Parallel Solid-Phase Synthesis and Evaluation of Inhibitors of Streptomyces coelicolor Type II Dehydroquinase Concepcio ´n Gonza ´ lez-Bello,* ,† Emilio Lence, † Miguel D. Toscano, ‡ Luis Castedo, † John R. Coggins, § and Chris Abell ‡ Departamento de Quı ´mica Orga ´ nica y Unidad Asociada al C.S.I.C., Facultad de Quı ´mica, Universidad de Santiago de Compostela, 15782 Santiago de Compostela, Spain, University Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW, UK, and Biochemistry and Molecular Biology, University of Glasgow, Glasgow G12 8QQ, UK Received August 5, 2003 A series of 1-substituted and 4-substituted benzyl analogues of the known inhibitor (1S,3R,4R)- 1,3,4-trihydroxy-5-cyclohexene-1-carboxylic acid has been synthesized and tested as inhibitors of Streptomyces coelicolor type II dehydroquinase. The solid-phase syntheses of 18 new analogues are reported. The most potent inhibitor, 2-nitrobenzyloxy analogue 5i, has K i of 8 μM, more than 30 times lower than the K M of the substrate and approximately 4 times more potent than the original inhibitor. The binding modes of the synthesized analogues in the active site were studied by molecular docking with GOLD 2.0. Introduction The shikimate pathway is the biosynthetic pathway to the aromatic amino acids phenylalanine, L-tryp- tophan, and L-tyrosine, as well as precursors to the folate coenzymes, alkaloids, and vitamins. 1 The pathway is present in bacteria, fungi, and plants and has been recently discovered in apicomplexan parasites. 2 Its absence from mammals has made this pathway an attractive target for the development of herbicides and antimicrobial agents. The successful herbicide Glypho- sate acts by specifically inhibiting the sixth enzyme on the pathway. 3 Dehydroquinase (3-dehydroquinate dehydratase) is the third enzyme on the shikimate pathway and is responsible for catalyzing the conversion of 3-dehydro- quinic acid (1) to 3-dehydroshikimic acid (2) (Scheme 1). There are two forms of the enzyme, type I (e.g. from Escherichia coli) 4 and type II (e.g. from Streptomyces coelicolor). 5 The type I enzyme mechanism involves covalent imine intermediates between the enzyme and the substrate and proceeds with syn stereochemistry. 6 In contrast, the type II reaction proceeds through an enolate intermediate with overall anti stereochemistry. 7 These mechanistic and stereochemical distinctions have allowed us to design and synthesize compounds that are specific to either the type I 8 or type II 9 enzymes. The enolate intermediate in the type II reaction is flattened relative to the 3-dehydroquinic acid (1). In addition, a negative charge is localized toward the enolate oxygen. Both of these features have been exploited in the design of first-generation inhibitors. 9 We have previously reported that analogue 3 (Scheme 2) showed a K i of 30 μM for S. coelicolor type II dehydroquinase. The crystal structure of S. coelicolor dehydroquinase with 3 bound in the active site has recently been solved. 10 This complex identifies a number of key interactions involved in inhibitor binding and sheds light on aspects of the catalytic mechanism of the enzyme. Also present in this structure was a molecule of glycerol, originated from the enzyme storage buffer, bound 3.7 Å away from the inhibitor 3. This fact was used to design bifuntional inhibitors that straddle the two binding sites identified in the crystal structure of the enzyme. 11 The important observation that com- pounds with a double bond between C5-C6 bind in the manner predicted for transition state mimics encour- aged us to design the next generation of inhibitors. In this paper we describe attempts to make more potent inhibitors by incorporating binding interactions onto the core structure 3. We describe the synthesis of 18 new analogues, 4a-i and 5a-i (Scheme 2), using solid-phase organic synthesis (SPOS). The inhibition studies and the molecular docking with these com- pounds against S. coelicolor type II dehydroquinase are also described. Synthesis of C-1 Substituted Analogues. The strategy used for making the analogues 4a-i involved the initial preparation of the resin 8 (Scheme 3). This was made from hydroxycarbolactone 7, which was synthesized from benzoate 6 using our previously reported protocol. 12 Treatment of bromo-Wang resin 13 with the sodium alkoxide of lactone 7 afforded the lactone resin 8. Although the reaction can be carried out using THF or DMF and at room temperature or 60 °C, the best results were obtained in DMF at 60 °C. The gel-phase FT-IR spectrum of the ether lactone resin 8 showed the lactone stretching band at 1797 cm -1 . Further evidence for the formation of 8 was obtained from examination of the gel-phase 13 C NMR spectrum. Distinctive signals for the TBS group were observed at 25.6 and -3.1 ppm. In addition, the signal for the methylene benzyl group moved from 34.0 ppm in bromo- Wang resin to 71.2 ppm in the lactone resin 8. 14 Deprotection of the tertiary hydroxyl group in 8 was * Corresponding author. Tel: +34 981 563100. FAX: +34 981 595012. E-mail: cgb1@lugo.usc.es. † Universidad de Santiago de Compostela. ‡ University Chemical Laboratory. § University of Glasgow. 5735 J. Med. Chem. 2003, 46, 5735-5744 10.1021/jm030987q CCC: $25.00 © 2003 American Chemical Society Published on Web 11/22/2003