Use of conformationally-restricted analogues and homology modeling to provide insight into the nature of the 5-HT2A receptor agonist binding site

Efforts to understand better the serotonin-2A (5-HT2A) receptor were concentrated on the characterization of the structural and chemical features of the agonist binding site. Design and synthesis of conformationally-restricted analogues of known 5-HT2A receptor agonists, coupled with a homology-based model of the 5-HT2A receptor that was developed during these studies, has enabled a more detailed understanding of the site. Conformationally-restricted analogues 6 and 7, designed to incorporate in their structure an oxygen-pattern that is atypical of hallucinogenic arylalkylamines, were synthesized and found to have Ki values for [125I]-DOI-labeled 5-HT2A cloned rat receptors of 6.3 and 1.8 nM, respectively. Next, a series of substituted tetrahydronaphthofurans was designed and synthesized in an attempt to constrain the 2-aminoalkyl substituent of arylalkylamines to an anti-periplanar orientation, hypothesized to be the physiologically relevant conformer for binding to the 5-HT2A receptor. These compounds were tested using in vitro assays, for their affinity and efficacy at selected 5-HT receptors. The benzofuran-containing analogues, 11 and 14, had significantly higher affinity for the serotonin receptors than the benzodihydrofuran-containing compounds, 9, 10, 12 and 13. The compound (8-bromo-6-methoxy-4,5-dihydro-3H-naphtho[1,8- bc]furan-5-yl)amino-methane, 14, had Ki values for [125I]-DOI-labeled 5-HT2A and 5-HT2C cloned rat receptors of 2.6 and 1.1 nM, respectively. Despite their high affinity, the compounds of this naphthofuran series were found to lack full intrinsic activity at the 5-HT2A receptor. A 5-HT2A receptor homology model was produced utilizing an in silico activated form of the rhodopsin protein as the structural core for the homology model. The computer model has offered distinct binding modes for ergolines, tryptamines, and arylalkylamines, the three main drug families that affect this receptor, that are compatible with the known SARs of these drug classes. A conformationally-restricted analogue of mescaline was designed based on analysis of the computer model. The resulting compound 17, (2,3-dihydro-4,5,6-trimethoxy-1H-inden-1-yl)aminomethane, was found to possess 3-fold higher affinity and equal efficacy at the 5-HT2A receptor when compared to the parent compound, mescaline. This compound, 17, was also tested in a drug discrimination paradigm and found to substitute fully for LSD, thus validating the predictive power of the receptor model