Interaction of Chlorisondamine with the Neuronal Nicotinic

Interaction of Chlorisondamine with the Neuronal Nicotinic. Acetylcholine Receptor. Amina S. Woods,* Susanne C. Moyer,† Hay-Yan J. Wang, and Roy A. ...
0 downloads 0 Views 307KB Size
Interaction of Chlorisondamine with the Neuronal Nicotinic Acetylcholine Receptor Amina S. Woods,* Susanne C. Moyer,† Hay-Yan J. Wang, and Roy A. Wise The National Institute on Drug Abuse, Intramural Research Program, National Institutes of Health, 5500 Nathan Shock Drive, Baltimore, Maryland, 21224 and the Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218 Received October 10, 2002

An epitope was found on the R2-nicotinic isoform of the neuronal nicotinic acetylcholine receptor that would likely form salt bridges with quaternary ammonium compounds and a cation-π interaction with the π-cloud of an aromatic ring. Chlorisondamine, a nicotinic antagonist, exerts a long-lasting, if not permanent, blockade of the ion channel gated by acetylcholine. Blocking of the ion channel prevents nicotine from exerting its rewarding effect on the CNS. Chlorisondamine contains two quaternary ammonium groups and a tetrachloroisoindoline ring. We propose that chlorisondamine interacts with an epitope on the R2 isoform of the rat neuronal nicotinic receptor (residues 388-402, GEREETEEEEEEEDE), where one or both of the quaternary ammonium groups of chlorisondamine form a salt bridge with either a glutamic acid side chain or a phosphate group, whereas the tetrachlorobenzene portion of the tetrachloroisoindoline ring interacts with the guanidinium group of arginine in a cation-π association. In this work, a new way of probing the interaction of a receptor epitope (R2) with organic molecules (chlorisondamine and hexachlorobenzene) was undertaken using matrix assisted laser desorption/ionization mass spectrometry. Keywords: noncovalent interactions • protein epitopes • quaternary amines • ganglionic blockers

Introduction The nicotinic acetylcholine receptor is the most widely studied and best characterized neuroreceptor.1,2 The neurotransmitter acetylcholine combines with this receptor to initiate excitatory postsynaptic potential in nerves or to initiate the end plate potential in muscles. The neuronal nicotinic cholinergic receptor has varied subunit composition. Some are composed of only one type of subunit (e.g., R7, R8, R9 homomers) and others of two or more kinds of subunits (e.g., pairs of R2, R3, or R4 with β2 or β4). The homology among the subunits is quite high; about 40% of the amino acid residues are identical. Ligand recognition sites are located at the amino termini and binding of agonists or competitive antagonists occurs at receptor subunit interfaces. To open the ion channel, two molecules of acetylcholine must bind to the receptor simultaneously.3-5 The EC50 of acetylcholine is 1mM.5 Most nicotinic cholinergic receptor blockers are quaternary or bis-quaternary ammonium compounds. Chlorisondamine (CHL, 4,5,6,7-tetrachloro-1,3-dihydroxy-2-methyl-2[2-trimethylammoniumethyl]-2H-isoindolium diiodide) a nicotinic antagonist, has a bis-quaternary ammonium structure. Its ability to cross the blood brain barrier is dose dependent.6 When given * To whom correspondence should be addressed. E-mail: awoods@intra. nida.nih.gov. † Department of Chemistry, Johns Hopkins University. 10.1021/pr025578h CCC: $25.00

 2003 American Chemical Society

in sufficiently high doses to cross the blood-brain barrier or when injected centrally, it is sequestered in the central nervous system, and it exerts a nicotinic blockade that lasts for months. However, it does not have the same long-lasting effect at skeletal muscle end plates. The administration of CHL does not interfere with acetylcholine or nicotine binding and does not result in neuronal degradation or in a decrease in the number of nicotinic cholinergic receptors. Extensive work by Clarke et al. suggests that CHL is a noncompetitive functional antagonist and that the binding does not occur at the acetylcholine binding site.6 Furthermore, the ability of chlorisondamine to block nicotinic receptors in the reward system suggests that the compound functions as a channel blocker rather than a surface blocker.7 The current work was undertaken to locate an epitope on the neuronal nicotinic acetylcholine receptor that could possibly interact with chlorisondamine. In addition, it suggests an explanation as to the chemistry and neurobiology involved. We identified an epitope on the R2-nicotinic isoform of the neuronal nicotinic acetylcholine receptor that can participate in multiple noncovalent interactions with chlorisondamine. This epitope can form salt bridges with the two quaternary ammonium groups and a cation-π interaction with the tetrachloroisoindoline ring of chlorisondamine. Dougherty and co-workers have accumulated several impressive experimental proofs showing that cation-π interactions Journal of Proteome Research 2003, 2, 207-212

207

Published on Web 01/25/2003

research articles occur between side chains of positively charged amino acid residues (arginine and lysine) or quaternary ammonium compounds and aromatic residues (tryptophan, tyrosine, and phenylalanine).8-11 Additionally, Schrader et al.12 have provided evidence of cation-π interactions between benzene rings and the guanidinium group of arginine. Several neurotransmitters and other biologically active amines associate with proteins via cation-π bonding. Sussman et al. have shown that in aminoaromatic interactions, the NH4+-π portion of the interaction contains a large nonelectrostatic component. This emphasizes the importance of both nonelectrostatic and electrostatic interactions in cation-π bonding.13 Numerous investigations have employed gas-phase techniques to better understand solution phase phenomena. In particular, mass spectrometry has found widespread use in the study of noncovalent interactions involving biological molecules. Recently, several researchers have implemented mass spectrometry in the study of cation-π interactions. Dunbar et al. performed extensive investigations on the gas-phase interactions of metal ions with aromatic molecules.14-18 RodriguezCruz and co-workers studied gas-phase dissociation reactions of hydrated alkaline earth metal ions with benzene.19 In addition, the binding energies of benzene to Na+ and K+ ions have been determined by CID experiments. Ryzhov and coworkers provided an elegant example of the gas phase determination of metal-ion affinities of the aromatic amino acids, providing insight into how these constituents might interact in biological systems.20,21 Although electrospray ionization (ESI) is the most widespread mass spectrometric technique for the study of noncovalent interactions, several researchers have recently utilized matrix-assisted laser desorption/ionization (MALDI).22-30 In previous work, it was shown that salt bridging occurred between peptides containing two or more adjacent basic residues (e.g., RR, RKR), and peptides containing two or more adjacent acidic residues (e.g., EE, DD).29,30 In peptides and proteins, the charge of the ionized groups of side chains of arginine, lysine, histidine, aspartic acid and glutamic acid residues, plus the R-amino and R-carboxyl groups are distributed over two or more hydrogen or oxygen atoms. The stable positive charge on quaternary ammonium compounds makes them readily available to participate in ionic interactions. In addition, these types of interactions are stable in the gas phase, making them amenable to mass spectrometric analysis. Clarke et al. used Sprague-Dawley rats to demonstrate the irreversible blockage of the neuronal nicotinic acetylcholine receptor function by chlorisondamine (MW ) 358.2).31,32 It is our contention that chlorisondamine interacts with a cytoplasmic epitope of the R2 isoform of the neuronal nicotinic receptor (residues 388-402, GEREETEEEEEEEDE). This epitope contains 12 acidic residues (11 Glu and 1 Asp) a Thr that has a 99.7% likelihood of being phosphorylated33 and an Arg. It was surmised that the two quaternary ammonium sites of CHL interact with the side chain of the Glu residues and/or the phosphate group on the Thr, whereas its aromatic ring interacts with the guanidinium group of Arg. To test this hypothesis, a study of the interaction of the nonphosphorylated (GEREETEEEEEEEDEN; MW ) 1981.8), and phosphorylated (GEREEpTEEEEEEEDEN; MW ) 2061.8) epitope (residues 388-403) of the R2 isoform with chlorisondamine and hexachlorobenzene was undertaken. Enzymatic digests of the epitope-chlorisondamine complexes were performed in order to assess the stability of the noncovalent interactions. 208

Journal of Proteome Research • Vol. 2, No. 2, 2003

Woods et al.

Experimental Section Peptides. Protein epitopes GEREETEEEEEEEDEN and GEREEpTEEEEEEEDEN (p ) phosphorylated residue) were synthesized at the Johns Hopkins School of Medicine Peptide Synthesis Core Facility. Peptides were used at a concentration of 100 pmol/µL in water. Chemicals. Hexachlorobenzene was purchased from Aldrich Chemical Co. (Milwaukee, WI) and chlorisondamine iodide was acquired from Tocris (St. Louis, MO). Both were diluted in water to a concentration of 1 nmol/µL. Sphingomyelin was purchased from Avanti (Alabaster, AL) and diluted to a concentration of 1 nmol/µL in water:methanol:chloroform, 2:1:1 (v/v/v). Solutions. Solutions (v/v) of each protein epitope with chlorisondamine (CHL), or both protein epitopes with chlorisondamine or each protein epitope and hexachlorobenzene were prepared. Enzyme. Pronase from Steptomyces griseus was purchased from Roche Molecular Biochemicals (Indianapolis, IN). The enzyme was used at a concentration of 0.06 ug/µL. Matrix. 6-aza-2-thiothymine (ATT) was purchased from Aldrich (Milwaukee, WI) and prepared fresh daily as a saturated solution in 50% ethanol. Pronase Digests. Enzymatic digestions were performed by combining 2 µL of the protein epitope-CHL solution, 4 µL of ammonium bicarbonate (25 mM, pH 8.0) and 2 µL of Pronase (0.06 µg/µL). A 0.3 µL aliquot portion of the digest was deposited on the sample plate, and the reaction was stopped at time intervals of 1, 2, 3, and 5 min and overnight by addition of 0.3 µL matrix. Blank digests were also evaluated, where water was substituted for the peptide. Sample Preparation. A mixture of 0.3 µL peptide + 0.3 µL matrix (ATT) was applied to the MALDI target and allowed to air-dry prior to introduction into the mass spectrometer. Instrument. Mass spectra were acquired on a PE-Biosystems DE-PRO MALDI time-of-flight mass spectrometer (Framingham, MA) equipped with a N2 laser (337 nm). A 20 kV extraction voltage was used. The instrument was operated in the linear mode and spectra were obtained in both the positive and negative ion modes. Each spectrum was the average of 50 laser shots.

Results and Discussion a. Addition of Chlorisondamine to the Synthetic Epitopes of the R2 Isoform of the Neuronal Nicotinic Receptor. The addition of chlorisondamine to each of the synthesized protein epitopes GEREETEEEEEEEDEN (Figure 1a) and GEREEpTEEEEEEEDEN (Figure 1b) was performed to assess the likelihood of complex formation. This resulted in complexation of two molecules of chlorisondamine to both the nonphosphorylated (m/z 2339.9, 2698.3) and the phosphorylated (m/z 2420.0, 2778.2) epitopes. Chlorisondamine was then added to an equimolar solution of both epitopes in order to determine if the phosphorylated Thr residue produces a more favorable complex. The mixture (Figure 1c) results in molecular ions of complexes located at m/z 2339.9, 2420.0, and 2778.2. This indicates that the presence of the phosphate group is more conducive to the formation of noncovalent complexes with the bis-quaternary amines on the CHL molecule, than the carboxyl groups of the Glu side chain. Other peptides containing two or more acidic adjacent residues or a phosphate group were tested and formed noncovalent complexes with chlorisondamine (data not shown).

research articles

Interaction of Chlorisondamine

Table 1. Fragments Resulting from the OvernightPronase Digest of the GEREETEEEEEEEDEN-Chlorisondamine (CHL) Complex fragment

sequence

MH+

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14

GEREET EEEEED EEEEEE GEREETE TEEEE + CHL GEREETEEEE EEEEED + CHL EEEEDEN + CHL EEEEEEEDEN GEREETEEEEEEE EEEEEEEDEN + CHL TEEEEEEEDEN + CHL PyrREETEEEEEE + CHL EREETEEEEEEE + CHL

720.7 779.8 793.7 850.0 993.9 1108.2 1136.9 1251.0 1281.1 1624.5 1638.3 1739.5 1777.7 1924.1

Table 2. Fragments Resulting from the Overnight Pronase Digest of the GEREEpTEEEEEEEDEN-Chlorisondamine (CHL) Complex

Figure 1. (a) Noncovalent complexes of the R-2 epitope and chlorisondamine (CHL). (b) Noncovalent complexes of the phosphorylated R-2 epitope and CHL. (c) Noncovalent complexes of a mixture of the R-2 epitope, the phosphorylated R-2 epitope and CHL.

Addition of Chlorisondamine to Sphingomyelin. Sphingomyelin is a molecule that has a heterogeneous fatty acid content, a phosphate group, and a quaternary amine. Mass spectrometry of the sphingomyelin-CHL mixture revealed that only one molecule of chlorisondamine added to sphingomyelin (m/z 1089.3, Figure 2). Several peaks are observed for both the uncomplexed and the complexed sphingomyelin due to the heterogeneity in the fatty acid chain. However, the relative intensities of these ions are rather low, suggesting that association between sphingomyelin and chlorisondamine is not as

fragment

sequence

MH+

P1 P2 P3 P4 P5 P6 P7 P8 P9

EEEEED EEEEDEN pTEEEEEEE EEEEED + CHL GEREEpTEEE GEREEpTE + CHL EEpTTEEEEEEED + CHL GEREEpTEEEEEEE + CHL GEREEpTEEEEEEE + (CHL-Cl)

779.8 893.8 1103.9 1137.0 1188.0 1287.0 1834.2 1932.6 2025.8

favorable as that between the protein epitope and chlorisondamine. Other glycolipids containing a phosphate group and/ or a quaternary amine were tested and resulted in complex formation (data not shown). Addition of Hexachlorobenzene. Hexachlorobenzene was added to both the phosphorylated and nonphosphorylated protein epitope. The relative intensity of the hexachlorobenzephosphorylated protein epitope is barely discernible from the noise. The nonphosphorylated epitope formed a better complex; however, it was not as prominent as the chlorisondamine-epitope complex. Other peptides containing one or more Arg, but devoid of adjacent acidic residues or phosphate groups displayed complexes with high relative intensities. These results seem to indicate that the guanidinium group of Arg does interact with the hexachlorobenzene ring, however the presence of the acidic carboxyl groups and the very acidic phosphate on the epitope can cause some Coulombic repulsion, thus minimizing the interaction between the epitopes and the hexachlorobenzene. Pronase Digest. The protein epitope and CHL were combined and the resulting complex was digested with Pronase. In the nonphosphorylated epitope-CHL digest, the most prominent peak at m/z 1136.9 corresponds to an epitope fragment complexed with CHL (P7: EEEEED + CHL, Table 1, Figure 3a). Other digest fragments formed complexes (P5, P8, P11, P12, P13, and P14) although their ion signals were not as prominent. The phosphorylated protein epitope-CHL digest (Table 2, Figure 3b) resulted in a total of nine fragments. The most prominent peak at m/z 1287.0 (P6: GEREEpTE + CHL) contains the phosphorylated Thr, 4 Glu and an Arg. The fragment EEEEED + CHL (P4), is present but its relative intensity is about 15% that of P6, suggesting that the formation of a complex is more likely in an epitope containing the more acidic phosphate Journal of Proteome Research • Vol. 2, No. 2, 2003 209

research articles

Woods et al.

Figure 2. Noncovalent complexes of sphingomyelin (sph) and chlorisondamine (CHL).

the interaction seems to indicate that it is strong enough to resist an attack by an aggressive enzyme such as Pronase. Ab Initio Calculations. Geometry optimization of CHL and hexachlorobenzene was carried out at the Hartree-Fock 6-31G** level of theory using Spartan ’02.34,35 The structure of CHL is shown in Figure 4a. Electrostatic potential surfaces were generated by mapping the 6-31G** electrostatic potentials. Briefly, red colors represent regions of negative potential, whereas blue colors represent areas of positive potential. Orange, yellow, and green represent regions of intermediate potential. Figure 4b and c displays the plots of the electrostatic potential surface of CHL. Figure 4b shows the electrostatic potential from above the plane of the ring, and Figure 4c shows the electrostatic potential from below the ring. Figure 4b shows that the quaternary ammonium group located on the CHL side chain is the region of highest positive electrostatic potential. The top surface of the ring is negatively charged, as indicated by the orange coloring. As seen in Figure 4c, the underside of the tetrachlorobenzene portion of the ring structure is the most negatively charged region of the molecule. The CHL molecule is positively charged at one end and has a negative region surrounding the plane of the tetrachlorobenzene portion of the tetrachloroisoindoline ring at the other. The peripheral quaternary ammonium has the potential to interact with a Glu, Asp, or phosphate groups, whereas the ring portion of CHL can possibly interact with a guanidinium group on an Arg residue. As indicated by the green color in the electrostatic potential plot, the quaternary ammonium in the ring structure is essentially neutral, which would suggest that interaction with an acidic residue is unlikely. Figure 3. Pronase digest of noncovalent complexes of the (a) R-2 epitope with chlorisondamine and (b) the phosphorylated R-2 epitope with chlorisondamine.

and the basic Arg, rather than in an epitope containing only acidic residues. The fact that an overnight digest did not disrupt 210

Journal of Proteome Research • Vol. 2, No. 2, 2003

The electrostatic potential surface of hexachlorobenzene was also generated (Figure 4d). The area above the aromatic ring is a region of slightly positive electrostatic potential that occurs because of the highly electron withdrawing nature of the six chlorine molecules on the ring structure. The region of high negative potential on this molecule is located around the

Interaction of Chlorisondamine

research articles Table 3. Cytoplasmic Epitopes from Sequences ofR2 Isoforms of Neuronal Nicotinic Acetylcholine Receptors

Figure 4. (a) Hartree-Fock 6-31G** optimized structure of chlorisondamine. The electrostatic potential surface of chlorisondamine: (b) view from above and (c) below the molecule. (d) The electrostatic potential surface of hexachlorobenzene.

chlorine atoms. This would suggest that hexachlorobenzene is able to interact with positively charged groups, such as the guanidinium group on an arginine side chain. However, due to the slightly positive electrostatic potential above the ring, the interaction can only occur with the negatively charged edge of the ring positioned to interact with the guanidinium group; thus making the overall nature of this interaction precarious. Cation-π interactions have been implicated in the binding of acetylcholine and other choline derivatives to the nicotinic receptor.36 Dougherty and co-workers performed ab initio calculations on Trp derivatives to assess the potential effect of modifications on cation-π interactions with acetylcholine.37

Other researchers have performed computational studies on tetramethylammonium and acetylcholine in order to predict potential interactions with other molecules.37,38 In addition, several groups have suggested that negatively charged amino acid residues as well as aromatic residues may be responsible for the binding of acetylcholine and other cholinergic ligands to receptor sites.39-43 The data indicates that the positively charged quaternary ammonium of CHL can form stable complexes with Glu and Asp, and negatively charged phosphate groups of posttranslationally modified Ser, Thr, and Tyr. The tetrachlorobenzene portion of CHL forms a stable noncovalent complex with the guanidinium group of Arg. The R2 subunit of Rattus norvegicus contains the epitope EREETEEEEEEEDE (amino acid residues 389-402) in its cytoplasmic domain. According to seven different modeling programs of the secondary and tertiary structure of the R2 subunit, this stretch of acidic residues forms an R-helix.44,45 The right-handed R-helix has 3.6 residues/turn, a translation of 1.5 Å/residue for a total translation of 5.4 Å/turn the φ and ψ are all less than 50°, whereas the internal angle of the chlorisondamine molecule is greater than 50°.46 It is, therefore possible that chlorisondamine surrounds the helix in a V-shaped vice that is made up on one side of the cation-π interaction with Arg 390, and on the other side of the peripheral quaternary ammonium with one or more of the acidic residues or with the phosphate group on Thr 393. However, it is also possible that the interactions could be coming from opposite sides of the chlorisondamine molecule, with the cation-π interaction proceeding from the underside of the ring and salt bridging at the opposite end of the molecule on the peripheral quaternary ammonium. The positive charge at one end of the molecule and negative charge at the other increases the potential for association. However, the likelihood of simultaneous dissociation of two different types of interactions is remote. Once CHL interacts with the epitope, a change in the conformation of the subunit could occur that might lead to an overall change in the conformation of the receptor. The conformational change in the receptor could result in the blockage of the channel, or in the prevention of dephosphorylation or phosphorylation of Thr 393 on the epitope, rendering the receptor nonfunctional. Chlorisondamine also interacts with phosphate groups found on membrane lipids, however the binding must be quite tenuous as the relative intensities of sphingomyelin-CHL complexes are rather small, compared with the intensities of the sphingomyelin-acetylcholine complexes (data not shown). A search of the NIH database (Table 3) revealed that in addition to the rat, the R2 isoform of the neuronal nicotinic Journal of Proteome Research • Vol. 2, No. 2, 2003 211

research articles receptors from human, mouse, monkey, and chicken have been sequenced. All of the sequences contain similar epitopes in their cytoplasmic region. The fact that the epitope is conserved attests to its importance and suggests that the presence of acidic residues might be essential for maintaining a structurefunction relationship of the R2 isoform of the neuronal nicotinic acetylcholine receptor and that the site might have a physiological role or function that is not yet known.

Acknowledgment. S.M. thanks the NSF/GOALI Program for a Graduate Research Fellowship. The authors gratefully acknowledge Andrew Taggi (Department of Chemistry, Johns Hopkins University) for helpful discussions regarding this work. References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20)

212

Karlin, A.; Akabas, M. H. Neuron 1995, 15, 1231. Galzi, J.-L.; Changeux, J.-P. Neuropharmacology 1995, 34, 563. Grutter, T.; Changeux J-P. Trends Biochem. Sci. 2001, 26, 459. Lindstrom, J. In Neuronal Nicotinic Receptors, Pharmacology and Therapeutic Opportunities; Arneric, S. P., Brioni, J. D., Eds.; WileyLis: New York, 1999; p 3. Taylor, P. In Goodman and Gillman The Pharmacological Basis of Therapeutics; Hardman, J. G., Limbird, L. E., Eds.; McGrawHill: New York, 2000; p 193. Clarke, P. B. S.; Chandieu, I.; El-Bizri, H.; Boksa, P.; Quick, M.; Esplin; B. A.; Capec, R. Br. J. Pharmacol. 1994, 111, 397. Wise, R.; Marcangione, C.; Bauco, P. Synapse 1998, 29, 72. Dougherty, D. A. Science 1996, 271, 163. Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303. Gallivan, J. P.; Dougherty, D. A. PNAS 1999, 96, 9459. Dougherty, D. A.; Lester, H. A. Nature 2001, 411, 252. Rensing, S.; Arendt, M.; Springer, A.; Grawe, T.; Schrader, T. J. Org. Chem. 2001, 66, 5814. Zhu, W.-L.; Tan, X.-J.; Puah, C. M.; Gu, J. D.; Jiang, H.-L.; Chen, K. X.; Felder, C. E.; Silman, I.; Sussman, J. L. J. Phys. Chem. A 2000, 104, 9573. Y-P.; Yang, Y.-C.; Klippenstein, S. J.; Dunbar, R. C. J. Phys. Chem. A. 1997, 101, 3338. Lin, Y.-C.; Dunbar, R. C. Organometallics 1997, 16, 2691. Gapeev, A.; Dunbar, R. C. J. Am. Chem. Soc. 2001, 123, 8360. Ho, Y.-P.; Dunbar, R. C. Int. J. Mass Spectrom. 1999, 182/183, 175. Gapeev, A.; Dunbar, R. C. J. Phys. Chem. A. 2000, 104, 4084. Rodriguez-Cruz, S. E.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2001, 12, 250. Ryzhov, V.; Yang, C.-N.; Klippenstein, S. J.; Dunbar, R. C. Int. J. Mass Spectrom. 1999, 185/186/187, 913.

Journal of Proteome Research • Vol. 2, No. 2, 2003

Woods et al. (21) Ryzhov, V.; Dunbar, R. C. J. Am. Chem. Soc. 1999, 121, 2259. (22) Lecci, P.; Pannell, L. K. J. Am. Soc. Mass Spectrom. 1995, 6, 972. (23) Woods, A. S.; Buchsbaum, J. C.; Worrall, T. A.; Cotter, R. J.; Berg, J. M. Anal. Chem. 1995, 67, 4462. (24) Juhasz, P.; Biemann, K. PNAS 1994, 91, 4333. (25) Tang, X.; Callahan, J. H.; Zhou, P.; Vertes, A. Anal. Chem. 1995, 67, 4542. (26) Farmer, T. B.; Caprioli, R. M. J. Mass Spectrom. 1998, 33, 697. (27) Lin, S.; Cotter, R. J.; Woods, A. S. Proteins Structure Function Genetics 1998, 2, 28. (28) Lin, S.; Long, S.; Ramirez, S. M.; Cotter, R. J.; Woods, A. S. Anal. Chem. 2000, 72, 2635. (29) Woods, A. S.; Huestis, M. A. J. Am. Soc. Mass Spectrom. 2001,12, 88. (30) Woods, A. S.; Koomen, J.; Ruotolo, B.; Gillig, K. J.; Russell, D. H.; Fuhrer, K.; Gonin, M.; Egan, T.; Schultz, J. A. J. Am. Soc. Mass Spectrom. 2002, 13, 166. (31) El-Bizri, H.; Clarke, P. B. S. Br. J. Pharmacol. 1994, 111, 406. (32) El-Bizri, H.; Clarke, P. B. S. Br. J. Pharmacol. 1994, 111, 414. (33) NetPhos Program, Blom, N.; Gammeltoft, S.; Brunak, S. J. Mol. Biol. 1999, 294, 1351. (34) Spartan ‘02, Wavefunction, Inc., Irvine, CA. (35) Kong, J.; White, C. A.; Krylov, A. I.; Sherrill, C. D.; Adamson, R. D.; Furlani, T. R.; Lee, M. S.; Lee, A. M.; Gwaltney, S. R.; Adams, T. R.; Ochsenfeld, C.; Gilbert, A. T. B.; Kedziora, G. S.; Rassolov, V. A.; Maurice, D. R.; Nair, N.; Shao, Y.; Besley, N. A.; Maslen, P. E.; Dombroski, J. P.; Daschel, H.; Zhang, W.; Korambath, P. P.; Baker, J.; Byrd, E. F. C.; Van Warshen, T.; Johnson, B. G.; Gill, P. M. W.; Head-Gordon, M.; Pople, J. A. J. Comput. Chem. 2000, 21, 1532. (36) Dougherty, D. A.; Stauffer, D. A. Science 1990, 250, 1558. (37) Zhong, W.; Gallivan, J. P.; Zhang, Y.; Li, L.; Lester, H. A.; Dougherty, D. A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 12 088. (38) Barrett, A. N.; Roberts, G. C.; Burgen, A. S.; Clore, G. M. Mol. Pharmacol. 1983, 24, 443. (39) Czajkowski, C.; Kaufmann, C.; Karlin, A. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 6285. (40) Satow, Y.; Cohen, G. H.; Padlan, E. A.; Davies, D. R. J. Mol. Biol. 1986, 190, 593. (41) Glockshuber, R.; Stadlmuller, J.; Pluckthun, A. Biochemistry 1991, 30, 3049. (42) Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.; Goldman, A.; Toker, L.; Silman, I. Science 1991, 253, 872. (43) Radic, Z.; Gibney, G.; Kawamoto, S.; MacPhee-Quigley, K.; Bongiorno, C.; Taylor, P. Biochemistry 1992, 31, 9760. (44) Lacroix, E.; Viguera, A. R.; Serrano, L. J. Mol. Biol. 1998, 284, 173. (45) Munoz, V.; Serrano, L. Nat Struct Biol. 1994, 1, 399. (46) Creighton, T. E. Proteins: Structure and Molecular Properties. W. H. Freeman and Company: New York, 1993; p 182.

PR025578H