10-Substituted 11-Oxygenated (R)-Aporphines:  Synthesis, Pharmacology, and Modeling of 5-HT_1A Receptor Interactions

Abstract

Derivatives of the selective serotonin 5-HT1A receptor agonist (R)-11-hydroxy-10-methylaporphine (2) having various substituents in the C10-position or at the nitrogen have been synthesized from natural morphine or 6-O-acetylcodeine, respectively. The C10-substituents were introduced using efficient Stille or Suzuki cross-coupling reactions. The compounds were evaluated for their affinities to 5-HT1A and dopamine (DA) D1 and D2A receptors in vitro. All compounds tested displayed low (micromolar) affinities to D1 and D2A receptors. In addition, changes in steric bulk and/or electronic properties of the C10-substituent as compared to a C10-methyl group, as well as substitution of the N-methyl group for a hydrogen or a larger N-alkyl group, produced a marked decrease in the affinities to 5-HT1A receptors. Selected compounds that displayed moderate to high affinities to 5-HT1A receptors were evaluated for their ability to stimulate 5-HT1A receptors in vivo. The evaluated compounds behaved as agonists at 5-HT1A receptors, except for the N-propyl analogue of 2, (R)-11-hydroxy-10-methyl-N-propylnoraporphine (23), which displayed weak DA receptor agonism at the doses tested. Hence, the substitution pattern of 2 (a C10-methyl, a C11-hydroxy, and an N-methyl group) appears to be optimal for potent interaction of 10,11-disubstituted (R)-aporphines with 5-HT1A receptors. Modeling of ligand-5-HT1A receptor interactions was performed in an attempt to rationalize the observed affinity data. The binding site model suggests the presence of a "methyl pocket" in the 5-HT1A receptor binding ste. The C11-methoxy-substituted aporphines appear to have a different binding mode compared to 2, implying a different accessibility of these compounds to the "methyl pocket".

Molecular Design Using the Minireceptor Concept

“…40,41 This minireceptor model agreed well with the previously described homology-based model, 39 the main difference being the use of Ser168 instead of Ser199 to hydrogen bond to the C11-oxygenated ligands. Ser199 is still close enough to interact with 4 but has to be rotated in order to make room for the larger C11 substituents.…”

“…It was shown in the previous binding site modeling of the aporphines that the space available for N-substituents is limited. 40,41 However, the aromatic substituent in i found a small pocket lined by several aromatic residues. Both ligands i and j interact with Asp116 and Ser168, while ligand j in addition partly occupies the phenyl pocket.…”

“…This model was refined by minireceptor optimization (Yak version 3.9) applied to the isolated active site in a stepwise manner using high-affinity ligands 3-5 in order of increasing size. 40 The receptor site was extracted from the homology-based model by deleting all residues further than 7.5 Å from any of the bound ligands. Only those receptor site residues which are within 5 Å of the ligands were optimized with Yak.…”

Synthesis of (R)‐(–)‐2‐Fluoronorapomorphine — A Precursor for the Synthesis of (R)‐(–)‐2‐Fluoro‐N‐[11C]propylnorapomorphine for Evaluation as a Dopamine D2 Agonist Ligand for PET Investigations

“…[20] Thus, codeine was O-acetylated followed by N-carbamoylation/N-demethylation by the use of α-chloroethyl chloroformate (ACE-Cl). [21] The resultant carbaScheme 3.…”

“…The secondary amine 9 was then selectively N-alkylated with benzyl bromide in the presence of K 2 CO 3 in DMF thus providing 7 which contains a convenient N-protecting group removable at a late stage in the synthesis. [20] Oxidation of 7 (Scheme 4) to ketone 6 was accomplished by the Swern protocol [DMSO, (COCl) 2 ]. Attempts to oxidise 7 by other methods such as TEMPO/re-oxidants [22,23] or activated MnO 2 [24] failed.…”