Synlett 2002(2): 0285-0289
DOI: 10.1055/s-2002-19783
LETTER
© Georg Thieme Verlag Stuttgart · New York

The Solid-Phase Parallel Synthesis of β-Strand Mimetic Templates via 1,3-Dipolar Cycloaddition with Resin-bound Vinylsulfone

Nobuhiro Fuchia, Takayuki Doia, Bolong Caob, Michael Kahnc,d, Takashi Takahashi*a
a Department of Applied Chemistry, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro, Tokyo 152-8552, Japan
b Molecumetics Ltd., 2023 120th Avenue NE, Bellevue, WA 98005-2199, USA
c University of Washington, Department of Pathobiology, SC-38, Seattle, WA 98195, USA
d Pacific Northwest Research Institute, 720 Broadway, Seattle, WA 98122, USA
e-Mail: ttakashi@o.cc.titech.ac.jp;
Further Information

Publication History

Received 17 November 2001
Publication Date:
02 February 2007 (online)

Abstract

A facile synthesis of a conformationally restricted β-strand mimetic library is described. Regioselective 1,3-dipolar cycloaddition with a resin-bound vinylsulfone provides a structurally diverse β-strand templated library. This library includes potent and selective inhibitors of serine proteases.

10

Fmoc-Rink-amide resin (0.52 mmol/g, watanabe chemical) was used as starting material. Deprotection of the Fmoc group (20% piperidine in DMF), followed by coupling to the Fmoc-protected acid with DIC in the presence of HOBt was carried out.

11

Reverse-phase HPLC analysis was performed with a Hewlett Packard 1100 series equipped with Inertsil-ODS-3 C18 column (GL Science Inc., 4.6 × 75 mm ) and a binary linear gradient (solvent A = water/0.1% formic acid, solvent B = acetonitrile/0.1% formic acid, 10% B-90% B in 10 min). Positive ion ESI-MS was measured on a Mariner TK3500 biospectrometry workstation.

12

Experimental Procedure for the Preparation of β-Strand mimetic Library 1a: To a 15 mL sealed-tube was added the vinylsulfone-bound resin 3a (150 mg, 0.52 mmol/g, 0.078 mmol), cyclic hydrazide 4 (5 equiv), aldehyde 5 (5 equiv) and 1,2-dichloroethane (2.5 mL). The tube was capped, magnetically stirred, and treated at reflux for 48 h. The reaction mixture was cooled to r.t., the resin was filtered, washed three times each with dichloromethane, DMF, dichloromethane, and methanol and the cycloadduct-bound resin 6 was dried in vacuo. To a syringe-shaped vessel was added cycloadduct-bound resin 6 (50 mg, 0.52 mmol/g, 0.026 mmol), DBU (5 equiv), and dichloromethane (1.5 mL) at r.t. and the mixture was shaken for 2 h. The reaction mixture was filtered and the resin was washed three times each with dichloromethane, DMF, dichloromethame, and methanol and the cycloadduct-bound resin was dried in vacuo. The resin was treated with a solution of 25% TFA in dichloromethane for 1 h at r.t. The reaction mixture was filtered and the filtrate was concentrated in vacuo to afford the cycloadduct 1a. Selected spectral data of 1a (R1 = H, R2 = H): 1H NMR (270 MHz, CD3OD): δ = 7.27-7.38 (m, 5 H, aromatic), 6.08 (t, J = 2.6 Hz, 1 H, C-8), 4.10 (dd, J = 2.6, 14.8 Hz, 1 H, C-7), 3.96 (dd, J = 2.6, 14.8 Hz, 1 H, C-7), 3.94 (s, 2 H, gly-CH2), 3.40 (d, J = 13.9 Hz, 1 H, Ph-CH2), 3.29 (d, J = 13.9 Hz, 1 H, Ph-CH2), 3.08-3.13 (m, 2 H, C-5), 2.47-2.52 (m, 1 H, C-4), 2.01-2.06 (m, 1 H, C-4); 13C NMR (67.8 MHz, CD3OD): δ = 174.8, 163.2, 163.1, 135.7, 134.8, 132.6, 130.9, 130.0, 118.7, 61.6, 52.2, 44.3, 44.1, 32.1, 27.3; IR(neat): 3034, 1779, 1668, 1203, 704 cm-1; ESI-TOF: 344.2 [M + H]+. 1a (R1 = CF3, R2 = H): 1H NMR (270 MHz, CD3OD): δ = 7.68 (d, J = 8.3 Hz, 2 H, aromatic), 7.49 (d, J = 8.3 Hz, 2 H, aromatic), 6.08 (t, J = 2.6 Hz, 1 H, C-8), 4.15 (dd, J = 2.6, 11.2 Hz, 1 H, C-7), 3.96 (dd, J = 2.6, 11.2 Hz, 1 H, C-7), 3.94 (s, 2 H, gly-CH2), 3.44 (s, 2 H, Ph-CH2), 3.09-3.20 (m, 2 H, C-5), 2.41-2.49 (m, 1 H, C-4), 2.06-2.11 (m, 1 H, C-4); ESI-TOF: 412.2 [M + H]+.

16

All cycloadducts were cleaved from resin (25% TFA in dichloromethane) and were analyzed by HPLC-MS (> 99% conversion, > 90% purity).

17

Experimental Procedure for the Preparation of β-Strand mimetic Library 1b: To a 15 mL sealed-tube was added the vinylsulfone-bound resin 3b (150 mg, 0.52 mmol/g, 0.078 mmol), cyclic hydrazide 4 (10 equiv), aldehyde 5 (10 equiv) and 1,2-dichloroethane (2.5 mL). The tube was capped, magnetically stirred, and treated at reflux for 48 h. The reaction mixture was cooled to r.t., the resin was filtered, washed three times each with dichloromethane, DMF, dichloromethane, and methanol and the cycloadduct-bound resin 9 was dried in vacuo. To a syringe-shaped vessel was added cycloadduct-bound resin 9 (50 mg, 0.52 mmol/g, 0.026 mmol), 0.1 M Dess-Martin reagent in dichloromethane (1.5 mL) at r.t. and the mixture was shaken for 2 h. The reaction mixture was filtered and the resin was washed three times each with dimethylsulfoxide, DMF, dichloromethame and the cycloadduct-bound resin was dried in vacuo. The resin was treated with a solution of 25% TFA in dichloromethane for 1 h at r.t. The reaction mixture was filtered and the filtrate was concentrated in vacuo to afford the cycloadduct 1b. Selected spectral data of 1b (R1 = H, R2 = H): 1H NMR (270 MHz, CD3OD): δ = 7.74-7.86 (m, 4 H), 7.19-7.50 (m, 9 H), 5.02-5.08 (m, 1 H), 4.16-4.45 (m, 6 H), 2.81-3.20 (m, 6 H), 2.35-2.46 (m, 4 H), 2.01-2.12 (m, 1 H), 1.55-1.88 (m, 6 H); ESI-TOF: 732.3 [M + H]+. 1b (R1 = CF3, R2 = H): 1H NMR (270 MHz, CD3OD): δ = 7.72-7.86 (m, 4 H), 7.56-7.71 (m, 2 H), 7.36-7.50 (m, 6 H), 5.02-5.10 (m, 1 H), 4.13-4.51 (m, 6 H), 2.78-3.25 (m, 6 H), 2.35-2.46 (m, 5 H), 1.55-1.88 (m, 6 H); ESI-TOF: 800.3 [M + H]+. 1b (R1 = H, R2 = i-Pr): 1H NMR (270 MHz, CD3OD): δ = 7.78-7.97 (m, 4 H), 7.26-7.54 (m, 9 H), 5.21-5.39 (m, 1 H), 4.08-4.43 (m, 5 H), 3.64-3.74 (m, 2 H), 2.68-2.85 (m, 5 H), 2.35-2.48 (m, 4 H), 0.84-1.48 (m, 13 H); ESI-TOF: 774.4 [M + H]+.

18

Serine protease Assays : All protease assays were performed at r.t. in 96-well microplates using Bio-Rad microplate reader (Model 3550), Molecular Devices SpectroMax 250, Labsystems Fluoroskan Ascent, or Wallac Victor fluorescence plate reader. Either 1 mM solutions of testing compounds in water or 10 mM solutions of testing compounds in DMSO served as the stock solutions for the inhibition assays. For thrombin and trypsin assays, the hydrolysis of chromogenic substrate, N-p-tosyl-Gly-Pro-Arg-pNA(Sigma) or fluorogenic substrate N-p-tosyl-Gly-Pro-Arg-AMC (Sigma) was monitored at 405 nm or at Ex: 355 nm, Em: 460 nm, respectively. For tryptase assay, the release of pNA from chromogenic substrate S-2366, pyroGlu-Pro-Arg-pNA (Chromogenix) was monitored at 405 nm. In Factor VIIa assays, the hydrolysis of fluorogenic substrate (D)Phe-Pro-Arg-AMC was monitored at Ex: 355 nm, Em: 460nm. The reaction progress curves were recorded by reading the plates, typically 80 times with 24 second intervals. Initial rates were determined by unweighted nonlinear least square fitting to a first order reaction in either GraFit (Erithacus Software Limited, London, England) or GraphPad Prism (GraphPad Software, Inc., San Diego, USA). The determined initial velocities were then nonlinear least square fitted against the concentrations of a tested compound using GraFit or Prism to obtain Ki. The general format of these assays were: 100 µL of inhibitor solution and 50 µL of enzyme solution were added in a microplate well, incubated at r.t. for 30 min, then 100 µL of a substrate solution was added to initiate the reaction.