Systematic LC-MS Analysis of Labile Post-Translational Modifications

Mar 14, 2009 - Abstract: Most proteins are post-translationally modified and the characterization of modified peptides in complex mixtures generated b...
0 downloads 0 Views 976KB Size
Download PDF
Systematic LC-MS Analysis of Labile Post-Translational Modifications in Complex Mixtures Christine Carapito,†,‡,# Clementine Klemm,†,‡,# Ruedi Aebersold,†,§,|,⊥ and Bruno Domon*,†,‡ Institute of Molecular Systems Biology, ETH Zurich, 8093 Zurich, Switzerland, Zu ¨ rich Glycomics Initiative (GlycoInit), ETH Zurich, 8049 Zurich, Switzerland, Institute for Systems Biology, Seattle, Washington 98103, Faculty of Science, University of Zurich, 8057 Zurich, Switzerland, and Competence Center for Systems Physiology and Metabolic Disease, ETH Zurich, 8093 Zurich, Switzerland Received October 17, 2008

Abstract: Most proteins are post-translationally modified and the characterization of modified peptides in complex mixtures generated by enzymatic digestion of multiple proteins remains a major analytical challenge. We describe an integrated LC-MS workflow implemented on a hybrid quadrupole time-of-flight (Q-ToF) instrument to detect modified peptides in a complex peptide sample and establish the nature of the modification. The method is based on the alternating acquisition of full mass spectra under different collision conditions inducing the cleavage of the substituents. Modified peptides are detected based on their specific fragmentation generating the nonmodified peptide backbone and reporter ions in the low mass region. The two mass analyzer stages of a Q-ToF instrument are used to eliminate the low mass chemical background in the quadrupole and thus facilitate the detection of low mass reporter ions in the ToF. Off-line data processing enables detection of one (or even multiple) modifications and the modified candidates are subsequently sequenced in a directed MS/MS mode. The technique was applied to the analysis of O-GlcNAc peptides, a very complex mixture of N-linked glycopeptides, and a phosphotyrosine peptide. Keywords: Post-translational modifications • LC-MS • reporter ions • tyrosine phosphorylation • O-glycosylation • N-glycosylation • LC-Chip-Q-ToF • specific detection • dual full mass spectrum acquisition

Introduction More than three hundred types of post-translational modifications (PTMs) of proteins have been described,1 and mass spectrometry is the method of choice for detecting, mapping the sites, and characterizing the nature of the modifications. Major efforts are invested in current proteomics studies toward glycosylation and phosphorylation, two biologically highly * Corresponding author: Dr. Bruno Domon. Tel: +41 44 633 20 88. E-mail: [email protected]. † Institute of Molecular Systems Biology. ‡ Zu ¨ rich Glycomics Initiative (GlycoInit). # Authors contributed equally to this work. § Institute for Systems Biology. | University of Zurich. ⊥ Competence Center for Systems Physiology and Metabolic Disease.

2608 Journal of Proteome Research 2009, 8, 2608–2614 Published on Web 03/14/2009

relevant types of modifications.2,3 These include, more specifically, N-linked glycosylation at asparagine residues within the consensus sequence motif Asn-X-Ser/Thr common in many cellular processes in eukaryotes,4 and O-linked glycosylation at serine and threonine residues. O-glycosylation by a single N-acetylglucosamine (GlcNAc) unit is known to play an antagonistic role to protein phosphorylation.5,6 Phosphorylation and particularly tyrosine phosphorylation is involved in receptor-mediated signal transduction affecting numerous biological processes and is of major interest.7 Despite the recent advances in mass spectrometry technologies in terms of sensitivity, resolving power and mass accuracy, the identification of modified peptides in complex mixtures remains a challenge, often requiring enrichment or selective detection methods for specific classes of modified peptides.8-12 Glycosylation and phosphorylation (on serine and threonine residues) are labile modifications and modified peptides can undergo loss of the substituent under conditions used in electrospray ionization. This effect is dramatically enhanced when collision activation is applied, either in the ion source or in the collision cell.13,14 Indeed, routine collision energies applied in quadrupole time of flight (Q-ToF) collision activated dissociation (CAD) experiments (20-40 eV) generally result in the cleavage of the modification and the formation of extensive series of b- and y-peptide fragment ions lacking the modification. Soft collision activation conditions can be adjusted to induce only preferential cleavage of the labile modification bonds without backbone fragmentation. Depending on the nature of the modification, the fragmentation of the modified peptides results either (i) in the loss of a neutral species, thus, generating only a signal corresponding to the nonmodified peptide backbone; or (ii) in the generation of two signals, one modification-specific fragment ion observed in the low mass region of the spectrum together with a signal corresponding to the nonmodified peptide backbone. The specific signals observed in the low molecular mass region are referred to as reporter ions for a given modification. Glycosylated and phosphorylated peptides undergo both fragmentation pathways: on one hand, they exhibit losses of the substituent as a neutral species, and on the other, they generate diagnostic reporter ions at m/z 204 and/or 366 (glycopeptides) or m/z 79 (phosphopeptides in negative mode). So far, several mass spectrometry-based approaches were proposed to take advantage of the specific fragmentation 10.1021/pr800871n CCC: $40.75

 2009 American Chemical Society

technical notes

Systematic LC-MS Analysis of Labile PTMs in Complex Mixtures behavior of modified peptides in order to detect them in complex mixtures. The appearance of characteristic reporter ions was exploited for the detection of PTMs using precursor ion scanning on triple quadrupole instruments,15 as well as on Q-ToF instruments, although less effectively.16 Alternatively, neutral loss scanning was successfully applied for the detection of pSer/pThr-peptides showing an easily induced loss of phosphoric acid on triple quadrupole instruments.17 In source induced fragmentation was also applied to generate reporter ions to enable the detection of glycosylated peptides in complex mixtures.13,14 Finally, a precursor ion discovery mode, showing similarities with our approach, was described on Q-ToF instruments for the detection of phosphopeptides in low complexity digests.12 In the present study, we present a generic integrated workflow for the detection of post-translationally modified peptides in complex mixtures based on their characteristic fragmentation induced by low collision activation. The identification of the modified peptides is based on the coincidence of two signals, that is, neutral loss and reporter ions. Our workflow includes the acquisition of alternating full mass spectra without and with low collision energy (in the collision cell) over the entire LC-MS run, without any precursor ion selection. Additionally, the two stages of a hybrid quadrupole time-of-flight instrument are exploited to remove low mass chemical background in the first quadrupole in order to allow proper detection of the reporter ions in the ToF. As full mass spectra are acquired, the extraction of the two (or even more) characteristic signals for a given modification (loss of the modification and reporter ion) enables unambiguous detection of modified peptide candidates. No a priori knowledge of the precursor ion masses or of the type of modification is required. Therefore, several types of modifications can be detected in one single analysis. This contrasts with conventional neutral loss or precursor ion experiments, which typically record only a single neutral loss or reporter ion. After acquisition of this first data set, our workflow integrates an off-line data analysis step in which the alternating full mass spectra are processed to determine potentially modified peptides. Those candidates are incorporated in an inclusion list to be subsequently sequenced in a directed CAD experiment.18 This MS/MS experiment enables us to confirm the nature of the modification, sequence the peptide, and localize the modification site. This technique was applied to the identification of various forms of glycosylation including O-GlcNAc containing peptides, a very complex mixture of partially deglycosylated N-linked glycopeptides from a yeast whole cell lysate. Finally, to illustrate the versatility of the presented approach, we applied the technique to a phosphotyrosine containing peptide.

Experimental Section Materials. All chemicals, if not otherwise mentioned, were bought at the highest available purity from Sigma-Aldrich, Buchs, Switzerland. Endo-β-N-acetylglucosaminidase H (Endo H) and porcine trypsin (sequencing grade) were purchased from Roche Diagnostics and Promega (Madison, WI), respectively. Sample Preparation. The protein samples (isolated proteins or whole yeast cell lysate) were solubilized in 0.1 M ammonium bicarbonate buffer containing 8 M urea at a final concentration of 3 mg/mL. After dithiothreitol reduction and iodoacetamide alkylation of cysteine residues, the solution was diluted to final 1 M urea concentration, and the proteins were digested with trypsin at 37 °C. After digestion, the peptide mixtures were

desalted using Sep-Pak tC18 cartridges (Waters, Milford, MA) and eluted with 80% acetonitrile. For invertase and yeast digests, after C18 cleanup, the peptides were evaporated to dryness, resolubilized in 50 mM sodium acetate buffer, pH 5.5, and digested with 10 mU Endo H overnight at 37 °C. A second C18 cleanup step was performed after partial deglycosylation. For the galactosylation of O-GlcNAc peptides, mutated Gal-T1 (Y289L) from the Click-iT O-GlcNAc enzymatic labeling kit Molecular Probes (Eugene, OR) was used. The enzymatic introduction of galactose was performed according to the manufacturer’s instructions except that UDP-Gal (Sigma) was used instead of UDP-GalNAz (azido-modified galactose). NanoLC-MS Setup. Reverse phase nanoLC separation was performed on a µ-fluidic HPLC chip cube system (C18 reversed phase column, 5 µm particle size, 75 µm i.d., 150 mm length) interfaced to a Q-ToF MS instrument (Agilent Technologies, Santa Clara, CA). Separation was performed at 300 nL/min flow rate using a 50 min gradient from 2 to 40% B (eluent A, 0.1% (v/v) formic acid in water and eluent B, 90% (v/v) acetonitrile, 0.1% (v/v) formic acid in water). Spray voltage was set to 1800 V and fragmentor voltage to 150 V. The instrument was run using a prototype software allowing the acquisition of full mass spectra without/with collision energy, without precursor isolation and with low mass chemical background filtering. In this special mode, a user-defined m/z cutoff value is given, that is, only ions with m/z-values above this cutoff pass into the collision cell (Q2) where they are subjected to collision activation. To maintain the mass accuracy, a short regular full mass range scan is performed in MS mode without filtering. Acquisition cycles were set as follows: first acquisition was run in regular MS mode, second acquisition was run in special mode with a cutoff value at m/z 300 and no collision energy followed by a third (or more) acquisition in special mode with a m/z cutoff value of 300 and a low collision energy. Acquisition time for MS and special mode scans was set to 67 ms and to 1 s, respectively.

Results and Discussion Concurrent Detection of Modified Peptides in Complex Mixtures by Alternating Full Mass Spectra without/with Collision Activation. This study aimed at developing an integrated mass spectrometry-based workflow capable of detecting different PTMs and their respective peptides in a single LC-MS analysis of complex mixtures. This was accomplished by the implementation of an innovative acquisition protocol on a hybrid quadrupole time-of-flight instrument consisting in the acquisition of alternating full mass spectra with and without collision activation. The method is schematically described in Figure 1. In a first stage, full mass spectra are acquired under alternating collision conditions (i.e., no or low collision energy) over the entire LC-MS run. To facilitate the detection of reporter ions in the low mass region of the spectrum, a user-defined m/z cutoff is applied to the first quadrupole ensuring that the low mass region of the spectrum is free of chemical background. The first ToF full mass spectrum is obtained by applying only translational energy to pass the ions through the collision cell (Q2) without fragmentation. In a second spectrum, additional kinetic energy is applied to induce soft collisions to cleave labile bonds, such as phosphate or glycan substituents attached to a specific amino acid. The two (or even more) characteristic signals for a given modification (loss of the Journal of Proteome Research • Vol. 8, No. 5, 2009 2609

technical notes

Carapito et al.

Figure 1. Principle of the alternating full mass spectra acquisition protocol without/with collision activation performed on a Q-ToF instrument and applied to the detection of modified peptides. Only ions with masses above a user defined m/z cutoff (here m/z 300) are transmitted through Q1 yielding a clean low mass region. In the first acquisition, the modified peptides remain intact, whereas in the second acquisition, the soft collision activation (in the range of 8-15 V) induces the loss of the modification. Modified peptides can thus be detected by the concomitant presence of (i) a signal corresponding to the loss of the modification from the peptide and (ii) the reporter ion in the low mass range of the spectrum. After off-line analysis of the data, the potentially modified peptides are included on an inclusion list before being subsequently sequenced by directed CAD.

modification and reporter ions) are extracted off-line using an in-house developed feature detection tool, SuperHirn,19 to establish a list of potentially modified peptides. Then, in order to confirm the nature of the modification, sequence the peptides and precisely localize the modification sites, a directed LC-MS/MS run is performed in a subsequent experiment. The list of potentially modified candidates is added to an inclusion list and a CAD spectrum is triggered on each peptide as soon as it is detected on the full mass spectrum above a specified threshold as previously described.20 The specificity of this approach is that the identification of potentially modified peptides is based on the coincidence of several specific signals: specific reporter ions, and the ion pair of the nonmodified and the modified precursor. This significantly increases the specificity of the method. Noteworthy, no a priori selection or definition of a specific neutral loss or reporter ion is required as acquisition is performed in full mass spectra mode. Therefore, several types of PTMs can be measured concurrently, in a single LC-MS run. Application to O-GlcNAc Containing Peptides. O-GlcNAc is an abundant modification present on many intracellular proteins involved in a variety of cellular functions such as gene expression, neuronal signaling and synaptic plasticity. Because of its broad biological importance, the modification has been attracting increasing interest.5 The labile glycosidic bond of the GlcNAc residue is readily cleaved off during the analytical process, which makes the detection of such peptides challenging using conventional LC-MS experiments.14 In spite of 2610

Journal of Proteome Research • Vol. 8, No. 5, 2009

significant efforts, a standard and robust analytical technique for the detection of O-GlcNAc modified peptides has yet to be developed. O-GlcNAc peptides show both fragmentation characteristics described earlier as they yield a characteristic oxonium ion at m/z 204.09 and the nonmodified peptide backbone. The reporter ion at m/z 204.09 is often difficult to detect due to low mass chemical background. For all these reasons, OGlcNAc peptides represent an ideal model system to evaluate the proposed approach. An initial experiment was performed on the synthetic O-GlcNAc peptide YSPTS#PSK to determine optimal soft collision conditions. Acquisitions were performed with increasing collision energies ranging from 5 to 25 V, and indicated that at 8 V extensive deglycosylation was induced without backbone fragmentation (Figure 4A). Since O-GlcNAc modified peptides are generally low abundant in real samples, an additional enrichment step may be required prior to the mass spectrometry analysis.5,21 The enrichment using wheat germ agglutinin (WGA) lectin presents low affinity and is typically not quantitative.21 Alternatively, the enzymatic derivatization of the GlcNAc residues with a galactose unit enables the isolation of N-acetyllactosamine peptides using the lectin Ricinus communis (RCA1) with higher affinity than WGA.22 We therefore applied our strategy on derivatized O-LacNAc-homologues (Gal-GlcNAc) to study their mass spectrometry behavior. Derivatization was performed on the model O-GlcNAc peptide YSPTS#PSK to generate the O-LacNAc homologue. The analysis of the resulting product yielded a full

Systematic LC-MS Analysis of Labile PTMs in Complex Mixtures

technical notes

Figure 2. Alternating full mass spectra acquired under no and soft collision conditions during the LC-MS analysis of alpha-crystallin A and B chains’ tryptic digest. In panel A, the O-GlcNac peptide 158AIPVS#REEKPSSAPSS173 from alpha-crystallin B chain is shown acquired under no and low collision conditions at a retention time of 7.9 min (m/z 615.65 corresponds to the triply charged glycosylated peptide 158 AIPVS#REEKPSSAPSS173; m/z 821.43 corresponds to doubly charged nonglycosylated peptide 158AIPVSREEKPSSAPSS173; m/z 204.09 corresponds to the singly charged GlcNAc-oxonium ion). In panel B, the phosphopeptide 13T*LGPFYPSR21 from alpha-crystallin A chain is shown acquired under no and low collision conditions at a retention time of 13.4 min (m/z 559.26 corresponds to the doubly charged phosphorylated peptide 13T*LGPFYPSR21; m/z 510.27 corresponds to the nonphosphorylated doubly charged peptide 13 T*LGPFYPSR21).

mass spectrum (measured at low collision energy) with signals at m/z 616.29, m/z 366.15, and m/z 204.09 corresponding to the doubly charged O-LacNAc peptide YSPTS∧PSK, the LacNAc and GlcNAc oxonium ions, respectively (Figure 4B). These results indicate that O-linked peptides with larger glycan moieties were also effectively detected by this technique. Furthermore, the formation of a second reporter ion at m/z 366.23 increases the specificity of the glycopeptide detection. In the following, the optimized experimental conditions were applied to LC-MS analysis of a tryptic digest of a mixture of alpha-crystallin (A and B chains), which is known to be O-glycosylated at serine 162 (A chain), and at threonine 170 (B chain).23,24 Both O-glycosylation sites were readily detected in the peptide mixture. Overall three O-GlcNAc peptides were identified at different retention times corresponding to the sequences 164EEKPAVT#AAPK174 (B chain), 158AIPVS#R162 (A chain) and 158AIPVS#REEKPSSAPSS173 (A chain, C-terminal peptide). Figure 2A shows two consecutive spectra acquired for the C-terminal fragment of the A-chain 158 AIPVS#REEKPSSAPSS173 acquired without and with low collision energy. While in the first spectrum a signal was detected at m/z 615.65, corresponding to the intact triply charged O-GlcNAc peptide 158AIPVS#REEKPSSAPSS173, under low collision activation, additional signals observed at m/z 821.43 and 204.09 could be assigned to the doubly charged nonglycosylated

analogue and the reporter ion, respectively. The subsequent directed MS/MS sequencing using an inclusion list allowed establishing the amino acid sequences of the peptides (data not shown). Furthermore, the technique is not restricted to the detection of one single type of modification during the LC-MS run as in addition to the O-GlcNAc peptides of alpha-crystallin; phosphorylated peptides were also simultaneously detected. Figure 2B presents two consecutive spectra acquired at no/ low collision energy at a different retention time in the same LC-MS run. The phosphorylated peptide 13T*LGPFYPSR21 from alpha-crystallin A chain was readily detected thanks to the neutral loss of phosphoric acid. These results illustrate the pertinence of the method for detecting various types of modifications concomitantly in the case of simple mixtures, but for complex mixtures, a prior modification-specific enrichment step might still be desirable. The data also show that the specificity is significantly enhanced by the generation of characteristic reporter ions by O-linked glycans. Application to N-Linked GlcNAc Containing Peptides. Nlinked glycosylation is one of the most commonly encountered modification associated with the NX[S/T] consensus sequence.4,25 To map N-glycosylation sites, the entire N-glycans are generally removed by peptide-N-glycosidase F (PNGase F) converting the glycosylated asparagine residue into aspartic acid. The deamiJournal of Proteome Research • Vol. 8, No. 5, 2009 2611

technical notes

Carapito et al.

Figure 3. LC-MS base peak chromatogram and extracted ion chromatogram of m/z 204.09 for a yeast whole cell lysate digest. After trypsin digestion, glycopeptides were enriched using Concanavalin A lectin affinity, deglycosylated with Endo H before being analyzed by LC-MS.

dated asparagine residues provide indirect evidence of Nglycosylation of the intact protein. Conclusive assignment of deamidation is not always obvious because of the only 1 Da mass shift between Asn and Asp, and of other artifactual deamidation reactions that might occur. Thus, we performed only partial deglycosylation of N-linked glycopeptides retaining one single GlcNAc unit on the Asn residue as direct unambiguous tag of glycosylation site.11 For this purpose, the enzyme endo-β-N-acetylglucosaminidase H (Endo H) was used to partially deglycosylate proteins from yeast. For the LC-MS analysis, collision energy of 15 V was determined as optimal to induce the selective loss of the GlcNAc moiety. The strategy described in Figure 1 was first applied to invertase from yeast after trypsin digestion and Endo H treatment. Glycosylation sites at Asn165, Asn356 and Asn369 in the tryptic sequences 159 351 NPVLAAN#STQFR170, AEPILN#ISNAGPVVSR365 and 366 374 FATN#TTLTK were unambiguously detected thanks to the coincidence of signals corresponding to the nonmodified peptide backbone and the reporter ion at m/z 204. The sequences of the peptides were further determined by directed MS/MS sequencing (data not shown). For the peptide 366 FATN#TTLTK374, the mass spectrum acquired at 15 V collision energy is shown in Figure 4C. Noteworthy is that this approach allowed identification of an additional glycosylation site on Asn23 in the quadruply charged semitryptic peptide 20 SMTN#ETSDRPLVHFTPNK37. Such peptides can be easily missed in classical shotgun experiments as only fully tryptic peptides are often considered in database searches in order to reduce the false positive identification rate. This example clearly demonstrates the potential of this method to identify additional glycosylation sites thanks to its high specificity. 2612

Journal of Proteome Research • Vol. 8, No. 5, 2009

Finally, the approach was applied to a complex mixture of N-linked glycopeptides extracted from a yeast whole cell lysate. After trypsin digestion of the whole cell lysate, glycopeptides were isolated by Concanavalin A lectin affinity before being deglycosylated by Endo H. The resulting fraction was analyzed by LC-MS: the base peak chromatogram and the extracted ion chromatogram of the reporter ion at m/z 204.09 are represented in Figure 3. In-house developed software tools implemented in SuperHirn19 were used to analyze the data off-line and to generate the list of potentially modified peptides based on the concomitant presence of a signal corresponding to the loss of the GlcNAc residue from the peptide, and to the complementary reporter ion. In a subsequent directed LC-MS/MS run, a total of 105 unique N-linked GlcNAc containing peptides was readily sequenced and identified with a false discovery rate below 2% using the Peptide Prophet validation tool (Supporting Information Table 1).26 Endo H is only applicable to high mannose type glycans; it is not suited for complex or hybrid type glycans. The enzyme endo-β-N-acetylglucosaminidase M (Endo-M) with much broader specificity was successfully applied to partially deglycosylate complex type glycans present in human serum.27 This application demonstrates that the proposed LC-MS method can rapidly and unambiguously determine N-glycosylation sites in very complex samples thanks to the specificity gained by tracking two concomitant signals for each modified peptide. Application to Phosphotyrosine Containing Peptides. Although phosphorylation of tyrosine is less frequent in eukaryotic cells compared to serine or threonine phosphorylation, it plays a crucial role in receptor-mediated signal transduction affecting numerous biological processes.7 The low stoichiom-

Systematic LC-MS Analysis of Labile PTMs in Complex Mixtures

technical notes

Figure 4. Full mass spectra with low collision energy for the detection of various types of modifications. Collision activated spectra acquired for (A) the O-GlcNAc peptide YSPTS#PSK at CE ) 8 V (m/z 204.09, GlcNAc oxonium ion; m/z 535.261, doubly charged O-GlcNAc peptide; m/z 866.437, doubly charged nonglycosylated peptide YSPTSPSK); (B) the galactosylated O-GlcNAc peptide YSPTS∧PSK at CE ) 8V (m/z 204.09 and m/z 366.15, GlcNAc and LacNAc oxonium ions; m/z 616.297 doubly charged O-LacNAc peptide YSPTS∧PSK; m/z 866.437, doubly charged nonglycosylated peptide YSPTSPSK); (C) the N-GlcNAc peptide from invertase 366FATN#TTLTK374 from yeast at CE ) 15 V (m/z 204.09, GlcNAc oxonium ion; m/z 600.315, doubly charged N-GlcNAc peptide FATN#TTLTK; m/z 996.540, singly charged nonglycosylated peptide FATNTTLTK); (D) the phosphorylated peptide DRVY*IHPF at CE )15 V (m/z 216.046, phosphotyrosine immonium ion; m/z 563.768, doubly charged phosphorylated peptide DRVY*IHPF).

etry of phosphotyrosine peptides makes their identification very challenging and more sensitive and selective methods are still required even though the use of phosphotyrosine specific antibodies has proven its efficiency.7,16,28 Incidentally, the phosphotyrosine residue generates a characteristic reporter ion at m/z 216.043 corresponding to the immonium fragment16 which makes this modification especially suitable for our workflow. Figure 4D shows the mass spectrum of the phosphopeptide DRVY*IHPF acquired at 15 V collision energy. Signals at m/z 563.768 and at m/z 216.046 represent the doubly charged species of the phosphorylated peptide and the phosphotyrosine immonium ion, respectively. This exemplifies the versatility of the method, whereas the high mass accuracy of the ToF ensures the correct assignment of the reporter ion as numerous peptide fragments having the same nominal mass of 216 were described.16

Conclusion The workflow presented is intended to facilitate the detection and identification of several types of PTMs and their associated peptides in complex mixtures using LC-MS analyses. The power of the acquisition of several (alternating) full mass spectra on a Q-ToF instrument under various collision conditions to detect

different types of modifications in very complex samples was demonstrated. The cleavage of a substituent strongly depends on the type of modification and to a certain extent on the peptide sequence. Therefore, the collision activation conditions were adjusted to induce the selective loss of labile substituents, such as glycan or phosphate residues in this work. The instrument control software is not restricted to the acquisition of two spectra (without and with fragmentation), but additional spectra can be recorded during each cycle at higher activation energies to detect various types of modifications in one single analysis. One major advantage of this technique, leveraging the two analyzer stages of the Q-ToF instrument, is the removal of the low mass background, often compromising proper detection of the reporter ions. In contrast to other triple quadrupole scanning techniques, the simultaneous extraction of two or even more specific signals associated with a given modification significantly increases the selectivity of the analysis. Most importantly, as no precursor isolation is performed, this approach is generically applicable to broad screens, with no a priori assumption on the types of modifications investigated. Further applications of the alternating full mass spectra acquisition will include determination of the glycosylation profiles of individual glycopeptides. Journal of Proteome Research • Vol. 8, No. 5, 2009 2613

technical notes Abbreviations: PTM, post-translational modification; Q-ToF, quadrupole time-of-flight instrument; CE, collision energy; CAD, collision activated dissociation; GlcNAc, N-acetylglucosamine; LacNAc, N-acetyllactosamine.

Acknowledgment. Drs. Rudi Grimm and Christine Miller (Agilent Technologies, Santa Clara, CA) are gratefully acknowledged for fruitful discussion. Dr. Ludovic Gillet and Dr. Lukas Mueller are acknowledged for technical assistance and Agilent Technologies is thanked for technical support. Supporting Information Available: List of the GlcNAc containing peptides identified from the yeast whole cell lysate, and demonstration of the low mass chemical background elimination. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Creasy, D. M.; Cottrell, J. S. Unimod: Protein modifications for mass spectrometry. Proteomics 2004, 4 (6), 1534–6. (2) Zhang, H.; Loriaux, P.; Eng, J.; Campbell, D.; Keller, A.; Moss, P.; Bonneau, R.; Zhang, N.; Zhou, Y.; Wollscheid, B.; Cooke, K.; Yi, E. C.; Lee, H.; Peskind, E. R.; Zhang, J.; Smith, R. D.; Aebersold, R. UniPepsa database for human N-linked glycosites: a resource for biomarker discovery. Genome Biol. 2006, 7 (8), R73. (3) Bodenmiller, B.; Malmstrom, J.; Gerrits, B.; Campbell, D.; Lam, H.; Schmidt, A.; Rinner, O.; Mueller, L. N.; Shannon, P. T.; Pedrioli, P. G.; Panse, C.; Lee, H. K.; Schlapbach, R.; Aebersold, R. PhosphoPepsa phosphoproteome resource for systems biology research in Drosophila Kc167 cells. Mol. Syst. Biol. 2007, 3, 139. (4) Helenius, A.; Aebi, M. Intracellular functions of N-linked glycans. Science 2001, 291 (5512), 2364–9. (5) Khidekel, N.; Ficarro, S. B.; Clark, P. M.; Bryan, M. C.; Swaney, D. L.; Rexach, J. E.; Sun, Y. E.; Coon, J. J.; Peters, E. C.; HsiehWilson, L. C. Probing the dynamics of O-GlcNAc glycosylation in the brain using quantitative proteomics. Nat. Chem. Biol. 2007, 3 (6), 339–48. (6) Wells, L.; Vosseller, K.; Hart, G. W. Glycosylation of nucleocytoplasmic proteins: signal transduction and O-GlcNAc. Science 2001, 291 (5512), 2376–8. (7) Hunter, T. The Croonian Lecture 1997. The phosphorylation of proteins on tyrosine: its role in cell growth and disease. Philos. Trans. R. Soc., B 1998, 353 (1368), 583–605. (8) Zhang, H.; Aebersold, R. Isolation of glycoproteins and identification of their N-linked glycosylation sites. Methods Mol. Biol. 2006, 328, 177–85. (9) Medzihradszky, K. F.; Guan, S.; Maltby, D. A.; Burlingame, A. L. Sulfopeptide fragmentation in electron-capture and electrontransfer dissociation. J. Am. Soc. Mass Spectrom. 2007, 18 (9), 1617– 24. (10) Bodenmiller, B.; Mueller, L. N.; Mueller, M.; Domon, B.; Aebersold, R. Reproducible isolation of distinct, overlapping segments of the phosphoproteome. Nat. Methods 2007, 4 (3), 231–7. (11) Hagglund, P.; Matthiesen, R.; Elortza, F.; Hojrup, P.; Roepstorff, P.; Jensen, O. N.; Bunkenborg, J. An enzymatic deglycosylation scheme enabling identification of core fucosylated N-glycans and O-glycosylation site mapping of human plasma proteins. J. Proteome Res. 2007, 6 (8), 3021–31. (12) Bateman, R. H.; Carruthers, R.; Hoyes, J. B.; Jones, C.; Langridge, J. I.; Millar, A.; Vissers, J. P. A novel precursor ion discovery method on a hybrid quadrupole orthogonal acceleration time-of-flight (Q-

2614

Journal of Proteome Research • Vol. 8, No. 5, 2009

Carapito et al.

(13)

(14) (15)

(16)

(17)

(18) (19)

(20)

(21)

(22)

(23) (24)

(25) (26)

(27)

(28)

TOF) mass spectrometer for studying protein phosphorylation. J. Am. Soc. Mass Spectrom. 2002, 13 (7), 792–803. Schindler, P. A.; Settineri, C. A.; Collet, X.; Fielding, C. J.; Burlingame, A. L. Site-specific detection and structural characterization of the glycosylation of human plasma proteins Lecithin: cholesterol acyltransferase and apolipoprotein D using HPLC/ electrospray mass spectrometry and sequential glycosidase digestion. Protein Sci. 1995, 4 (4), 791–803. Chalkley, R. J.; Burlingame, A. L. Identification of GlcNAcylation sites of peptides and alpha-Crystallin using Q-TOF mass spectrometry. J. Am. Soc. Mass Spectrom. 2001, 12 (10), 1106–13. Huddleston, M. J.; Bean, M. F.; Carr, S. A. Collisional fragmentation of glycopeptides by electrospray ionization LC/MS and LC/MS/ MS: methods for selective detection of glycopeptides in protein digests. Anal. Chem. 1993, 65 (7), 877–84. Steen, H.; Kuster, B.; Fernandez, M.; Pandey, A.; Mann, M. Detection of tyrosine phosphorylated peptides by precursor ion scanning quadrupole TOF mass spectrometry in positive ion mode. Anal. Chem. 2001, 73 (7), 1440–8. Hoffman, M. D.; Sniatynski, M. J.; Rogalski, J. C.; Le Blanc, J. C.; Kast, J. Multiple neutral loss monitoring (MNM): a multiplexed method for post-translational modification screening. J. Am. Soc. Mass Spectrom. 2006, 17 (3), 307–17. Domon, B.; Aebersold, R. Mass spectrometry and protein analysis. Science 2006, 312 (5771), 212–7. Mueller, L. N.; Rinner, O.; Schmidt, A.; Letarte, S.; Bodenmiller, B.; Brusniak, M. Y.; Vitek, O.; Aebersold, R.; Muller, M. SuperHirnsa novel tool for high resolution LC-MS-based peptide/protein profiling. Proteomics 2007, 7 (19), 3470–80. Schmidt, A.; Gehlenborg, N.; Bodenmiller, B.; Mueller, L. N.; Campbell, D.; Mueller, M.; Aebersold, R.; Domon, B. An integrated, directed mass spectrometric approach for in-depth characterization of complex peptide mixtures. Mol. Cell. Proteomics 2008, 7 (11), 2138–50. Vosseller, K.; Trinidad, J. C.; Chalkley, R. J.; Specht, C. G.; Thalhammer, A.; Lynn, A. J.; Snedecor, J. O.; Guan, S.; Medzihradszky, K. F.; Maltby, D. A.; Schoepfer, R.; Burlingame, A. L. O-linked N-acetylglucosamine proteomics of postsynaptic density preparations using lectin weak affinity chromatography and mass spectrometry. Mol. Cell. Proteomics 2006, 5 (5), 923–34. Haynes, P. A.; Aebersold, R. Simultaneous detection and identification of O-GlcNAc-modified glycoproteins using liquid chromatography-tandem mass spectrometry. Anal. Chem. 2000, 72 (21), 5402–10. Roquemore, E. P.; Chevrier, M. R.; Cotter, R. J.; Hart, G. W. Dynamic O-GlcNAcylation of the small heat shock protein alpha B-Crystallin. Biochemistry 1996, 35 (11), 3578–86. Roquemore, E. P.; Dell, A.; Morris, H. R.; Panico, M.; Reason, A. J.; Savoy, L. A.; Wistow, G. J.; Zigler, J. S., Jr.; Earles, B. J.; Hart, G. W. Vertebrate lens alpha-crystallins are modified by O-linked Nacetylglucosamine. J. Biol. Chem. 1992, 267 (1), 555–63. Apweiler, R.; Hermjakob, H.; Sharon, N. On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochim. Biophys. Acta 1999, 1473 (1), 4–8. Keller, A.; Nesvizhskii, A. I.; Kolker, E.; Aebersold, R. Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Anal. Chem. 2002, 74 (20), 5383–92. Carapito, C.; Klemm, C.; Mueller, L. N.; Aebersold, R.; Domon, B. Integrated methodology for proteomics analysis of glycoproteins. In Proceedings of the 56th ASMS Conference on Mass Spectrometry, Denver, CO, June 1-5, 2008, TOG-0310. Blagoev, B.; Ong, S. E.; Kratchmarova, I.; Mann, M. Temporal analysis of phosphotyrosine-dependent signaling networks by quantitative proteomics. Nat. Biotechnol. 2004, 22 (9), 1139–45.

PR800871N