Combined Electron Capture and Infrared ... - ACS Publications

on-axis dispenser cathode electron source inside the magnet bore of a 9.4-T Fourier ... An MS3 experiment involving IR laser irradiation of the ch...
12 downloads 0 Views 191KB Size
Download PDF
Anal. Chem. 2003, 75, 3256-3262

Combined Electron Capture and Infrared Multiphoton Dissociation for Multistage MS/MS in a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer Kristina Håkansson, Michael J. Chalmers, John P. Quinn, Melinda A. McFarland,† Christopher L. Hendrickson,† and Alan G. Marshall†,*

Ion Cyclotron Resonance Program, National High Magnetic Field Laboratory, Tallahassee, Florida

We have mounted a permanent on-axis dispenser cathode electron source inside the magnet bore of a 9.4-T Fourier transform ion cyclotron resonance mass spectrometer. This configuration allows electron capture dissociation (ECD) to be performed reliably on a millisecond time scale. We have also implemented an off-axis laser geometry that enables simultaneous access to ECD and infrared multiphoton dissociation (IRMPD). Optimum performance of both fragmentation techniques is maintained. The analytical utility of performing either ECD or IRMPD on a given precursor ion population is demonstrated by structural characterization of several posttranslationally modified peptides: IRMPD of phosphorylated peptides results in few backbone (b- and y-type) cleavages, and product ion spectra are dominated by neutral loss of H3PO4. In contrast, ECD provides significantly more backbone (c- and z•-type) cleavages without loss of H3PO4. For N-glycosylated tryptic peptides, IRMPD causes extensive cleavage of the glycosidic bonds, providing structural information about the glycans. ECD cleaves all backbone bonds (except the N-terminal side of proline) in a 3-kDa glycopeptide with no saccharide loss. However, only a charge-reduced radical species and some side chain losses are observed following ECD of a 5-kDa glycopeptide from the same protein. An MS3 experiment involving IR laser irradiation of the charge-reduced species formed by electron capture results in extensive dissociation into c- and z-type fragment ions. Mass-selective external ion accumulation is essential for the structural characterization of these low-abundance (modified) peptides. The combination of electron capture dissociation (ECD)1-3 with either infrared multiphoton dissociation (IRMPD)4,5 or collisionally * Correspondence should be addressed to Dr. Alan G. Marshall, Ion Cyclotron Resonance Program, National High Magnetic Field Laboratory, Florida State University, 1800 East Paul Dirac Dr., Tallahassee, FL 32310-3706. E-mail: [email protected]. † Also members of the Department of Chemistry and Biochemistry, Florida State University, Tallahassee, FL. (1) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1998, 120, 3265-3266. (2) McLafferty, F. W.; Horn, D. M.; Breuker, K.; Ge, Y.; Lewis, M. A.; Cerda, B.; Zubarev, R. A.; Carpenter, B. K. J. Am. Soc. Mass Spectrom. 2001, 12, 245-249.

3256 Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

activated dissociation (CAD)6 can provide extensive and complementary information about biomolecule primary structure in electrospray ionization (ESI) Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometry.7-17 In many cases, IRMPD is preferred over CAD because no collision gas needs to be introduced into the ICR cell. Thus, the pump-down delay required for CAD is eliminated so that analyses can be performed much faster. For peptides and proteins, ECD generally provides more random cleavage than IRMPD, resulting in more extensive sequence coverage.1,7,8,11 The only exception is cleavage on the N-terminal side of proline residues, which is not observed in ECD.8 However, that location is the preferred cleavage site in slow heating techniques, such as IRMPD18 Generally, slow heating techniques result in cleavage of the weakest bonds, such as loss of labile posttranslational modifications.19 ECD, on the other hand, preferentially cleaves the peptide backbone with retention of (3) Zubarev, R. A.; Haselmann, K. F.; Budnik, B.; Kjeldsen, F.; Jensen, F. Eur. Mass Spectrom. 2002, 8, 337-349. (4) Woodlin, R. L.; Bomse, D. S.; Beauchamp, J. L. J. Am. Chem. Soc. 1978, 100, 3248-3250. (5) Little, D. P.; Speir, J. P.; Senko, M. W.; O’Connor, P. B.; McLafferty, F. W. Anal. Chem. 1994, 66, 2809-2815. (6) McLuckey, S. A. J. Am. Soc. Mass Spectrom. 1992, 3, 599-614. (7) Kruger, N. A.; Zubarev, R. A.; Horn, D. M.; McLafferty, F. W. Int. J. Mass Spectrom. 1999, 185/186/187, 787-793. (8) Kruger, N. A.; Zubarev, R. A.; Carpenter, B. K.; Kelleher, N. L.; Horn, D. M.; McLafferty, F. W. Int. J. Mass Spectrom. 1999, 182/183, 1-5. (9) Kelleher, N. L.; Zubarev, R. A.; Bush, K.; Furie, B.; Furie, B. C.; McLafferty, F. W.; Walsh, C. T. Anal. Chem. 1999, 71, 4250-4253. (10) Mirgorodskaya, E.; Roepstorff, P.; Zubarev, R. A. Anal. Chem. 1999, 71, 4431-4436. (11) Zubarev, R. A.; Horn, D. M.; Fridriksson, E. K.; Kelleher, N. L.; Kruger, N. A.; Lewis, M. A.; Carpenter, B. K.; McLafferty, F. W. Anal. Chem. 2000, 72, 563-573. (12) Stensballe, A.; Norregaard-Jensen, O.; Olsen, J. V.; Haselmann, K. F.; Zubarev, R. A. Rapid Commun. Mass Spectrom. 2000, 14, 1793-1800. (13) Shi, S. D.-H.; Hemling, M. E.; Carr, S. A.; Horn, D. M.; Lindh, I.; McLafferty, F. W. Anal. Chem. 2001, 73, 19-22. (14) Hakansson, K.; Cooper, H. J.; Emmett, M. R.; Costello, C. E.; Marshall, A. G.; Nilsson, C. L. Anal. Chem. 2001, 73, 4530-4536. (15) Olsen, J. V.; Haselmann, K. F.; Nielsen, M. L.; Budnik, B. A.; Nielsen, P. E.; Zubarev, R. A. Rapid Commun. Mass Spectrom. 2001, 15, 969-974. (16) Leymarie, N.; Berg, E. A.; McComb, M. E.; O’Connor, P. B.; Grogan, J.; Oppenheim, F. G.; Costello, C. E. Anal. Chem. 2002, 74, 4124-4132. (17) Hakansson, K.; Hudgins, R. R.; Marshall, A. G.; O’Hair, R. A. J. J. Am. Soc. Mass Spectrom. 2003, 14, 23-41. (18) Vekey, K. Mass Spectrom. Rev. 1995, 14, 195-225. (19) McLuckey, S. A.; Goeringer, D. E. J. Mass Spectrom. 1997, 35, 461-474. 10.1021/ac030015q CCC: $25.00

© 2003 American Chemical Society Published on Web 05/15/2003

modifications, allowing their localization.9,10,12-14,20 Similarly, the combination of ECD and CAD can provide complementary structural information about peptide nucleic acids.15 More recently, the combination of ECD and IRMPD generated complementary cleavage in gas-phase oligonucleotide dications.17 The first application of the complementary capabilities of ECD and IRMPD was reported by Hakansson et al. for the structural characterization of an N-glycosylated tryptic peptide.14 Those experiments were performed on a 7-T FT-ICR mass spectrometer employing a standard heated filament as the electron source. Encouraged by those results, we proceeded to implement a similar configuration on a higher performance 9.4-T instrument.21 However, it is difficult to configure a single instrumental geometry that allows for simultaneous optimization of both ECD and IRMPD. Our next design placed an electron dispenser cathode slightly off-axis in the fringe field of the magnet outside the iron shield. That configuration allowed the infrared laser to remain on-axis and provided routine ECD spectra, but a long (20-30 s) irradiation period was required, and the ECD efficiency was relatively low. We later evaluated a novel (temporary) design in which the electron source was placed inside the magnet bore and allowed for variation of both its radial and axial position.22 The optimum radial position of the dispenser cathode was found to be on-axis. However, rapid ECD (5-ms irradiation) was still obtained when the axial distance from the ICR cell was increased (up to 33 cm from the rear trapping cylinder). We therefore decided to keep the electron source on-axis and try an off-axis geometry for the laser. In this paper, we report the performance of a permanent dispenser cathode electron source mounted on-axis inside the bore of the magnet. We also describe the implementation of off-axis IRMPD, allowing simultaneous access to both techniques. The utility of this configuration is illustrated by structural characterization of several phospho- and glycopeptides. METHODS Sample Preparation. Bovine ubiquitin, bradykinin, substance P, acetic acid, and formic acid (FA) were purchased from Sigma (St. Louis, MO) and used without further purification. Peptide stock solutions (100 pmol/µL) were prepared by dissolving the lyophilized samples in HPLC grade water (J. T. Baker, Philipsburg, NJ). For electrospray, the stock solutions were diluted to the required concentration (100 fmol/µL-10 pmol/µL) in either 1:1 methanol (Baker)/water with 2.5% acetic acid, or 1:1 acetonitrile (Baker)/water with 0.1% FA. Phosphorylated peptides (50 µg), AKRRRL(pS)SLRASTS and TFRPRTS(pS)NASTIS, were obtained from Upstate Biotechnology (Lake Placid, NY) and dissolved in 50 µL of water. Those stock solutions were then diluted to 1 pmol/ µL in 1:1 CH3CN/H2O with 0.1% FA. Glycopeptides were generated by dissolving 10 nmol of the lectin from Erythrina corallodendron (Sigma) in 500 µL of buffer containing 0.1 mM CaCl2 and (20) Haselmann, K. F.; Budnik, B. A.; Olsen, J. V.; Nielsen, M. L.; Reis, C. A.; Clausen, H.; Johnsen, A. H.; Zubarev, R. A. Anal. Chem. 2001, 73, 29983005. (21) Quinn, J. P.; Hakansson, K.; Hudgins, R. R.; Hendrickson, C. L.; Marshall, A. G. In 49th ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, 2001; CD-ROM. (22) Quinn, J. P.; Hakansson, K.; McFarland, M. A.; Hendrickson, C. L.; Marshall, A. G. In 50th ASMS Conference on Mass Spectrometry and Allied Topics, Orlando, FL, 2002; CD-ROM.

Figure 1. Schematic diagram (not to scale) of the current configuration of the NHMFL 9.4-T ESI-Q-FT-ICR instrument, showing the electrospray ion source, ion optics, ICR cell, and infrared laser and electron source geometry.

0.1 M NH4HCO3, to which 100 pmol of modified trypsin (Promega, Madison, WI) had been added. Digestion of the protein proceeded at 38 °C for 4 h. The mixture was dried in a Speedvac (Savant, Inc., Farmingdale, NY), and the dried residue was dissolved in 100 µL of water to yield a concentration of 100 pmol/µL. That stock solution was then diluted to a final concentration of 5 pmol/ µL in 50:50 methanol/water with 2.5% acetic acid. Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. All experiments were performed with a passively shielded 9.4-T ESI-Q-FT-ICR mass spectrometer. The magnet bore diameter was 220 mm and the distance from the magnet shield to the rear trapping cylinder of the ICR cell was 760 mm. A microelectrospray23 emitter consisting of a 50-µm i.d. fused-silica capillary, which had been mechanically ground to a uniform thinwalled tip,24 was used to infuse the peptides at a flow rate of 300350 nL/min. The emitter was operated in the positive ion mode. Ions were externally accumulated25 in the front octopole (see Figure 1) for 0.2-10 s. A quadrupole mass filter located behind the front octopole was employed to mass-selectively accumulate the desired peptide ions in the middle octopole (Figure 1).26,27 The event series “accumulation, mass selection/accumulation” was looped 1-50 times, for a total accumulation period of 4-10 s, to build up a precursor ion signal-to-noise ratio of ∼1000:1. After mass-selective accumulation, the ions were transferred from the middle octopole (modified to allow improved ejection of ions along the z axis28) through an octopole ion guide and captured by gated trapping in an open cylindrical cell.29 Stored-waveform inverse Fourier transform (SWIFT)30,31 ejection was applied to further isolate the peptide ion under investigation. After ECD or IRMPD (23) Emmett, M. R.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 1994, 5, 605613. (24) Quinn, J. P.; Emmett, M. R.; Marshall, A. G. In 46th ASMS Conference on Mass Spectrometry and Allied Topics, Orlando, FL, 1998; 1388-1388. (25) Senko, M. W.; Hendrickson, C. L.; Emmett, M. R.; Shi, S. D.-H.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 1997, 8, 970-976. (26) Hendrickson, C. L.; Quinn, J. P.; Emmett, M. R.; Marshall, A. G. In 48th ASMS Conference on Mass Spectrometry and Allied Topics, Long Beach, CA, 2000; CD-ROM. (27) Hendrickson, C. L.; Quinn, J. P.; Emmett, M. R.; Marshall, A. G. In 49th ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, 2001; CD-ROM. (28) Wilcox, B. E.; Hendrickson, C. L.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 2002, 13, 1304-1312. (29) Beu, S. C.; Laude, D. A., Jr. Int. J. Mass Spectrom. Ion Processes 1992, 112, 215-230.

Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

3257

(see below), product ions were subjected to chirp excitation (48 to 720 kHz or 72 to 720 kHz at 150 Hz/µs) and direct-mode broadband detection (512 Kword data points). Hanning apodization and one zero-fill were applied prior to fast Fourier transform, followed by magnitude calculation. Frequency-to-m/z conversion was performed with a two-term calibration equation.32,33 The experimental event sequence was controlled by a MIDAS data acquisition system.34 Each displayed spectrum represents a sum of 1-50 time-domain transients. Electron Capture Dissociation. A 10-mm-diameter dispenser cathode35 (Catalogue no. 1109, Heat Wave, Watsonville, CA) was mounted on the central axis of the system, with the emitting surface displaced 73 cm axially from the center of the ICR cell (i.e., within the magnet bore). That distance was chosen by optimizing both the ECD performance (millisecond time-scale ECD was obtained) and the accessibility of the ICR cell for the off-axis laser beam. A molybdenum grid of ∼80% transparency was placed 8 mm in front of the emitting surface. ECD was performed with the trap plates at 5-10 V by biasing the cathode at - 5.5 V and pulsing the grid potential to +100 V for 3-51 ms. Immediately following the ECD event, the trap plates were set to 2 V, and potentials of +5 V for the grid and +10 V for the cathode were applied for 10 ms to remove the remaining electrons. At all other times, the cathode bias voltage was -0.1 V, and the grid potential was -200 V. The cathode heating power was 11 W, corresponding to an emission current of ∼300 nA. Infrared Multiphoton Dissociation. A 40-W, 10.6-µm CO2 laser (Synrad, Mukilteo, WA) fitted with a 2.5× beam-expander was used for IRMPD. The laser beam was directed to the center of the cell through either an on- or off-axis BaF2 window with a clear aperture of ∼30 mm (see Figure 1 for the off-axis configuration). In the on-axis configuration, the window was located 132 cm from the center of the cell. In the off-axis configuration, the distance between the window and the center of the cell was 136 cm; the central axis of the window lay at an angle of 2.5° relative to the center axis of the system, and intersected the center axis of the system approximately at the center of the ICR cell. Photon irradiation was performed for 450-800 ms at 30-90% laser power. For MS3, precursor ion isolation and electron irradiation was performed as described above. Following electron irradiation, the precursor ions were single-frequency-ejected, and charge-reduced product ions were irradiated with the IR laser for 450 ms at 70% laser power. RESULTS AND DISCUSSION Rapid ECD of Standard Peptides and Proteins. Figure 2 shows a single-scan ECD FT-ICR mass spectrum of the neuropeptide, substance P, with a product ion signal-to-noise ratio of ∼100:1 (30) Marshall, A. G.; Wang, T.-C. L.; Ricca, T. L. J. Am. Chem. Soc. 1985, 107, 7893-7897. (31) Guan, S.; Marshall, A. G. Int. J. Mass Spectrom. Ion Processes 1996, 157/ 158, 5-37. (32) Ledford, E. B., Jr.; Rempel, D. L.; Gross, M. L. Anal. Chem. 1984, 56, 2744-2748. (33) Shi, S. D.-H.; Drader, J. J.; Freitas, M. A.; Hendrickson, C. L.; Marshall, A. G. Int. J. Mass Spectrom. 2000, 195/196, 591-598. (34) Senko, M. W.; Canterbury, J. D.; Guan, S.; Marshall, A. G. Rapid Commun. Mass Spectrom. 1996, 10, 1839-1844. (35) Tsybin, Y. O.; Hakansson, P.; Budnik, B. A.; Haselmann, K. F.; Kjeldsen, F.; Gorshkov, M.; Zubarev, R. A. Rapid Commun. Mass Spectrom. 2001, 15, 1849-1854.

3258 Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

Figure 2. Single-scan MS/MS spectrum obtained from ECD (20 ms irradiation) of doubly protonated substance P. All backbone bonds are cleaved, except the N-terminal side of proline.

after 20 ms of electron irradiation. Cleavage at all expected backbone bonds is observed. Note that ECD mainly cleaves the backbone amine bonds, resulting in product ions of the c and z• types.1 The short irradiation period makes ECD compatible with experiments such as on-line liquid chromatography36,37 and hydrogen/deuterium exchange. Figure 3 shows a 40-scan ECD spectrum of the small protein, ubiquitin, after 51 ms irradiation (five-min duration for the total experiment). In that case, 67 of 75 backbone bonds are cleaved. For comparison, 55 cleavages were observed in a previous 40-scan spectrum.21 However, the total duration of that experiment was ∼25 min. We have recently obtained ECD MS/MS spectra similar to that shown in Figure 3 with an irradiation period as short as 3 ms. On-Axis vs Off-Axis IRMPD. To generate an IRMPD performance benchmark, we collected a number of product ion spectra with the laser in its original, on-axis position. The laser was then moved to the off-axis position and aligned with respect to a previously sited HeNe laser beam. That procedure directed the CO2 laser such that the beam (∼1 cm diameter) crosses the z axis at the center of the ICR cell (see Figure 1). Similar product ion spectra were obtained from both on-axis and off-axis IRMPD of equivalent populations of bradykinin [M + 2H]2+ ions (data not shown). However, for that peptide, a slightly longer irradiation period was required when the laser was positioned off-axis. Note that those experiments were performed with the instrument in an intermediate configuration in which the rear flange contained two BaF2 windows but no ECD dispenser cathode probe. We were therefore able to compare the relative performance of the on- and off-axis IRMPD under identical experimental conditions within a few hours of each other. Following independent demonstration of millisecond-duration ECD and efficient off-axis IRMPD, we converted the instrument to the final configuration shown in Figure 1. All results presented above subsequently proved to be equivalent and reproducible (data not shown). The experiments described below were all performed with the instrument in its final configuration. (36) Palmblad, M.; Tsybin, Y. O.; Ramstrom, M.; Bergquist, J.; Hakansson, P. Rapid Commun. Mass Spectrom. 2002, 16, 988-992. (37) Davidson, W.; Frego, L. Rapid Commun. Mass Spectrom. 2002, 16, 993998.

Figure 3. Product ion spectrum (bottom) and a 100-Th window with fragment ion labels (top) obtained from ECD (51-ms irradiation, 40 scans) FT-ICR MS/MS of the 9+ charge state (quadrupole isolated) of bovine ubiquitin. All observed isotopic distributions in the displayed window could be assigned as sequence-specific fragment ions. In the complete spectrum, cleavage at 67 of 75 bonds was observed.

ECD and Off-axis IRMPD of Phosphorylated Peptides. It is often beneficial to perform both ECD and IRMPD on the same sample, because the derived structural information is typically complementary. For example, we have examined the fragmentation behavior of two serine-phosphorylated peptides. In both cases, the most abundant product ion following IRMPD corresponds to loss of H3PO4. Losses of H2O and NH3 are also observed from multiply protonated precursor ions during IRMPD. Figure 4 shows the product ion spectrum obtained from off-axis IRMPD of quadrupole and SWIFT isolated [AKRRRL(pS)SLRASTS + 3H]3+ phosphopeptide ions. During an IRMPD experiment, product ions are not removed from the path of the laser beam. Therefore, it should be possible to further fragment the [M - H3PO4 + 3H]3+ product ions. However, although higher laser fluence somewhat increased the abundance of sequence-specific product ions, the spectrum shown in Figure 4 represents the most informative IRMPD data obtained from this phosphopeptide ion. The peptide was cleaved at 5 of 13 amide bonds, yielding five y-type ions and one b-type ion.38,39 All product ions containing the originally phosphorylated serine residue were detected after neutral loss of H3PO4, as evidenced by a residue mass corresponding to serine minus 18 Da. An unmodified y7 ion was observed, whereas the y8 ion could be assigned only after taking into account phosphate (38) Roepstorff, P.; Fohlman, J. Biomed. Mass Spectrom. 1984, 11, 601-601. (39) Biemann, K. Biomed. Environ. Mass 1988, 16, 99-111.

Figure 4. Product ion spectrum obtained from off-axis IRMPD FTICR MS/MS of a population of quadrupole- and SWIFT-isolated [AKRRRL(pS)SLRASTS + 3H]3+ phosphopeptide ions. The spectrum is dominated by ions resulting from neutral loss of H3PO4, NH3, and H2O. Five (of 13) peptide backbone bonds are broken, and the location of the phosphorylation site is identified only by observation of the (y8 - H3PO4 - NH3) ion (present at very low magnitude) and the singly and doubly charged (b7 - H3PO4) ion. Irradiation was for 500 ms at ∼36-W laser power, and the data represent a sum of 10 scans.

and ammonia loss, thereby identifying the site of phosphorylation as serine residue number seven. Additional evidence is provided Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

3259

Figure 5. Product ion spectrum obtained from ECD (20-ms irradiation) FT-ICR MS/MS of the same quadrupole- and SWIFTisolated phosphopeptide as in Figure 4. Twelve of 13 peptide backbone bonds are cleaved, and the location of the phosphate is readily assigned by observation of the abundant c7 ions.

by the observation of a b7 ion with a mass deviation of 18 Da. The corresponding unmodified b7 ion was not seen. However, the location of the modification was not immediately obvious and would have been difficult to assign in a de novo sequencing experiment. The product ion spectrum following ECD of the [M + 3H]3+ phosphopeptide ion is shown in Figure 5. The peptide backbone was cleaved at 12 of the 13 possible sites. Three z•-type and 11 c-type ions were detected. No loss of H3PO4 was observed during ECD, and the site of modification can be easily located by the appearance of abundant c7 ions. In combination, the IRMPD and ECD data reveal phosphorylation (on the basis of loss of H3PO4) as well as the entire peptide sequence (including the site of phosphorylation). Similar fragmentation behavior was observed for the [M + 2H]2+ ion of a second phosphorylated peptide (TFRPRTS(pS)NASTIS, data not shown). The product ion spectrum following IRMPD of that peptide was dominated by neutral loss of H3PO4 from the precursor ion and by dephosphorylated product ions. Five of 13 backbone bonds were cleaved, generating one y ion and five b ions. Following ECD of the same peptide ion species, 7 of 13 backbone bonds were broken, forming six c ions and one z• ion. No losses of H3PO4 were observed. These results highlight the benefits of a combined IRMPD/ECD approach for localization of protein phosphorylation. ECD and Off-Axis IRMPD of N-Glycosylated Peptides. The lectin from Erythrina corallodendron (28 kDa, 255 amino acid residues) was chosen as a model glycoprotein. The protein consists of two identical subunits, each with two N-glycosylation sites, asparagine 1740 and asparagine 113.41,42 Both sites carry the same glycan structure (xylose type,43 see Figure 6) and are fully occupied and homogeneous. Tryptic digestion results in two (40) Adar, R.; Richardson, M.; Lis, H.; Sharon, N. FEBS Lett. 1989, 257, 8185. (41) Young, N. M.; Watson, D. C.; Yaguchi, M.; Adar, R.; Arango, R.; RodriguezArango, E.; Sharon, N.; Blay, P. K. S.; Thibault, P. J. Biol. Chem. 1995, 270, 2563-2570. (42) Kelly, J. F.; Locke, S. J.; Ramaley, L.; Thibault, P. J. Chromatogr., A 1996, 720, 409-427. (43) Kamerling, J. P. Pure Appl. Chem. 1991, 63, 465-472.

3260 Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

Figure 6. Product ion FT-ICR mass spectrum (50 scans) following ECD (40-ms irradiation) of a quadrupole- and SWIFT-isolated, triply protonated N-glycosylated tryptic peptide of mass 3013 Da. GlcNAc ) N-acetyl glucosamine, Fuc ) fucose, Man ) mannose, Xyl ) xylose. All backbone amine bonds, except the N-terminal side of proline, are cleaved, but no cleavage is observed within the glycan. Peaks corresponding to glycosylated fragment ions are designated by #. ν2 and ν3 refer to harmonics of the parent ion signal.

glycopeptides: one containing amino acid residues [100-116] (mass 3013.344, glycopeptide 1), and one containing amino acid residues [1-36] (mass 5024.380, glycopeptide 2). Both glycopeptides are observed as triply protonated ions after ESI (m/z 1005.455 ( 0.005 and m/z 1675.801 ( 0.005).14 In previous experiments with a 7-T FT-ICR instrument equipped with on-axis IRMPD and a filament electron source, we found that ECD and IRMPD provide strictly complementary information for glycopeptide 1: ECD (30-s irradiation, 100 scans) cleaved 11 of 16 peptide backbone bonds (nine c-type and two z-type ions were observed) with retention of the glycan structure. In contrast, on-axis IRMPD cleaved all glycosidic bonds, thereby allowing structural charac-

Figure 8. Product ion FT-ICR mass spectrum (50 scans) following MS3 of the same glycopeptide as in Figure 7. The charge-reduced [M + 3H]2+• ion formed upon electron irradiation was further isolated and irradiated with IR photons (450 ms, ∼28-W laser power). Extensive backbone fragmentation forms c- and z-type ions.

Figure 9. Summary of the structural information obtained from ECD and IRMPD MS/MS and MS3 of a 5-kDa N-glycosylated tryptic peptide. IRMPD of the even-electron [M + 3H]3+ ion resulted in cleavage of all glycosidic bonds, whereas ECD followed by IRMPD of the [M + 3H]2+• radical ion formed upon electron irradiation resulted in cleavage of 25 of 35 backbone bonds.

Figure 7. Product ion FT-ICR mass spectrum (50 scans) following off-axis IRMPD (450 ms, ∼12-W laser power) of a quadrupole- and SWIFT-isolated, triply protonated N-glycosylated tryptic peptide of mass 5024 Da. Note the extensive fragmentation of the glycosidic bonds.

terization of the glycan.14 However, we could not build up enough precursor ion signal for glycopeptide 2 to allow for detection of any product ions. Figure 6 shows the product ion spectrum (50 scans) following ECD of glycopeptide 1 obtained with the 9.4-T instrument in its final configuration (i.e., on-axis dispenser cathode, Figure 1). In that experiment, 15 of 16 backbone bonds were cleaved, and 14 c-type and 12 z-type ions were detected after 40 ms of irradiation (corresponding to all expected cleavages, because one amino acid residue is proline). As indicated in the Figure, both product ion types were observed as either radical or even-electron species.

Such hydrogen migration behavior has been observed previously.7,10,11,44 In addition to the backbone fragments, we observed side chain cleavages, such as loss of NH3, and loss of 45 Da, which corresponds to fragmentation within an asparagine or glutamine side chain.45 The off-axis IRMPD spectrum of glycopeptide 1 was very similar to the one obtained from the previous on-axis experiment and contained equivalent information (data not shown). The quadrupole mass filter of our 9.4-T FT-ICR mass spectrometer (see Figure 1) allows for mass-selective external ion accumulation26,27,46,47 in a middle octopole (see the Methods Section). We used that capability to build up sufficient precursor (44) Hakansson, K.; Emmett, M. R.; Hendrickson, C. L.; Marshall, A. G. Anal. Chem. 2001, 73, 3605-3610. (45) Cooper, H. J.; Hudgins, R. R.; Hakansson, K.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 2002, 13, 241-249. (46) Belov, M. E.; Nikolaev, E. N.; Anderson, G. A.; Auberry, K. J.; Harkewicz, R.; Smith, R. D. J. Am. Soc. Mass Spectrom. 2001, 12, 38-48.

Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

3261

ion signal for ECD and IRMPD of glycopeptide 2. As for glycopeptide 1, IRMPD resulted in extensive fragmentation of the glycosidic bonds (see Figure 7), providing structural information for the glycan. However, upon electron irradiation, only chargereduction from the even-electron [M + 3H]3+ ion to the radical ion [M + 3H]2+• and some side chain fragmentation occurred (data not shown). No hydrogen atom ejection to form even-electron [M + 2H]2+ ions was observed. ECD and Off-Axis IRMPD MS3 of an N-Glycosylated Peptide. McLafferty and co-workers have suggested that the scarce fragmentation observed in ECD of large (>20 kDa) proteins is due to cleavage of backbone covalent bonds without disruption of the weaker intramolecular noncovalent bonds responsible for secondary and tertiary structure. Such noncovalent interactions may prevent separation of the ECD products.48-50 Consequently, the number of fragment ions observed in “activated ion ECD”, in which the parent ions are collisionally or IR-heated during electron irradiation, is substantially higher than in regular ECD.48,51,52 Encouraged by these results, we performed MS3 by isolating and irradiating with the IR laser the [M + 3H]2+• ions formed following electron irradiation of glycopeptide 2 (see above). The resulting spectrum is shown in Figure 8. Extensive c- and z-type fragmentation occurs. (Note, however, that all fragment ions are evenelectron ions.) We were surprised to find that no significant carbohydrate loss was observed at the laser fluence used to achieve optimum fragmentation efficiency of the [M + 3H]2+• radical species (450-ms irradiation, ∼28-W laser power). A lower laser fluence (450 ms, ∼12-W laser power) was used to obtain extensive fragmentation of glycosidic bonds in IRMPD of the evenelectron [M + 3H]3+ ion, as discussed above. A possible explanation is that the [M + 3H]2+• ions are excited during precursor ion isolation as a result of an imperfect isolation waveform. Excitation results in a larger cyclotron radius so that the precursor (47) Belov, M. E.; Nikolaev, E. N.; Anderson, G. A.; Udseth, H. R.; Conrads, T. P.; Veenstra, T. D.; Masselon, C. D.; Gorshkov, M. V.; Smith, R. D. Anal. Chem. 2001, 73, 253-261. (48) Horn, D. M.; Ge, Y.; McLafferty, F. W. Anal. Chem. 2000, 72, 4778-4784. (49) Horn, D. M.; Breuker, K.; Frank, A. J.; McLafferty, F. W. J. Am. Chem. Soc. 2001, 123, 9792-9799. (50) Breuker, K.; Oh, H.; Horn, D. M.; Cerda, B. A.; McLafferty, F. W. J. Am. Chem. Soc. 2002, 124, 6407-6420. (51) Ge, Y.; Lawhorn, B. G.; EINaggar, M.; Strauss, E.; Park, J. H.; Begley, T. P.; McLafferty, F. W. J. Am. Chem. Soc. 2002, 124, 672-678. (52) Sze, S. K.; Ge, Y.; Oh, H.; McLafferty, F. W. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 1774-1779.

3262

Analytical Chemistry, Vol. 75, No. 13, July 1, 2003

ions might move in and out of the area covered by the laser beam. Consequently, higher fluence is required to activate a majority of the precursor ions. While tuning the IR laser fluence, we did not observe any c or z ions at lower fluence, establishing that the mechanism of ion activation is, in fact, IR heating. A summary of the structural information obtained for glycopeptide 2 is given in Figure 9. IRMPD of the even-electron [M + 3H]3+ ion is clearly complementary to IRMPD of the [M + 3H]2+• radical ion: The former results in dissociation of all glycosidic bonds, whereas the latter cleaves 25 of 35 backbone bonds. CONCLUSION Our current ECD configuration with the electron source located on-axis, inside the magnet bore, achieves reliable ECD in milliseconds. The off-axis geometry of the infrared laser provides the same IRMPD performance as our earlier on-axis configuration. Thus, both tandem mass spectrometry techniques are optimized and available simultaneously. ECD and IRMPD are applied to phosphopeptide and glycopeptide structural characterization. Evidence for phosphorylation is observed in the IRMPD data through the neutral loss of H3PO4 from precursor and product ions; however, a limited number of backbone cleavages are obtained. For the equivalent ECD data, increased peptide backbone cleavages and no loss of H3PO4 are observed, allowing for simple and direct localization of the modification. For N-glycosylated peptides, ECD or ECD followed by IRMPD provides extensive backbone fragmentation, whereas IRMPD results in extensive fragmentation of glycans, for complementary structural information. Mass-selective external ion accumulation builds up sufficient precursor ion signal for ECD and IRMPD MSn of phosphorylated and glycosylated peptides. ACKNOWLEDGMENT This work was supported by the NSF National High Field FT-ICR Facility (CHE-99-09502), Florida State University, and the National High Magnetic Field Laboratory in Tallahassee, Florida. We thank Dr. Carol L. Nilsson for providing the lectin digest and Dr. Mark R. Emmett for helpful discussions.

Received for review January 7, 2003. Accepted April 8, 2003. AC030015Q