mol−1) when formed by electron transfer.37 These features allow us to estimate the range of internal vibrational energies in (GtK + 2H)+• as Eint(X) = 213−630 kJ mol−1, Eint(A) = 95−510 kJ mol−1, and Eint(B) = 60−476 kJ mol−1 for the X, A, and B states, respectively, and likewise diminishing for higher excited states of (GtK + 2H)+•. The rather broad internal energy ranges span the possibility of fluoranthene formation in the S0 through S2 electronic states. Relaxed GtK Cation-Radicals and Potential Energy Surfaces. Geometry relaxation in the X state of (GtK + 2H)+• leads to a zwitterionic structure GtK1+•, which is a local energy minimum (Scheme 1). Structure GtK1+• shows a pyramidized thioxoamide carbon center and an elongated C−S bond at 1.750 Å compared to 1.659 Å in GtK12+. The sulfur atom carries a negative charge (−0.44). The unpaired electron density is mostly localized in the thioxoamide group (0.88) with a major component on the carbon atom (0.59). The Lys side chain in GtK1+• refolded upon gradient geometry optimization to allow the charged ammonium group to reach to the thioxoamide sulfur atom at an NH3···S distance of 2.24 Å (Scheme 1). Zwitterion GtK1+• is a reactive intermediate that can undergo radical and charge induced reactions. The latter is represented by proton migration from the N-terminal ammonium to the thioxoamide sulfur atom. The migration requires a very low activation energy in TS1 (ETS1 = 17 kJ mol−1) and leads to an aminothioketyl radical (GtK1a+•). The GtK1+• → GtK1a+• isomerization is nearly isoenergetic

a

In units of eV. b From B3LYP/6-31++G(2d,p) single-point calculations. cAdiabatic recombination energies include B3LYP/631++G(2d,p) zero-point energy corrections.

conformation. The electronic states calculated by TDDFT with the B3LYP and M06-2X functionals for vertical electron attachment to GtK12+ are depicted in Figure 6. This shows that

Figure 6. Low-energy electronic states in (GtK + 2H)+• after vertical electron attachment to (GtK + 2H)2+. Orbitals and excitation energies in electron volts are from TD-DFT calculations with B3LYP/6-311+ +G(2d,p) (left panel) and M06-2X/6-311++G(2d,p) (right panel).

the lowest-energy state (X) is formed by the electron entering the thioxoamide group to form a transient zwittterionic state consisting of a thioketyl anion radical, which is represented by a π* orbital and shows a charge of −0.86 and a spin density of 0.93 by natural population analysis32 of the M06-2X/6-311+ +G(2d,p) electron densities. The two ammonium groups carry substantial positive charge (0.55 and 0.64 for the N-terminal and Lys NH3 groups, respectively) but have a negligibly low unpaired electron density ( 108 s−1) to induce complete dissociation of the charge-reduced ions on the time scale of the experiment (100 ms). However, the calculated relative rate constants (Figure 7b) unequivocally prefer the N−Cα bond cleavage. Furthermore, since GtK1c+• is more stable than GtK1a+•, the H loss is further disfavored by the low population of the GtK1a+• reactant. Figure S6a, Supporting Information, shows that GtK1a+• and GtK1c+• are expected to rapidly equilibrate by backbone rotations (krot >1010 s−1), and the calculated equilibrium constant consistently favors the more stable GtK1c+• (Figure S6b, Supporting Information). Hence, the energetics of the ground electronic state of (GtK + 2H)+• and the calculated reaction kinetics are incompatible with the experiment, even if dissociations at high internal energies are considered. The clue to a qualitative interpretation of the ETD mass spectra is provided by the properties of excited electronic states, as illustrated for (GtK + 2H)+• (Figure 6). The A and B excited states consist of 3s Rydberg-like orbitals at the respective Lys and N-terminal ammonium groups. Analogous hypervalent ammonium radicals are known from previous studies to dissociate by loss of N−H and C−N bonds, whereby the branching ratios for these dissociations depend on the electronic state of the radical.8a,37 In particular, states characterized by ammonium 3s Rydberg orbitals undergo preferential N−H bond dissociations, resulting in the loss of hydrogen atoms.8a,37 The 3s states of hypervalent ammonium radicals have lifetimes that have been estimated in the nanosecond range, due to both very low barriers to N−H bond cleavage37 and H-atom tunneling.38 The same argument can be made for dissociations of (GtK + 2H)+• and (AtK + 2H)+• where the loss of ammonium H atoms predominates. We cannot map the potential energy surface of the A and B states within the DFT formalism without resorting to Ehrenfest dynamics, which requires a major

Table 4. Relative and Dissociation Energies of (AtR + 2H)+• Cation-Radicals relative energy (kJ/mol)a,b species/reaction

B3LYP/6-31+ +G(2d,p)

B3-MP2/6-311+ +G(2d,p)

AtR1+• AtR1a+• AtR1b+• AtR1c+• TS16 TS17 TS18 AtR1+• → AtR1d+ + H• AtR1+• → AtR1e+ + H• TS19 TS20 TS21 AtR1+• → AtR1f + AtR1g+• AtR1+• → AtR1h + AtR1g+•

0.0 −27 −38 −567 −47 45 67 52 69 54 30 24 13 23

0.0 −24 −36 −52 −3 46 62 38 55 66 41 35 33 43

Relative energy (relative to AtR1+• energy) in units of kJ mol−1 including B3LYP/6-31++G(2d,p) zero-point vibrational energies and referring to 0 K. bFrom spin-projected MP2 energies wherever it applies. a

unstable with respect to isomerization to AtR1a+•, which is expected to be very fast.10 In contrast, N−Cα bond cleavage in AtR1+• requires overcoming a moderate barrier in TS19 (Scheme 3, ETS19 = 66 kJ mol−1 relative to AtR1+•). This indicates that the preferred reaction channel of AtR1+• is isomerization to AtR1a+•, not N−Cα bond cleavage. The aminothioketyl radicals AtR1b+• and AtR1c+• are the presumed reactants for the cleavage of the sulfhydryl S−H bond, resulting in loss of the H atom. The pertinent transition states are TS17 (ETS17 = 82 kJ mol−1 relative to AtR1b+•) and TS18 (ETS18 = 114 kJ mol−1 relative to AtR1c+•) (Table 4). The higher energy required for TS18 can be explained by ring I



Article

Scheme 4

by electron-transfer, such as loss of ammonia and N−Cα bond dissociation, is peculiar and indicates that the X state is not populated by electron transfer from fluoranthene anion radical. Conversely, the nearly exclusive loss of an H-atom in ETD indicates predominant formation of the A and or B states upon electron transfer. An interesting corollary concerns the ETD mass spectrum of (GK + 2H)+• (Figure 1b) that also did not show fragments from N−Cα bond dissociation. In contrast, the ETD mass spectrum of (GK + 2H)+• obtained by collisional electron transfer from Na atoms did show major fragments by the loss of ammonia and N−Cα bond dissociation.40 This implies that the nature of the electron donor and the collision process may play a larger role in determining the electronic states accessed by electron transfer than hitherto presumed. In particular, these factors should be considered, as theoretical models for electron transfer and excited state involvement are further refined.8b−d,41

■

ASSOCIATED CONTENT

S Supporting Information *

Complete references 16, 17, and 35; time courses; mass spectra; RRKM rate constants;isomerizations and dissociations; relative and dissociation energies. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address †

Institute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland.

Figure 7. Top panel: RRKM rate constants for S−H bond dissociation in GtK1a+• (kS−H) and N−Cα bond dissociation in GtK1c+• (kN−Cα). Bottom panel: ratio of rate constants from the top panel.

Notes

The authors declare no competing financial interest.

■

computational effort that would exceed the scope of this work.10 However, a previous study of the related (GK + 2H)+• ion indicated a very low barrier (15 kJ mol−1) to H atom loss from the Lys ammonium group.39 This indicates that the A excited state of (GtK + 2H)+• is also likely to dissociate by the loss of an ammonium H atom. The properties of the X state and the RRKM dissociation kinetics further indicate that N−Cα bond dissociation should be competitive if the X state was accessed. Thus, the absence of the typical dissociations caused

ACKNOWLEDGMENTS

Dedicated to Peter Armentrout on the occasion of his 60th birthday. This research was supported by a grant from the National Science Foundation (CHE-1055132). Instrumentation support for the ETD mass spectra measurements was provided by the University of Washington Proteomics Resource (Dr. Priska von Haller) along with access to the ETD-equipped Thermo-LTQ Orbitrap Velos by the Institute for Systems Biology (Drs. Richard Johnson and Rob Moritz). J



■

Article

(18) Sugita, Y.; Okamoto, Y. Chem. Phys. Lett. 1999, 314, 141−151. (19) Stewart, J. J. P. J. Mol. Model. 2007, 13, 1173−1213. (20) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (c) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623− 11627. (21) Møller, C.; Plesset, M. S. Phys. Rev. 1934, 46, 618−622. (22) Tureček, F. J. Phys. Chem. A 1998, 102, 4703−4713. (23) Č ížek, J.; Paldus, J.; Šroubková, L. Int. J. Quantum Chem. 1969, 3, 149−167. (24) Purvis, G. D.; Bartlett, R. J. J. Chem. Phys. 1982, 76, 1910−1918. (25) Curtiss, L. A.; Raghavachari, K.; Pople, J. A. J. Chem. Phys. 1993, 98, 1293−1298. (26) Furche, F.; Ahlrichs, R. J. Chem. Phys. 2002, 117, 7433−7447. (27) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241. (28) Gilbert, R. G.; Smith, S. C. Theory of Unimolecular and Recombination Reactions; Blackwell Scientific Publications: Oxford, U.K., 1990; pp 52−132. (29) Zhu, L.; Hase, W. L. Quantum Chemistry Program Exchange; Indiana University: Bloomington, IN, 1994; Program No. QCPE 644. (30) Frank, A. J.; Sadílek, M.; Ferrier, J. G.; Tureček, F. J. Am. Chem. Soc. 1997, 119, 12343−12353. (31) (a) Bleiholder, C.; Suhai, S.; Paizs, B. J. Am. Soc. Mass Spectrom. 2006, 17, 1275−1281. (b) Harrison, A. G. Mass Spectrom. Rev. 1997, 16, 201−217. (32) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735−748. (33) (a) Hayakawa, S.; Minami, K.; Iwamoto, K.; Toyoda, M.; Ichihara, T.; Nagao, H. Int. J. Mass Spectrom. 2007, 266, 122−128. (b) Moss, C. L.; Chung, T. W.; Wyer, J. A.; Nielsen, S. B.; Hvelplund, P.; Tureček, F. J. Am. Soc. Mass Spectrom. 2011, 22, 731−751. (34) Chaudhuri, J.; Jagur-Grodzinski, J.; Szwarc, M. J. Phys. Chem. 1967, 71, 3063−3065. (35) (a) Maddams, W. F.; Schnurmann, R. J. Chem. Phys. 1951, 19, 973−974. (b) Michl, J. J. Mol. Spectrosc. 1969, 30, 66−76. (c) Ruth, A. A.; Wick, M. T. Chem. Phys. Lett. 1997, 266, 206−216. (36) Tureček, F.; Chung, T. W.; Moss, C. L.; Wyer, J. A.; Ehlerding, A.; Holm, A. I. S.; Zettergren, H.; Nielsen, S. B.; Hvelplund, P.; Chamot-Rooke, J.; et al. J. Am. Chem. Soc. 2010, 132, 10728−10740. (37) (a) Boldyrev, A. J.; Simons, J. J. Chem. Phys. 1992, 97, 6621− 6627. (b) Nguyen, V. Q.; Sadílek, M.; Frank, A. J.; Ferrier, J. G.; Tureček, F. J. Phys. Chem. A 1997, 101, 3789−3799. (c) Yao, C.; Tureček, F. Phys. Chem. Chem. Phys. 2005, 7, 912−920. (d) Holm, A. I. S.; Larsen, M. K.; Panja, S.; Hvelplund, P.; Nielsen, S. B.; Leib, R. D.; Donald, W. A.; Williams, E. R.; Hao, C.; Tureček, F. Int. J. Mass Spectrom. 2008, 276, 116−126. (38) (a) Gellene, G. I.; Cleary, D. A.; Porter, R. F. J. Chem. Phys. 1982, 77, 3471−3477. (b) Gellene, G. I.; Porter, R. F. Acc. Chem. Res. 1983, 16, 200−207. (39) Tureček, F.; Chen, X.; Hao, C. J. Am. Chem. Soc. 2008, 130, 8818−8833. (40) Chakraborty, T.; Holm, A. I. S.; Hvelplund, P.; Nielsen, S. B.; Poully, J.-C.; Worm, E. S.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2006, 17, 1675−1680. (41) (a) Anusiewicz, I.; Berdys-Kochanska, J.; Simons, J. J. Phys. Chem. A 2005, 109, 5801−5813. (b) Anusiewicz, I.; BerdysKochanska, J.; Skurski, P.; Simons, J. J. Phys. Chem. A 2006, 110, 1261−1266. (c) Sobczyk, M.; Simons, J. Int. J. Mass Spectrom. 2006, 253, 274−280. (d) Sobczyk, M.; Simons, J. J. Phys. Chem. B 2006, 110, 7519−7527. (e) Skurski, P.; Sobczyk, M.; Jakowski, J.; Simons, J. Int. J. Mass Spectrom. 2007, 265, 197−212. (f) Sobczyk, M.; Neff, D.; Simons, J. Int. J. Mass Spectrom. 2008, 269, 149−164. (g) Swierszcz, I.; Skurski, P.; Simons, J. J. Phys. Chem. A 2012, 116, 1828−1837.

REFERENCES

(1) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1998, 120, 3265−3266. (2) Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528−9533. (3) (a) Shaffer, S. A.; Sadílek, M.; Tureček, F. J. Org. Chem. 1996, 61, 5234−5245. (b) Jones, J. W.; Sasaki, T.; Goodlett, D. R.; Tureček, F. J. Am. Soc. Mass Spectrom. 2007, 18, 432−444. (c) Chamot-Rooke, J.; Malosse, C.; Frison, G.; Tureček, F. J. Am. Soc. Mass Spectrom. 2007, 18, 2146−2161. (d) Sohn, C. H.; Chung, C. K.; Yin, S.; Ramachandran, P.; Loo, J. A.; Beauchamp, J. L. J. Am. Chem. Soc. 2009, 131, 5444−5459. (e) Jones, A. W.; Mikhailov, V. A.; Iniesta, J.; Cooper, H. J. J. Am. Soc. Mass Spectrom. 2010, 21, 268−277. (4) (a) Belyayev, M. A.; Cournoyer, J. J.; Lin, C.; O’Connor, P. B. J. Am. Soc. Mass Spectrom. 2006, 17, 1428−1436. (b) Tureček, F. J. Mass Spectrom. 2010, 45, 1280−1290. (5) (a) Kleinnijenhuis, A. J.; Mihalca, R.; Heeren, R. M. A.; Heck, A. J. R. Int. J. Mass Spectrom. 2006, 253, 217−224. (b) Tureček, F.; Jones, J. W.; Holm, A. I. S.; Panja, S.; Nielsen, S. B.; Hvelplund, P. J. Mass Spectrom. 2009, 44, 707−724. (c) Tureček, F.; Holm, A. I. S.; Panja, S.; Nielsen, S. B.; Hvelplund, P. J. Mass Spectrom. 2009, 44, 1518−1531. (d) Afonso, C.; Tabet, J.-C.; Giorgi, G.; Tureček, F. J. Mass Spectrom. 2012, 47, 208−220. (6) (a) Kjeldsen, F.; Zubarev, R. A. J. Am. Soc. Mass Spectrom. 2011, 22, 1441−1452. (b) Hansen, T. A.; Jung, H. R.; Kjeldsen, F. J. Am. Soc. Mass Spectrom. 2011, 22, 1953−1957. (7) (a) Sargaeva, N. P.; Lin, C.; O’Connor, P. B. J. Am. Soc. Mass Spectrom. 2011, 22, 480−491. (b) Ben Hamidane, H.; Vorobyev, A.; Larregola, M.; Lukaszuk, A.; Tourwe, D.; Lavielle, S.; Karoyan, P.; Tsybin, Y. O. Chem.Eur. J. 2010, 16, 4612−4622. (c) Moss, C. L.; Tureček, F. Int. J. Mass Spectrom. 2012, 316-318, 57−67. (8) (a) Syrstad, E. A.; Tureček, F. J. Am. Soc. Mass Spectrom. 2005, 16, 208−224. (b) Sobczyk, M.; Anusiewicz, I.; Berdys-Kochanska, J.; Sawicka, A.; Skurski, P.; Simons, J. J. Phys. Chem. A 2005, 109, 250− 258. (c) Neff, D.; Smuczynska, S.; Simons, J. Int. J. Mass Spectrom. 2009, 283, 122−134. (d) Simons, J. J. Am. Chem. Soc. 2010, 132, 7074−7085. (9) (a) Tureček, F.; Syrstad, E. A. J. Am. Chem. Soc. 2003, 125, 3353−3369. (b) Tureček, F. J. Am. Chem. Soc. 2003, 125, 5954−5963. (c) Chung, T. W.; Tureček, F. Int. J. Mass Spectrom. 2011, 301, 55−61. (10) Moss, C. L.; Liang, W.; Li, X.; Tureček, F. J. Am. Soc. Mass Spectrom. 2012, 23, 446−459. (11) (a) Iavarone, A. T.; Paech, K.; Williams, E. R. Anal. Chem. 2004, 76, 2231−2238. (b) Ben Hamidane, H.; Vorobyev, A.; Tsybin, Y. O. Eur. J. Mass Spectrom. 2011, 17, 321−331. (c) Vorobyev, A.; Ben Hamidane, H.; Tsybin, Y. O. J. Am. Soc. Mass Spectrom. 2009, 20, 2273−2283. (d) Jensen, C. S.; Holm, A. I. S.; Zettergren, H.; Overgaard, J. B.; Hvelplund, P.; Nielsen, S. B. J. Am. Soc. Mass Spectrom. 2009, 20, 1881−1889. (e) Chan, W. Y. K.; Chan, T. W. D. J. Am. Soc. Mass Spectrom. 2010, 21, 1235−1244. (f) Ben Hamidane, H.; He, H.; Tsybin, O. Y.; Emmett, M. R.; Hendrickson, C. L.; Marshall, A. G.; Tsybin, Y. O. J. Am. Soc. Mass Spectrom. 2009, 20, 1182−1192. (12) Zimnicka, M.; Gregersen, J. A.; Tureček, F. J. Am. Chem. Soc. 2011, 133, 10290−10301. (13) Syrstad, E. A.; Stephens, D. D.; Tureček, F. J. Phys. Chem. A 2003, 107, 115−126. (14) Moss, C. L.; Chamot-Rooke, J.; Brown, J.; Campuzano, I.; Richardson, K.; Williams, J.; Bush, M.; Bythell, B.; Paizs, B.; Tureček, F. J. Phys. Chem. B 2012, 106, 3445−3456. (15) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. J. Comput. Chem. 2005, 26, 1781−1802. (16) MacKerell, A. D., Jr.; Bashford, D.; Bellott, M.; Dunbrack, R. L., Jr.; Evanseck, J. D.; Field, M. J.; Fischer, S.; Gao, J.; Guo, H.; Ha, S.; et al. J. Phys. Chem. B 1998, 102, 3586−3616. (17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, revision A.02; Gaussian, Inc.: Wallingford CT, 2009. K


Perturbing Peptide Cation-Radical Electronic States by Thioxoamide

Recommend Documents