Selective Inhibition of Human r-Thrombin by Cobalt(III) Schiff Base Complexes Toshihiko Takeuchi, ² Arnd Bo ¨ttcher, ‡ Cindy M. Quezada, Melvin I. Simon, Thomas J. Meade,* and Harry B. Gray* DiVision of Biology and the Beckman Institute California Institute of Technology Pasadena, California 91125 ReceiVed April 9, 1998 Human R-thrombin, a 34 kD protease associated with the blood coagulation cascade, converts fibrinogen into fibrin, which ultimately forms blood clots. 1 Thrombin activity has been linked to reocclusion of coronary arteries after thrombolytic therapy following heart attacks and therefore is an attractive choice for the development of new inhibitors. 2 Cobalt(III) Schiff base complexes of class 1 (acacen) (Figure 1) 3 bind histidine residues in active sites and on enzyme surfaces in a random fashion. 4 Spectroscopic and chromatographic evidence indicates that the binding of these complexes is controlled by axial ligand substitution. With several targets, including thrombin, binding of 1 to one or more histidines reduces enzymatic activity. 4c To increase inhibitor specificity and potency, a short peptide (-dFPR-) that is known to have a high affinity for the human R-thrombin active site was attached to the chelate (3, Figure 1). 5-7 A triglycine linker was inserted between the Schiff base complex and the active-site targeting sequence; the peptide was coupled to the cobalt complex by modification of the Schiff base ligand, 8 followed by activation and coupling to the peptide (NH 2 -GGG- d-FPR-CO-NH 2 ; 5) via standard solution and solid-phase peptide synthesis techniques. 9 The inhibition of thrombin by 3 was evaluated by assaying the activity of the peptide and cobalt complex in both the uncoupled and coupled forms (Figure 2). 10,11 The targeting peptide, 5, was found to be a reversible inhibitor with K i ) 290 μM. Complex 2 does not inhibit thrombin, although it does inhibit other enzymes such as thermolysin. 4c This lack of inhibition is likely due to unfavorable interactions of the carboxylic acid of 2 with the S1 aspartate located in the arginine binding pocket. In contrast, the 2-5 conjugate (3) strongly inhibits thrombin: In the reaction of 3 (1 μM) with the enzyme, a reversible interaction (K i ) 1.2 μM) is followed by an irreversible binding step with a pseudo-first-order rate constant (k app ) of 3 × 10 -5 s -1 . Alanine was substituted for arginine in the peptide-coupled complex (4) to test whether the increased inhibition is a result of a specific peptide sequence. Alanine is uncharged, and therefore the resulting peptide lacks a critical interaction required for binding to the thrombin active site. Indeed, 4 was found to be a ² Present address: Department of Pharmaceutical Chemistry, University of California, San Francisco 94143-0446. ‡ Present address: BASF AG, Ammoniaklaboratorium, 67056 Ludwig- shafen, Germany. (1) (a) Mann, K. G. Trends Biochem. Sci. 1987, 12, 229-233. (b) Shuman, M. A. Ann. N.Y. Acad. Sci. 1986, 485, 228-239. (c) Bode, W.; Huber, R.; Rydel, T. J.; Tulinsky, A. Thrombin: Structure and Function; Berliner, L. J., Ed.; Plenum: New York, 1992; pp 3-51. (d) Agnelli, G.; Pascucci, C.; Cosmi, B.; Nenci, G. G. Thromb. Haemostasis 1991, 66, 592-597. (2) Work on the inhibition of proteases by metal complexes has been reported previously: (a) Brothers, H. M.; Kostic, N. M. Biochemistry 1990, 29, 7468-7474. (b) Higaki, J. N.; Haymore, B. L.; Chen, S.; Fletterick, R. J.; Craik, S. S. Biochemistry 1990, 29, 8582-8586. (c) Bo ¨ttcher, A.; Takeuchi, T.; Simon, M. I.; Meade, T. J.; Gray, H. B. J. Inorg. Biochem. 1995, 59, 221. (d) Halfon, S.; Craik, C. S. J. Am. Chem. Soc. 1996, 118, 1227-1228. (e) Katz, B. A.; Clark, J. M.; Finer-Moore, J. S.; Jenkins, T. E.; Johnson, C. R.; Ross, M. J.; Luong, C.; Moore, W. R.; Stroud, R. M. Nature 1998, 391, 608- 612. (3) Bo ¨ttcher, A.; Takeuchi, T.; Hardcastle, K. I.; Cwikel, D.; Kapon, M.; Dori, Z.; Meade, T. J.; Gray, H. B. Inorg. Chem. 1997, 36, 2498-2504. (4) (a) Louie, A.; Meade, T. J. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 6663-6668. (b) Blum, O.; Haiek, A.; Cwikel, D.; Dori, Z.; Meade, T. J.; Gray, H. B. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 6659-6662. (c) Takeuchi, T. Doctoral Dissertation, California Institute of Technology, 1996. (5) (a) Claeson, G.; Elgendy, S.; Cheng, L.; Chino, N.; Goodwin, C. A.; Scully, M. F.; Deadman, J. The Design of Synthetic Inhibitors of Thrombin; Claeson, G., Scully, M. F., Kakkar, V. V., Deadman, J., Eds.; Plenum Press: New York, 1993. (b) Gru ¨ tter, M. G.; Priestle, J. P.; Rahuel, J.; Grossenbacher, H.; Bode, W.; Hofsteenge, J.; Stone, S. R. EMBO J. 1990, 9, 2361-2365. (6) G ) glycine, F ) phenylalanine, P ) proline, R ) arginine, A ) alanine, and d refers to the dextro isomer of phenylalanine. The peptide was prepared by the Beckman Institute Biopolymer Synthesis Group (Caltech) on p-methylbenzhydrylamine (MBHA) resin using N-tert-butyloxycarbonyl (Boc) amino acid derivatives for Merrifield solid-phase synthesis on an ABI Model 430A peptide synthesizer. (7) Human R-thrombin contains a Ser195-His57-Asp102 catalytic triad and another aspartate (Asp189) in the substrate-binding region. 5b. (8) Two equivalents of ethylenediamine were added to a solution of 1 equiv of 4,6-dioxoheptanoic acid in CH2Cl2. An insoluble condensation product was isolated and characterized. This material was reacted with 2 equiv of 2,4- pentanedione in absolute ethanol for 6 h and filtered. Successive reactions were performed as previously described and the final product purified by ion exchange or flash chromatography. 4c The structure of the desired product, 2, was confirmed by 1 H and 13 C NMR, and X-ray crystallographic analysis. (9) The starting materials (1, 2) and products (3-5) were characterized by HPLC, MALDI/TOF, and electrospray mass spectrometry. (10) Human R-thrombin (3037 NIH units/mg) and the substrate Spec- trozyme TH (H-D-hexahydrotyrosyl-L-alanyl-L-arginine-p-nitroanilide-diacetate salt) were purchased from American Diagnostica. Enzyme-catalyzed hydrolysis rates were measured spectrophotometrically by using a Hewlett-Packard HP8452A diode array spectrophotometer equipped with a HP89090A peltier temperature control accessory. The peptidase activities of thrombin were determined by following the increase in absorption at 406 nm due to the enzymatic hydrolysis of Spectrozyme TH. Initial velocities were determined for e10% of the reaction. All assays were performed in 10 mM Tris, 10 mM HEPES, 0.1% poly(ethylene glycol), 0.5 M NaCl, pH 7.8 (run buffer), using 3.07 × 10 -9 M thrombin and 40 μM substrate concentration. Irreversible thrombin assays were performed by incubating thrombin in run buffer to yield a final enzyme concentration of 1.54 × 10 -7 M thrombin and a cobalt concentration (3) of 1 μM. These samples were incubated at 25 °C together with controls where the cobalt complex was absent. 20 μL aliquots of these solutions were assayed periodically for residual enzyme activity by dilution of the solution to 3.07 × 10 -9 M thrombin, followed by addition of substrate to 40 μM. The percent activities reported are in reference to a thrombin control, which did not lose activity during the course of the experiments. (11) Ki for the peptide was determined by plotting V0/Vi versus [I] (V0/Vi ) [I]/Ki(1 + [S]/Km) + 1), using a Km value of 2.45 × 10 -6 M for Spectrozyme TH (from: Sonder, S. A.; Fenton, J. W., II. Clin. Chem. 1986, 32, 934-937). The determinations were performed with three different concentrations of Spectrozyme (10, 40, and 100 μM) and peptide concentrations ranging from 1 to 20 mM. The IC50 value for 3 at time 0 is 1.3 μM. The pseudo-first-order rate constant for inactivation (kapp) was calculated from the initial rates of inhibition according to the following: ln(E1/E2) ) kapp(T2 - T1), where E1 and E2 are the observed activities at times T1 and T2. The Ki of 3 is an upper limit based upon IC50 measurements, where IC50 ) Ki (1 + {[S]/Km}) (from: Cheng, Y.-C.; Prusoff, W. H. Biochem. Pharm. 1973, 22, 3099-3108). Figure 1. Structure of cobalt(III) Schiff base complexes: X ) NH3, 2-MeIm; acacen ) bis(acetylacetone)ethylenediimine. E + CoLX 2 y \ z K i E‚CoLX + X 9 ' k app E-CoL + X 8555 J. Am. Chem. Soc. 1998, 120, 8555-8556 S0002-7863(98)01191-3 CCC: $15.00 © 1998 American Chemical Society Published on Web 08/05/1998