Tetrahydrobenzodifuran functionalities were employed as conformationally restricted bioisosteres of the aromatic methoxy groups in prototypical hallucinogenic phenylalkylamines 1 and 2. Thus, a series of 8-substituted 1-(2,3,6,7-tetrahydrobenzo[1,2-b:4,5-b']difuran-4-yl)-2-aminoal kanes (7a-e) were prepared and evaluated for activity in the two-lever drug discrimination paradigm in rats trained to discriminate saline from LSD tartrate (0.08 mg/kg) and for the ability to displace [3H]ketanserin from rat cortical homogenate 5-HT2A receptors and [3H]-8-OH-DPAT from rat hippocampal homogenate 5-HT1A receptors. In addition, 1-(8-bromo-2,3,6,7-tetrahydrobenzo[1,2-b:4,5-b']difuran-4-yl)-2-am inopropane (7b), which was found to be extremely potent in the rat in vivo assays, was evaluated for its ability to compete with [125I]DOI and [3H]ketanserin binding to cells expressing cloned human 5-HT2A, 5-HT2B, and 5-HT2C receptors. All of the dihydrofuranyl compounds having a hydrophobic substituent para to the alkylamine side chain had activities in both the in vitro and in vivo assays that equaled or surpassed the activity of the analogous conformationally flexible parent compounds. For example, 7b substituted for LSD in the drug discrimination assay with an ED50 of 61 nmol/kg and had Kj values in the nanomolar to subnanomolar range for the displacement of radioligand from rat and human 5-HT2 receptors, making it one of the most potent hallucinogen-like phenylalkylamine derivatives reported to date. The results suggest that the dihydrofuran rings in these new analogues effectively model the active binding conformations of the methoxy groups of the parent compounds 1 and 2. In addition, the results provide information about the topography and relative orientation of residues involved in agonist binding in the serotonin 5-HT2 receptors.

The discriminative stimulus effect of LSD in rats occurs in two phases, and these studies provide evidence that the later temporal phase is mediated by D2 dopamine receptor stimulation. A second temporal phase that involves dopaminergic pathways would be consistent with the widespread belief that excessive dopaminergic activity may be an underlying cause of paranoid psychosis.

A series of fluorinated analogues of the hallucinogenic tryptamines N,N-diethyltryptamine (DET), 4-hydroxy-N,N-dimethyltryptamine (4-OH-DMT, psilocin), and 5-methoxy-DMT was synthesized to investigate possible explanations for the inactivity of 6-fluoro-DET as a hallucinogen and to determine the effects of fluorination on the molecular recognition and activation of these compounds at serotonin receptor subtypes. The target compounds were evaluated using in vivo behavioral assays for hallucinogen-like and 5-HT(1A) agonist activity and in vitro radioligand competition assays for their affinity at 5-HT(2A), 5-HT(2C), and 5-HT(1A) receptor sites. Functional activity at the 5-HT(2A) receptor was determined for all compounds. In addition, for some compounds functional activity was determined at the 5-HT(1A) receptor. Hallucinogen-like activity, evaluated in the two-lever drug discrimination paradigm using LSD-trained rats, was attenuated or abolished for all of the fluorinated analogues. One of the tryptamines, 4-fluoro-5-methoxy-DMT (6), displayed high 5-HT(1A) agonist activity, with potency greater than that of the 5-HT(1A) agonist 8-OH-DPAT. The ED(50) of 6 in the two-lever drug discrimination paradigm using rats trained to discriminate the 5-HT(1A) agonist LY293284 was 0.17 micromol/kg, and the K(i) at [(3)H]8-OH-DPAT-labeled 5-HT(1A) receptors was 0.23 nM. The results indicate that fluorination of hallucinogenic tryptamines generally has little effect on 5-HT(2A/2C) receptor affinity or intrinsic activity. Affinity at the 5-HT(1A) receptor was reduced, however, in all but one example, and all of the compounds tested were full agonists but with reduced functional potency at this serotonin receptor subtype. The one notable exception was 4-fluoro-5-methoxy-DMT (6), which had markedly enhanced 5-HT(1A) receptor affinity and functional potency. Although it is generally considered that hallucinogenic activity results from 5-HT(2A) receptor activation, the present results suggest a possible role for involvement of the 5-HT(1A) receptor with tryptamines.

A series of conformationally restricted analogues of the hallucinogenic phenethylamine 1 (2,5-dimethoxy-4-bromophenethylamine, 2C-B) was synthesized to test several hypotheses concerning the bioactive conformation of phenethylamine ligands upon binding to the 5-HT(2A) receptor. These benzocycloalkane analogues were assayed for their receptor binding affinity and ability to activate downstream signaling pathways, and one exceptional compound was selected for testing in an in vivo drug discrimination model of hallucinogenesis. All compounds were examined in silico by virtual docking into a homology model of the 5-HT(2A) receptor. On the basis of these docking experiments, it was predicted that the R enantiomer of benzocyclobutene analogue 2 would be the most potent. Subsequent chemical resolution and X-ray crystallography confirmed this prediction, as (R)-2 proved to be equipotent to LSD in rats trained to discriminate LSD from saline. Thus, we propose that the conformation of 2 mimics the active binding conformation of the more flexible phenethylamine type hallucinogens. In addition, (R)-2 is one of the most potent and selective compounds yet discovered in the in vivo drug discrimination assay. Further, 2 was found to be a functionally selective agonist at the 5-HT(2A) receptor, having 65-fold greater potency in stimulating phosphoinositide turnover than in producing arachidonic acid release. If hallucinogenic effects are correlated with arachidonic acid production, such functionally selective 5-HT(2A) receptor agonists may lack the intoxicating properties of hallucinogens such as LSD.

A major focus of our research for a number of years has been to understand the structure-activity relationship (SAR) of classical hallucinogens and their derivatives, the pharmacological action of which is believed to be mediated primarily by agonist activity at the serotonin 5-HT 2A receptor. 2,3 In particular we have sought to understand the relationship between the modes of binding of the hallucinogens containing an indole nucleus, including LSD and psilocybin, and those containing a benzene nucleus, such as mescaline and its potent structural analogue 1 (DOB). 4 In pursuing this goal, we recently reported the synthesis of the tetrahydrobenzodifuran 2 as a rigid analogue of DOB. 5 Rat behavioral data and human receptor binding data showed that 2 is more potent than 1, a finding that was consistent with a binding site model first postulated by Westkaemper and Glennon, 6 in which two hydrogenbond-donor residues in the receptor, possibly serine residues, lie on opposite sides of the molecule and form hydrogen bonds with the ether oxygen lone pairs. It was conjectured that the increase in activity of 2 relative to 1 is due to the fact that the lone pairs in 2 are fixed in an orientation favorable for forming such hydrogen bonds.It occurred to us that a protected precursor of 2 could be aromatized in one step, leading after deprotection to the fully aromatic benzodifuran 3. We anticipated that this compound would provide valuable information about the electronic requirements for binding to 5-HT 2 receptor subtypes, since sterically it would be almost identical to 2 but electronically it would differ dramatically. Thus, 3 was synthesized and evaluated in a rat behavioral assay and in binding assays using rat and cloned human 5-HT 2A receptor preparations.Chemistry. Benzodifuran 3 was prepared as depicted in Scheme 1. The aminopropane 4, available from our previous synthetic work, 5 was protected as the trifluoroacetamide 5. Bromination with Br 2 gave 6, which was then oxidized with dichlorodicyanobenzoquinone (DDQ) in toluene, yielding the aromatized derivative 7. Deprotection produced the free amine that was neutralized with ethanolic HCl and precipitated from ether as fine white crystals of the hydrochloride salt of 3. Pharmacology. Compound 3 was evaluated in the two-lever drug discrimination assay in two groups of rats, each of which was trained to discriminate the effects of ip injections of saline from those of either LSD or DOI; the methods have been described previously. 5 Potencies were calculated as ED 50 values with 95% confidence intervals. 7 In addition, 3 was assayed for its ability to displace the highly selective 5-HT 2A antagonist [ 3 H]MDL 100,907 8 from rat prefrontal cortex and to compete for the nonselective 5-HT 2 agonist [ 125 I]DOI at cloned human 5-HT 2A and 5-HT 2C receptors and for [ 3 H]-serotonin at cloned human 5-HT 2B receptors; the methods employed here have also been described previously. 5Results and Discussion. Compound 3 possessed a K i of 0.23 ( 0.03 nM in competition with the 5-...

Lysergic acid amides were prepared from (R,R)-(-)-, (S,S)-(+)-, and cis-2,4-dimethyl azetidine. The dimethylazetidine moiety is considered here to be a rigid analogue of diethylamine, and thus, the target compounds are all conformationally constrained analogues of the potent hallucinogenic agent, N,N-diethyllysergamide, LSD-25. Pharmacological evaluation showed that (S,S)-(+)-2,4-dimethylazetidine gave a lysergamide with the highest LSD-like behavioral activity in the rat two lever drug discrimination model that was slightly more potent than LSD itself. This same diastereomer also had the highest affinity and functional potency at the rat serotonin 5-HT(2A) receptor, the presumed target for hallucinogenic agents, and a receptor affinity profile in a panel of screens that was most similar to that of LSD itself. Both cis- and the (R,R)-trans-dimethylazetidines gave lysergamides that were less potent in all relevant assays. The finding that the S,S-dimethylazetidine gave a lysergamide with pharmacology most similar to LSD indicates that the N,N-diethyl groups of LSD optimally bind when they are oriented in a conformation distinct from that observed in the solid state by X-ray crystallography. The incorporation of isomeric dialkylazetidines into other biologically active molecules may be a useful strategy to model the active conformations of dialkylamines and dialkylamides.

A method was found to synthesize 1-(2,5-dimethoxy-4-(trifluoromethyl) phenyl)-2-aminopropane, 5, and its des-alpha-methyl congener 2-(2,5-dimethoxy-4-(trifluoromethyl)phenyl)aminoethane, 6, the trifluoromethyl analogs of substituted hallucinogenic phenethylamine derivatives such as 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (3, DOI) that are potent 5-HT2A/2C agonists. In our hands, 5 and 6 have proven to have affinity for [3H]ketanserin or [125I]-3-labeled 5-HT2A/2C sites in rat cortex comparable to or higher than the analogous bromo or iodo analogs. Similarly, 5 and 6 had potency comparable to or slightly greater than that of their bromo or iodo congeners in the two-lever drug discrimination assay in rats trained to discriminate saline from LSD tartrate. The agonist properties of 5 and 6 were evaluated by measuring the accumulation of [3H]inositol monophosphate in cultured cells selectively expressing either 5-HT2A or 5-HT2C receptors. In comparison to serotonin (5-HT), compounds 3 (DOI), 5, and 6 were equally efficacious and full agonists at the 5-HT2C receptor. Similarly, 3 and 5 produced equivalent responses at the 5-HT2A receptor as compared to 5-HT. In contrast, 6, the alpha-desmethyl analog of 5, was only half as potent at stimulating inositol monophosphate accumulation at the 5-HT2A receptor. In conclusion, the title compound 5 and its alpha-desmethyl congener 6 appear to be the most potent of the so-called hallucinogenic amphetamine 5-HT agonists reported to date. Further, the reduced efficacy of 6 at the 5-HT2A receptor may offer at least a partial explanation for the observed higher in vivo potencies of alpha-methyl-substituted compounds in this series.

We have demonstrated that the behavioral effects of low doses of lisuride are clearly mediated by stimulation of 5-HT(1A) receptors.