Etoxadrol-meta-isothiocyanate: a potent, enantioselective, electrophilic affinity ligand for the phencyclidine-binding site.

Etoxadrol-meta-isothiocyanate (2S,4S,6S-2-ethyl-2-(3-isothiocyanatophenyl)-2-piperidyl)1,3-dioxolane, 4a) has been synthesized and characterized as an irreversible ligand for the phencyclidine (PCP)-binding site. It is the first chiral electrophilic affinity ligand for this site to have been described. This affinity ligand is based upon etoxadrol, a 1,3-dioxolane known to have PCP-like effects in vivo and in vitro. Etoxadrol-meta-isothiocyanate was found to be four-five times more potent in vitro than metaphit (1-[1-(3- isothiocyanatophenyl)cyclohexyl]piperidine), the only previously known electrophilic affinity ligand for the PCP-binding site. The binding was shown to be highly enantioselective for etoxadrol-meta-isothiocyanate (4a). The 2R,4R,6R-enantiomer of 4a was essentially inactive. The ability of the 2S,4S,6S-enantiomer (4a) to interact with the benzodiazepine, muscarinic, and mu opioid receptor systems was also examined, and it was found not to interact with these receptor systems. It seems likely that 4a will prove to be a valuable tool in the study of structure and function of the PCP-binding site.

Excitatory amino acid antagonists induce a phencyclidine-like catalepsy in pigeons: structure-activity studies.

The excitatory amino acid antagonists D,L-2-amino-5-phosphonovalerate (D,L-AP5), its isomers D-(-)-AP5 and L-(+)-AP5, D,L-2-amino-4-phosphonobutyrate (AP4), D,L-2-amino-7-phosphonoheptanoate (AP7), beta-D-aspartylaminomethylphosphonic acid (ASP-AMP), cis-2,3-piperidinedicarboxylic acid (cis-PDA), and gamma-D-glutamylaminomethylsulphonic acid (GAMS) were tested for their ability to produce a phencyclidine (PCP)-like catalepsy in pigeons when administered intracerebroventricularly. Each of the antagonists produced catalepsy, although L-AP5, and the non-selective antagonists GAMS and cis-PDA, produced the effect only at toxic doses. The rank order of potency to produce catalepsy was AP7 greater than D-AP5 greater than D,L-AP5 greater than cis-PDA greater than ASP-AMP greater than AP4 greater than L-AP5 greater than GAMS; there was a strong positive correlation between this rank order of potency in vivo and the potency order of these compounds in vitro as NMDA antagonists. The antagonists did not displace significant amounts of [3H]N-[1-(2-thienyl)cyclohexyl]piperidine (a congener of phencyclidine) from its recognition site in the brain of pigeon. Thus, the PCP-like catalepsy that is produced by the excitatory neurotransmission at NMDA-preferring receptors that are distinct from, but related to, PCP receptors. The results strongly support the hypothesis that a reduction of neurotransmission at excitatory synapses, utilizing NMDA-preferring receptors, may underlie catalepsy in pigeons induced by PCP.

Specificity of phencyclidine-like drugs and benzomorphan opiates for two high affinity phencyclidine binding sites in guinea pig brain.

Recently, the presence of two high affinity binding sites for phencyclidine were described in guinea pig brain, with one site coupled to the glutamate excitatory amino acid receptor, specifically activated by N-methyl-D-aspartate (NMDA) (site 1) and the other site associated with the dopamine (DA) reuptake carrier (site 2). Phencyclidine and its analogs, as well as the benzomorphan opiates, are known to interact with binding sites for phencyclidine. In this study, the equilibrium dissociation constants (Kd) of these compounds for the two binding sites for phencyclidine were determined. Phencyclidine and 1-[1-(2-thienyl)cyclohexyl]piperidine (TCP), an analog of PCP, were essentially non-selective between the two sites and also were the two drugs of the group observed to have the highest affinity for site 2. (+)-5-Methyl-10,11-dihydro-5H-dibenzo[a,d]cycloheptene-5,10-imine [(+)MK801] was the most selective agent for site 1, while none of the drugs tested showed selectivity for site 2. In humans, phencyclidine produces psychotomimetic effects, while (+)MK801 has been reported to produce minimal, if any, psychotomimetic effects, at doses sufficient to reduce seizures. These clinical observations, in conjunction with the present biochemical binding data, suggest that (+)MK801 may serve as a "marker" for site 1 and that the psychotomimetic effects of phencyclidine might be mediated by site 2.

Further studies of the structure-activity relationships of 1-[1-(2-benzo[b]thienyl)cyclohexyl]piperidine. Synthesis and evaluation of 1-(2-benzo[b]thienyl)-N,N-dialkylcyclohexylamines at dopamine uptake and phencyclidine binding sites.

We previously reported (J. Med. Chem. 1993, 36, 1188-1193) that changes to the ring size of the piperidine and cyclohexyl rings of the high-affinity and selective dopamine (DA)-uptake inhibitor 1-[1-(2-benzo[b]thienyl)cyclohexyl]piperidine (BTCP, 2) caused different, and in some cases opposite, changes in affinity for sites on the DA transporter labeled by [3H]BTCP and [3H]-cocaine. These results suggested that the radioligands label different sites on the transporter. In the present study, we extend the structure-activity relationships (SAR) of BTCP by studying the binding characteristics of a series of N,N-disubstituted 1-(2-benzo[b]thienyl)cyclohexylamines 7-32 at the DA transporter. Cyclohexyl was selected as opposed to other ring sizes since it corresponds to BTCP. The binding results indicate that a considerable degree of structural variation is permitted for the N-substituents, while still retaining nanomolar affinity for sites on the transporter (studied in rat forebrain homogenates). As observed in our earlier study, the differential effects of structural change on binding to sites on the DA transporter labeled by these radioligands suggests that they are different and distinct binding sites. In general, and up to a point, increasing the size and lipophilicity of the N substituents resulted in improvements in binding but appeared to have less predictable effects on DA-uptake inhibition (as measured in rat brain synaptosomes). The binding of these compounds to sites labeled by [3H]BTCP appeared to correlate best with IC50 for DA-uptake inhibition. To our surprise, the monoalkyl N-substituted BTCP derivatives displayed the highest affinity for the DA transporter of all the compounds in this series. For example, the N-(cyclopropylmethyl) derivative 14 displayed IC50's = 23 nM ([3H]cocaine) and 1 nM ([3H]-BTCP), and the N-butyl derivative 10 showed IC50's = 60 nM ([3H]cocaine) and 0.3 nM ([3H]-BTCP). BTCP exhibited IC50's of 39 nM ([3H]cocaine) and 5 nM ([3H]BTCP) in this assay. The observation that N,N-dibutyl derivative 31 exhibited low ratios of IC50 [3H]cocaine/IC50 DA reuptake and IC50 [3H]BTCP/IC50 DA reuptake suggests that it may be a potential candidate for cocaine antagonism studies. The effect of additional amino, amide, and aromatic groups on the N-substituents was examined, and the results are discussed. The failure of all of the compounds in this series to bind phenycyclidine receptors coupled with their high affinity and range of selectivities at the DA transporter identifies many of them as useful tools for probing the mode of action of BTCP at this site.

Synthesis and biological evaluation of 1-[1-(2-benzo[b]thienyl)cyclohexyl]piperidine homologues at dopamine-uptake and phencyclidine- and sigma-binding sites.

Piperidine and cyclohexyl ring homologues of the high-affinity dopamine (DA) uptake inhibitor 1-[1-(2-benzo[b]thienyl)cyclohexyl]piperidine (BTCP, 3) were each prepared in four steps from the appropriate cycloalkanones. These compounds were tested for their ability to displace [3H]BTCP and [3H]cocaine and to inhibit [3H]DA uptake in rat striatal homogenates. The ratios IC50([3H]cocaine)/IC50([3H]BTCP) ranged from 62 for BTCP to 1.5 for 1-[2-(benzo[b]thienyl)-cyclopentylamine (17); cocaine gave a ratio of 0.6. This indicates that BTCP is the most selective of all the compounds tested for sites labeled by [3H]BTCP whereas cocaine is most selective for sites labeled by [3H]cocaine. The wide differences in the relative abilities of these compounds to displace [3H]BTCP and [3H]cocaine suggests that these two radioligands are labeling different sites on the transporter. In general, the compounds structurally related to BTCP exhibited greater selectivity for sites labeled by [3H]BTCP. However, several of the BTCP-related derivatives showed greater (compared with BTCP and cocaine) ability to displace [3H]cocaine. Most notably, 1-[1-(2-benzo[b]thienyl)cyclohexyl]pyrrolidine (7) exhibited a 3.4-fold greater affinity for these sites compared with BTCP and a 9-fold greater affinity at these sites than cocaine. Most of the BTCP homologues displayed greater ability to inhibit [3H]DA uptake in rat forebrain synaptosomes than cocaine. BTCP and 7 were the most potent of all the compounds tested in terms of their ability to inhibit uptake of [3H]DA. IC50 ratios for [3H]cocaine binding/[3H]DA uptake ranged from 0.47 for 1-[1-(2-benzo[b]thienyl)cyclopentyl]homopiperidine (11) to 8.8 for 1-(2-benzo[b]thienyl)cyclohexylamine (4). The importance of this ratio remains unclear in terms of identification of potential cocaine antagonists. As for BTCP, all of the compounds tested showed Ki values > 10,000 nM for displacement of [3H]TCP from rat brain homogenates. These compounds were able to displace the highly selective sigma receptor probe [3H]-(+)-pentazocine from guinea pig brain homogenates with Ki values ranging from 125 to 9170 nM. The significance of their sigma-binding activity in light of their dopaminergic properties is unclear. The diverse binding properties of these compounds at the DA-uptake site and their spectrum of inhibitory activities for [3H]DA uptake identifies them as a useful base for the development of subtype selective probes at this site. These compounds will allow further study of the structure and function of the "cocaine" receptor as well as the development of potential cocaine antagonists.

Synthesis, stereochemistry, and biological activity of the 1-(1-phenyl-2-methylcyclohexyl)piperidines and the 1-(1-phenyl-4-methylcyclohexyl)piperidines. Absolute configuration of the potent trans-(-)-1-(1-phenyl-2-methylcyclohexyl)piperidine.

The (-)- and (+)-isomers of the cis- and trans-Ph/Me 1-(1-phenyl-2-methylcyclohexyl)piperidines have been synthesized and the achiral cis- and trans-Ph/Me 1-(1-phenyl-4-methylcyclohexyl)piperidines were prepared, and their in vitro [displacement of [3H]TCP (1-[1-(2-thienylcyclohexyl)]piperidine) from the PCP (1-(1-phenylcyclohexyl)piperidine) binding site] and in vivo (rotarod assay) activities determined. The 1-(1-phenyl-2-methylcyclohexyl)piperidine isomers were resolved by classical crystallization procedures, through the diastereomeric salts obtained with d- and l-10-camphorsulfonic acid. The relative stereochemistry of the cis- and trans-Ph/Me 1-(1-phenyl-2-methylcyclohexyl)piperidines and the achiral cis- and trans-Ph/Me 1-(1-phenyl-4-methylcyclohexyl)piperidines was established by using 13C and 1H NMR. Both (-)-trans-1-(1-phenyl-2-methylcyclohexyl)piperidine ((-)-2) and (+)-trans-1-(1-phenyl-2-methylcyclohexyl)piperidine ((+)-2) were examined by single-crystal X-ray analysis, and the absolute configuration of (-)-2 was determined to be 1S,2R. The (-)-2 was found to be about five times more potent than PCP in vitro and twice as potent in vivo. It is the most potent of all of the simple methyl-substituted cyclohexyl PCP isomers and is among the most potent PCP-like compounds which have been synthesized. It was nine times more potent in vitro and four times more potent in vivo than (+)-2. The racemic cis-1-(1-phenyl-2-methylcyclohexyl)piperidine (3), and its enantiomers ((+)-3 and (-)-3), were essentially inactive in vitro and in vivo. The cis-Ph/Me 1-(1-phenyl-4-methylcyclohexyl)piperidine (18) was more potent than trans-Ph/Me 1-(1-phenyl-4-methylcyclohexyl)piperidine (17), but considerably less potent than (-)-2. The enantioselectivity observed at the PCP binding site for (-)-2 could indicate that this site can discriminate between enantiotopic edges of the achiral PCP (choosing the pro-1-S edge), as does the mu-opioid receptor in the prodine series of opioids. Benzimidoyl or benzoyl group replacement of the phenyl ring in the 1-(1-phenyl-2-methylcyclohexyl)piperidine series gave compounds which showed little in vitro and in vivo activity.

Rigid phencyclidine analogues. Binding to the phencyclidine and sigma 1 receptors.

Three phencyclidine (PCP) analogues possessing a highly rigid carbocyclic structure and an attached piperidine ring which is free to rotate were synthesized. Each analogue has a specific fixed orientation of the ammonium center of the piperidinium ring to the centrum of the phenyl ring. The binding affinities of the rigid analogues 1-piperidino-7,8-benzobicyclo[4.2.0]octene (14), 1-piperidinobenzobicyclo[2.2.1]heptene (16), and 1-piperidinobenzobicyclo[2.2.2]octene (13) for the PCP receptor ([3H]TCP) and th-receptor (NANM) were determined. The three analogues show low to no affinity for the PCP receptor but good affinity for the th-receptor and can be considered th-receptor selective ligands with PCP/th ratios of 13, 293, and 368, respectively. The binding affinities for the th-receptor are rationalized in terms of a model for the th-pharmacophore.

Synthesis, pharmacological action, and receptor binding affinity of the enantiomeric 1-(1-phenyl-3-methylcyclohexyl)piperidines.

The cis and trans enantiomers of the 1-(1-methylcyclohexyl)piperidines were prepared from either 3(R)- or 3(S)-methylcyclohexanone through the Bruylents reaction or a modified azide route, respectively. Separation of the intermediate amines 5 and 6 was achieved through chromatography or selective crystallization of a fumarate salt. The cis isomer 2b had about one-third of the affinity of phencyclidine for the PCP receptor. The other isomers were less potent. There was a 40-fold difference between the binding affinity of the cis enantiomers 2a and 2b and a fourfold difference between the affinities of the trans enantiomers 1a and 1b. None of the compounds antagonized the stereotypy induced by phencyclidine in the rotorod assay in mice, after intraperitoneal introduction.

Synthesis and structure-activity relationship of C5-substituted analogues of (+-)-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine [(+-)-desmethyl-MK801]: ligands for the NMDA receptor-coupled phencyclidine binding site.

A series of eight C5-substituted analogues of (+-)-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine (1) have been prepared by the directed lithiation-alkylation (and acylation) of its (+-)-N-tert-butylformamidinyl derivative 2 followed by formamidine solvolysis. An additional 10 analogues were prepared by elaboration of the C5-ethyl ester derivative. Analogues possessing large (e.g. propyl and larger) lipophilic substituents displace [3H]-1-(1-thienylcyclohexyl)piperidine [( 3H]TCP) from the high-affinity phencyclidine (PCP) binding site in rat brain homogenates only at high concentrations (Ki greater than 1000 nM); however, the presence of a polar amino functionality (e.g. 2-aminoethyl) offsets this effect (Ki = 20 nM). Thus, the boundary condition for lipophilic substituents larger than ethyl appears to be polar in nature. Interaction of the 11 relatively small (MR less than 14) C5-substituted analogues of 1 with the high-affinity PCP binding site associated with the N-methyl-D-aspartate (NMDA) receptor is best described by the equation log (1/Ki) = -5.83F + 0.64 pi + 7.41 (r = 0.90).

Synthesis, configuration, and activity of isomeric 2-phenyl-2-(N-piperidinyl)bicyclo[3.1.0]hexanes at phencyclidine and sigma binding sites.

The novel semirigid derivatives (+)-cis-1-[2-phenyl-2-bicyclo[3.1.0]hexyl]piperidine [(+)-8], its enantiomer (-)-8, and (+-)-trans-1-[2-phenyl-2-bicyclo[3.1.0]hexyl]piperidine [(+/-)-9] were synthesized as probes to investigate the mode of interaction of phencyclidine (PCP) with its binding site on the N-methyl-D-aspartate receptor complex. Each target compound was obtained in five steps starting from cyclopent-2-enone. (+)- and (-)-8 were obtained in greater than 98% optical purity through three recrystallizations from ethanol of the (S)-(+)- and (R)-(-)-mandelate salts of intermediate (+-)-cis-2-phenyl-2-bicyclo[3.1.0]hexylamine ([(+/-)-16]. Crystallization of the (R)-(-)-mandelate salt afforded (1R,2R,5S)-(-)-16, whereas the (S)-(+)-mandelate salt afforded (1S,2S,5R)-(+)-16; the absolute configuration was determined by single-crystal X-ray analysis of (-)-16.(R)-(-)-mandelate. Single-crystal X-ray analysis of (+/-)-9-picrate confirmed its trans configuration and provided conformational data. (+)- and (-)-8 and (+/-)-9 were examined for their ability to interact with PCP and sigma binding sites in vitro using [3H]TCP and [3H]pentazocine as radioligands. The binding was compared with that of PCP and contrasted with the rigid symmetrical phencyclidine derivatives cis- and trans-1-[3-phenyl-3-bicyclo[3.1.0]hexyl]piperidines (6 and 7). The results of the study indicated that the conformations of PCP represented by 6-9 are not optimal for potent interaction at either of these sites. Affinities ranged from 582 nM [(+/-)-9] to 29,000 nM [(+)-8] at PCP binding sites and from 1130 nM [(-)-8] to 16,300 nM (7) at sigma sites. In this assay, PCP exhibited affinities of 64.5 nM at PCP and 1090 nM at sigma sites. Qualitative correlation between the sigma and PCP binding data suggests some similarities between these binding sites. An axial phenyl and equatorial piperidine ring with the nitrogen lone pair of electrons antiperiplanar to the phenyl ring has been postulated as the receptor-active conformation of PCP-like ligands at the PCP binding site. Comparison of the binding data of 7-9 with that of the previously described methylcyclohexyl-PCP derivatives allowed its rationalization in terms of this model. It is likely that the lowered affinity in this bicyclo[3.1.0]hexane series is a consequence of nonoptimal geometry (pseudoequatorial phenyl or pseudoboat) for binding as opposed to the presence of steric bulk which proved deleterious in the methylcyclohexyl-PCP derivatives.

A marked change of receptor affinity of the 2-methyl-5-(3-hydroxyphenyl)morphans upon attachment of an (E)-8-benzylidene moiety: synthesis and evaluation of a new class of sigma receptor ligands.

The (E)-8-benzylidene and (E)-8-(3,4-dichlorobenzylidene), 7-ketone derivatives, 5 and 6, of the synthetic opiate 2-methyl-5-(3-hydroxyphenyl)morphan [5-(3-hydroxyphenyl)-2-methyl-2-azabicyclo[3.3.1]nonane, 1], were synthesized from the 7-ketone derivatives 2 or 4 via the Claisen-Schmidt reaction. The corresponding enantiomers of 5 and 6 were obtained in > 99% optical purity from the optical isomers of 4, resolved with the O,O'-dibenzoyltartaric acids. The absolute configurations of the enantiomers of 4 were determined by conversion, via Clemmensen reduction, to the enantiomers of 1, the configurations of which are known. The determination of the regioisomer and configurational isomer of 5, with respect to the introduced benzylidene group, was determined from a single-crystal X-ray analysis. 1H NMR data was used to confirm that 6 possessed the same configuration as 5. Radioreceptor binding studies in rat and guinea pig brain preparations revealed that (-)-(1S,5S)-5 displayed an 11-fold decrease in affinity for the opioid mu receptor and an increase in affinity for sigma receptors of 81-fold (low nanomolar affinity) relative to the ketone precursor (+)-(1S,5S)-4. An analogous, albeit less dramatic, trend was seen with compound (-)-(1S,5S)-6. Compounds (-)-(1S,5S)-5 and (-)-(1S,5S)-6 are distinct from the typical sigma-opiates in that they have very low affinity for either PCP sites or muscarinic receptors. The high affinity and selectivity of these novel sigma receptor ligands suggests that they will be valuable for the elucidation of the functional roles of sigma receptors.

Antipodal alpha-N-(methyl through decyl)-N-normetazocines (5,9 alpha-dimethyl-2'-hydroxy-6,7-benzomorphans): in vitro and in vivo properties.

The enantiomeric (-)- and (+)-N-(methyl through decyl) normetazocines (5,9 alpha-dimethyl-2'-hydroxy-6,7-benzomorphans) were synthesized and their in vitro and in vivo activities determined. Increasingly bulky enantiomeric N-alkyl homologs were prepared until their interaction with the sigma 1 receptor decreased and their insolubility became a hindrance to their evaluation in vivo and/or in vitro. The (-)-methyl, -pentyl, -hexyl, and -heptyl homologs were essentially as potent as, or more potent than, morphine in the tail-flick, phenylquinone, and hot-plate assays for antinociceptive activity; the (-)-propyl homolog had narcotic antagonist activity between that of nalorphine and naloxone in the tail-flick vs morphine assay, and it also displayed antagonist properties in the single-dose suppression assay in the rhesus monkey. The antinociceptively potent (-)-heptyl homolog did not substitute for morphine in monkeys but did show morphine-like properties in a primary physical-dependence study in continuously infused rats. All five potent compounds showed high affinity for the mu opioid receptor from both rat and monkey preparations and the kappa opioid receptor (< 0.05 microM), and all except the (-)-methyl homolog interacted reasonably well at the delta receptor (K(i) < 0.1 microM). The (-)-propyl compound was equipotent (K(i) 1.5-2.0 nM) at mu and kappa receptors. The pattern of interaction of the (-)-enantiomeric homologs with mu receptors from rat and monkey preparations was similar, but not identical. The enantioselectivity of the homologs for mu receptors was greater in the rat than in the monkey preparation for all but the N-H and butyl compounds, and the enantioselectivity of the lower homologs (methyl through butyl) for the mu (monkey) receptor was greater than for the kappa or delta receptors. However, bulkier homologs (hexyl through decyl) displayed higher enantioselectivity at kappa or delta receptors than at the mu (monkey) receptor. The (+)-butyl through (+)-octyl homologs were essentially equipotent with, or more potent than, (+)-pentazocine at the sigma receptor. Only the (+)-H and (+)-methyl homologs had high affinity (< 0.05 microM) at PCP binding sites.

Synthesis and binding properties of MK-801 isothiocyanates; (+)-3-isothiocyanato-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten- 5,10-imine hydrochloride: a new, potent and selective electrophilic affinity ligand for the NMDA receptor-coupled phencyclidine binding site.

Three new site-directed irreversible (wash-resistant) ligands for the high-affinity phencyclidine (PCP) binding site associated with the N-methyl-D-aspartate (NMDA) receptor were synthesized and their binding characteristics were studied. (+)-3- And (+)-2-isothiocyanato-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cycl ohepten-5,10 - imine hydrochloride ((+)-8a,b.HCl) were prepared in four steps from the corresponding nitro derivatives (+)-4a,b, which were obtained by nitration of (+)-3 (MK-801). In the same way the optical antipode (-)-8a.HCl was synthesized from (-)-3. At a concentration of 100 nM, the 3-isothiocyanate derivative (+)-8a irreversibly labeled approximately 50% of the (+)-[3H]-3 binding sites, compared to 20 microM needed for its optical antipode (-)-8a and the 2-isothiocyanate (+)-8b. The apparent Ki values for reversible inhibition of (+)-[3H]-3 binding by (+)- and (-)-8a and (+)-8b were 37,838, and 843 nM, respectively. In contrast, metaphit (1b) and etoxadrol m-isothiocyanate (2b), two previously reported irreversible ligands for the PCP binding site, label about 50% of the (+)-[3H]-3 binding sites at 100 microM and 250 nM, respectively, with apparent Ki values for reversible inhibition of 535 and 94 nM. Compound (+)-8a is also a selective affinity ligand, displaying little or no irreversible in vitro affinity at 100 microM for opioid, benzodiazepine, muscarinic, and dopamine receptors. At a 25 microM concentration, (+)-8a caused an irreversible 52% reduction of binding to sigma 1-receptors. Compound (+)-8a is the most potent known electrophilic affinity label for the PCP binding site. Its potency and selectivity should enable it to be a valuable tool for the elucidation of the structure and function of the NMDA receptor-associated PCP binding site in the mammalian central nervous system.

Novel 1-phenylcycloalkanecarboxylic acid derivatives are potent and selective sigma 1 ligands.

Carbetapentane (1, 2-[2-(diethylamino)ethoxy]ethyl 1-phenyl-1-cyclopentanecarboxylate) binds with high affinity to sigma sites and is a potent antitussive, anticonvulsant, and spasmolytic agent. However, carbetapentane interacts at muscarinic binding sites as well, and it is not clear whether either of these receptor systems is involved in the mechanism(s) of action(s) of this drug. In an attempt to determine whether these psychoactivities can be attributed to interaction at sigma sites, a series of carbetapentane analogs were prepared. Phenyl ring substitution; contraction, expansion, and replacement with a methyl group of the cyclopentyl ring; replacement of the carboxylate function with an amide, methyl ether, and methylamine; and replacement of the N,N-diethyl substituent with a morpholino or piperidino moiety were investigated. All of these novel analogs were evaluated for binding to sigma 1 and sigma 2 sites, and comparison of binding at muscarinic m1 and m2 and PCP (1-(1-phenylcyclohexyl)piperidine) receptors was performed. All of the compounds were selective for sigma 1 over sigma 2 sites, with the three most selective analogs being compounds 34 (65-fold), 35 (78-fold), and 39 (51-fold). None of the compounds were active at PCP sites, and chemical modification including (1) replacing the ester function, (2) replacing the cyclopentyl ring with a smaller ring system (cyclopropyl) or a methyl group, and (3) replacing the diethylamino moiety with a morpholino group resulted in > 220-fold selectivity over muscarinic receptor binding. Therefore, several of these novel compounds are potent, sigma 1-selective ligands which can now be investigated as potential antitussive, anticonvulsant, and antiischemic agents. These studies may reveal whether sigma 1 sites play a role in the pharmacological actions of these drugs.

Analogues of the dioxolanes dexoxadrol and etoxadrol as potential phencyclidine-like agents. Synthesis and structure-activity relationships.

A series of dioxolane analogues based on dexoxadrol ((4S,6S)-2,2-diphenyl-4-(2-piperidyl)-1,3-dioxolane) and etoxadrol ((2S,4S,6S)-2-ethyl-2-phenyl-4-(2-piperidyl)-1,3-dioxolane) were prepared and tested for their ability to displace [3H]TCP (1-[1-(2-thienyl)cyclohexyl]piperidine) from PCP (1-(1-phenylcyclohexyl)piperidine) binding sites in rat brain tissue homogenates. Qualitative structure-activity relationships within this series were explored through modifications of the three major structural units of dexoxadrol, the piperidine, 1,3-dioxolane, and aromatic rings of the molecule. N-Alkyl derivatives of dexoxadrol were found to be inactive, as were those analogues where the dioxolane ring was modified. Phenyl-substituted etoxadrol analogues were compared to similarly substituted PCP analogues and distinct differences were found in their structure-activity relationships suggesting that the aromatic rings in these two drug classes interact differently with the PCP binding sites. The replacement of the phenyl ring in etoxadrol by either a 2- or 3-thienyl ring led to compounds with affinity comparable to etoxadrol, and the replacement of the ethyl moiety on etoxadrol's dioxolane ring with propyl (7) or isopropyl (8) led to compounds which were more potent than etoxadrol or PCP. The most potent compound was (2S,4S,6S)-2-ethyl- 2-(1-chlorophenyl)-4-(2-piperidyl)-1,3-dioxolane (11), where a chlorine moiety was placed in the ortho position in the aromatic ring of etoxadrol. Its potency was comparable with TCP in vitro.

Synthesis and in vitro and in vivo activity of (-)-(1R,5R,9R)- and (+)-(1S,5S,9S)-N-alkenyl-, -N-alkynyl-, and -N-cyanoalkyl-5, 9-dimethyl-2'-hydroxy-6,7-benzomorphan homologues.

Two of the synthesized (-)-(1R,5R,9R)-N-homologues (N-but-3-enyl- and N-but-3-ynyl-5,9-dimethyl-2'-hydroxy-6,7-benzomorphan (9, 13)) were found to be about 20 times more potent than morphine in the mouse tail-flick assay (ED(50) = 0.05 mg/kg), and (-)-(1R,5R, 9R)-N-but-2-ynyl-5,9-dimethyl-2'-hydroxy-6,7-benzomorphan ((-)-(1R, 5R,9R)-N-but-2-ynylnormetazocine, 12) was about as potent as the opioid antagonist N-allylnormetazocine (AD(50) in the tail-flick vs morphine assay = 0.3 mg/kg). All of the homologues examined had higher affinity for the kappa-opioid receptor than the mu-receptor except (-)-N-but-2-ynyl-normetazocine (12), which had a kappa/mu ratio = 7.8 and a delta/mu ratio = 118. The (-)-N-2-cyanoethyl (3), -allyl (8), and -but-3-ynyl (13) analogues had good affinity (<10 nM) for delta-opioid receptors. Two homologues in the (+)-(1S,5S,9S)-normetazocine series, N-pent-4-enyl (24) and N-hex-5-enyl (25), were high-affinity and selective sigma(1)-ligands (K(i) = 2 nM, sigma(2)/sigma(1) = 1250, and 1 nM, sigma(2)/sigma(1) = 750, respectively); in contrast, N-allylnormetazocine (22) had relatively poor affinity at sigma(1), and its sigma(1)/sigma(2) ratio was <100.

Hexahydro-1H-1-pyrindines from acid rearrangement of 9-alkylidene-5-(m-methoxyphenyl)-2-methylmorphans. A new structural type of narcotic antagonists.

9-Methylene- and 9-ethylidene-5-(m-methoxyphenyl)-2-methylmorphans (1, 2) and refluxing 48% HBr have given rearrangement products 3 and 4, respectively. The structure of 4 [4a-ethyl-2,4a,5,6,7,7a-hexahydro-4-(3-hydroxyphenyl)-1-methyl-1H-1- pyrindine] was determined by X-ray crystallography and that of 3 [1,4a-dimethyl-2,4a,5,6,7,7a-hexahydro-4-(3-hydroxyphenyl)-1-methyl-1H- pyrindine] follows from analogy and NMR data. Compounds 3 and 4 are opioid antagonists of about the potency of nalorphine in the tail-flick vs. morphine assay and precipitate a complete abstinence syndrome in morphine-dependent monkeys. Both are nearly devoid of antinociceptive activity and they have about 0.025 times the affinity of nalorphine for the mu opioid receptor.

Synthesis and anticonvulsant activity of 1-phenylcyclohexylamine analogues.

Thirty-eight analogues of 1-phenylcyclohexylamine (PCA), a phencyclidine (PCP) derivative, were examined for their activities in the mouse maximal electroshock (MES) seizure test and in a motor-toxicity assay. In addition, we determined the binding affinities of the compounds for PCP acceptor sites in rat brain membranes labeled with [3H]-1-[1-(2-thienyl)cyclohexyl]piperidine. Many of the analogues were protective against MES seizures (ED50s of 5-41 mg/kg, ip) and all of these compounds caused motor toxicity. The potencies in the motor toxicity and MES seizure tests showed a moderate correlation with the affinities for PCP sites. Several analogues exhibited a greater separation of potencies in the motor toxicity and MES seizure tests than did the parent compound PCA. These were obtained by (i) 3-methylation of the cyclohexyl ring trans to the phenyl ring, (ii) methoxylation at the ortho position on the phenyl ring, and (iii) contraction of the cyclohexane ring to form the corresponding cyclopentane.

(E)-8-benzylidene derivatives of 2-methyl-5-(3-hydroxyphenyl)morphans: highly selective ligands for the sigma 2 receptor subtype.

The determination of the structure and function of the sigma receptor subtypes and their physiological role(s) has been impeded by the unavailability of selective ligands. We have developed a new class of sigma subtype selective receptor ligands that are (E)-8-benzylidene derivatives of the synthetic opioid (+/-)-, (+)-, and (-)-2-methyl-5-(3-hydroxyphenyl)morphan-7-one (1). The derivatives can be prepared by reaction of 1, (+)-1, and (-)-1 with the appropriate benzaldehyde under Claisen-Schmidt conditions. Incorporation of substituted (E)-8-benzylidene moieties onto the 7-keto precursor of (+)-2-methyl-5-(3-hydroxyphenyl)morphan, (+)-1, produces compounds (-)-2 through (-)-7 (5.8-32.0 nM, sigma 1), which have between a 25- and 131-fold increase in affinity for the sigma 1 receptor subtype relative to the keto precursor (+)-1 (Ki = 762 nM, sigma 1). Compound (-)-2 is the most selective of this group (16-fold) for the sigma 1 subtype versus sigma 2. Substitution of an (E)-8-benzylidene moiety onto the 7-keto precursor of (-)-2-methyl-5-(3-hydroxyphenyl)morphan, (-)-1, produces compounds (+)-2-(+)-9 (6.4-52.6 nM, sigma 2), which have at least a 475-3906-fold increase in affinity for the sigma 2 receptor subtype relative to the keto precursor (-)-1 (Ki = 25 x 10(3) nM). This enhancement of sigma 2 receptor affinity is accompanied by substantial selectivity of all of these dextrorotatory products for the sigma 2 relative to the sigma 1 subtype (32-238-fold), and thus, they are among the most sigma 2 selective compounds currently known. Furthermore, the sigma 1 subtype is highly enantioselective for the levorotatory isomers, (-)-2-(-)-7 (41-1034-fold), whereas the sigma 2 subtype is only somewhat enantioselective for the dextrorotatory isomers, (+)-2-(+)-7 (2.6-9.3-fold). All of these derivatives retain substantial affinity for the mu opioid receptor. Despite the high affinity of the dextrorotatory derivatives for the mu opioid receptor, the high affinity and selectivity for sigma 2 over sigma 1 sites will surely prove beneficial as tools for the delineation of the function and physiological role of sigma 2 receptors.

Synthesis, phencyclidine-like pharmacology, and antiischemic potential of meta-substituted 1-(1-phenylcyclohexyl)-1,2,3,6-tetrahydropyridines.

A series of 1-[1-arylcyclohexyl]-1,2,3,6-tetrahydropyridines were prepared by the reaction between 1-(1-cyanocyclohexyl)-1,2,3,6-tetrahydropyridine (1) and an appropriately substituted Grignard reagent. The resulting compounds were tested for their phencyclidine binding site affinities. Selected compounds were then tested for their ability to produce ketamine appropriate responding in monkeys and/or to show neuroprotective effects in a baby rat hypoxia/ischemia model. While it was found that binding site affinity correlated well with discriminative stimulus effects, it was found to be a poor indicator of neuroprotective efficacy within this series.