Identification of a New Scaffold for Opioid Receptor Antagonism Based on the 2-Amino-1,1-dimethyl-7-hydroxytetralin Pharmacophore Peter Grundt, Ian A. Williams, John W. Lewis, and Stephen M. Husbands* Department of Pharmacy and Pharmacology, University of Bath, Bath, BA2 7AY, U.K. Received March 9, 2004 The trans-(3,4)-dimethyl-4-(3-hydroxyphenyl)piperidines are a unique class of opioid antagonists that have recently provided selective antagonists for µ-opioid receptors (MOR) and κ-opioid receptors (KOR). Molecular modeling indicated a strong structural similarity between the parent of this series and 2-amino-1,1-dimethyl-7-hydroxytetralin. In binding and in vitro functional assays, the aminotetralin derivatives displayed some overlap in SAR with that previously reported for the phenylpiperidine series, providing evidence for a common binding mode for the two series at opioid receptors. Introduction of a methoxy group in the 3-position increased potency at MOR and KOR receptors, suggesting that this aminotetralin skeleton can be utilized as a new scaffold for the design of selective opioid receptor antagonists. Introduction There has been considerable interest over many years in the development of selective ligands with which to study the function of opioid receptors. 1,2 Significant advances have been made with selective agonists and antagonists available for each of the three opioid recep- tors (µ, MOR; δ, DOR; κ, KOR). 1,3-5 Portoghese has developed both KOR- and DOR-selective antagonists by applying the message-address concept of Schwyzer to the opioid antagonist naltrexone (1, Chart 1), which itself is slightly MOR-selective. 6 The prototypic KOR- antagonist, developed in this way, is norBNI (2). 7 Portoghese has since shown that the large bimorphinan structure of 2 can be significantly simplified while retaining KOR selectivity and antagonist potency. This has ultimately led to the development of GNTI (3). 8 The undoubted success of this approach means a large number of KOR and DOR antagonists have been synthesized on the basis of the oxymorphone framework. It is now desirable for the range of scaffolds to be increased because this could provide ligands with, for example, differing pharmacokinetic and pharmacody- namic properties or differing receptor subtype selectiv- ity. In this regard, the provision of KOR antagonists that lack the extremely long duration of activity of 2 and 3 would be of particular value. Carroll and co-workers have been successful in de- veloping selective MOR and KOR antagonists based on an alternative to the oxymorphone framework, trans- (3,4)-dimethyl-4-(3-hydroxyphenyl)piperidines, which pro- duced a unique class of opioid antagonists, discovered by Zimmerman. 9-11 It was shown that modification of the N-substituent provided a means to control both the selectivity and potency of the ligands without introduc- ing efficacy. By this approach, the selective MOR antagonist (4a) 10 and the highly selective KOR antago- nist JDTic (4b) 11 were discovered. In these cases the phenylpiperidine unit acts as the message while the cinnamyl phenyl group and 7-hydroxyterahydroisoquin- oline moieties provide the MOR and KOR address components, respectively. We recognized the structural equivalence of the 14- amino group in the morphinone series (5) 12 and the piperidine basic center in the phenylpiperidine series (4) in their spatial relationships to the respective phenolic binding centers. The 14-alkylaminomorphi- none series (5a), particularly those members with a side chain terminal aryl group, consistently provides potent opioid receptor antagonism with only low-level agonist activity. This SAR is characteristic also of the phenylpi- peridine series (4). Though we investigated the 14- cinnamylaminomorphinone (5a) and showed it to have potent noncompetitive MOR antagonist activity, 12 we were not immediately attracted to further investigation of the 14-amino skeleton as a new message scaffold, particularly for introducing the address component for KOR selectivity because its synthesis from thebaine is multistep and very low yielding. 13 We considered the simpler but closely related aminobenzomorphan struc- ture (6), but that too failed our criterion of ready accessibility. 14 We therefore turned to an even simpler structure, 2-amino-1,1-dimethyl-7-hydroxytetralin (7). The 2-ami- notetralin skeleton has previously been employed in the development of opioid analgesics. 15-17 A primary amino group was required for good in vivo analgesic activity with dezocine (8) the most well characterized example. 18 Because of their reduced analgesic potency, little atten- * To whom correspondence should be addressed. Phone: 44-1225- 383103. Fax: 44-1225-386114. E-mail: s.m.husbands@bath.ac.uk. Figure 1. Overlay of 4a (green) with 9k (red). 5069 J. Med. Chem. 2004, 47, 5069-5075 10.1021/jm040807s CCC: $27.50 © 2004 American Chemical Society Published on Web 09/08/2004