Phosphorous acid analogs of novel P2–P4 macrocycles as inhibitors of HCV–NS3 protease Marco Pompei a, * , Maria Emilia Di Francesco a , Uwe Koch a , Nigel J. Liverton b , Vincenzo Summa a a IRBM (Merck Research Laboratories), Rome, Italy b Merck Research Laboratories, West Point, PA 19486, United States article info Article history: Received 11 February 2009 Revised 3 March 2009 Accepted 5 March 2009 Available online 17 March 2009 Keywords: Hepatitis C HCV NS3 Protease inhibitor Bioisosteres Phosphonate MK7009 abstract HCV–NS3 protease is essential for viral replication and NS3 protease inhibitors have shown proof of con- cept in clinical trials. Novel P2–P4 macrocycle inhibitors of NS3/4A comprising a P1 C-terminal carboxylic acid have recently been disclosed. A series of analogs, in which the carboxylic residue is replaced by phos- phorous acid functionalities were synthesized and found to be inhibitors of the NS3 protease. Among them the methylphosphinate analogue showed nanomolar level of enzyme inhibition and sub-micromo- lar potency in the replication assay. Ó 2009 Elsevier Ltd. All rights reserved. Hepatitis C virus (HCV) infection is the major causative agent of chronic liver disease and it is estimated that more than 170 million individuals worldwide are infected with this pathogen. 1 Currently, the only FDA-approved regimen for the treatment of HCV infection is a combination of alpha interferon and ribavirin, a therapy which is poorly tolerated and has a suboptimal response particularly in patients infected with HCV-genotype 1 (gt1), the most prevalent virus genotype. 2 HCV is a small, enveloped, single stranded positive RNA virus belonging to the Flaviviridae family. 3 To date, amongst the inhibitors of the HCV–NS3 serine protease so far discovered, 4 the structurally optimized product-derived analogs represent the most promising category. 5 Clinical proof of concept was achieved with BILN-2061, 6 (Fig. 1) a rapidly reversible inhibitor. The search for NS3 protease inhibi- tors incorporating new structural motifs was reinforced by discon- tinuation of clinical evaluation of BILN-2061 due to its cardiac toxicity. 7 This prompted several groups to investigate optimization of both the peptide backbone 8 and the C-terminal P1 carboxylate region. 9 Through this approach, our laboratories identified a novel class of P2–P4 macrocycle analogues as inhibitors of HCV–NS3 protease. 10 Among them, MK-7009 (1), a potent inhibitor bearing a cyclopropylacylsulfonamide in P1, showed strong antiviral activ- ity in HCV infected chimpanzees and is currently being evaluated in clinical trials. 11 Previous studies, focused on product based inhibitors, have extensively investigated the impact of different carboxylic acid replacements (e.g., tetrazole, acylcyanamide, acy- lhydrazine and acylsulfonamide) 22 and led to the discovery of the cycloproprylacylsulfonamide fragment as the preferred P1 carbox- ylic acid replacement. 12 As part of our strategy to identify novel P2–P4 macrocyclic inhibitors of NS3/4A, we planned to evaluate alternative carboxylic acid bioisosters and focused specifically on phosphonates/phosphinates as potential acid replacements. These moieties have been widely studied as acid bioisosters and have been exploited to develop novel series of biologically active com- pounds such as enzyme inhibitors, 13 but very few examples are described in the specific case of HCV–NS3 protease. 14 Phosphonic acids, classified as non-planar bioisosteric surrogates of the car- boxylic acid function, 15 can be involved both in electrostatic inter- actions and hydrogen bonds. Therefore, although differences might exist between the two functional groups in terms of key interac- tions with the enzyme active site, 16 we became interested in exploring the incorporation in our class of P2–P4 macrocycles of acidic phosphorous groups in the P1 position. In this letter we report the synthesis and in vitro evaluation of a series of novel P2–P4 macrocycles bearing a phosphonic or phosphinic moiety as a replacement of the P1 carboxylic acid functionality (Fig. 2). The synthesis of the P1 phosphorous building blocks 5 and 9 is shown in Scheme 1. Commercially available diethyl phosphonate 3 was converted in two steps into diethyl-(N-benzylidenaminomethyl)-phosphonate (4). Reaction with (2E)-1,4-dibromobut-2-ene in the presence of 0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.bmcl.2009.03.038 * Corresponding author. Tel.: +39 06 91 09 32 83; fax: +39 06 91 09 36 54. E-mail address: marco_pompei@merck.com (M. Pompei). Bioorganic & Medicinal Chemistry Letters 19 (2009) 2574–2578 Contents lists available at ScienceDirect Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl