Selective Extraction and Characterization of a Histidine

ways, similar protocols for peptides containing phosphor- .... (Milwaukee, WI) unless otherwise noted. Acetoni- ... Phosphopeptide Extraction Protocol...
0 downloads 0 Views 164KB Size
Anal. Chem. 2003, 75, 1741-1747

Selective Extraction and Characterization of a Histidine-Phosphorylated Peptide Using Immobilized Copper(II) Ion Affinity Chromatography and Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry Scott Napper,*,†,‡ Jason Kindrachuk,† Douglas J. H. Olson,§ Stephen J. Ambrose,§ Carmen Dereniwsky,§ and Andrew R. S. Ross†,§

Department of Biochemistry, Health Sciences Building, University of Saskatchewan, 107 Wiggins Road, Saskatoon, SK S7N 5E5 Canada, Veterinary Infectious Disease Organization, University of Saskatchewan, 120 Veterinary Road, Saskatoon, SK S7N 5E3 Canada, and Plant Biotechnology Institute, National Research Council of Canada, 110 Gymnasium Place, Saskatoon, SK, S7N 0W9, Canada

Phosphorylation is the predominant posttranslational modification involved in regulating enzymatic activity and mediating signal transduction in prokaryotic and eukaryotic cells. Selective enrichment of phosphorylated peptides prior to mass spectrometric analysis facilitates identification of phosphorylated proteins, determination of specific phosphorylated residues, and characterization of the conditions under which phosphorylation occurs. Such protocols have been established for peptides containing residues that form phosphoesters, such as serine and threonine, using immobilized metal-ion affinity chromatography. Despite the importance of histidine phosphorylation in two-component signal transduction pathways, similar protocols for peptides containing phosphorylated histidine (P-His) residues have proven elusive, due to the instability of these modifications and the propensity of unphosphorylated histidines to interact with immobilized metals ions. We describe a method for the selective extraction of a P-His-containing peptide using immobilized copper(II) ions and disposable metal-chelating pipet tips (ZipTipMC, Millipore). The method is contingent upon pHdependent interactions between the phosphate group and immobilized copper(II) ions. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry with postsource decay confirms the identity and phosphorylation state of the extracted peptide. Peptides containing unphosphorylated histidine residues or other phosphorylated amino acids are not retained, demonstrating the specificity of the method for P-His-containing peptides. Proteomics, which involves identifying and quantifying the spatial and temporal expression of specific proteins, is effective * Corresponding author. Phone: (306) 966-4377. Fax: (306) 966-4390. E-mail: [email protected]. † Department of Biochemistry, University of Saskatchewan. ‡ Veterinary Infectious Disease Organization, University of Saskatchewan. § National Research Council of Canada. 10.1021/ac026340f CCC: $25.00 Published on Web 03/05/2003

© 2003 American Chemical Society

in understanding cell and organelle function and in identifying cellular responses to changing stimuli and conditions. However, current methodology is limited in its ability to elucidate cellular mechanisms and events that are independent of changes in protein expression. For example, classical proteomics allows for visualization of the dynamic patterns of protein expression initiated in response to a stimulus but not for the information-transfer events that initiate the transcriptional process. To understand information transfer through proteomic analysis, one must consider the interactions and communications mediated through a stable network of cellular proteins. Information is transmitted through these networks by posttranslation modifications, which result in the initiation of cellular responses. Phosphorylation is one of the most common and important posttranslational modifications for proteins.1 Based on physiological function, there are two primary classes of protein phosphorylation. The first encompasses phosphorylation for the purpose of regulating enzymatic activity and typically involves the modification of serine, threonine, and tyrosine residues. The resulting phosphoesters are stable entities and generally serve to regulate enzymatic catalysis without any direct involvement in the catalytic mechanism. Dephosphorylation of phosphoesters usually requires a specific phosphatase enzyme, and the reversible phosphorylation of serine, threonine, and tyrosine regulates a wide variety of signal transduction pathways. The second class encompasses phosphorylation for the purpose of phosphate group transfer and is largely restricted to phosphorylation of histidine residues. Phosphohistidines generally act as high-energy intermediates for the purpose of transferring the phosphate group from a phosphodonor to a phosphoacceptor molecule.2 Examples of phosphohistidines include the bacterial histidine kinases of the two-component systems3 and the phosphoenolpyruvate-sugar phosphotransferase system (PTS).4,5 Phosphate-histidine linkages are acid-labile and (1) Cohen, P. Trends Biochem. Sci. 2000, 25, 596-601. (2) Stock, J.; Ninfa, A.; Stock, A. M. Microbiol. Rev. 1989, 53, 450-490. (3) Parkinson, J. S.; Kofoid, E. C. Annu. Rev. Genet. 1992, 26, 71-112.

Analytical Chemistry, Vol. 75, No. 7, April 1, 2003 1741

difficult to preserve experimentally.6 Consequently, the importance of these modifications is often underestimated. Nevertheless, phosphorylation of histidine may account for 6% of phosphorylation events in eukaryotes7 as well as prokaryotes.8 Estimates suggest that the concentrations of phosphohistidine-containing proteins in PTS-utilizing bacteria might exceed 0.2 mM under certain conditions.9 One of the roles in which the importance of histidine phosphorylations is rapidly emerging is that of histidine kinase enzymes in two-component signal transduction systems. Such systems are utilized by bacteria,3 plants,10 and lower eukaryotes11 to sense and adapt to changing environmental conditions. These signal transduction pathways mediate a diverse range of responses, from rapid changes in motility to long-term reorganizations of gene expression and cell morphology. They are also involved in establishment of virulence12 and antibiotic resistance.13 In response to changing conditions, the histidine kinase autophosphorylates at a conserved histidine residue, using ATP as the phosphodonor molecule. The phosphorylated histidine kinase then communicates with the associated response regulator through phosphorylation of an aspartyl residue, which acts as a switch to control output activity.14 At present, the identification of novel histidine kinases is performed largely on the basis of sequence similarities to known members of this class of enzymes. The establishment of a protocol for the selective extraction of peptides containing phosphohistidine (P-His) residues would provide an alternate approach for the identification of histidine kinase enzymes, as well as other proteins that contain these residues. Protocols have been established for the selective extraction of peptides containing phosphoester residues using immobilized iron(III) or gallium(III) ion affinity chromatography.15,16 Quantitative analysis of phosphopeptides, based upon the derivatization of phosphoesters to thiol groups and subsequent implementation of isotope-coded affinity tagging (ICAT) procedures, has also been the subject of recent investigations.17 However, the development of similar protocols for P-His residues has been hindered by the much greater instability of these attachments. Here, we report on the development of a new procedure for selective enrichment and identification of P-Hiscontaining peptides based upon immobilized copper(II) ion affinity chromatography (Cu(II)-IMAC), matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry (MALDI-TOF MS), and use of appropriate pH conditions. (4) Meadow, N. D.; Fox, D. K.; Roseman, S. Annu. Rev. Biochem. 1990, 59, 497-542. (5) Meadow, N. D.; Roseman, S. J. Biol. Chem. 1982, 257, 14526-14537. (6) Hultquist, D. Biochim. Biophys. Acta 1968, 153, 329-340. (7) Matthews, H. R. Pharmacol. Ther. 1995, 67, 323-350. (8) Waygood, E. B.; Mattoo, R. L.; Peri, K. G. J. Cell. Biochem. 1984, 25, 139159. (9) Mattoo, R. L.; Khandelwal, R. L.; Waygood, E. B. Anal. Biochem. 1984, 139, 1-16. (10) Urao, T.; Yamaguchi-Shinozaki, K.; Shinozaki, K. Trends Plant Sci. 2000, 5, 67-74. (11) Loomis, W. F.; Shaulsky, G.; Wang, N. J. Cell Sci. 1997, 110, 1141-1145. (12) Uhl, M. A.; ller, J. F. EMBO J. 1996, 15, 1028-1036. (13) Hakenbeck, R.; Grebe, T.; Zahner, D.; Stock, J. Mol. Microbiol. 1999, 33, 673-678. (14) Robinson, V. L.; Buckler, D. R.; Stock, A. M. Nat. Struct. Biol. 2000, 7, 626-633. (15) Andersson, L.; Porath, J. Anal. Biochem. 1986, 154, 250-254. (16) Posewitz, M. C.; Tempst, P. Anal. Chem. 1999, 71, 2883-2892.

1742

Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

The well-characterized histidine-containing protein (HPr) from the Escherichia coli PTS was chosen as a model P-His-containing protein for these studies. HPr occupies a central role in the bacterial PTS, which mediates phosphorylation-dependent sugar uptake4 as well as numerous regulatory roles in bacterial metabolism.18,19 HPr undergoes phosphorylation at a conserved histidine residue (His-15) and functions as a phosphotransfer protein between enzyme I, the initiating enzyme of the PTS, and a sugarspecific enzyme IIA protein.20,21 The kinetics of HPr phosphorylation, and the phosphohydrolysis properties of the protein, are well established.22 MATERIALS AND METHODS All reagents were analytical grade and obtained from Sigma Chemical Co. (Milwaukee, WI) unless otherwise noted. Acetonitrile (HPLC grade) was supplied by EM Science (Gibbstown, NJ), and deionized water was prepared on site using the Milli-Q system from Millipore (Bedford, MA). Enzyme I, enzyme IIAglc, and HPr proteins from E. coli were purified to homogeneity using previously published protocols,23-25 as was the CheY protein from Salmonella typhimurium.26 Phosphorylation Reactions. Phosphorylation of HPr was achieved by combining 20 ng of enzyme I with 10 µg (1.1 nmol) of HPr in 20-µL volumes of a reaction mixture containing 2 mM MgCl2, 5 mM phosphoenolpyruvate, and 20 mM HEPES buffer (pH 7.0). Complete phosphorylation of HPr was achieved after incubation at 37 °C for 15 min. For nonphosphorylating controls, the phosphodonor (phosphoenolpyruvate) was omitted from the mixture. Proteolytic Digests. HPr digestion reactions were performed by adding 2.5 µL of 0.4 mg/mL Staphlococcus aureus V8 protease in 0.1 M bicine (pH 8.6) and 1 mM EDTA to each 20-µL volume of phosphorylation reaction mixture. Digests were incubated at 37 °C for 30 min, and the extent of proteolysis was estimated at 90%, based upon SDS-PAGE of both digested and undigested samples. To obtain P-Ser phosphopeptides for method development and validation, 1 mg/mL standard phosphoprotein bovine β-casein (Sigma) was dissolved in 25 mM NH4HCO3 and digested overnight at 37 °C in the presence of 10 µg/mL porcine trypsin (Promega sequencing grade). Phosphopeptide Extraction Protocol. Disposable metalchelating pipet tips (ZipTipMC) were obtained from Millipore Corp. The tips were washed 10 times with 10 µL of 0.1% acetic acid in deionized water and then charged with metal ions by aspirating (17) Conrads, T. P.; Haleem, J. I.; Veenstra, T. D. Biochem. Biophys. Res. Commun. 2002, 290, 885-890. (18) Saier, M. H., Jr. J. Cell. Biochem. 1993, 51, 69-74. (19) Postma, P. W.; Lengeler, J. W.; Jacobson, G. R. Microbiol. Rev. 1993, 57, 543-594. (20) Anderson, B.; Weigel, N.; Kundig, W.; Roseman, S. J. Biol. Chem. 1971, 246, 7023-7033. (21) Waygood, E. B.; Erikson, E. E.; El-Kabbani, O. A. L.; Delbaere, L. T. J. Biochemistry 1985, 24, 6938-6945. (22) Anderson, J. W. Biochem. Cell. Biol. 1995, 73, 219-222. (23) Anderson, J. W.; Bhanot, P.; Georges, F.; Klevit, R. E.; Waygood, E. B. Biochemistry 1991, 30, 9601-9607. (24) Brokx, S. J.; Talbot, J.; Georges, F.; Waygood, E. B. Biochemistry 2000, 39, 3624-3625. (25) Napper, S.; Brokx, S. J.; Pally, E.; Kindrachuk, J.; Delbaere, L. T. J.; Waygood, E. B. J. Biol. Chem. 2001, 276, 41588-41593. (26) Stock, A.; Koshland, D. E., Jr.; Stock, J. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 7989-7993.

Table 1. MALDI-TOF MS Instrumental Parameters for Different Modes of Operation operating mode

accelerating voltage (V)

grid voltage (%)

guide wire voltage (%)

extraction delay (ns)

laser intensity

mirror voltage ratio

linear reflector PSD

20 000 20 000 20 000

95.0 73.5 80.0

0.050 0.001 0.010-0.020

200 150 280

2570 2440 2590

1.12 1.12

Figure 1. (A) Positive ion reflectron MALDI-TOF mass spectrum of the Cu(II)-IMAC extract from a V8 digest of the histidine-phosphorylated protein P-HPr. (B) Linear MALDI-TOF MS mass spectrum of the Cu(II)-IMAC extract from a 25-fold diluted P-HPr digest. (C) Linear MALDI-TOF MS mass spectrum of the Cu(II)-IMAC extract from a V8 digest of the nonphosphorylated HPr protein.

and dispensing 10 µL of 100 mM metal salt solution 15 times. The charged tips were then rinsed five times with 10 µL of deionized water and five times with 10 µL of 0.1% acetic acid in

50% acetonitrile. Phosphopeptides were loaded onto the tips by aspirating and dispensing 10 µL of sample 10 times and then washing 10 times with 10 µL of 0.1% acetic acid in 50% acetonitrile. Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

1743

Figure 2. Positive ion reflectron MALDI-TOF mass spectrum of the Cu(II)-IMAC extract from a V8 digest of a mixture of proteins.

Phosphopeptides were eluted in 2 µL of 1% NH4OH and immediately neutralized with 1 µL of 2% trifluoroacetic acid (TFA). Samples were then mixed with an equal volume of R-cyano-4hydroxycinnamic acid (CHCA) matrix solution (5 mg/mL in 75% acetonitrile with 0.1% TFA) and applied directly to the MALDI target plate. The unextracted P-HPr digest was desalted using conventional C18 ZipTips according to the manufacturer’s instructions and then combined with an equal volume of the same matrix solution prior to MS analysis. Mass Spectrometry. Phosphopeptides were analyzed by MALDI-TOF MS using a Voyager DE-STR instrument (Perseptive Biosystems, Framingham, MA) operating in the linear, reflectron or postsource decay (PSD) modes with positive or negative ionization. Typical operating conditions are shown in Table 1. Fullscan spectra were obtained by combining and processing 100 scans. For PSD analysis, the reflectron mirror ratio was adjusted incrementally, the resulting spectra were combined, and fragment ion peaks were assigned using the Data Explorer instrument software. RESULTS AND DISCUSSION P-HPr Digestion Efficiency. Protein digestion times were limited to 30 min in an attempt to preserve the phosphorylated state of the histidine residues. To quantify the actual concentration of free peptides present in the digestion mixture at the time of IMAC extraction, aliquots of the digestion mixture were separated by SDS-PAGE and the intensities of digested (peptide) and undigested (protein) bands quantified using a phosphoimager. The intensity ratio of undigested to digested bands was 10:1, corresponding to 1 nmol (44 pmol/µL) of potentially phosphorylated peptide in the digestion solution. The actual concentration of phosphorylated peptide is considerably lower due to spontaneous loss of the phosphoryl group during the incubation period. Under the conditions of the digest, HPr has a rate constant of phosphohydrolysis of 0.13 min-1 and exhibits first-order decay.30 Accordingly, only 2% of the initially phosphorylated histidines would (27) Janek, K.; Wenschuh, H.; Bienert, M.; Krause, E. Rapid Commun. Mass Spectrom. 2001, 15, 1593-1599. (28) Chalmers, M.; Ross, A. R. S.; Olson, D.; Gaskell, S. J. Proceedings of the 48th ASMS Conference, Long Beach, CA, June 12-15, 2000. (29) Hultquist, D.; Moyer, R. W.; Boyer, P. D. Biochemistry 1966, 5, 322-331

1744 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

Figure 3. Negative ion linear MALDI-TOF mass spectrum of the Cu(II)-IMAC extract from a V8 digest of the histidine-phosphorylated protein P-HPr.

remain modified after the 30-min digestion period. This corresponds to an actual concentration of phosphorylated peptide of ∼880 fmol/µL. The following discussion applies to results obtained using undiluted protein digests. However, analysis of serial dilutions indicated that recovery and detection of P-His peptides can be achieved at concentrations of 35 fmol/µL or less. Evaluation of Different Metal Ions. Several metal ions were evaluated for their ability to selectively extract and recover P-Hiscontaining peptides by IMAC, including gallium(III), copper(II), and iron(III). Solutions containing 100 mM concentrations of each metal ion were prepared by dissolving the appropriate amount of soluble salt in deionized water. Extractions were performed using each metal ion according to the protocol described above. Of the metals investigated, only copper proved effective in retaining peptides that contain P-His residues. Selectivity for Phosphorylated HPr Peptides. Positive ion reflectron MALDI-TOF MS analysis of the unextracted P-HPr digestion mixture (not shown) confirmed the presence of all peptides hypothesized by a theoretical V8 digestion. The P-His peptide VTITAPNGL(pH)TRPAAQFVKE (6-25) was detected in both phosphorylated (m/z 2230) and nonphosphorylated (m/z 2150) forms with a relative abundance of ∼1:25, indicating that phosphorylation of the histidine residue is unstable under experi(30) Anderson, J. W.; Pullen, K.; Georges, F.; Klevit, R. E.; Waygood, E. B. J. Biol. Chem. 1993, 268, 12325-12333.

Figure 4. PSD mass spectrum of the protonated P-His molecular ion (m/z 2230) detected in the V8 digest of P-HPr.

mental conditions. Likewise, analysis of the Cu(II)-IMAC extract of the P-HPr digest solution contained a single sharp peak at m/z 2150 (Figure 1A), corresponding to the predicted mass of the 6-25 peptide without the histidine phosphate group (HPO3). This ion was also observed in the linear mode at up to 25-fold dilution, corresponding to 350 fmol of the peptide in 10 µL of extracted sample (Figure 1B). However, when the same protocol was applied to a V8 digest of the nonphosphorylated HPr protein, no peptide ions were detected, even when operating in the linear mode (Figure 1C). This suggests that Cu(II)-IMAC extraction is selective for the phosphorylated form of the histidine-containing peptide but that the phosphate group is subsequently lost before or during MALDI-TOF MS analysis. To verify that peptides containing nonphosphorylated histidine residues are not recovered using this procedure, a mixture containing four purified proteins (enzyme I, enzyme IIAglc, and HPr from E. coli and CheY from S. typhimurium), each at a concentration of 1 mg/mL, was digested to completion with V8 protease and subjected to the Cu(II)-IMAC extraction protocol. This mixture contains 11 unique peptides bearing histidine residues; however, none of these was recovered using the Cu(II)-IMAC procedure (Figure 2). Recovery of Intact P-His Peptides. Previous studies have shown that switching from the positive to the negative ionization mode increases MS detection sensitivity for phosphorylated peptides relative to the nonphosphorylated form.27 Negative ion MALDI-TOF MS analysis of the Cu(II)-IMAC extract detected an ion of m/z 2228, corresponding to the intact P-His peptide from P-HPr, although the dephosphorylated form of this peptide (m/z 2148) was also observed (Figure 3). This further suggests that the intact P-His peptide is recovered using the Cu(II)-IMAC procedure but is susceptible to dephosphorylation during analysis. The same extract was subsequently analyzed in positive ion mode with PSD, after setting the ion gate to m/z 2230. The resulting spectrum (Figure 4) showed sequential loss of 80 and 18 Da from a m/z 2230 precursor, corresponding to loss of the histidine phosphate group and water from the intact phosphopeptide. Similar results are reported for collision-induced dissociation

(CID) of P-His-containing peptide ions generated by MALDI,27 confirming that the intact phosphopeptide had been recovered. Specificity for Histidine-Phosphorylated Peptides. To determine whether the protocol was specific for phosphopeptides containing phosphorylated histidine residues, we applied the Cu(II)-IMAC procedure to a tryptic digest of bovine β-casein. This protein yields two tryptic peptides that are phosphorylated at one and four serine residues, respectively, and that produce ions of m/z 2062 and 3112 when analyzed by positive ion MALDI-TOF MS.16,28 The Cu(II)-charged IMAC ZipTips did not appear to retain either of these P-Ser-containing peptides (Figure 5A). However, extraction of the β-casein digest using a Ga(III)-IMAC protocol optimized for the recovery of peptides containing phosphoester residues16 selectively retained the singly phosphorylated P-Ser peptide of m/z 2062, as confirmed by positive ion MALDI-TOF and PSD analysis (Figure 5B and C). We have previously reported that the monophosphorylated (m/z 2062) peptide is detectable to 129 fmol using the Ga(III)-IMAC protocol.28 That the Ga(III)-IMAC protocol did not recover any of the P-His peptides from the P-HPr digest (see Evaluation of Different Metal Ions, above) suggests that it may be possible to use Ga(III)- and Cu(II)-IMAC sequentially to extract peptides modified with phosphoester and phosphohistidine groups, respectively, from mixed protein digests. pH Considerations. The extraction of P-His-containing peptides using an IMAC approach is compromised by the inherent affinity of histidine residues for immobilized metal ions. This property can be exploited for other procedures, such as the purification of His-tagged proteins and peptides. However, it can also lead to false positives when one attempts to extract and identify P-His peptides, since facile loss of HPO3 following IMAC renders the phosphorylated peptide indistinguishable from the nonphosphorylated form. The extraction of His-tagged proteins by IMAC is based on the interaction between certain divalent metal ions and the imidazole group of histidine. Under basic conditions, the uncharged imidazole ring forms a complex with the immobilized metal ion. Elution of His-tagged proteins and peptides is then achieved by lowering pH and protonating the imidazole nitrogen (pKa 6.0), generating a positive charge that is Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

1745

Figure 5. (A) Negative ion linear MALDI-TOF mass spectrum of the Cu(II)-IMAC extract from a trypsin digest of β-casein. (B) Positive ion reflectron MALDI-TOF mass spectrum of the Ga(III)-IMAC extract from a trypsin digest of β-casein. (C) PSD mass spectrum of the protonated P-Ser molecular ion (m/z 2062) detected in the same extract.

repelled by the immobilized metal cation. In our protocol, the binding of nonphosphorylated histidines is prevented by loading the sample under acidic conditions, which apparently result from mixing of the weakly buffered sample with residual acetic acid on the IMAC column. This reduces the sample pH from an initial value of 7.0 (20 mM HEPES) to ∼3.5 (0.1% acetic acid), at which point protonation of the imidazole ring negates interaction with the immobilized copper(II) ions. Subsequent washing with 0.1% acetic acid in 50% acetonitrile serves to maintain the imidazole in protonated form. Histidine is unique among amino acids in that it can be phosphorylated at two different positions; namely, at atoms Nδ1 and N2 of the imidazole ring. For free histidine, phosphorylation at the Nδ1 position is much less stable than at the N2 position due to the interaction of the Nδ1 phosphoryl group with the positively charged amino group.6,29 In peptides and proteins, the amino group of histidine is involved in peptide bonding, and the 1746 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

stability of the phosphate linkage may differ considerably from that of free histidine. Moreover, the local structural and electrostatic environment of the protein may have significant influence on the chemical properties of histidine and may vary in different protein contexts.21,30 For native HPr in E. coli, the rates of phosphohydrolysis for P-Nδ1-histidine are 3 times higher than that for a free P-Nδ1-hisitidine,29 indicating that the protein serves to destabilize the linkage, perhaps a necessary feature to mediate efficient phosphotransfer. The pH dependence of phosphohydrolysis for the intact P-His-HPr is also quite distinctive, exhibiting a bell-shaped curve with greatest instability of the phosphoamidate bond between pH 5 and 8. The phosphorylated peptide generated from V8 digest of P-HPr has phosphohydrolysis rates 1 order of magnitude less than for the intact protein, showing greatest stability at neutral to basic pH.29 This allows for retention of the phosphate moiety during base elution of the chelating pipet tips used in our protocol.

CONCLUSION Selective extraction of histidine-phosphorylated peptides from HPr protein digests can be achieved using immobilized copper(II) ion affinity chromatography. On-column acidification of the sample by residual acetic acid apparently inhibits binding of nonphosphorylated histidine residues during Cu(II)-IMAC, while recovery of the intact phosphopeptide via base elution is confirmed using negative ion MALDI-TOF MS. Subsequent PSD analysis of the protonated molecular ion shows characteristic loss of the HPO3 moiety present in histidine-phosphorylated peptides. The specificity of Cu(II)-IMAC for phosphohistidine residues, and of Ga(III)IMAC for phosphoesters, may assist in establishing the unique role of histidine phosphorylation in two-component and other signal transduction systems.

ACKNOWLEDGMENT We thank Mark Kavonian of Millipore Corp. for providing early access to the metal-chelating ZipTips, and Michael Chalmers for his initial involvement in evaluating this product for IMAC enrichment of phosphopeptides. This research was supported in part by the University of Saskatchewan and National Research Council Summer Student programs. S.N. also thanks the Health Services Utilization Research Council and the National Science and Engineering Research Council for financial support. Received for review November 22, 2002. Accepted January 30, 2003. AC026340F

Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

1747