A high throughput screening campaign of the Lilly

A high-throughput screening campaign of the Lilly corporate compound collection led to the discovery of as potent and selective inhibitor of GlyT1 (IC=45nM, GlyT2 IC>30,000nM). In addition to the characteristic -methyl glycine moiety as acidic head group and the novel ethanolamine linker, also contained a 2-acyl biphenyl system which was the focus of our initial SAR studies (). Herein we will report on the impact of changes in Domains 1 and 2 on glycine transporter binding affinity and selectivity. Our synthetic strategy was based on the use of advanced intermediates from which the targets should be accessible in one or two steps using robust synthetic protocols amenable to a rapid parallel synthesis (RPS) approach (). We envisaged the use of cross-coupling chemistry for the preparation of analogues containing a biphenyl system either (Domain 2) or (Domain 1) to the ethanolamine linker. A Weinreb amide intermediate could be used for the introduction of different acyl substituents (addition elimination). To allow efficient exploration of a variety of aromatic substituents in Domain 2 of hit we used -bromo analogue as late stage intermediate suitable for a rapid parallel synthesis approach (). Readily accessible phenol was coupled with [(2-hydroxyethyl)-methylamino]acetic Cy7 carboxylic acid (non-sulfonated) -butyl ester under modified Mitsunobu conditions using 1,1′-(azodicarbonyl)dipiperidine (ADDP) to give . Introduction of the -aryl group was achieved following the Suzuki–Miyaura protocol. Ester hydrolysis under acidic or basic conditions gave the final compounds. Both the coupling and the deprotection step were carried out in a parallel fashion. Replacement of the thiophene group in by phenyl and cyclohexyl was achieved using the aforementioned ADDP-mediated coupling of the requisite phenol with the -butyl protected side chain. The phenol coupling partners were prepared following established literature procedures. Changes to the acyl group in Domain 1 required a variety of different chemical approaches (). Reduction of the carbonyl precursors, and , respectively, using triethylsilane and trifluoroacetic acid provided convenient access to the methylene linked phenyl and cyclohexyl analogues and . Attempted reduction of the carbonyl group in to access under similar conditions led to decomposition of the starting material. Instead, Wolff–Kishner reduction of phenol followed by ADDP-mediated coupling and Suzuki–Miyaura coupling gave . In Domain 1, a biaryl system to the ethanolamine linker was introduced by Pd-mediated cross-coupling methodology employing advanced aryliodide intermediate and a variety of aryl and heteroaryl boronic acids (). In Domain 2, analogues containing the preferred 2-chloro-phenyl motif were obtained according to . Suzuki–Miyaura coupling of commercially available 2-iodo anisole gave followed by regioselective bromination, demethylation and ADDP-mediated coupling gave advanced intermediate for the preparation of final targets by RPS methods. In the 2-thiophene series Domain 2 analogues were prepared in a similar fashion from advanced intermediate . Glycine reuptake inhibition was determined in whole cell assays for both GlyT1 and GlyT2. All members of this series had no significant activity at the GlyT2 (IC>30μM). The effects of changes in Domain 2 while maintaining the thiophenacyl moiety in Domain 1 are summarized in . Both 2-chlorophenyl ( and 2-methylphenyl () caused a significant decrease in activity, suggesting an important role of the biphenyl system orientation for maintaining potency. In general, electron-releasing groups appeared to be preferred as Domain 2 substituents. Combination of a 3-methyl and 4-methoxy group as in resulted in increased potency. This motif was found to be preferred over 3,4-dialkyl substitution as in . Interestingly, 3,4-dimethoxy analogue was significantly less active, whereas the methylenedioxy analogue was more potent than the original hit . Replacement of Domain 2 by bulky non-aromatic groups as in led to a significant drop in activity.