Anal. Chem. 2003, 75, 1164-1172
Development of a Methodology Based on Metal-Catalyzed Oxidation Reactions and Mass Spectrometry To Determine the Metal Binding Sites in Copper Metalloproteins Jihyeon Lim and Richard W. Vachet*
Department of Chemistry, University of Massachusetts, Amherst, Massachusetts 01003
Efforts have been made to develop a method that uses metal-catalyzed oxidation (MCO) reactions and mass spectrometry (MS) to identify the binding site of copper in metalloproteins. This method uses MCO reactions to oxidize the amino acids in the metal-binding site and MS to identify the amino acids that have been oxidized. Several reaction conditions, including Cu(II)/ascorbate/ O2, Cu(II)/O2/H2O2, and Cu(II)/ascorbate/O2/H2O2, have been tested at varying concentrations to find the optimum conditions for specific oxidation of only the amino acids bound to copper. For small peptides, such as angiotensin I (Agt I) and [Gln11]-amyloid-β-protein fragment 1-16 (Aβ1-16), the optimum conditions for specific modification involve the use of Cu(II)/ascorbate/O2. For a larger protein, azurin, the speed and specificity of the MCO reactions are enhanced by the presence of a relatively high concentration of ascorbate (100 mM) and a small concentration of H2O2 (1 mM). Optimized reaction conditions combined with MS/MS and MSn analysis on a quadrupole ion trap mass spectrometer allow the copper-binding sites to be specifically identified. For Agt I and Aβ1-16, the amino acids bound to copper can be identified without any false positives. For azurin, four of the five amino acids bound to copper are identified with one false positive. This false positive, however, corresponds to the oxidation of Met44, which is probably due to its susceptibility to oxidation and its proximity to the only residue not identified (i.e., Gly45). The results altogether suggest that MCO reactions and MS provide a very promising approach for identifying the amino acid residues bound to copper in metalloproteins. Metalloproteins play an important role in the chemistry of life processes, such as dioxygen transport, electron transfer, catalysis, etc. Metal ions in these proteins are used to perform a wide variety of specific functions related with these processes. How a particular metal ion executes these distinct functions highly depends on finetuning by the protein, which is accomplished by the coordination structure around the metal. The type, orientation, and number of * Corresponding author address: Department of Chemistry, LGRT 701, 710 N. Pleasant St., University of Massachusetts, Amherst, MA 01003. Fax: 413545-4490. E-mail:
[email protected].
1164 Analytical Chemistry, Vol. 75, No. 5, March 1, 2003
amino acids or ligands around the metal very much influence the general reactivity of the metal center. With this in mind, an important step in understanding the chemistry of a metalloprotein is determining the coordination environment around the metal. Several techniques are available to provide this information, such as various X-ray techniques, NMR, EPR, and Raman spectroscopy, but these techniques are usually not very sensitive and often require pure samples. Because of its sensitivity and structural characterization capability, MS has had a well-recognized impact on the analysis of peptides and proteins. The structural information that this technique has provided for metalloproteins so far, however, has not extended much to coordination structure. MS, however, has been used by some investigators to determine metal-ligand stoichiometry,1-4 and others have used it to determine transition metal oxidation states.4-7 As for other structural information, several studies have used tandem MS (MS/MS) to study the metal binding of small peptides.8-15 In general, these studies show that dissociation occurs along the backbone at sites near the residues to which the metal is bound. Although in each of these cases a correlation between gas-phase and solution-phase binding of transition metal-peptide complexes can be inferred from the collision-induced dissociation (CID) spectra, little evidence has (1) Yu, X.; Wojciechowski, M.; Fenselau, C. Anal. Chem. 1993, 65, 1355-1359. (2) Hu, P.; Ye, Q.; Loo, J. A. Anal. Chem. 1994, 66, 4190-4194. (3) Lei, Q. P.; Cui, X. Y.; Kurtz, D. M.; Amster, I. J.; Chernushevich, I. V.; Standing, K. G. Anal. Chem. 1998, 70, 1838-1846. (4) Johnson, K. A.; Verhagen, M. F. J. M.; Brereton P. S.; Adams, M. W. W.; Amster, I. J. Anal. Chem. 2000, 72, 1410-1418. (5) Li, Y. T.; Hsieh, Y. L.; Henion, J. D.; Ganem, B. J. Am. Soc. Mass Spectrom. 1993, 4, 631-637. (6) Johnson, K. A.; Verhagen, M. F. J. M.; Adams, M. W. W.; Amster, I. J. Int. J. Mass Spectrom. 2001, 204, 77-85. (7) Johnson, K. A.; Amster, I. J. J. Am. Soc. Mass Spectrom. 2001, 12, 819825. (8) Hu, P.; Gross, M. L. J. Am. Chem. Soc. 1993, 115, 8821-8828. (9) Reiter, A.; Adams, J.; Zhao, H. J. Am. Chem. Soc. 1994, 116, 7827-7838. (10) Loo, J. A.; Hu, P.; Smith, R. D. J. Am. Soc. Mass Spectrom. 1994, 5, 959965. (11) Hu, P.; Loo, J. A. J. Am. Chem. Soc. 1995, 117, 11314-11319. (12) Hu, P.; Sorensen, C.; Gross, M. L. J. Am. Soc. Mass Spectrom. 1995, 6, 1079-1085. (13) Nemirovskiy, O. V.; Gross, M. L. J. Am. Soc. Mass Spectrom. 1996, 7, 977980. (14) Li, H.; Siu, K. W. M.; Guevremont, R.; Le Blanc, J. C. Y. J. Am. Soc. Mass Spectrom. 1997, 8, 781-792. (15) Nemirovskiy, O. V.; Gross, M. L. J. Am. Soc. Mass Spectrom. 1998, 9, 12851292. 10.1021/ac026206v CCC: $25.00
© 2003 American Chemical Society Published on Web 01/25/2003
Figure 1. Generation of reactive oxygen species during MCO reactions.
been presented to suggest coordination structure can be determined a priori. Furthermore, despite the advances being made to gather structural information from intact proteins using “topdown” strategies,16-19 it seems unlikely that CID will be suitable for gathering coordination structure from large intact metalloproteins. Thus, using MS to gather information about the amino acids bound to transition metal ions in metalloproteins requires a different approach. Metal-catalyzed oxidation (MCO) reactions are known to cause the oxidative inactivation of enzymes. MCO reactions generate reactive oxygen species (O2•- or OH•) through the cycle shown in Figure 1. The reactive oxygen species can interact with nearby amino acid residues and can oxidize or cleave the polypeptide at sites where the reactive oxygen species are generated.20 MCO reactions are caged processes in which amino acid residues at the metal binding sites are specific targets.21 The reactive oxygen species are hindered from diffusing into the surrounding medium because they react quickly with amino acid residues near the metal binding site. A few studies have demonstrated that some amino acids in the metal-binding site of certain metalloenzymes can be identified by treating the enzyme with its native transition metal center, which must be redox-active, and a reducing agent.22-24 In these studies, oxidative modifications seem to occur at the amino acids bound to the metal. Complete elucidation of the amino acid modifications, however, has been difficult because the modified proteins have been analyzed mainly by traditional methods, such as amino acid analysis and gel electrophoresis. Though much has been learned from these studies, specific sites and the degree of oxidation in these proteins have been difficult to clarify. Recently, though, Kurahashi et al. used MS to identify the amino acids oxidized upon treatment of Cu,Zn superoxide dismutase (SOD) with hydrogen peroxide.25 The oxidation reac(16) Kelleher, N. L.; Lin, H. Y.; Valaskovic, G. A.; Aaserud, D. J.; Fridriksson, E. K.; McLafferty, F. W. J. Am. Chem. Soc. 1999, 121, 806-812. (17) Ge, Y.; Lawhorn, B. G.; El Naggar, M.; Strauss, E.; Park, J. H.; Begley, T. P.; McLafferty, F. W. J. Am. Chem. Soc. 2002, 124, 672-678. (18) Sze, S. K.; Ge, Y.; Oh, H.; McLafferty, F. W. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 1774-1779. (19) Reid, G. E.; Shang, H.; Hogan, J. M.; Lee, G. U.; McLuckey, S. A. J. Am. Chem. Soc. 2002, 124, 7353-7362. (20) Hlavaty, J. J.; Nowak, T. Biochemistry 1997, 36, 15514-15525. (21) Stadtman, E. R. Annu. Rev. Biochem. 1993, 62, 797-821. (22) Zhang, Z.; Barlow, J. N.; Baldwin, J. E.; Schofield, C. J. Biochemistry 1997, 36, 15999-16007. (23) Cao, W.; Barany F. J. Biol. Chem. 1998, 273, 33002-33010. (24) Hlavaty, J. J.; Benner, J. S.; Hornstra, L. J.; Schildkraut, I. Biochemistry 2000, 39, 3097-3105. (25) Kurahashi, T.; Miyazaki, A.; Suwan, S.; Isobe, M. J. Am. Chem. Soc. 2001, 123, 9268-9278.
tions facilitated by H2O2 seemed to be predominantly metalcatalyzed, occurring at most of the Cu-binding amino acids. A significant degree of nonspecific oxidation reactions was observed as well, though, with Val, Pro, Cys, and Met residues undergoing oxidation despite not being coordinated to Cu. These investigators suggested that this approach could be useful for identifying the metal-binding sites of other metalloproteins too. Similarly, Scho¨neich and co-workers used MS and MCO reactions to identify three suspected Cu-binding residues in bovine growth hormone.26 Because the Cu-binding site of this protein is unknown, complete confidence that the modified amino acid residues were the actual residues bound to Cu was not possible; nor could the possibility of other amino acids in the coordination sphere be ruled out. Nonetheless, MCO reactions combined with MS may provide a powerful means of characterizing the metal binding site of proteins, especially given the sensitivity of MS. To test the general applicability of this approach, we have set out to find the optimum conditions that allow elucidation of the metal-binding site of several peptides and proteins with known coordination structures. In particular, we have sought conditions that allow the MCO reactions to specifically modify only the amino acids bound to the metal in these peptides/proteins. Such conditions should enable the amino acid residues bound to the metal to be unambiguously identified. The general approach of combining MCO reactions and MS involves first oxidizing the protein under the appropriate reaction conditions. The oxidatively modified sites are then identified using a combination of proteolytic enzymes and the peptide sequencing ability of MS. In particular, a quadrupole ion trap mass spectrometer is used, which enables not only tandem MS (MS/MS) analysis but also multiple stages of MS analysis (MSn) to be performed. EXPERIMENTAL SECTION Materials. Human angiotensin I (Agt I), [Gln11]-amyloid-βprotein fragment 1-16 (Aβ1-16), azurin from Pseudomonas aeruginosa, dithiothreitol (DTT), L-ascorbate, CuSO4, and ammonium acetate were purchased from Sigma (St. Louis, MO). The solvents used, including acetonitrile and methanol, were HPLC grade (EM Science, Gibbstone, NJ). Hydrogen peroxide (30%), formic acid, tris(hydroxymethyl)aminomethane (Tris), and tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl) were also obtained from EM Science. Trypsin and chymotrypsin were obtained from Roche Diagnostics (Indianapolis, IN). Acetic acid was from Fischer Scientific (Fair Lawn, NJ). Distilled, deionized water was obtained from a Millipore (Burlington, MA) Simplicity 185 water purification system. Metal Catalyzed Oxidation Reactions. MCO reactions with Agt I were performed at room temperature in aqueous solutions containing 35 µM peptide, 0-35 µM CuSO4, 0-40 mM ascorbate, 0-1 mM H2O2, and 50 mM Tris-HCl/Tris, buffered to a pH of 7.4. MCO reactions of Aβ1-16 were performed at room temperature with 0.5-1.0 mM peptide, 0-0.5 mM CuSO4, 10-50 mM ascorbate, 0-1 mM H2O2, and 50 mM Tris-HCl/Tris, buffered to a pH of 7.4. MCO reactions of azurin were performed at room temperature with 35 µM holo-protein, 0-100 mM ascorbate, 0-5 (26) Hovorka, S. W.; Williams, T. D.; Scho ¨neich, C. Anal. Biochem. 2002, 300, 206-211.
Analytical Chemistry, Vol. 75, No. 5, March 1, 2003
1165
Figure 2. Mass spectrum of oxidized Agt I under condition C. The insets show the expanded mass spectra around the triply charged ion of Agt I for (a) condition A, 15-min reaction; (b) condition B, 15-min reaction; (c) condition C, 15-min reaction; (d) condition D, 15-min; and (e) condition E, 5-min reaction.
mM H2O2, and 50 mM Tris-HCl/Tris, buffered to a pH of 7.4. In all cases, the MCO reactions were initiated by the addition of L-ascorbate H2O2, or both, and were stopped by adjusting the pH to 2 by the addition of 1 µL of 10% (w/v) acetic acid. Reaction times ranged from 30 min to 6 h. When present in the reaction mixture, residual H2O2 was removed by lyophilization before analysis of the reaction products. In some cases, the MCO reactions were monitored by removing an aliquot of the reaction mixture at selected time intervals and analyzing the product by HPLC-ESI-MS. When the reaction products were to be analyzed directly by ESI-MS, the solutions were first lyophilized and then redissolved in 48.5/48.5/3.0 water/methanol/acetic acid so that the concentrations of ascorbate (
