Protonation Sites and Conformations of Peptides of Glycine (Gly

J. Phys. Chem. B 2009, 113, 8767–8775

8767

Protonation Sites and Conformations of Peptides of Glycine (Gly1-5H+) by IRMPD Spectroscopy Ronghu Wu and Terry B. McMahon* Department of Chemistry, UniVersity of Waterloo, Waterloo, Ontario, Canada N2L 3G1 ReceiVed: December 29, 2008; ReVised Manuscript ReceiVed: April 19, 2009

The protonation sites and conformations of protonated glycine and its peptides (Gly1-5) have been investigated using infrared multiple photon dossociation (IRMPD) spectroscopy in combination with theoretical calculations. For small peptides, protonation is generally presumed to occur at the amine nitrogen of the N-terminus or a nitrogen of a basic side chain. However, for triglycine, the experimental and calculated results indicate that one of the main species is an isomer in which the proton is bound to an amide oxygen. The amide II vibrational mode is found to be very sensitive to the protonation site. When the protonation site is at the amine nitrogen, the amide II mode appears around 1540 cm-1 for diglycine, tetraglycine, pentaglycine, and one of the main isomers of triglycine (GGGH02). When the proton is bound to an amide oxygen, the amide II mode is blueshifted to 1590 cm-1, as seen in GGGH01. IR spectra have been obtained to provide direct evidence that an amide oxygen may serve as the protonation site in a peptide. An analogous result is found for the tripeptide of alanine. In the progression from glycine to pentaglycine, the corresponding conformations of the most stable isomers vary from linear to cyclic structures. Both glycine and diglycine are linear structures, while the most stable isomers of the tetra- and pentapeptides are both cyclic structures. For triglycine, the linear and cyclic isomers are found to coexist. The carbonyl stretches also directly reflect the conformational changes. For the linear isomers of the di- and tripeptides of glycine, two well-separated bands are observed. The amide I modes appear slightly above 1700 cm-1, but as a result of the fact that the CdO bond in the carboxylic acid moiety is stronger than those of the amide carbonyls, the corresponding band appears near 1800 cm-1. However, for the cyclic isomers of the tri-, tetra-, and pentapeptides, the carbonyl oxygen in the carboxylic acid group acts as a proton acceptor to form a very strong intramolecular hydrogen bond with the protonated amine terminus. This results in a weakening of the CdO bond, such that the amide I modes are nearly identical in frequency to the carbonyl stretch of the carboxylic acid group. 1. Introduction Peptides and proteins play a wide variety of roles in living organisms and display a range of properties. Their functions and properties are directly related to the structures and conformations. Protonation and proton transfer are among the most important factors in determining structures, properties, and functions of biological molecules,1-9 such as energy transfer, charge distribution, acid-base reactions, and enzyme catalysis.10-15 For amino acids and small peptides, based on the proton affinities of the various possible sites, protonation is generally presumed to occur at the amine nitrogen of the N-terminus or a nitrogen of a basic side chain. It is also observed that proton transfer along the peptide chain apparently becomes more facile as the size of the peptide increases. This has also given rise to some continuing speculation that amide oxygen may serve as a protonation site in larger peptides.16,17 However, direct experimental evidence to support this conjecture is still lacking. Proteins may be characterized by four levels of structure. Examination of three-dimensional representations suggests that complex tertiary and quaternary structures can be deconstructed into a limited number of secondary structural elements, such as R-helices, β-sheets, and R-, β-, and γ-turns. It is therefore fundamentally very important to understand the secondary structures of peptides. 18 There have been * Corresponding author. Telephone: (519) 888-4591. Fax: (519) 7460435. E-mail: [email protected].

numerous studies of the secondary structures of peptides in solution.19-21 Unfortunately, structures in solution are inevitably affected by the surrounding medium. However, recent advances in mass spectrometric ionization sources, such as electrospray ioniztion (ESI)22 and matrix-assisted laser desorption/ionization (MALDI),23,24 now make it possible to study large biological molecules in the gas phase.25-30 Gasphase experiments provide the ideal means for understanding the intrinsic structure and characteristics of peptides and proteins in the absence of any interference from solvents or other species present in solution. Infrared spectroscopy is a very important tool for the study of the secondary structures of peptides. In particular, the amide A (NH stretching), amide I (carbonyl stretching coupled with in-plane NH bending and CN stretching mode), amide II (NH bending coupled with CN stretching), and amide III (CN stretching coupled with NH bending) modes are very revealing of elements of secondary structure.18 For gaseous ionic species, conventional absorption IR spectroscopy is not generally feasible due to the fact that Coulombic repulsion limits the ionic number density. Variable wavelength infrared multiple photon dissociation (IRMPD) spectroscopy is a very powerful and sensitive technique to study the structure of ions in the gas phase.31-39 Recently the combination of a very powerful free electron laser (FEL) and an ion trap or a Fourier transfer ion cyclotron resonance (FT-ICR) mass spectrometer has been used to

10.1021/jp811468q CCC: $40.75  2009 American Chemical Society Published on Web 06/01/2009

8768

J. Phys. Chem. B, Vol. 113, No. 25, 2009

Wu and McMahon

examine many structures of gaseous ion and ionic clusters, together with theoretical calculation.40-57 In the present work, the protonation sites and conformations of protonated glycine and its oligomers (Gly1-5) have been investigated using IRMPD spectroscopy in combination with electronic structure calculations. IRMPD spectra of the protonated species have been obtained, and the energetics and structures of different possible isomers of these protonated species have been computed at the B3LYP/6-311+G(d,p) level of theory. The IR spectra have also been calculated using the same computational method. From the combination of experimental and calculated spectra, the protonation sites, structures, and conformations of the protonated peptides may be determined unambiguously. In addition, valuable insight is provided into the IRMPD mechanism for these species. 2. Experimental Section and Theoretical Calculations 2.1. Experimental Section. Experiments have been carried out using the free electron laser (FEL) at the Centre de Laser Infrarouge d’Orsay (CLIO) facility in Orsay (France) coupled to an electrospray ionization-ion trap mass spectrometer (Bruker Esquire3000+). This experimental configuration has been described in detail previously.36,58 The FEL facility is based on emission from a 10-50 MeV electron beam. For the high energy electron beam, the emission photon wavelength can be tuned by adjusting the gap of an undulator which is placed in the optical cavity. In the present work, the electron energy was set to 48 MeV to continuously scan from 1000 to 2000 cm-1. The IR-FEL output consists of macropulses 8 µs in length with a repetition rate of 25 Hz. Each macropulse involves approximately 500 micropulses with the width of a few picoseconds. For a typical average IR power of 500 mW, the corresponding micropulse and macropulse energies are about 40 µJ and 20 mJ, respectively. In order to produce protonated species of glycine and its oligomers, a 10-5 M solution of the sample dissolved in a mixture of water and methanol, with a very small amount of formic acid, was used. The solution was introduced into the mass spectrometer using electrospray ionization, and the desired ionic species was isolated and confined in the ion trap. The IRFEL beam with a desired wavelength was focused and introduced into the center of the ion trap of the mass spectrometer. Mass spectra were recorded following irradiation, and 10 mass spectra were accumulated at each wavelength. IRMPD spectra were obtained by scanning the wavelength in steps of ∼4 cm-1. The spectra reported here are expressed as the fragmentation efficiency, Pfrag (Pfrag ) -ln(Iparent/(Iparent + ∑Ifragment))), as a function of the photon energy, in cm-1. The experimental IRMPD spectra of protonated glycine and its oligomers are shown in Figure 1. 2.2. Theoretical Calculations. Theoretical calculations have been carried out using the Gaussian 03 program package.59 The structures of protonated glycine and its oligomers (GlyH+, Gly2H+, Gly3H+, Gly4H+, and Gly5H+) were optimized at the density functional theory (DFT) level, employing the B3LYP exchange-correlation functional and the 6-311+G(d,p) basis set. An exhaustive search has been carried out. Many different possible isomers have been considered, and a multitude of initial structures have been optimized. For the species studied, many calculated structures have previously been reported,16,17,60-62 which have been used as references of the current structural optimizations. The calculated harmonic vibrational frequencies have been determined using the same computational method, with a scaling factor of 0.99. The calculated frequencies and

Figure 1. IRMPD spectra of protonated glycine and its oligomers (a, GlyH+; b, GlyGlyH+; c, GlyGlyGlyH+; d, GlyGlyGlyGlyH+; e, GlyGlyGlyGlyGlyH+).

intensities were convoluted assuming a Lorentzian profile with a 35 cm-1 full width at half-maximum. Single point energies have been calculated using a MP2(full)/6-311++G(2d,2p)// B3LYP/6-311+G(d,p) protocol for GlyH+, Gly2H+, and Gly3H+, and MP2/6-311+G(d,p)//B3LYP/6-311+G(d,p) for Gly4H+ and Gly5H+, because the more accurate energy can be obtained by single point energy calculation with MP2 and a large basis set. The same conclusion was drawn after different methods and basis sets were compared to calculate many protonated cytosine tautomers.63 Zero point energy and thermal energy corrections at 298 K were also included. The harmonic oscillator model was used in thermochemical calculations. The relative entropy changes were obtained from the B3LYP/6-311+G(d,p) level of theory. 3. Results 3.1. Gly. Glycine is the simplest amino acid among the 20 naturally occurring amino acids. It is well-known that the protonation site is at the amine nitrogen, which is consistent with the current calculated results. The computed structures of the four most stable isomers of GlyH+ are shown in Figure S1 in the Supporting Information, with their relative energetics given in Table S1 in the Supporting Information. Only the structure of the most stable isomer, GH01, is given in Figure 2. In GH01 the proton is bound to the amino nitrogen, with the formation of an intramolecular hydrogen bond between the protonated amino group and the carbonyl oxygen. Conversely, if the protonated amino group acts as proton donor to form an hydrogen bond with the hydroxyl oxygen, the isomer (GH02) is formed, which has an energy 4.6 kcal mol-1 higher than that of GH01. If proton is bound to the carbonyl oxygen, as in GH03 and GH04, isomers having energies 26.3 and 26.8 kcal mol-1 greater than that of the most stable isomer at the MP2(full)/6311++G(2d,2p)//B3LYP/6-311+G(d,p) level of theory are found. Thus, under the current experimental conditions, these isomers will be virtually nonexistent. In order to confirm whether both GH01 and GH02 might exist under the IRMPD conditions, the experimental IRMPD spectrum of protonated glycine may be compared to the calculated vibrational spectra of these isomers. The experimental spectrum and the calculated vibrational spectra of the two most stable

Conformations of Gly1-5H+ by IRMPD Spectroscopy

J. Phys. Chem. B, Vol. 113, No. 25, 2009 8769

Figure 2. Calculated structures of the most stable isomers of the oligomers of glycine obtained at the 6-311+G(d,p) level of theory.

Figure 3. Experimental IRMPD spectrum of protonated glycine and calculated spectra of the two most stable isomers (GH01 and GH02).

isomers are displayed in Figure 3. As seen, the IRMPD spectrum may be well assigned exclusively to the calculated spectrum of the most stable isomer, GH01. The band at 1165 cm-1 is assigned as the hydroxyl bending vibration. The band between 1400 and 1500 cm-1 involves several vibrations, i.e., the CH2 wagging and scissors and the NH3 umbrella, with the corresponding calculated values of 1419, 1458, and 1492 cm-1, respectively. The carbonyl stretching vibration appears at 1800 cm-1, which is in very good agreement with the calculated value of 1799 cm-1 for GH01. However, it is significantly different from that of GH02 (1869 cm-1). Compared to GH02, the carbonyl stretching vibration of GH01 is red-shifted by 70 cm-1, because the intramolecular hydrogen bond between the carbonyl group and the protonated amino group results in a weakening of the CdO bond, with the bond length increasing from 1.19 Å in GH02 to 1.21 Å in GH01. When the proton is bound to the carbonyl oxygen, the carbonyl stretching band disappears in this range for both GH03 and GH04, which is completely different from the observed IRMPD spectrum. According to the

calculated and experimental results, it is very clear that only GH01 exists under experimental conditions. 3.2. GlyGly. The calculated structures of the several most stable isomers of protonated GlyGly are shown in Figure S2 in the Supporting Information, and their relative energies are summarized in Table S2 in the Supporting Information. Only the most stable isomer, GGH01, is displayed in Figure 2. In GGH01, the proton is bound to the amino nitrogen and a hydrogen bond is formed between one of the acidic hydrogens of the resulting ammonium group and the amide oxygen. The amide NH then forms the additional hydrogen bond with the carbonyl oxygen at the C-terminus. If the carboxylic acid group rotates out of the backbone plane, a second isomer, GGH02, is formed which is 1.0 kcal mol-1 higher in electronic energy at 298 K than GGH01 at the MP2(full)/6-311++G(2d,2p)// B3LYP/6-311+G(d,p) level of theory. This is similar to calculated results previously reported.64 In GGH03 a proton is bound to the amide oxygen, with an energy 1.8 kcal mol-1 higher than that of GGH01. The protonation site is also at the amide oxygen in GGH04 with the formation of two intramolecular hydrogen bonds and an energy 3.2 kcal mol-1 higher than that of the most stable isomer. The experimental IRMPD spectrum is shown in Figure 4, together with the calculated spectra of the three most stable isomers. The structure of GGH02 is very close to that of GGH01, and their calculated IR spectra are also very similar. The IRMPD spectrum is in very good agreement with those calculated for both GGH01 and GGH02. The weak band at 1077 cm-1 corresponds to the amide CN stretch, and the strongest band at 1166 cm-1 can be assigned to the bending vibration of the hydroxyl group. They are consistent with the calculated values of 1077 and 1172 cm-1 for GGH01, respectively. The umbrella vibration of -NH3 appears at 1418 cm-1, which fits very well with the calculated value of 1409 cm-1. The band at 1540 cm-1 is due to the amide II mode based on the calculated value of 1542 cm-1 for the NH bending vibration of the amide group.

8770

J. Phys. Chem. B, Vol. 113, No. 25, 2009

Wu and McMahon TABLE 1: Relative Calculated Enthalpy and Entropy Changes (298 K) for the Different Isomers of Protonated Triglycine B3LYP/6311+G(d,p)

MP2(full)/ 6-311++G(2d,2p)// B3LYP/6-311+G(d,p)

∆∆H298 ∆∆S ∆∆H298a ∆∆G298b (kcal mol-1) (cal mol-1 K-1) (kcal mol-1) (kcal mol-1) GGGH01 GGGH02 GGGH03 GGGH04

0 3.3 3.7 2.9

0 -1.9 4.7 3.1

0 -1.1 2.7 3.2

0 -0.5 1.3 2.3

a Zero point energy (ZPE) and thermal energy correction at 298 K obtained at B3LYP/6-311+G(d,p). b ∆H is from the single point energy, and ∆S is from B3LYP/6-311+G(d,p).

Figure 4. Experimental IRMPD spectrum of protonated GlyGly and calculated spectra of the three most stable isomers (GGH01, GGH02, and GGH03).

The bands between 1650 and 1850 cm-1 are usually due to the carbonyl stretching vibrations and are the most important modes for the identification of conformation of peptides. The observed bands at 1722 and 1788 cm-1 are in excellent agreement with the calculated values of 1729 and 1786 cm-1 for the amide I mode and the carbonyl stretch of the terminal carboxylic acid groups, respectively. The higher frequency of the carbonyl stretching vibration in the carboxylic acid group is consistent with the calculated CdO bond lengths of 1.21 Å in the carboxylic acid group and 1.24 Å in the amide group. According to the calculated energetics and presuming a Boltzmann distribution of possible isomers, the amounts of GGH01, GGH02, and GGH03 are 75.0, 23.0, and 1.7%, respectively. According to the experimental IRMPD spectrum and the calculated results, both GGH01 and GGH02 are likely present under the experimental conditions. Based on the vibrational spectrum, the presence of GGH03 cannot be excluded. However, the energetic data would suggest that it would not constituent a substantial contribution to the population. These results are also consistent with those of the dipeptide of alanine reported in the literature.65,66 3.3. GlyGlyGly. The calculated structures of the two most stable isomers of protonated GlyGlyGly are shown in Figure 2, and a number of other, higher energy, isomers are given in Figure S3 in the Supporting Information. The relative energetics of the different isomers are summarized in Table 1. GGGH01, in which the proton is bound to the amide oxygen at the N-terminus, is found to be the most stable isomer at the B3LYP/ 6-311+G(d,p) level of theory. Formation of this structure is promoted by the additional strong hydrogen bond to the second amide oxygen with a short contact of only 1.42 Å. The cyclic structure, GGGH02, in which a proton is bound to the N-terminal nitrogen, with formation of two intramolecular hydrogen bonds to the adjacent amide oxygen and the carbonyl oxygen of the C-terminus, is 3.3 kcal mol-1 higher in energy than that of GGGH01 at the B3LYP/6-311+G(d,p) level of theory. However, at the MP2(full)/6-311++G(2d,2p)//B3LYP/ 6-311+G(d,p) level of theory, it is more stable than GGGH01. With entropy considerations, the free energy of GGGH02 is still 0.5 kcal mol-1 lower than that of GGGH01. Zhang et al. reported that GGGH02 is the most stable isomer calculated at the HF/6-31G(d)//HF/3-21G level of theory.16 However, GGGH01

Figure 5. Experimental IRMPD spectrum of protonated GlyGlyGly and calculated spectra of the two most stable isomers (GGGH01 and GGGH02) and their mixture.

has been calculated to be 1.2 kcal mol-1 more stable than GGGH02 using B3LYP/6-31++G(d,p).17 The computational results are thus ambiguous, and some doubt remains as to which isomer exists and at which site protonation occurs. Another isomer, GGGH03, also involves protonation at the amine nitrogen of the N-terminus, but without hydrogen bond induced cyclization. However, GGGH03 is 2.7 kcal mol-1 higher in energy than GGGH01. Proton transfer from the amino group in GGGH03 to the adjacent amide oxygen gives rise to a new isomer, GGGH04, which is 0.5 kcal mol-1 less stable than GGGH03. An examination of the experimental IRMPD spectrum leads to the conclusion that the bands may be best assigned based on a combination of the calculated spectra of GGGH01 and GGGH02, as shown in Figure 5. The strong band at 1166 cm-1 is assigned as the bending vibration of the free OH of the carboxylic acid group, calculated to occur at 1184 cm-1. However, the calculated intensity is weaker than that observed. The weak band at 1108 cm-1 corresponds to the bending vibration of the hydrogen bonded OH, which is in good agreement with the calculated value of 1109 cm-1 for GGGH01. The very weak broad band around 1280 cm-1 corresponds to the amide III mode (CN stretching), whose lower intensity is a result of the fact that this mode is not IR active. The broad band from 1410 to 1480 cm-1 is related to the CH2 and NH2 rocking and scissors modes. Each CH2 group has a somewhat

Conformations of Gly1-5H+ by IRMPD Spectroscopy

J. Phys. Chem. B, Vol. 113, No. 25, 2009 8771

TABLE 2: Relative Calculated Enthalpy and Entropy Changes (298 K) for the Different Isomers of Protonated Tetraglycine B3LYP/6311+G(d,p)

MP2/6-311+G(d,p)// b3LYP/6-311+G(d,p)

∆∆S ∆∆H298a ∆∆G298b ∆∆H298 (kcal mol-1) (cal mol-1 K-1) (kcal mol-1) (kcal mol-1) GGGGH01 GGGGH02 GGGGH03 GGGGH04 GGGGH05

0 -2.0 1.3 2.8 0.6

0 -0.9 4.2 5.2 11.8

0 0.3 3.3 6.0 13.7

0 0.5 2.0 4.5 10.2

a ZPE and thermal energy correction at 298 K obtained at B3LYP/6-311+G(d,p). b ∆H is from the single point energy, and ∆S is from B3LYP/6-311+G(d,p).

different environment, resulting in the broad band. This is also consistent with the corresponding simulated band. The amide II mode appears at 1589 cm-1, which matches very well with the calculated NH bending and CN stretching vibration of 1585 cm-1 for GGGH01. The band at 1550 cm-1 cannot be assigned to GGGH01. However, it corresponds very well with the amide II mode of GGGH02. Compared to this band in GGGH02, the frequency for that in GGGH01 is blue-shifted about 40 cm-1. This is also consistent with the difference of the calculated values due to the different protonation sites. The amide II mode of GGGH02 is similar to that in protonated GlyGly (∼1540 cm-1), in which the protonation site is also at the amine nitrogen. In the experimental spectrum, the bands at 1703 and 1803 cm-1 are due to the carbonyl stretching vibrations in the amide and carboxylic acid groups of GGGH01, respectively. The band at 1745 cm-1 cannot be assigned to any band of GGGH01. However, it matches very well with the carbonyl stretches of GGGH02. The three corresponding calculated modes closely overlap around 1745 cm-1. According to the calculated structure, the three CdO bond lengths are very similar at 1.23 Å for the amide carbonyl in the N-terminus (1730 cm-1), 1.22 Å for the carboxyl group (1744 cm-1), and 1.21 Å for the internal amide carbonyl (1764 cm-1). However, for GGGH01 the two carbonyl groups are markedly different due to the proton being bound to one of the amide oxygens and the formation of a linear structure. The corresponding CdO bond lengths are 1.20 and 1.25 Å in the carboxyl and amide groups (1805 and 1704 cm-1), respectively. These two bands of GGGH01 fit very well with the corresponding bands in the experimental IRMPD spectrum. According to the experimental and computational results, both GGGH01 and GGGH02 thus exist under the experimental conditions, although minor amounts of other isomers cannot be completely excluded. Therefore, the present work constitutes the first experimental confirmation of the existence of a protonated peptide in which the proton is bound to the amide oxygen. 3.4. GlyGlyGlyGly. A large number of isomers have been calculated at the B3LYP/6-311+G(d,p) level of theory based on previous calculated results.62 The two most stable isomers are shown in Figure 2, and three additional stable isomers are shown in Figure S4 in the Supporting Information, with their relative energetics summarized in Table 2. The single point energies for the tetra- and pentapeptides of glycine have been calculated at the MP2/6-311+G(d,p)//B3LYP/6-311+G(d,p) level of theory. GGGGH02 with a cyclic hydrogen bonded structure is the most stable isomer at the B3LYP/6-311+G(d,p) level of theory. A proton bound to the amino group at the N-terminus forms a

Figure 6. Experimental IRMPD spectrum of GlyGlyGlyGlyH+ and calculated spectra of the two most stable isomers (GGGGH01 and GGGGH02) and their mixture.

very strong hydrogen bond with the carbonyl oxygen at the C-terminus with a hydrogen bond length of 1.65 Å. The hydroxyl group at the C-terminus forms another strong hydrogen bond with the adjacent carbonyl oxygen. In addition, several relatively weak intramolecular H-bonds exist in the isomer. GGGGH01 is also a cyclic structure, in which protonated amino group forms a very strong H-bond with the second carbonyl oxygen from the C-terminus and a relatively weak H-bond exists between the carboxylic acid oxygen and the amide group adjacent to the N-terminus. GGGGH01 is 2.0 kcal mol-1 higher in energy at 298 K than GGGGH02 at the B3LYP/6311+G(d,p)) level of theory. However, it becomes the more stable isomer at the MP2/6-311+G(d,p)//B3LYP/6-311+G(d,p) level of theory by 0.3 kcal mol-1. In the case of GGGGH03, the protonated amino group forms three H-bonds with the three carbonyl oxygens, and it is 3.3 kcal mol-1 higher in energy than GGGGH01 according to the single point energy calculations. Unlike the protonated tripeptide of glycine, isomers with a linear structure have markedly higher energies than the cyclic hydrogen bonded isomers. For example, the most stable linear isomer is GGGGH05, whose structure is similar to that of GGGH01, where the proton is bound to an amide oxygen. However, it is 13.7 kcal mol-1 higher in energy than the most stable isomer. The IRMPD spectrum of GlyGlyGlyGlyH+ is shown in Figure 6, together with the calculated spectra of the two most stable isomers and a mixture of the two. As for the tripeptide of glycine, the simulated spectrum of the mixture of 50% GGGGH01 and 50% GGGGH02 fits very well with the experimental IRMPD spectrum. The band at 1165 cm-1 is assigned to the bending vibration of the free hydroxyl group in GGGGH01, which does not exist in GGGGH02 because of the formation of the intramolecular H-bond. The weak band between 1200 and 1300 cm-1 corresponds to the wagging and rocking vibrations of the several CH2 and NH groups. The shoulder at 1436 cm-1 is due to the bending vibration of the hydrogen bonded hydroxyl group in GGGGH02. The broad band between 1500 and 1600 cm-1 is due to the amide II modes and umbrella vibrations of the NH3 group in GGGGH01 and GGGGH02. This band is quite broad because because the eight vibrational frequencies in this range, while similar, are not identical. For the calculated spectrum of GGGGH01, the values are 1520,

8772

J. Phys. Chem. B, Vol. 113, No. 25, 2009

Wu and McMahon

TABLE 3: Relative Calculated Enthalpy and Entropy Changes (298 K) for the Different Isomers of Protonated Pentaglycine B3LYP/6311+G(d,p)

MP2/6-311+G(d,p)// b3LYP/6-311+G(d,p)

∆∆S ∆∆H298a ∆∆G298b ∆∆H298 (kcal mol-1) (cal mol-1 K-1) (kcal mol-1) (kcal mol-1) GGGGGH01 GGGGGH02 GGGGGH03 GGGGGH04 GGGGGH05 GGGGGH06

0 -0.4 -1.3 2.5 5.0 5.1

0 3.4 -1.8 -0.1 -4.0 16.7

0 0.7 2.0 3.4 6.9 17.5

0 -0.3 1.5 3.4 8.1 12.5

a ZPE and thermal energy correction at 298 K obtained at B3LYP/6-311+G(d,p). b ∆H is from the single point energy, and ∆S is from B3LYP/6-311+G(d,p).

1535, 1559, and 1575 cm-1, while those for GGGGH02 are 1520, 1531, 1544, and 1557 cm-1, respectively. Similarly, the carbonyl stretching modes also appear as a broad strong band with a peak at 1740 cm-1 and shoulders at 1725, 1755, and 1775 cm-1. According to the calculated positions and intensities of the carbonyl stretching vibrations, the mixture of GGGGH01 and GGGGH02 fits very well with the experimental IRMPD spectrum. The calculated carbonyl stretching vibration in the carboxylic acid group of GGGGH01 appears at 1774 cm-1, which is in very good agreement with one of the shoulders at 1775 cm-1. The calculated bands at 1736, 1713, and 1676 cm-1 are due to the three amide I modes. These vibrational frequencies are consistent with the corresponding relative CdO bond lengths. The amide oxygen adjacent to the carboxylic acid group forms a very strong H-bond with the protonated amino group with a bond length of 1.67 Å, which weakens this CdO bond. The associated CdO bond length of 1.24 Å is longest of the four, with the others being 1.21, 1.22, and 1.23 Å. For the GGGGH02, the four calculated carbonyl stretching modes at 1734, 1726, 1707, and 1676 cm-1 correspond to the carboxylic acid moiety, the amide group at the N-terminus, the middle amide group, and the amide group adjacent to the C-terminus, respectively. The formation of the very strong H-bond between the carboxylic acid group and the protonated amino group results in similar bond lengths of 1.22 or 1.23 Å for the four different CdO bonds, also giving rise to a broad band in the calculated spectrum. The experimental and calculated results thus indicate that both GGGGH01 and GGGGH02 exist under the experimental conditions although minor amounts of other isomers cannot be excluded. 3.5. GlyGlyGlyGlyGly. The calculated structures of a number of stable isomers of protonated pentaglycine are shown in Figure 2 and Figure S5 in the Supporting Information, and their relative energies are summarized in Table 3. As found for the tetrapeptide, isomers with a linear structure have notably higher energies than the most stable cyclic hydrogen bonded isomers. For example, GGGGGH06 is 17.5 kcal mol-1 higher in energy than GGGGGH01. GGGGGH01 is the most stable isomer at the MP2/6-311+G(d,p)//B3LYP/6-311+G(d,p) level of theory. The protonated amino group forms three intramolecular H-bonds with the adjacent amide oxygen, the carboxylic acid carbonyl oxygen, and the amide oxygen adjacent to the C-terminus. In addition, the amide NH at the C-terminus forms a H-bond with the adjacent amide oxygen and the associated amide NH forms a further H-bond with the next amide oxygen. This then results in the formation of two β-turns. In GGGGGH02, the protonated amino group forms three H-bonds with the

Figure 7. Experimental IRMPD spectrum of protonated GlyGlyGlyGlyGly and the calculated spectra of the two most stable isomers (GGGGGH01 and GGGGGH02).

adjacent amide, the middle amide, and the carboxylic acid carbonyl oxygens. In comparison to GGGGGH01, this isomer is entropically favored. At the B3LYP/6-311+G(d,p) level of theory, it is 0.4 kcal mol-1 lower in energy than GGGGGH01. However, its energy is 0.7 kcal mol-1 higher from the single point energy calculations. The structures of GGGGGH03 and GGGGGH04 are similar to those of GGGGH01 and GGGGH02, respectively. However, their energies are 2-3 kcal mol-1 higher than that of the most stable isomer. As the chain length increases, peptides become more flexible, which results in more facile formation of intramolecular H-bonds between the protonated amino group and the amide oxygens. The IRMPD spectrum of GlyGlyGlyGlyGlyH+ is shown in Figure 7, together with the calculated spectra of the two most stable isomers. The experimental spectrum is in good agreement with that calculated for GGGGGH01. The band at 1165 cm-1 is assigned to the bending vibration of the free hydroxyl group. The weak band between 1200 and 1300 cm-1 corresponds to the wagging and rocking vibrations of the CH2 and NH groups. The band around 1540 cm-1 is due to the amide II modes and umbrella vibrations of the NH3 group in GGGGGH01, which is consistent with the simulated band at 1550 cm-1. The amide I mode appears as a broad band with several shoulders between 1650 and 1800 cm-1. The two adjacent peaks are at 1725 and 1740 cm-1 may be assigned to the carbonyl stretching vibrations of the middle amide group and that at the N-terminus with calculated values of 1718 and 1735 cm-1, respectively. The clear shoulder peak at 1770 cm-1 in the experimental spectrum corresponds to the carbonyl stretching of the carboxylic acid group, whose calculated value is 1765 cm-1. According to the computational results, GGGGGH02 has an energy comparable to that of GGGGGH01. However, the calculated spectrum of GGGGGH02 resembles the experimental IRMPD spectrum to a lesser degree. Due to the much larger size of this peptide, the calculated energies and frequencies may be somewhat lesser reliable. Based on the experimental and calculated results, GGGGGH01 appears to be the dominant species, but other isomers cannot be excluded. 4. Discussion 4.1. Fragment Pathways and Mechanism of IRMPD. Fragmentation of protonated peptides and proteins is one of the most important methods for characterization of the amino acid sequence.67-71 The most common activation method is collision induced dissociation (CID).72,73 IRMPD is a relatively new activation method,74-88 which has the strong potential to reveal

Conformations of Gly1-5H+ by IRMPD Spectroscopy both structural and energetic features for peptides and proteins. Therefore, it is of considerable interest to understand the IRMPD pathways and mechanisms for the dissociation of protonated peptides. For protonated glycine, the main IRMPD fragment peak is at m/z 48, corresponding to the loss of CO. The other commonly observed fragmentation, due to loss of CO + H2O, is weaker. This fragmentation pattern is markedly different from that obtained by CID,89 in which the main fragmentation corresponds to the loss of 46 u. According to the most extensive computational study,90 the loss of 46 u under low energy collision conditions is due to the sequential loss CO and water. From the calculated dissociation energy barrier of 36.6 kcal mol-1, at least eight photons with a wavelength of 1750 cm-1 are required to dissociate the protonated glycine. In the current IRMPD spectrum, the main fragment corresponds to the loss of CO. This means that, after the loss of CO, a substantial fraction of these primary fragment ions do not have sufficient internal energy remaining to lose water, although this energy barrier is only 6 kcal mol-1,67 which is more than the energy of one 1750 cm-1 photon. This indicates that IRMPD is a very soft activation and dissociation method. For a protonated peptide, when fragmentation occurs at the C-N bond of a peptide linkage, a “b” ion is produced if the charge is retained on the N-terminal fragment. Alternatively, for a peptide protonated at the N-terminus, a “y” ion is produced after proton migration from the N-terminus to the C-terminus followed by C-N bond cleavage. Other product ions, including “a” ions, and ions that result from the loss of small neutral molecules, e.g., CO, H2O, and NH3, may also be formed from the fragmentation of the protonated peptide or from its charged primary fragments. The dominant IRMPD fragment of protonated GlyGly is a y1 ion. For the protonated GlyGlyGly, the main fragment is a b2 ion. The formation of a y1 ion is the result of the cleavage of the same amide bond as in the formation of the b2 ion. The presence of a substantially stronger b2 peak can be attributed to a thermal dynamic equilibrium process. The main fragment of the protonated GlyGlyGlyGly is a y2 ion. The y2 and b2 ions are due to the cleavage of the same central amide bond. In contrast to the triglycine, here the y2 ion is obviously stronger, which is also the result of a thermal dynamic equilibrium process, since the proton affinity of the corresponding neutral for the b2 ion is less than that for the y2 ion, i.e., diglycine. For the protonated GlyGlyGlyGlyGly, peaks due to the b2, b3, b4, y2, and y3 ions all appear. The b2/y3 and b3/y2 pairs are each due to the cleavage of the same amide bonds, respectively. In the CID spectrum of pentaglycine,91 the main peaks are b4 and the loss of water. Other peaks, such as b2, y2, b3, and y3, are much weaker than b4. In our spectrum, b2, b3, b4 and y2, y3 have almost the same intensities. From these phenomena, it may be inferred that the barriers for these different dissociation channels, i.e., the cleavage of the amide bonds except the first amide bond adjacent to the N-terminus, are very similar, with an energy difference of less than the energy of one 1750 cm-1 photon (∼5 kcal mol-1). Reid et al. have investigated the fragmentation of oligomers of glycine using CID.91 For tetra- and pentaglycines, the MS/ MS spectra reveal major neutral losses of water. However, although the peak corresponding to the loss of water does appear via IRMPD, the intensity is extremely weaker and the main fragment pathways are due to the cleavage of the amide bonds. This indicates that IRMPD forms more structurally informative sequence ions, the b and y ions, and fewer nonsequence ions

J. Phys. Chem. B, Vol. 113, No. 25, 2009 8773 formed by the loss of small neutral molecules such as water and ammonia. IRMPD is thus, in principle, more useful in the identification of sequences in peptides and proteins. 4.2. Protonation Sites. The protonation site plays a very important role in biological systems, because protonation at different sites may result in markedly different structures and properties for biological molecules. For glycine, the experimental and calculated results indicate that the most stable isomer is GH01, in which the protonation site is at the amino nitrogen. If a proton is bound to oxygen, the isomers GH03 and GH04 result, and, although they are local minima on the potential energy surface, their energies are more than 20 kcal mol-1 higher than that of the most stable isomer. Thus, under thermal equilibrium conditions, their relative abundances will be negligible. For the dipeptide of glycine, when protonation occurs at the amide oxygen, the corresponding isomers (GGH03 and GGH04) have much lower relative energies in contrast to those of oxygenprotonated glycine. Although the experimental IRMPD spectrum cannot distinguish whether GGH03 exists as a minor species, the population of this isomer would only be ∼2% according to the calculated energies and presuming a Boltzmann distribution. As the peptide chain length increases, as for the tripeptide of glycine, the calculated results are ambiguous with regard to which of the two isomers, GGGH01 or GGGH02, is the more stable. In GGGH01 the proton is bound to an amide oxygen, while in GGGH02 amino nitrogen protonation occurs. The experimental IRMPD spectrum indicates very clearly that both isomers are present in similar abundance under the experimental conditions. In GGGH01, the proton is shared by the two adjacent amide oxygens very effectively; however, the tripeptide chain is not sufficiently long to permit formation of a cyclic isomer involving an intramolecular H-bond between the N-terminus and the C-terminus. In contrast, such a cyclization is possible in GGGH02, although again the limited chain length constrains the intramolecular hydrogen bond to being relatively weak. These factors result in the two isomers, GGGH01 and GGGH02, coexisting as the main species. For the tetra- or pentapeptides of glycine, a proton is bound to the amino nitrogen in the most stable isomer. Proton binding to an amide oxygen results in isomers which are local minima on the potential energy surface; however, the associated energies are higher than that of the corresponding most stable isomer. For the current IRMPD spectra, the amide I, II, and III modes are all in the range 1000-2000 cm-1 although, unfortunately, the amide III mode is not IR active. Both amide I and II modes are very important for the structural identification of the peptides. An examination of the IRMPD spectra of protonated glycine and its oligomers in Figure 1 shows that the amide II mode is particularly sensitive to the protonation site. When a proton is bound to the amino nitrogen, the amide II mode appears at ∼1540 cm-1 for the di-, tetra-, and pentapeptides as well as for one of the tripeptide isomers (GGGH02). When a proton is bound to the amide oxygen in GGGH01, the amide II mode is blue-shifted to 1590 cm-1. This is reasonable, since the protonation site at an amide oxygen gives rise to a marked change in the CN and NH bond lengths in the corresponding amide group. The experimental and calculated results indicate that the protonation site is related not only to the proton affinities of the nitrogen and oxygen atoms in the peptide, but also to the strength of the resulting intramolecular H-bonds which determine the secondary structure. These IR spectra provide direct experimental evidence that an amide oxygen may serve as the

8774

J. Phys. Chem. B, Vol. 113, No. 25, 2009

protonation site for the tripeptide of glycine in the gas phase. This is also true for the tripeptide of alanine.48 4.3. Conformational Changes. These experimental and computational results provide a basis for comparison of the conformational changes in the series of peptides of glycine examined here. The most stable isomers of protonated Gly and GlyGly are linear structures. For the tetra- and pentapeptides of glycine, the linear isomers are less stable than their cyclic analogues, in which the protonated amino group forms a very strong H-bond with the carboxyl oxygen at the C-terminus. However, for the tripeptide of glycine, the linear and cyclic isomers coexist as the main species under the current experimental conditions. When the series of IRMPD spectra for Gly to pentaglycine in Figure 1 are compared, the amide II modes appear at similar positions near 1540 cm-1, except for the linear isomer of triglycine, in which a proton is bound to the amide oxygen. This means that the amide II mode is not very sensitive to the conformations of the peptides. For the linear isomers of di- and tripeptides of glycine, the amide I modes appear as two separated bands. The stretching vibrations of the carbonyl moiety in the carboxylic acid group appear at the higher frequency at ∼1800 cm-1, similar to that of protonated glycine. The free CdO stretch of the carboxylic acid group appears at higher frequency than that of an amide group. Although an intramolecular H-bond may form between this carbonyl oxygen and the adjacent amide group in the linear structure, the H-bond remains relatively weak due to the steric effect and the relatively weak acidity of the amide N-H. Thus the CdO bond remains relatively strong with a bond length of ∼1.20 Å. Conversely, for the cyclic isomers of tri-, tetra-, and pentapeptides, the corresponding modes appear as broad bands. In the cyclic structure, the carbonyl oxygen in the carboxylic acid group forms a very strong H-bond with the protonated amino group. This strong H-bond weakens the CdO bond in the carboxylic acid group, which results in a bond length of ∼1.23 Å. This is very similar to the corresponding bond length in the other amide groups. Ultimately, the amide I modes are nearly identical in frequency to that of the carbonyl stretch of the carboxylic acid group and a broad band appears. In contrast to the amide II mode, the amide I mode is very sensitive to the conformation of peptide.

Wu and McMahon evidence that an amide oxygen may serve as the protonation site for the tripeptide of glycine in the gas phase. In proceeding from glycine through pentglycine, the corresponding conformations of the most stable isomers are changed from linear to cyclic. For glycine and diglycine, these are linear structures, in which the protonated amino group forms an intramolecular hydrogen bond with the adjacent carbonyl oxygen. Cyclic structures are the most stable isomers for the tetra- and pentaglycines. Linear and cyclic isomers coexist for triglycine under the present experimental conditions. The amide I mode is directly indicative of conformational changes. For the linear isomers of di- and triglycine, the amide I modes appear as two separated bands. In addition, the stretching vibration of the carbonyl group at the C-terminus appears at relatively high frequency, because the free CdO bond in the carboxylic acid group is usually stronger than that in an amide group. Conversely, for the cyclic isomers of tri-, tetra-, and pentaglycines, the corresponding modes appear as broad bands. In the cyclic structure, the carbonyl oxygen of the carboxylic acid group forms a very strong H-bond with the protonated amino group. This strong H-bond weakens the CdO bond, which results in the bond length increasing from 1.20 to 1.22 or 1.23 Å, which is very similar to that found in an amide group. Even though glycine oligomers represent the simplest peptides, since they lack any side chains, they do serve as excellent models for backbone structure and, in this way, their study provides an excellent model for understanding peptide and protein structure and properties. Acknowledgment. The generous financial support by the Natural Sciences and Engineering Research Council of Canada (NSERC) is gratefully acknowledged as is the financial support of the European Commission through the NEST/ADENTURE program (EPITOPES, Project No. 15637). We are very grateful for the valuable assistance of the CLIO team, P. Maitre, J. Lemaire, T. Besson, D. Scuderi, J. M. Bakker, and J. M. Ortega. Supporting Information Available: Tables S1 and S2, Figures S1-S5, and the complete ref 59. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes

5. Conclusions IRMPD spectra of the protonated oligomers of glycine have been recorded, which proves to be an excellent way of interrogating ionic structure in combination with the theoretical calculations. IRMPD gives rise to more structurally informative sequence ions, the b and y ions, and fewer nonsequence ions through the loss of small neutral molecules such as water and ammonia, which means that IRMPD is a soft activation method. For small peptides, protonation is generally presumed to occur at the amine nitrogen of the N-terminus or a nitrogen of a basic side chain. However, for the tripeptide of glycine, one of the main species is the isomer with a proton bound to an amide oxygen (GGGH01). The amide II mode is very sensitive to the protonation site. When the protonation site is the amino nitrogen, the amide II mode appears at ∼1540 cm-1 for diglycine, tetraglycine, pentaglycine, and one of triglycine isomers in which a proton is bound to the amino nitrogen (GGGH02). When a proton is bound to the amide oxygen, the amide II mode is blueshifted by ∼50 cm-1, as in GGGH01. This is a logical consequence of the fact that protonation at an amide oxygen gives rise to a marked change in the CN and NH bond strengths in the corresponding amide group. IR spectra provide direct

(1) Kearley, G. J.; Fillaux, F.; Baron, M.-H.; Bennington, S.; Tomkinson, J. Science 1994, 264, 1285. (2) Green, M. K.; Lebrilla, C. B. Mass Spectrom. ReV. 1997, 16, 53. (3) Stowell, M. H. B.; McPhillips, T. M.; Rees, D. C.; Soltis, S. M.; Abresch, E.; Feher, G. Science 1997, 276, 812. (4) Wu, R. H.; McMahon, T. B. J. Am. Chem. Soc. 2007, 129, 569. (5) Cukier, R. I.; Nocera, D. G. Annu. ReV. Phys. Chem. 1998, 49, 337. (6) Jarrold, M. F. Annu. ReV. Phys. Chem. 2000, 51, 179. (7) Dongre, A. R.; Jones, J. L.; Somogyi, A.; Wysocki, V. H. J. Am. Chem. Soc. 1996, 118, 8365. (8) Wu, R. H.; McMahon, T. B. J. Am. Chem. Soc. 2008, 130, 3065. (9) Wu, R. H.; McMahon, T. B. J. Am. Chem. Soc. 2008, 130, 12554. (10) Chang, C. J.; Chang, M. C. Y.; Damrauer, N. H.; Nocera, D. G. Biochim. Biophys. Acta: Bioenerg. 2004, 1655, 13. (11) Huynh, M. H. V.; Meyer, T. J. Chem. ReV. 2007, 107, 5004. (12) Sibert, R.; Josowicz, M.; Porcelli, F.; Veglia, G.; Range, K.; Barry, B. A. J. Am. Chem. Soc. 2007, 129, 4393. (13) Makowska, J.; Baginska, K.; Kasprzykowski, F.; Vila, J. A.; Jagielska, A.; Liwo, A.; Chmurzynski, L.; Scheraga, H. A. Biopolymers 2005, 80, 214. (14) Holliday, G. L.; Almonacid, D. E.; Mitchell, J. B. O.; Thornton, J. M. J. Mol. Biol. 2007, 372, 1261. (15) Smith, M. D.; Collins, R. A. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 5818. (16) Zhang, K.; Cassady, C. J.; Chung-Phillips, A. J. Am. Chem. Soc. 1994, 116, 11512. (17) Rodriquez, C. F.; Cunje, A.; Shoeib, T.; Chu, I. K.; Hopkinson, A. C.; Siu, K. W. M. J. Am. Chem. Soc. 2001, 123, 3006.

Conformations of Gly1-5H+ by IRMPD Spectroscopy (18) Vass, E.; Hollosi, M.; Besson, F.; Buchet, R. Chem. ReV. 2003, 103, 1917. (19) Haris, P. I.; Chapman, D. Biopolymers 1995, 37, 251. (20) Nesloney, C. L.; Kelly, J. W. Bioorg. Med. Chem. 1996, 4, 739. (21) Kubelka, J.; Huang, R.; Keiderling, T. A. J. Phys. Chem. B 2005, 109, 8231. (22) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64. (23) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass Spectrom. 1988, 2, 151. (24) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299. (25) Yates, J. R., III. J. Mass Spectrom. 1998, 33, 1. (26) Domon, B.; Aebersold, R. Science 2006, 312, 212. (27) Cravatt, B. F.; Simon, G. M.; Yates, J. R. Nature 2007, 450, 991. (28) Wyttenbach, T.; Bowers, M. T. Annu. ReV. Phys. Chem. 2007, 58, 511. (29) Benesch, J. L. P.; Ruotolo, B. T.; Simmons, D. A.; Robinson, C. V. Chem. ReV. 2007, 107, 3544. (30) Grabenauer, M.; Bernstein, S. L.; Lee, J. C.; Wyttenbach, T.; Dupuis, N. F.; Gray, H. B.; Winkler, J. R.; Bowers, M. T. J. Phys. Chem. B 2008, 112, 11147. (31) Peiris, D. M.; Riveros, J. M.; Eyler, J. R. Int. J. Mass Spectrom. 1996, 159, 169. (32) Valle, J. J.; Eyler, J. R.; Oomens, J.; Moore, D. T.; van der Meer, A. F. G.; von Helden, G.; Meijer, G.; Hendrickson, C. L.; Marshall, A. G.; Blakney, G. T. ReV. Sci. Instrum. 2005, 76, 023103. (33) Oh, H. B.; Lin, C.; Hwang, H. Y.; Zhai, H.; Breuker, K.; Zabrouskov, V.; Carpenter, B. K.; McLafferty, F. W. J. Am. Chem. Soc. 2005, 127, 4076. (34) von Helden, G.; van Heijnsbergen, D.; Meijer, G. J. Phys. Chem. A 2003, 107, 1671. (35) Moore, D. T.; Oomens, J.; Eyler, J. R.; von Helden, G.; Meijer, G.; Dunbar, R. C. J. Am. Chem. Soc. 2005, 127, 7243. (36) Maitre, P.; Le Caer, S.; Simon, A.; Jones, W.; Lemaire, J.; Mestdagh, H.; Heninger, M.; Mauclaire, G.; Boissel, P.; Prazeres, R.; Glotin, F.; Ortega, J.-M. Nucl. Instrum. Methods A 2003, 507, 541. (37) Kong, X. L.; Tsai, I. A.; Sabu, S.; Han, C. C.; Lee, Y. T.; Chang, H. C.; Tu, S. Y.; Kung, A. H.; Wu, C. C. Angew. Chem., Int. Ed. 2006, 45, 4130. (38) Kamariotis, A.; Boyarkin, O. V.; Mercier, S. R.; Beck, R. D.; Bush, M. F.; Williams, E. R.; Rizzo, T. R. J. Am. Chem. Soc. 2006, 128, 905. (39) Vaden, T. D.; de Boer, T. S. J. A.; Simons, J. P.; Snoek, L. C. Phys. Chem. Chem. Phys. 2008, 10, 1443. (40) Asmis, K. R.; Pivonka, N. L.; Santambrogio, G.; Brummer, M.; Kaposta, C.; Neumark, D. M.; Woste, L. Science 2003, 299, 1375. (41) Oomens, J.; Sartakov, B. G.; Meijer, G.; von Helden, G. Int. J. Mass Spectrom. 2006, 254, 1. (42) Fridgen, T. D.; MacAleese, L.; McMahon, T. B.; Lemaire, J.; Maitre, P. Phys. Chem. Chem. Phys. 2006, 8, 955. (43) Polfer, N. C.; Oomens, J.; Moore, D. T.; von Helden, C.; Meijer, G.; Dunbar, R. C. J. Am. Chem. Soc. 2006, 128, 517. (44) Dunbar, R. C.; Moore, D. T.; Oomens, J. J. Phys. Chem. A 2006, 110, 8316. (45) Wu, R. H.; McMahon, T. B. J. Am. Chem. Soc. 2007, 129, 4864. (46) Dunbar, R. C.; Polfer, N. C.; Oomens, J. J. Am. Chem. Soc. 2007, 129, 14562. (47) Forbes, M. W.; Bush, M. F.; Polfer, N. C.; Oomens, J.; Dunbar, R. C.; Williams, E. R.; Jockusch, R. A. J. Phys. Chem. A. 2007, 111, 11759. (48) Wu, R. H.; McMahon, T. B. J. Am. Chem. Soc. 2007, 129, 11312. (49) Wu, R. H.; McMahon, T. B. Angew. Chem., Int. Ed. 2007, 46, 3668. (50) Dunbar, R. C.; Moore, D. T.; Oomens, J. Int. J. Mass Spectrom. 2007, 265, 182. (51) Forbes, M. W.; Bush, M. F.; Polfer, N. C.; Oomens, J.; Dunbar, R. C.; Williams, E. R.; Jockusch, R. A. J. Phys. Chem. A 2007, 111, 11753. (52) Bush, M. F.; Forbes, M. W.; Jockusch, R. A.; Oomens, J.; Polfer, N. C.; Saykally, R. J.; Williams, E. R. J. Phys. Chem. A 2007, 111, 7753.

J. Phys. Chem. B, Vol. 113, No. 25, 2009 8775 (53) Bush, M. F.; Oomens, J.; Saykally, R. J.; Williams, E. R. J. Am. Chem. Soc. 2008, 130, 6463. (54) Gabelica, V.; Rosu, F.; De Pauw, E.; Lemaire, J.; Gillet, J. C.; Poully, J. C.; Lecomte, F.; Gregoire, G.; Schermann, J. P.; Desfrancois, C. J. Am. Chem. Soc. 2008, 130, 1810. (55) Correia, C. F.; Balaj, P. O.; Scuderi, D.; Maitre, P.; Ohanessian, G. J. Am. Chem. Soc. 2008, 130, 3359. (56) Armentrout, P. B.; Rodgers, M. T.; Oomens, J.; Steill, J. D. J. Phys. Chem. A 2008, 112, 2248. (57) Rodgers, M. T.; Armentrout, P. B.; Oomens, J.; Steill, J. D. J. Phys. Chem. A 2008, 112, 2258. (58) Lemaire, J.; Boissel, P.; Heninger, M.; Mauclaire, G.; Bellec, G.; Mestdagh, H.; Simon, A.; Le Caer, S.; Ortega, J.-M.; Glotin, F.; Maitre, P. Phys. ReV. Lett. 2002, 89, 273002. (59) Frisch, M. J.; et al. Gaussian 03, revision B.03; Gaussian Inc.: Pittsburgh, PA, 2003. (60) Zhang, K.; Chung-Phillips, A. J. Comput. Chem. 1998, 19, 1862. (61) Zhang, K.; Zimmerman, D. M.; Chung-Phillips, A.; Cassady, C. J. J. Am. Chem. Soc. 1993, 115, 10812. (62) Chung-Phillips, A. J. Phys. Chem. A 2005, 109, 5917. (63) Yao, C. X.; Turecek, F.; Polce, M. J.; Wesdemiotis, C. Int. J. Mass Spectrom. 2007, 265, 106. (64) Kohtani, M.; Breaux, G. A.; Jarrold, M. F. J. Am. Chem. Soc. 2004, 126, 1206. (65) Lucas, B.; Gregoire, G.; Lemaire, J.; Maitre, P.; Ortega, J. M.; Rupenyan, A.; Reimann, B.; Schermann, J. P.; Desfrancois, C. Phys. Chem. Chem. Phys. 2004, 6, 2659. (66) Vaden, T. D.; de Boer, T. S. J. A.; Simons, J. P.; Snoek, L. C.; Suhai, S.; Paizs, B. J. Phys. Chem. A 2008, 112, 4608. (67) Turecek, F.; Syrstad, E. A. J. Am. Chem. Soc. 2003, 125, 3353. (68) Bogdanov, B.; Smith, R. D. Mass Spectrom. ReV. 2005, 24, 168. (69) Wysocki, V. H.; Tsaprailis, G.; Smith, L. L.; Breci, L. A. J. Mass Spectrom. 2000, 35, 1399. (70) Aebersold, R.; Mann, M. Nature 2003, 422, 198. (71) Han, X. M.; Breuker, K.; McLafferty, F. W. Science 2006, 314, 109. (72) Laskin, J.; Futrell, J. H. Mass Spectrom. ReV. 2003, 22, 158. (73) Paizs, B.; Suhai, S. Mass Spectrom. ReV. 2005, 24, 508. (74) Shin, S. K.; Beauchamp, J. L. J. Am. Chem. Soc. 1990, 112, 2057. (75) Little, D. P.; Speir, J. P.; Senko, M. W.; Oconnor, P. B.; McLafferty, F. W. Anal. Chem. 1994, 66, 2809. (76) Stephenson, J. L.; Booth, M. M.; Shalosky, J. A.; Eyler, J. R.; Yost, R. A. J. Am. Soc. Mass Spectrom. 1994, 5, 886. (77) Longfellow, C. A.; Lee, Y. T. J. Phys. Chem. 1995, 99, 15532. (78) Colorado, A.; Shen, J. X. X.; Vartanian, V. H.; Brodbelt, J. Anal. Chem. 1996, 68, 4033. (79) Tonner, D. S.; McMahon, T. B. Anal. Chem. 1997, 69, 4735. (80) Hakansson, K.; Cooper, H. J.; Emmett, M. R.; Costello, C. E.; Marshall, A. G.; Nilsson, C. L. Anal. Chem. 2001, 73, 4530. (81) Xie, Y. M.; Lebrilla, C. B. Anal. Chem. 2003, 75, 1590. (82) Dodds, E. D.; Hagerman, P. J.; Lebrilla, C. B. Anal. Chem. 2006, 78, 8506. (83) Polfer, N. C.; Valle, J. J.; Moore, D. T.; Oomens, J.; Eyler, J. R.; Bendiak, B. Anal. Chem. 2006, 78, 670. (84) Wilson, J. J.; Brodbelt, J. S. Anal. Chem. 2007, 79, 2067. (85) Stearns, J. A.; Boyarkin, O. V.; Rizzo, T. R. J. Am. Chem. Soc. 2007, 129, 13820. (86) Wu, R. H.; McMahon, T. B. Chem.sEur. J. 2008, 14, 7765. (87) Wu, R. H.; McMahon, T. B. ChemPhysChem 2008, 9, 2826. (88) Wu, R. H.; McMahon, T. B. Mass Spectrom. ReV. 2009, in press. (89) Rogalewicz, F.; Hoppilliard, Y.; Ohanessian, G. Int. J. Mass Spectrom. 2000, 195/196, 565. (90) Rogalewicz, F.; Hoppilliard, Y. Int. J. Mass Spectrom. 2000, 199, 235. (91) Reid, G. E.; Simpson, R. J.; O’Hair, R. A. J. Int. J. Mass Spectrom. 1999, 190/191, 209.

JP811468Q