characterization of the interaction of n-acyl-l-tryptophan benzyl

THE JOURNAL OF BIOILXXCAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 269, No. 9, Issue of March 4, pp. 6587-6591, 1994 Printed in U.S.A.

Characterization of the Interaction of N-Acyl-L-Tryptophan Benzyl Ester Neurokinin Antagonists with the Human Neurokinin4 Receptor*

(Received for publication, September 27, 1993, and in revised form, November 22, 1993)

Margaret A. CascieriS, A n g u s M. Macleods, Dennis Underwoodn, Lin-Lin Shiao, Elzbieta Ber, Sharon Sadowski, Hong Yu, Kevin J. Merchants, Christopher J. Swains, Catherine D. Strader, and Tung Ming Fong From the Department of Molecular Pharmacology and Biochemistry, and Wepartment of Molecular Systems, Merck Research Laboratories, Rahway, New Jersey 07065 and the $Neuroscience Research Center, Merck Sharp & Dohme Research Laboratories, Harlow, Essex CM20 ZP?: United Kingdom

We have recently shown that a series of N-acyl-L-tryptophan benzyl esters are potent substance P antagonists (Macleod, A. M., Merchant, K. J., Cascieri, M. A, Sad- owski, S., Ber, E., Swain, C. J., and Baker, R. (1993) J. Med Chern. 14,2044-2046). We now report the detailed characterization of the interaction of N-acetyl-L-tryptopha- 3,S-bistrifluoromethy1 benzyl ester (L-732,138) with the human neurokinin-1 (NK-1) receptor. L-732,138 inhibits the binding of 12sI-substance P to the cloned human NK1 receptor expressed in Chinese hamster ovary cells with an ICao of 2.3 t 0.7 n ~ . In contrast, it has 200-fold lower affinity for the cloned rat NK-1 receptor and has >1000- fold lower affinity for the human NK-2 and NK-3 receptors. L-732,138 acts as a competitive antagonist of substance P, as shown by functional Schild analysis of the inhibition of substance P-induced inositol phosphate synthesis, by kinetic analysis of the dissociation rate, and by thermodynamic analysis of the equilibrium binding of ‘=I-substance P to the NK-1 receptor. L-732,138 also competitively inhibits the binding of the quinuclidine amine antagonist, [12sI]L-703,606, to the receptor. The compound has 230- and 10-fold reduced affinity for mutant NK-1 receptors in which histidine 266 or histidine 197, respectively, are replaced with alanine. We have previously shown that these residues play key roles in the binding of quinuclidine antagonists to the NK-1 receptor. These results suggest that the tryptophan and quinuclidine series of NK-1 antagonists bind to similar binding sites on the human NK-1 receptor.

The neurokinin 1 (NK-1)’ receptor mediates the actions of substance P and other tachykinins in the central and periph- eral nervous systems and is a member of the family of G- protein-coupled receptors. NK-1 antagonists have been shown

~

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “uduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Laboratories, Bldg. 80M-213, P. 0. Box 2000, Rahway, N J 07065. Tel.: $To whom correspondence should be addressed: Merck Research

The abbreviations used are: NK, neurokinin; CHO, Chinese hamster ovary; SP, substance P Gpp(NH)p, guanyl-5”yl imidodiphosphate; L-732,138, N-acetyl-~-tryptophan-3,5-bistrifluoromethylbenzyl ester; CP-96,345-(2S,3S~-cis-2~diphenylmethyl~-N-~~2-methoxyphenyl)- methyll-l-azabicyclo[2,2,2loctan-3-amine; L-703,606-(cis)-2-(diphenylmethyl)-N-~~2-iodophenyl~methyll-l-azabicyclo~2,2,2loctan-3-amine; L-709,210-~2S,3S~-cis-2~diphenylmethyl~-3-~3,5,-bistrifluoromethyIben- zyl-l-azabicyclo[2,2,2lodane; L-708,568-N-acetyl-~-tryptophan-3,5-di- methylbenzyl ester. L-709,400, N-acyl-benzothiophene benzyl ester.

~ ~ ~~

908-594-4609; Fax: 908-594-3337.

to attenuate C-fiber conditioning-induced reflex facilitation in rats (11, to slow excitatory postsynaptic potential in dorsal horn neurons induced by noxious stimuli in cats (2) and to inhibit capsaicin-induced plasma extravasation in rats (3) and guinea pigs (4). These data suggest that NK-1 antagonists may be useful as novel analgesic and antiinflammatory agents.

Several reports have appeared describing the synthesis and characterization of non-peptidal antagonists of the NK-1 receptor (5-8). The quinuclidine amine CP 96,345 was described as a potent and selective antagonist of the bovine and human but not the rat NK-1 receptors (5) (Fig. 1). Subsequent work in our laboratories has shown that appropriately substituted quinuclidine ethers are also potent human NK-1 receptor antagonists (7, 8 ) (Fig. 1). Using an iodinated analog of CP 96,345, [12511L-703,606, we have demonstrated that these compounds are competitive antagonists of the cloned human NK-1 receptor expressed in CHO cells (9). In addition, site-directed mutagen- esis of the human NK-1 receptor suggests that the benzhydryl group a t position 2 of the quinuclidine ring in CP 96,345 par- ticipates in an amino-aromatic interaction with histidine 197 of the receptor (10) and that the substituted aromatic ring at position 3 of certain quinuclidine amine and quinuclidine ether antagonists interacts with histidine 265 (11).

We have recently shown that a series of N-acyl-L-typtophan benzyl esters are potent antagonists of the human NK-1 receptor (12). Initial structure-activity studies in this series demonstrate that the optimal substitutions of the benzyl ester moiety are the same as the optimal substitutions of the benzyl ether moiety of the quinuclidine ether series (121, suggesting that the interaction of the benzyl group with the NK-1 receptor is similar in these two series. However, in contrast to the requirement for a basic nitrogen in the quinuclidine series, acylation of the amine increases receptor affinity in the tryptophan series (12). The present report characterizes the interaction of a potent representative of this series, N-acetyl-~-tryptophan-3,5-bistrifluoromethylbenzyl ester (L-732,138; Fig. l), with the human NK-1 receptor.

MATERIALS AND METHODS

L-732,138 was synthesized as previously described (12). Receptor assays and functional assays were performed using a stable CHO cell line expressing 1 x lo5 human NK-1 receptordcell that was selected and maintained as previously described (9, 13). Mutants of the NK-1 receptor were prepared as previously described and assayed after transient expression in COS cells (10, 11).

Receptor Binding Assays-Human NK-UCHO cells (5 x lo4) or membranes (1-2 pg) were incubated with 1251-Tyra-substance P (lZ5I-SP, 0.1 I”, 2200 CUmmol, DuPont NEN) at room temperature for 45 min as previously described and then filtered over GF/C filters that had been

6587

6588 Interaction of L-732,138 with Neurokinin-1 Receptor

CF3

Ph Ph Ph CP 96.345 L-703,606 L-709,210

dine amine (CP 96,345, L-703,606), FIG. 1. Structures of the quinucli-

quinuclidine ether (L-709,210), N- acyl-tryptophan benzyl ester (L-732, 138, G708,568), and N-acyl-benzothiophene benzyl ester (L-703,400) NK-1 receptor antagonists.

0 0

presoaked in 0.1% polyethylenimine, using a Tomtec 96-well harvester. Experiments with mutant receptors were carried out under the same conditions after transient expression in COS cells, and were performed in parallel with the wild type receptor expressed in COS cells. Assays using ['2511L-703,606 were performed in a similar fashion as described in detail previously (9).

SP-induced Inositol Phosphate Production-"he assay was performed as described by Berridge et al. (14) using cells grown to conflu- ence in 12-well tissue culture dishes. Cells were prelabeled with rnyo- [2-3Hlinositol for 24 h, washed, and incubated with substance P in the presence or absence of L-732,138 as previously described (9). Inositol

tography (9). monophosphate was isolated after extraction and ion exchange chroma-

RESULTS

L-732,138 inhibits the binding of 1251-SP to the human NK-1 receptor stably expressed in CHO cells with an IC5,-, = 2.3 2 0.7 I ~ M (mean 2 S.D., n = 6) (Fig. 2). In contrast, it has 200-fold lower affinity for the rat NK-1 receptor (Fig. 2). L-732,138 has >1000-fold lower affinity for the homologous NK-2 and NK-3 receptors, since 1 L-732,138 does not inhibit binding of lZ5I-neurokinin A to the human NK-2 receptor or of 1251-Bolton Hunter-labeled eledoisin to the human NK-3 receptor.

The addition of increasing concentrations of L-732,138 results in an increase in the EC50 of substance P for stimulating inositol phosphate synthesis in human NK-YCHO cells with- out altering the maximal stimulation observed (Fig. 3). Schild analysis of these data gives a slope of 1.15 suggesting that L-732,138 functions as a competitive antagonist of substance P at the human NK-1 receptor.

The dissociation of lZ5I-SP from the NK-1 receptor is bipha- sic, with dissociation rates of 0.014 0.001 min-l and 0.7 2 0.2 min-l for the high and low affinity binding sites, respectively (Fig. 4). Addition of excess L-732,138 (100 ma) does not alter the rates of dissociation of lZ5I-SP from the human NK-1 receptor (Fig. 4). Thus, kinetic analysis suggests that L-732,138 is a competitive antagonist of lZ5I-SP binding. In contrast, the non- hydrolyzable guanine nucleotide analog, Gpp(NH)p, which acts as an allosteric antagonist to uncouple the receptor from the G-protein, increases the rate of dissociation of 12'I-SP by de- creasing the number of receptors in the G-protein coupled, high affinity state from 81 e 1 to 48 2 4%. Scatchard analysis of the binding of lZ5I-SP to the receptor in the presence or absence of

L-732,138 CF3

0

L-708,568 CH3

L-709,400 CF3

80 -

60 -

40 -

20 -

I I I \ , I 1 10-9 10-8 1 0 7 10-6

L-732,138 (M) FIG. 2. Inhibition of 1261-SP binding to the human (0) and rat

(B) NK-1 receptors and of 1aaI-L-703,606 binding to the human NK-1 receptor (0) by G732,138. Assays were performed using membranes prepared from CHO cells stably expressing the cloned receptors.

L-732,138 (1 nM) shows that the compound increases the apparent Kd for substance P 2-fold, from 0.12 2 0.02 to 0.24 e 0.06 m, while the apparent number of binding sites is unaltered, again consistent with a competitive mechanism of antagonism with substance P (Fig. 5).

Interaction of L-732,138 with the Quinuclidine Binding Site-L-732,138 inhibits the binding of the quinuclidine amine antagonist, [1251]L-703,606 to the human NK-1 receptor with an IC50 = 3 ma (Fig. 2). The rate of dissociation of [lZ5I]L- 703,606 is the same in the absence (0.21 min") or presence (0.27 min") of excess L-732,138, suggesting that L-732,138 is a competitive inhibitor of the binding of quinuclidine antagonists to the human NK-1 receptor.

L-732,138 has 230-fold reduced affinity for the mutant NK-1 receptor in which histidine 265 is replaced by alanine (H265A), and 10-fold reduced affhity when histidine 197 is replaced with alanine (H197A) (Table I). These mutant receptors have normal binding affinity for lZ5I-SP, suggesting that the mutant receptors are synthesized and processed normally (10, 11). As was

Interaction of ,5732,138 with Neurokinin-1 Receptor 6589

I 1 I I

Substance P (M) 10-9 10-8 10-7 10-6 10-5

(ZPI) synthesis in human NK-l/CHO cells by G732,138. Control FIG. 3. Inhibition of substance P-induced inositol phosphate

(O), 100 m L-732,138 ( O ) , or 1 L-732,138 (0). Inset, Schild analysis of these data.

I I

Time Imin) 10 20 30 40 50 60

FIG. 4. Effects of C732,138 on the dissociation of 1261-SP from the human NK-1 receptor. Ligand and NK-WCHO cell membranes were incubated until equilibrium and dissociation were initiated by the addition of 100 m substance P in the absence (0) or presence of 100 m L-732,138 (0) or 10 p Gpp(NH)p ( + ). Data were analyzed as described by Duggleby (18), and theoretical curves were generated using the formula Y=R,e"xXk-l'+R,e-eYx"Z', where Y is the ratio lZ5I-SP x R/1251-SP x R,, S is time, R1 and Rz are the percentages of receptors in the low and high affinity states, and k-l and k-z are the dissociation rates from the low and high affinity sites, respectively.

previously observed with the quinuclidine antagonists, the binding affinity of L-732,138 is maintained if glutamine or phenylalanine but not serine are substituted for histidine 197. Simi- larly, the binding affinity is partially restored if glutamine but not tyrosine is substituted for histidine 265 (Table I). An analog of L-732,138 in which the indole is substituted by a benzothiophene moiety (L-709,400) has 5-fold reduced affinity for the human NK-1 receptor. As with L-732,138, this compound has 45- and 50-fold reduced affinity for the H197A and H265A mutant receptors, respectively. In contrast to L-732,138, phenylalanine will not completely substitute for histidine 197 in its interaction with L-709,400 (Table I). The 3,5-dimethyl-substituted analog of L-732,138 (L-708,568) has 16-fold reduced affinity for the human NK-1 receptor as compared to L-732,138 (Table I). The affinity of L-708,568 for the H265Amutant is reduced 80-fold compared to its affinity for the wild type receptor, while its affinity for the H197A mutant is reduced by 20-fold.

. L L m

.05

Bound (pM)

FIG. 5. Scatchard analysis of the binding of lasI-SP to the human NK-1 receptor in the absence (0) or presence (0) of 1 m L-732,138. Human NK-WCHO cell membranes were incubated with various concentrations of 1251-SP (1 PM to 1.5 rm) as described except that 10 p substance P was used to determine the amount of nonspecific binding. The data were analyzed using LIGAND (19), as purchased from BiosoR.

TABLE I Interaction of N-acyl tryptophan benzyl esters with histidines 197

and 265 of the human NKl receptor Mutants were prepared and expressed in COS cells as previously

described (10, 11). Binding assays were performed 4 days post-trans- fection as described under "Materials and Methods." The data are presented as the mean t S.D. for ( n ) determinations.

Cells ICs0

L-732,138 L-708,568 L709,400

nM hNK1R (Wild type) 3.2 f 2.6 (5) 53 f 26 (5) 15 f 7 (4) H197A H197Q

31 f 9 (2) 1000 2 10 (2) 670 f 320 (2) 2 f 1 (2)

H197F 53 f 13 (2) 12 f l ( 2 )

H197S 4 f 2 (2) 78 f 13 (2) 100 f l ( 2 )

H265A 83 2 67 (2) 950 f 50 (2) 820 f 170 (2)

H265Q 766 -c 205 (3) 4133 5 1472 (3) 770 -c 25 (2)

H265Y 26 -c 4 (2) 580 f 320 (2) 43 -c 18 (2)

3800 f 800 (2) 8500 f 3500 (2) 250 -c 10 (2)

DISCUSSION

L-732,138 is a potent, selective, and competitive inhibitor of the binding of substance P and the quinuclidine NK-1 antagonists to the human NK-1 receptor. We have previously shown that the conserved histidine residues at positions 197 and 265 in transmembrane domains 5 and 6 of the human NK-1 receptor are important determinants of the binding of quinuclidine antagonists to the receptor (10, 11). The results of the current study demonstrate that L-732,138 also specifically interacts with these two residues, suggesting that the quinuclidine and tryptophan binding sites overlap.

It is interesting to note that the H197A and H265A mutants have normal binding affinity for izsI-SP, suggesting that these 2 residues do not directly interact with the peptide agonist ligand (10, 11). However, both the quinuclidine (9) and tryptophan antagonists are competitive antagonists of substance P according to all the pharmacological criteria normally used to characterize binding and functional interactions. These pharmacological data indicate that substance P and the antagonist cannot both bind to the receptor at the same time, and that increasing concentrations of substance P will overcome the antagonism. However, the present data demonstrate that phar- macologically competitive binding does not imply that the two ligands utilize the same intermolecular binding interactions with the receptor.

Analysis of the structure activity relationships of the quinu-

6590 Interaction of L-732,138 with Neurokinin-1 Receptor

formers of L-732,138 (bold) with L-709,210 (thin). Conformational FIG. 6. Modeling studies comparing minimized energy con-

studies were camed out using the random search facility in SYBYL (Tripos Associates Inc.) with generation of 250 structures for each compound followed by full energy minimization of each structure. A minimized energy conformer for each compound had the 3,5-disubstituted phenyl rings in an offset face-to-face configuration with the benzydxyl or indole moieties. The lowest energymnformers with these rings remote from each other were higher in energy by 3.5 kcaVmol for L-732,138.

clidine antagonists with mutant receptors in which histidine 265 in transmembrane helix 6 of the NK-1 receptor is replaced by other amino acid residues suggests that this histidine forms an amino-aromatic or aromatic-aromatic interaction with the 3,5-disubstituted benzyl substituent at the 3-position of the quinuclidine antagonists (11). The data show that the affinity of quinuclidine antagonists for the H265A mutant receptor is dependent on the pattern of substitution of the benzyl ether moiety. Substitution of histidine 265 with alanine results in a 230-fold loss in affinity for L-732,138, and as we have previously shown with the quinuclidine antagonists, glutamine is the only residue tested that can partially substitute for histidine 265 in the binding interaction with the tryptophan antagonists. Since 3,5-disubstitution of the benzyl ester substituent of the tryptophan antagonists optimizes receptor affinity in a manner that is similar to that of the benzyl ether substituent of the quinuclidine antagonists (7,121, we propose that a similar interaction with histidine 265 is important for the binding of the tryptophan antagonists to the receptor. However, in contrast to the quinuclidine ether antagonists in which compounds having 3,5-dimethyl and 3,5-bistrifluoromethyI substitutions of the benzyl ether have similar affinities for the receptor (7, €0, in the tryptophan series the bistrifluoromethyl-substituted benzyl ester is 20-40-fold more potent than the 3,5-dimethyl analog (12). The electronegativity of the 3,5-disubstituted benzyl ester does not effect the interaction of the compound with histidine 265, since both L-732,138 and L-708,568 exhibit the same relative loss in affinity for the H265A mutant receptor.

An analysis of the structure activity relationships for the quinuclidine antagonists for mutant receptors in which histidine 197 in transmembrane helix 5 is replaced by other amino acids led us to propose a specific amino-aromatic interaction between this residue and the benzhydryl substituent at position 2 of the quinuclidine ring (10). In the present study we observed that substitution of histidine 197 with alanine, thereby removing the imidazole side chain from that residue, resulted in a 10-fold loss in affinity for G732,138, suggesting that histidine 197 also interacts with the tryptophan antagonist. The ability of glutamine or phenylalanine, but not serine, to substitute for histidine 197 in the interaction with L-732,138

r llular Rec I

I FIG. 7. Model for the interaction of L-732,138 with histidine 197

in helix 5 and histidine 265 in helix 6 of the human NK-1 receptor. L-732,138 and histidines 197 and 265 are shown in CPK format. Only the upper parts of helices 4, 5, 6, and 7 are shown. The three- dimensional model of the receptor reflects the propensity of each residue to prefer a lipid, water, or protein environment. The arrangement of the helices is based on the bacteriorhodopsin footprint (20).

is consistent with an amino-aromatic or aromatic-aromatic interaction rather than a hydrogen bonding interaction. In contrast to the indole-containing antagonists L-732,138 and L-708,568, phenylalanine does not substitute for the interaction of histidine 197 with the benzothiophene antagonist, L-709,400. These data suggest that histidine 197 interacts with the indole substituent of these antagonists, since a simple change in the heteroatom of this ring from nitrogen to sulfur weakens the interaction with the H197F mutant receptor relative to the wild type receptor. Modeling studies comparing the minimized energy conformers of L-732,138 with those of the quinuclidine antagonist L-709,210 also support the conclusion that the indole moiety of the tryptophan antagonists occupies the same space as the benzhydryl moiety of the quinuclidine antagonists, and that the substituted quinuclidine benzyl ether and the substituted tryptophan benzyl ester overlap (Fig. 6). Based on these data, we propose that the indole substituent of the tryptophan antagonists may specifically interact with histidine 197 via either an amino-aromatic or an aromatic-aromatic interaction.

Amino-aromatic interactions, in which the 6(+) hydrogens of histidine or glutamine would form an electrostatic interaction with the 6(-) pi-electrons of an aromatic group, have been observed in many protein structures (15-17). Amino-aromatic and aromatic-aromatic interactions are postulated to contribute up to 3 kcaVmol of binding energy. The relative potencies of L-732,138 for the wild type, H265A, and H197A mutant receptors suggest that the interactions of this antagonist with histidine 265 and histidine 197 contribute 3.3 and 1.4 kcaVmol of binding energy, respectively, which is consistent with the pro- posed interactions.

The data presented here support the model for the binding of tryptophan benzyl ester antagonists to the human NK-1 receptor shown in Fig. 7. In this model, histidine 197 and histidine 265 are located near the extracellular side of the transmembrane region on the inner face of helices 5 and 6, respectively. This model suggests that the indole and the bistrifluoromethyl benzyl ester moieties are appropriately spaced to interact with the side chains of histidine 197 and histidine 265, respectively. In addition, we have determined that bulky substituents on the amino moiety of the tryptophan antagonists are well tolerated (data not shown). The model shown in Fig. 7 suggests that the acylated amine moiety of L-732,138 faces the extracellular re-

Interaction of ,5732,138 with Neurokinin-1 Receptor 6591

gion of the receptor, and predicts that there would be a great deal of tolerance for substitution at this position.

In summary, L-732,138 is an example of an important new structural class of NK-1 antagonists. The localization of specific binding interactions for L-732,138 and the quinuclidine NK1 antagonists within helices 5 and 6 of the NK-1 receptor in positions analogous to residues previously identified to be in- volved in small molecule binding to the P-adrenergic receptor' suggests that a similar binding site suitable for antagonist interaction may be common to many G-protein-coupled receptors.

REFERENCES

1. Xu, X-J., Dalsgaard, C.-J., and Weisenfeld-Hallin, Z. (1992) Euz JPharmacol.

2. De Koninck, Y., and Henry, J . L. (1991) Proc. Natl Acad. Sci. U. S. A. 88, 216,337-344

3. Eglezos, A,, Giuliani, S., Viti, G., and Maggi, C. A. (1991) Eur J. Pharmacol. 11344-11348

4. Nagahisa,A., Kanai, Y., Suga, O., Taniguchi, K., Tsuchiya, M., Lowe, J. A,, and 209,277-279

5. Snider, R. M., Constantine, J. W., Lowe, J. A., Longo, K. P., Lebel, W. S., Woody, Hess, H-J. (1992) Eur J. Pharmacol. 217, 191-195

H. A., Drozda, S. E., Desai, M. C., Vinick, F. J., Spencer, R. W., and Hess, H.-J. (1991) Science 251, 43-37

6. Garret, C., Carruette, A., Fardin, V., Moussaoui, S. , Peyronel, J.-F., Blanchard, J.-C., and Laduron, P. M. (1991) Proc. Natl. Acad. Sci. U. S. A. 66, 10208-

7. Seward, E. M., Swain, C. J., Merchant, K., Owen, S. N., Sabin, V., Cascieri, M. 10212

A,, Sadowski, S. , Strader, C., and Baker, R. (1993) Bioorg. & Med. Chem. Lett. 3, 1361-1366

8. Swain, C. J., Seward, E. M., Owen, S. , Baker, R., Cascieri, M. A,, Sadowski, S., Strader, C., and Ball, R. G. (1993)Bioorg. & Med. Chem. Lett. 3,1703-1706

9. Cascieri, M. A,, Ber, E., Fong, T. M., Sadowski, S., Bansal, A,, Swain, C., Seward, E., Frances, B., Bums, D., and Strader, C. D. (1992) Mol. Pharma-

10. Fong, T. M., Cascieri, M. A., Yu, H, Bansal, A,, Swain, C., and Strader, C. D. col. 42, 458-463

11. Fong, T. M., Yu, H., Cascieri, M. A., Underwood, D., Swain, C. J., and Strader, (1993) Nature 362,35&353

12. Macleod, A. M., Merchant, K. J., Cascieri, M. A,, Sadowski, S., Ber, E., Swain, C. D. (1994) J. Biol. Chem., in press

13. Fong, T. M.,Anderson, S. A,, Yu, H., Huang, R.-R. C., and Strader, C. D. (1992) C. J., and Baker, R. (1993) J. Med Chem. 14, 2044-2045

14. Berridge, M. J., Downes, C. P., and Hanley, M. R. (1982) Biochem J. 206, Mol. Pharmacol. 41, 24-30

15. Perutz, M. F., Feme , G., Abraham, D. J., Poyart, C., and Bursaux. E., (1986) 587-596

16. Burley, S . K., and Petsko, G . A. (1986) FEBS Lett. 203, 139-143 J. Am. Chem. Soc. 108, 1064-1078

17. Burley, S. K., and Petsko, G. A. (1988)Adu. Protein Chem. 39, 125189 18. Duggleby, R. G. (1984) Biol. Med. 14,447455 19. Munson, P. J., and Rodbard, D. (1980) Anal. Biochem. 107,220-239 20. Henderson, R., Baldwin, J. M., Ceska, T. A,, Zemlin, F., Beckmann, E., and

Downing, K. H., (1990) J. Mol. Biol. 213, 899-929

characterization of the interaction of n-acyl-l-tryptophan benzyl

Documents