J. Phys. Chem. B 2006, 110, 16111-16116
16111
Infrared Multiphoton Dissociation Spectroscopic Analysis of Peptides and Oligosaccharides by Using Fourier Transform Ion Cyclotron Resonance Mass Spectrometry with a Midinfrared Free-Electron Laser Kazuhiko Fukui,*,† Yasutoshi Takada,‡ Tetsumi Sumiyoshi,‡ Takayuki Imai,§ and Katsutoshi Takahashi† Computational Biology Research Center, National Institute of AdVanced Industrial Science and Technology, 2-41-6 Aomi, Koto-ku, Tokyo 135-0064, Japan, Cyber Laser, Telecom Center Building East Tower 2F, 2-38 Aomi, Koto-ku, Tokyo 135-8070, Japan, and IR FEL Research Center, Research Institute for Science and Technology, Tokyo UniVersity of Science, Yamazaki 2641, Noda, Chiba 278-8510, Japan ReceiVed: February 7, 2006; In Final Form: June 12, 2006
The fragmentation of peptides and oligosaccharides in the gas phase was investigated by means of electrospray ionization Fourier transform ion cyclotron resonance (FTICR) mass spectrometry coupled with dissociation by a laser-cleavage infrared multiphoton dissociation (IRMPD) technique. In this technique, an IR freeelectron laser is used as a tunable source of IR radiation to cause cleavage of the ionized samples introduced into the FTICR cell. The gas-phase IRMPD spectra of protonated peptides (substance P and angiotensin II) and two sodiated oligosaccharides (sialyl Lewis X and lacto-N-fucopentaose III) were obtained over the IR scan range of 5.7-9.5 µm. In the IRMPD spectra for the peptide, fragment ions are observed as y/b-type fragment ions in the range 5.7-7.5 µm, corresponding to cleavage of the backbone of the parent amino acid sequence, whereas the spectra of the oligosaccharides have major peaks in the range 8.4-9.5 µm, corresponding to photoproducts of the B/Y type.
I. Introduction There has been significant interest in the fragmentation analysis of polypeptides, proteins, oligosaccharides, and glycopeptides using a variety of mass spectrometric techniques that involve collisions with gaseous atoms, surfaces, electrons, or photons. The development of soft ionization methods, such as matrix-assisted laser desorption (MALDI)1-4 and electron-spray ionization (ESI),5-7 in combination with mass spectrometry (MS) has made it possible to investigate many biologically relevant molecules in the gas phase. In particular, Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS) in combination with soft ionization methods is useful in highsensitivity analysis to identify small quantities of peptides/ proteins, oligosaccharides, or glycopeptides/proteins and to identify which amide bonds and glycosyl bonds are cleaved in the charged state. FTICR-MS is a suitable tool for the photochemical analysis of biopolymers, as it permits the storage of mass-selected ions in a collision-free environment for up to several minutes and is sufficiently adaptable to allow the introduction of a laser beam into the cell; it also gives accurate and high-resolution values of the molecular masses of fragments. Infrared multiphoton dissociation (IRMPD) studies with FTICRMS employ photons at an IR wavelength, typically 10.6 µm as radiated by a CO2 laser, to activate trapped ions in the gas phase.8 Because the amount of energy transferred to large ions in the collisional activation dissociation (CAD) process is limited, the efficiency of CAD decreases with increasing mass* Author to whom correspondence should be addressed. E-mail: k-fukui@ aist.go.jp. † National Institute of Advanced Industrial Science and Technology. ‡ Cyber Laser. § Tokyo University of Science.
to-charge ratios. In contrast, the uptake of energy by photoabsorption is independent of mass.9 The technique of photoinduced dissociation in a cell has been shown to be effective in providing information on sequence and structure for peptide mixtures and protein ions. In recent years, FTICR-MS in combination with the IRMPD technique, employing a powerful and wavelength-tunable laser pulse from a free-electron laser (FEL), has been used to investigate laser-induced chemical reactions in metal ions and peptides.10-17 Moore et al. determined the binding sites for complexes of a transition metal ion with aromatic molecules by using FTICR-MS coupled with a FEL laser.15 In this study, we investigated peptides and oligosaccharides by using IRMPD with a tunable IR FEL in conjunction with nano-ESI FTICR-MS. This study addresses fundamental questions concerning the photochemical dynamics of peptides and oligosaccharides and will lead ultimately to a better understanding of the photochemistry of biopolymer systems. The photoproduct ions induced by IRMPD can be probed by monitoring the fragmentation of the ions in the IR wavelength range of 5.7-9.5 µm. Infrared spectroscopic observations provide detailed information on the dependence of the fragmentation efficiency at a certain wavelength. In the IRMPD process, internal energy, which increases as a result of the absorption of a number of IR photons, is transformed into internal vibrational and rotational states of the ion. When the ion has sufficient energy to exceed its dissociation limit, fragment ions are produced and observed in the mass spectra. The FEL can be tuned over the various vibrational absorption lines for peptides and oligosaccharides, and the efficiency of IRMPD can be probed by monitoring the fragment ions. To investigate the normal modes that are activated by IR excitation energy, the absorption
10.1021/jp0607824 CCC: $33.50 © 2006 American Chemical Society Published on Web 07/22/2006
16112 J. Phys. Chem. B, Vol. 110, No. 32, 2006
Fukui et al. FEL laser beam was directed to the center of the cell through a ZnSe window and mildly focused with a concave mirror of focal length 15 m. For instance, when the wavelength was 6.1 µm, the diameter of the laser beam was estimated to be 3 mm at the front of the power meter. The beam diameter at the center of the FTICR ion trap was estimated to be 0.8 mm. The FEL irradiating time was controlled with a high-speed electrical shutter (Melles Griot, CA). The IRMPD products were investigated following laser irradiation, which was performed for 180 s in the case of peptides and 60 s in the case of oligosaccharides; the MS data at the IR wavelengths were recorded by integrating the mass spectra three times. III. Results and Discussion
Figure 1. FEL beam profile. The upper circle shows the structure of micropulses in a macropulse.
frequencies for a sodiated oligosaccharide ion (sialyl Lewis X) were calculated by ab initio quantum mechanics calculations at the HF/6-31G* level. II. Experimental Section The FEL experiments were carried out by using the facilities at the IR FEL Research Center, Research Institute for Science and Technology, Tokyo University of Science (TUS). The FELSUT delivers macropulses at a pulse repetition rate of 5 macropulses/s, as shown in Figure 1. Each macropulse is about 2 µs long and consists of about 6000 2-ps-long micropulses; each pair of micropulses is separated by an interval of 350 ps. In the case of experiments designed to obtain IR spectra, the FEL was tuned over the range of 5.8-9.5 µm with energies of 1050 mW. The spectral width of the beam (full width at halfmaximum) at a wavelength of 6.1 µm was 0.08 µm. Details regarding the principles of operation and characteristics of the FEL are described elsewhere.18-20 The FEL system was coupled to a BioAPEX III FTICR mass spectrometer equipped with a 4.7 T superconducting magnet (Bruker Daltonics, Billerica, MA). By combination of the generation of fragment ions by irradiation with the FEL beam and accurate mass determination of the IRMPD product ions, IR spectra were recorded for two peptides (the neuropeptide substance P and the hormone angiotensin II) and two oligosaccharides (the human blood-group antigen sialyl Lewis X and the immunodominant differentiation antigen lacto-N-fucopentaose III (LNFP III)). The samples were obtained from Sigma-Aldrich (St. Louis, MO) and used without additional purification. A 40 nmol portion of the peptide or oligosaccharide was dissolved in 1 mL of HPLC-grade 50:50 methanol/water mixture to give a concentration of 40 µM. Each aliquot of this solution was diluted to a final concentration of 10 pmol/µL. The protonated peptide and sodiated oligosaccharide ions were delivered to the ICR cell by means of a nanoelectrospray ionization source. A 7 µL aliquot of the sample was pipetted into a metal-coated fused silica needle (New Objective, MA). Figure 2 shows the experimental setup, which was designed to permit the use of the FEL laser and a CO2 laser, with both irradiating the samples on the same axis. The
IRMPD of Peptides and Oligosaccharides. The mass spectrum of substance P (amino acid sequence, Arg-Pro-LysPro-Gln-Gln-Phe-Phe-Gly-Leu-Met) in solution shows the presence of singly, doubly, and triply protonated ions, identified by resolving the isotopic peak spacing. The most abundant ion produced by the nano-electrospray is a doubly charged peptide ion at m/z ) 674.864 Da, corresponding to [Sub P+ 2H]2+. To investigate the dependence of its fragmentation on the IR wavelength, the most abundant parent ions (the whole group of substance P at charge +2) were isolated in the FTICR ion-trap cell by using a correlated sweep excitation method before the IRMPD event. Figure 3a shows the IRMPD-MS spectra, and Figure 3b shows the relative intensities of fragment ions for substance P as a function of the FEL wavelength. It is noted that relative intensities (all ion signals) are normalized to the sum of all ions. The highest fragmentation was observed at a wavelength of 6.1 µm (1655 cm-1), corresponding to the amide I band: This is discussed later, in the section on fragment analysis. The dominant fragment ion is assigned as b2/y9 (primary fragment ions), resulting from cleavage of the parent peptide at Pro-Lys. Figure 3b shows the IRMPD efficiency curves obtained from the fragmentation spectra for varying FEL laser wavelengths; these permit the determination of which bond in the ionized peptide is more effectively and easily cleaved at a specific wavelength. The effective IR wavelength for cleaving the peptide bonds was in the range of 5.9-8.3 µm. The error bars for the data points were estimated at (5% by measuring the signals from product ions at a fixed wavelength of 6.1 µm. Figure 4 shows the power profile for the FEL-TUS recorded as a function of the IR wavelength and measured using a power meter in front of the FTICR mass spectrometer. The power profile of the FEL-TUS shows minima at around 6.5 and 8.0 µm; it is difficult to increase the FEL power at this region in our experiment. The parent ion of [Sub P+2H]2+ may not absorb around 6.5 µm and in the 8.5-9.5 µm range. It should be noted that the fragment ions originating from an internal cleavage assigned as y9-Met (secondary fragment ion) were observed as a result of further absorption of photons by the y9 fragment ion. The IR wavelength that produced the highest intensity for the internal fragment ion corresponded to the amide I band at 6.1 µm. IRMPD spectroscopic analysis was also performed for angiotensin II (amino acid sequence, Asp-Arg-Val-Tyr-Ile-HisPro-Phe) as a function of the FEL IR wavelength. The most abundant fragment ions were the y2/b6 pair of ions from cleavage at His-Pro. The rich abundance of primary y2/b6 product ions formed by the amide bond cleavage between the His and Pro residues was due to the structural features of proline, which lacks amide protons participating in hydrogen bonding and the presence of a rigid cyclic structure that imposes distinct
IRMPD Spectroscopy of Biopolymers with FTICR-MS
J. Phys. Chem. B, Vol. 110, No. 32, 2006 16113
Figure 2. Optical layout for the combination of the free-electron laser and FTICR-MS at FEL-TUS.
Figure 4. Rower profile of the FEL-TUS measured in front of the FTMS.
Figure 3. (a) IRMPD mass spectra of doubly charged substance P at an irradiation time of 180 s. Each of the mass spectra represents a 0.2 µm interval in the range from 5.7 to 9.5 µm. (b) Relative intensities of the fragment ions of [Sub P + 2H]2+ as a function of the IR wavelength. The micropulse energy at a wavelength of 6.1 µm and a power of 28 mW is about 1 µJ.
conformational restrictions. The internal fragment ion b6-Ile (secondary fragment) was also observed (Supporting Information). It is noted that the dependence of the fragmentation on the charge state is observed for the peptides having a specific
relationship between the number of ionizing protons and acidic/ basic residues (e.g., angiotensin II).8 Figure 5 shows the IRMPD spectra for sialyl Lewis X (sLex), which consists of four monosaccharide units, including Fuc and Neu5Ac, and is known to form a terminal saccharide unit in oligo- and polysaccharides and glycoproteins. The parent ion was a singly charged Na-adduct ion at m/z ) 674.864 Da, corresponding to [sLex+ Na]+. The parent ion was the major ion in the mass spectra recorded between 5.7 and 8.4 µm. Figure 6 shows the relative intensities of the different fragments as a function of wavelength for [sLex + Na]+. The observed major product ions were formed by cleavage at the glycosyl bonds of Gal-Neu5Ac and GalNAc-Fuc. The highest relative intensities of the observed fragment ions corresponded to a loss of 291 Da ([M + Na]+ - Neu5Ac), assigned as Y2R. The range of irradiation wavelengths effective in cleaving the glycosyl bond between Gal and Neu5Ac was broader than that for the bond between Fuc and GalNAc. The loss of sialic acid appears to be the most frequent fragmentation reaction, indicating that the energy required to cleave the glycosyl bond of Gal-Neu5Ac is lower than that required to cleave GalNAc-Fuc. It is known that Neu5Ac and Fuc residues are easily cleaved in low-energy collisionally induced dissociation (CID).21, 22 Because Neu5Ac
16114 J. Phys. Chem. B, Vol. 110, No. 32, 2006
Fukui et al.
Figure 5. IRMPD mass spectra of sodiated sialyl Lewis X obtained with an irradiation time of 60 s. Each mass spectrum represents a 0.1 µm interval in the range of 5.7-9.5 µm. The micropulse energy at a wavelength of 9.1 µm and a power of 23 mW is about 0.8 µJ.
Figure 7. Relative intensities of the fragment ions of [LNFP III + Na]+ as a function of the IR wavelength.
Figure 6. Relative intensities of the fragment ions of [sLex + Na]+ as a function of the IR wavelength. The abbreviations used are Gal ) galactose, GlcNAc ) N-acetylglucosamine, Fuc ) fucose, Neu5Ac ) N-acetylneuraminic acid, and anFuc ) anhydrofucose (Fuc - H2O).
has a carboxyl group at C-1, the sodium cation has a tendency to interact with the oxygen atoms of the carboxyl group as a result of both their terminal location and negative charge (acidic functionality). The loss of acidic hydrogen from the carboxyl group of the sialic acid residue gives rise to the reactive proton that is involved in the Y-type cleavage. Internal fragment ions observed at m/z ) 388.1 and 406.1 Da were assigned to Z1β/ Y2R [(M - Fuc - Neu5Ac) + Na]+ and Y1β/Y2R [(M - anFuc - Neu5Ac) + Na]+. The population of the internal fragments gradually increased above 8.5 µm, and the relative intensities
of the internal fragments showed a maximum around 9.2 µm. The irradiation wavelength most effective for cleavage of the glycosyl bonds of GalNAc-Fuc was in the range of 8.6-9.5 µm. The population of cleaved ions is greater than that of the parent ion [sLex + Na]+ at wavelengths above 8.5 µm. Figure 7 shows relative intensities of the ions obtained after irradiation of [LNFP II + Na]+ ions as a function of wavelength. It can be clearly seen that no fragmentation is observed until 9 µm where the abundance of fragment ions starts to increase. This result suggests that the wavelength effective in breaking the glycosyl bonds of oligosaccharides containing Neu5Ac has a broader range than that for oligosaccharides that lack this saccharide residue. Fragmentation Analysis. Peptides and proteins have three dominant absorptions in the fingerprint region (400-1800 cm-1) of their IR absorbance spectra: These are named amide I, amide II, and amide III. The amide I band occurs at around 6.1 µm (1655 cm-1) and arises predominately as a result of CdO stretching vibrations of the peptide backbone (local mode). The amide II and III bands are mixed vibrational modes (global modes). The amide II band occurs at around 6.45 µm (1550 cm-1) and is the result of an out-of-plane combination of N-H bending and C-N stretching of the backbone. The amide III band is the result of an in-phase combination of C-N stretching and N-H bending and produces a broad absorbance at 7.78.3 mm (1200-1300 cm-1). As shown in Figure 3, the fragment ions of protonated substance P, [Sub P + H]+, were observed in the 5.9-8.3 µm region, which includes the amide I, II, and III bands. The internal energy increases through the absorption of multiple IR photons in the process of IRMPD when the tuned wavelength from FEL coincides with that of the fundamental frequency, ν, of a vibrational mode i. The resonantly absorbed energy (hνi) is redistributed among other modes as a result of the coupling of vibrational modes and is then transferred into
IRMPD Spectroscopy of Biopolymers with FTICR-MS the internal vibrational and rotational states of the ion. The time scale for intramolecular vibrational redistribution (IVR) is typically on the order of picoseconds.11,23 When an ion has gained sufficient energy from multiple photon absorptions, parent ions dissociate, and fragments can be observed in the mass spectra. It is difficult to estimate how many photons are absorbed during the multiple photon excitation process, since the IVR efficiency must be taken into consideration. It should be noted that the laser irradiation time is merely an instrumental parameter in recording IRMPD product spectra. To obtain sufficient product yields, irradiation times of several minutes were required in our experimental setup. However, the time scale of the dissociation processes is much shorter than the irradiation time and may depend on the laser power. Under extremely slow activation conditions, like those applied in these studies, a higher laser power may result in a shorter unimolecular dissociation time and a higher ionic temperature in the equilibrium state. The IR absorption of [sLex + Na]+ was calculated to analyze the vibrational modes in the gas phase. The geometry of the neutral oligosaccharide was first optimized by means of a semiempirical calculation. Quantum chemical calculations were then performed by using the Hartree-Fock method (HF/6-31+G*) on an IBM Regatta supercomputer at the Computational Biology Research Center. In the calculations, the alkali-metal-cationized complex ions [sLex + Na]+ were fully optimized by placing Na+ at the appropriate position relative to the oligosaccharide and obtaining the geometry of the preferential binding position of Na+.22,24 Figure 8a shows the most stable conformation of [sLex + Na]+. In the [sLex + Na]+ complex, the preferred position of the metal cation is near the GlcNAc residue. Figure 8b shows the IR spectrum calculated from the HF/6-31G(d) vibrational frequency analysis (frequency scaling factor ) 0.8). The vibrational bands for the amide groups are found in the regions near 6.1 µm (1640 cm-1), which corresponds to the CdO stretching vibration, and 6.5 µm (1540 cm-1), which corresponds to the mixed modes of C-N stretching and N-H bending in the saccharide units of GlcNAc and Neu5Ac. Figure 8c shows the IRMPD spectrum obtained by monitoring the relative intensity of the fragment ions as a function of FEL wavelength. The frequency scaling factor of 0.8 was used to obtain a good agreement of the spectrum. In the region of the amide group (amide bands I, II, and III), the fragmentation reaction is mainly a loss of Neu5Ac, leading to the cleavage of the glycosyl bond of Neu5Ac-GlcNAc. The effectiveness of wavelengths in the 8.5-9.5 µm range in cleaving the glycosyl bonds is attributed to the highly coupled modes (global modes) in the calculated normal coordinate analysis. These modes involve the glycosyl bond C-1-O-C-1 stretching and bending. The parent intensity of [sLex + Na]+ decreases above 8.5 µm. The reason is that the photoabsorption yield of the saccharide ion in this range is higher than that around 7-8 µm. The average power in this region was 23 mW (micropulse energy ) 0.8 µJ). This indicates that the energy required to cleave the glycosyl bonds is less than that needed to cleave the peptide bonds. IR radiation in this wavelength region is more effectively absorbed by the oligosaccharides than that in the amide band regions. Ab initio calculations of the bond energy of the glycosyl bonds for sialyl Lewis X were also performed. In these calculations, the geometry was fully optimized by the HF/6-31+G* method used for zero-point energy (ZPE) corrections. The calculated glycosyl bond energies of GlcNac-Glc, GlcNAc-Fuc, and Neu5Ac-Glc were 254, 226, and 180 kJ/mol, respectively. The energy of the glycosyl bond between Neu5Ac and Glc, the cleavage of which produces B/Y-type ions, is significantly lower than that
J. Phys. Chem. B, Vol. 110, No. 32, 2006 16115
Figure 8. (a) Optimized structure of [sLex + Na]+ at the HF/6-31+G(d). (b) Calculated vibrational frequencies for sodiated sialyl Lewis X. The blue lines show the experimental FTIR spectrum for sLex. (c) The red lines show the IRMPD spectrum for [sLex + Na]+ and the calculated vibrational frequencies.
of the other bonds. The calculated bond energies (GlcNac-Glc > GlcNAc-Fuc > Neu5Ac-Glc) are directly related to the order of stability of the glycosyl linkages and correlate with the abundance of the experimentally observed fragment ions. IV. Conclusions IRMPD-MS spectroscopy was applied to the study of fragment ions of peptides and oligosaccharides in the gas phase by using a combination of FTICR-MS and FEL irradiation. This study was prompted by the need to gain an understanding of the sensitivity of the fragmentation products obtained from protonated peptides and sodiated oligosaccharides in the gas
16116 J. Phys. Chem. B, Vol. 110, No. 32, 2006 phase to the wavelength of incident IR laser radiation. In the spectroscopic study of peptides, photoproduct ions arising from the cleavage of peptide bonds in the parent amino acid sequence were observed in the 5.7-7.5 µm range, which includes the amide I, II, and III band regions. If the fragment curves of the peptides and oligosaccharides ions are compared, then a dependence on the IR wavelength of the fragment ions produced from the cleavage of peptide and of glycosyl bonds, respectively, is clearly observed. The IR wavelengths effective for cleaving the glycosyl bonds are mainly in the 8.6-9.5 µm range, which corresponds to the global modes of oligosaccharides. For oligosaccharides containing a Neu5Ac sugar residue, cleavage is observed over a broader wavelength range. The rich abundance of product ions formed by cleavage of the glycosyl bond between Neu5Ac and GlcNAc is a consequence of the acidity of Neu5Ac. We are in the process of experimentally investigating a number of polysaccharides and glycopeptides in a series of photodissociation studies by using coupled nano-ESI FTICRMS and FEL instrumentation. IRMPD with FEL can be advantageous, since the reactions involved are more controllable than those produced by other fragmentation methods such as CID and surface-induced dissociation. One of the applications of this technique is the spectroscopic molecular characterization of peptides and oligosaccharides. A disadvantage of our experimental setup is that long irradiation times are needed to observe the fragment ions. A shorter period of FEL irradiation is desirable to achieve more efficient IRMPD spectroscopy. Acknowledgment. This work was supported by the New Energy and Industrial Technology Development Organization as a part of Structural Glycomics Project in Japan. K.F. thanks Professor K. Nakai, Professor A. Iwata, Professor K. Tsukiyama, and T. Morotomi for supporting the FEL at TUS. K.F. also thanks H. Suzuki at the University of Tokyo for the FTIR spectra. Supporting Information Available: Fragmentation analysis for angiotensin II, [AngII + 2H]2+, as a function of the FEL power, dependence of the fragment ions on the FEL laser power, and fragmentation analysis for the singly protonated parent ion, [AngII + H]+. This material is available free of charge via the Internet at http://pubs.acs.org.
Fukui et al. References and Notes (1) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass Spectrom. 1988, 2, 151-153. (2) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299. (3) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Anal. Chem. 1991, 63, A1193. (4) Little, D. P.; Speir, J. P.; Senko, M. W.; O’Connor, P. B.; McLafferty, F. W. Anal. Chem. 1994, 66, 2809-2815. (5) Fenn, J.; Mann, M.; Meng, C.; Wong, S.; Whitehouse, C. Science 1989, 246, 64-71. (6) Loo, J. A.; Edmonds, C. G.; Smith, R. D. Science 1990, 238, 201204. (7) Whitehouse, C. M.; Dreyer, R. N.; Yamashita, M.; Fenn, J. B. Anal. Chem. 1985, 57, 675. (8) Fukui, K.; Naito, Y.; Akiyama, Y.; Takahashi, K. Int. J. Mass Spectrom. 2004, 235, 25-32. (9) Gabryelski, W.; Li, L. ReV. Sci. Instrum. 1999, 70, 4192-4199. (10) Lemaire, J.; Boissel, P.; Heninger, M.; Mauclaire, G.; Bellec, G.; Mestdagh, H.; Simon, A.; Caer, S. L.; Ortega, J. M.; Glotin, F.; Maitre, P. Phys. ReV. Lett. 2002, 89, 273002. (11) 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 Phys. Res., Sect. A 2003, 507, 541-546. (12) Fridgen. T. D.; MacAleese. L.; Maitre. P.; McMahon. T. B.; Boissel. P.; Lemaire. J Phys. Chem. Chem. Phys. 2005, 7, 2747-2755. (13) van Heijnsbergen. D.; von Helden. G.; Meijer. G.; Maitre. P.; Duncan. M. A. J. Am. Chem. Soc. 2002, 1562-1563. (14) Oomens, J.; Moore, D. T.; von Helden, G.; Meijer, G.; Dunbar, R. C. J. Am. Chem. Soc. 2005, 126, 724-725. (15) Moore, D. T.; Oomens, J.; Eyler, J. R.; von Helden, G.; Meijer, G.; Dunbar, R. C. J. Am. Chem. Soc. 2005, 127, 7243-7254. (16) Lucas, B.; Lecomte, F.; Reimann, B.; Barth, H. D.; Gregoire, G.; Bouteiller, Y.; Schermann, J. P.; Desfrancois, C. Phys. Chem. Chem. Phys. 2004, 6, 2600-2604. (17) Lucas, B.; Gregoire, G.; Lemaire, J.; Maitre, P.; Glotin, F.; Schermann, J. P.; Desfrancois, C. Int. J. Mass Spectrom. 2005, 243, 105113. (18) Kuroda, H. Jpn. J. Appl. Phys. 2002, 41, 1-9. (19) Nomaru, K.; Kawai, M.; Yokoyama, M.; Oda, F.; Nakayama, A.; Koike, H.; Kuroda, H. Nucl. Instrum. Methods Phys. Res., Sect. A 2000, 445, 379-383. (20) Yokoyama, M.; Oda, F.; Nomaru, K.; Koike, H.; Sobajima, M.; Kawai, M.; Kuroda, H.; Nakai, K. Nucl. Instrum. Methods Phys. Res., Sect. A 2003, 507, 261-264. (21) Powell, A. K.; Harvey, D. J. Rapid Commun. Mass Spectrom. 1996, 10, 1027-1032. (22) Fukui, K.; Kameyama, A.; Mukai, Y.; Takahashi, K.; Ikeda, N.; Akiyama, Y.; Narimatsu, H. Carbohydr. Res. 2006, 341, 624-633. (23) Fukui, K.; Naito, Y.; Akiyama, Y.; Takahashi, K. Eur. J. Mass Spectrom. 2004, 10, 639-647. (24) Yamagaki, T.; Fukui, K.; Tachibana, K. Anal. Chem. 2006, 78, 1015-1022.
