Slow-onset, long-duration 3-(3',4'-dichlorophenyl)-1-indanamine monoamine reuptake blockers as potential medications to treat cocaine abuse

A series of 3-(3',4'-dichlorophenyl)-1-indanamine monoamine reuptake blockers have been synthesized in an effort to develop a compound that could be used as a maintenance therapy to treat cocaine abuse. Since the effects of cocaine on dopamine (DA) and serotonin (5HT) transporters are important components of its pharmacological activity, the focus was on nonselective inhibitors of monoamine transport. To reduce or eliminate the abuse potential of a DA reuptake blocker, the compounds were designed to be slow-onset, long-duration prodrugs whose N-demethylated metabolites would have increased activity over the parent compound with the ideal being a parent compound that has little or no activity. To achieve this, pairs of compounds with different groups on the amine nitrogen and with and without an additional N-methyl group were synthesized. All of the synthesized compounds were screened for binding and reuptake at the cloned human DA, 5HT, and norepinephrine (NE) transporters. As previously found, trans isomers are nonselective blockers of DA, 5HT, and NE reuptake, cis isomers with small N-alkyl groups are selective blockers of 5HT reuptake, and tertiary amines of the trans compounds are less potent than the corresponding N-demethylated secondary amines as blockers of DA reuptake. Larger N-alkyl groups in both the trans and cis series were found to reduce activity for the 5HT and NE transporters with less effect at DA transporters. Selected trans compounds were also screened for locomotor activity in mice and generalization to a cocaine-like profile in rats. With intraperitoneal administration, all of the trans isomers showed a slow onset of at least 20 min and an extremely long duration of action in the locomotor assays. Several of the trans compounds also fully generalized to a cocaine-like pharmacological profile. An initial lead compound, the N,N-dimethyl analogue trans-1b, was resolved into chirally pure enantiomers. Surprisingly, both enantiomers were found to have significant affinity for the DA transporter and to cause locomotor activation. This is in contrast to the N-methyl compound in which only the (+)-enantiomer had significant activity. The absolute configuration of the more active enantiomer was determined by X-ray crystallography to be 3R,1S.

Conformational preferences of the potent dopamine reuptake blocker BTCP and its analogs and their incorporation into a pharmacophore model

Molecular mechanics calculations using MM3-92 and ab initio quantum mechanical calculations using SPARTAN 5.0 were performed on the structurally similar PCP and BTCP, in which only the latter has a cocaine-like pharmacological profile as a dopamine reuptake blocker. Calculations were also performed on BTCP analogs with a methyl group in various positions of the cyclohexane ring. The results for the cis-2-methyl compound, which retains good pharmacological activity, allowed us to determine that an aryl-axial conformer is the biologically active form for at least some of the compounds in this series. However, an aryl-equatorial conformer presents the identical pharmacophore, as shown by superposition of the two conformers. X-ray crystallographic structures were also obtained for BTCP and related compounds with a 2-methyl group on the cyclohexane ring, with reasonable agreement between the computational and experimental results. Superposition studies were performed with two rigid analogs of cocaine which illustrate the optimal orientations of the ammonium hydrogen for monoamine transporters. There is excellent agreement between a 'back-bridged' cocaine analog that is optimal as a dopamine reuptake blocker and the previously proposed biologically active conformer of methylphenidate. However, BTCP is found to be a better fit to the 'front-bridged' cocaine analog that is optimal for a serotonin reuptake blocker.

Conformational analysis and a crystal structure of bupropion, an antidepressant with dopamine reuptake blocking activity

A conformational analysis has been performed on the antidepressant bupropion using the MM3-92 program. In addition, the structure of the compound in the crystal state was obtained. There is good agreement between the computed global minimum and the structure observed by crystallography. The three-dimensional structure of the preferred conformer of bupropion is consistent with the three-dimensional structures of other dopamine reuptake blockers such as cocaine, CFT, and methylphenidate.

The incorporation of butyrophenones and related compounds into a pharmacophore for dopamine D2 antagonists

This study is an attempt to incorporate the butyrophenones, an important class of nontricyclic antipsychotic drugs, into a previously proposed pharmacophore model of tricyclic dopamine D2 receptor antagonist ligands. Conformational energy calculations were performed using the MM3-92 program on spiperone, as a representative butyrophenone, and milenperone and R48455, as related compounds with more limited conformational freedom. Twenty seven conformers were evaluated for spiperone with MM3-92 calculations and nine of these were within 1.1 kcal/mole of the global minima indicating the flexibility of the compound. A conformational analysis of twenty crystal structures of butyrophenones was also performed and six distinct conformers were represented. All of the energy minimized conformers of spiperone were superimposed in a least squares sense onto loxapine as a relatively rigid, typical D2 antagonist and a pair of mirror image conformers, which are observed in one crystal structure of spiperone, were found to be the best fit. However, it was not possible to discriminate between these two conformers since they fit the pharmacophore model equally well. The para-fluoro and carbonyl group of the butyrophenones were found to correspond best to the oxygen and chlorine atoms of loxapine, respectively. The conformations of milenperone and R48455 were also consistent with the two putative biologically active forms of spiperone and the pharmacophore model. Conformational energy calculations were also performed on molindone, an antipsychotic drug in clinical use, which can be related to the butyrophenones since both have a carbonyl group adjacent to an aromatic ring. A putative biologically active form was proposed for molindone and this was related to the structure of piquindone, a rigid analog of molindone. All of the compounds were found to be entirely consistent with the pharmacophore model. However, as previously found, there is great variability in the distance between the ammonium nitrogen and the center of the relevant aromatic ring with the most extreme case in the present study being R48455 where the distance is 7.2 A. The results of the present study should also be relevant to the structures of novel, atypical antipsychotic drugs such as risperidone which appear to be analogs of the butyrophenones.

Conformational analysis of methylphenidate and its structural relationship to other dopamine reuptake blockers such as CFT

PURPOSE

This work was performed 1) to determine the conformational preferences of the threo and erythro isomers of the dopamine reuptake blocker methylphenidate, 2) to determine the crystal conformation of the threo isomer, 3) to confirm the absolute configuration of the more active threo enantiomer, and 4) to incorporate the compound into a previously determined pharmacophore for dopamine reuptake blockers.

METHODS

A conformational analysis was performed with the MM2-87 program, a crystal of the (-)-threo HCl salt was analyzed by x-ray crystallography, and the global minima of the (+)-threo isomer and the potent dopamine reuptake blocker CFT were superimposed.

RESULTS

In the global minimum of the threo isomer, the carbonyl oxygen of the ester group is oriented toward the ammonium group as was also found in the crystal state. In the erythro isomer, the ester group prefers an extended conformation relative to the piperidine group. The absolute configuration of the biologically active (+)-threo enantiomer was confirmed to be R,R. The atomic sequence from the amine group through the ester group is identical in the active enantiomers of methylphenidate and CFT.

CONCLUSIONS

The dopamine reuptake protein requires a precise orientation of the ammonium and ester groups but allows considerable leeway in the position of the phenyl ring. The pKa of the threo isomer is predicted to be higher than that of the erythro isomer.

Dopaminergic (4aR,10bS)-cis- and (4aS,10bS)-trans-octahydrobenzo[F]quinolines have similar pharmacophores

The structures and absolute configurations of two N-phenethyl substituted cis- and trans-octahydrobenzo[f]quinolines were determined by X-ray crystallography. The absolute configurations of the enantiomers that have high affinity for dopaminergic receptors were found to be (4aR,10bS) and (4aS,10bS) for the (-)-cis- and (-)-trans-8,9-dihydroxy substituted compounds. This is consistent with previous results for a dopamine agonist pharmacophore. MM2-87 calculations for a cis isomer, which has two alternative chair conformations of the piperidine ring, indicated that the preferred conformer is the same as that observed in the crystal structure. Superposition of the more active cis and trans enantiomers showed that the three dimensional orientations of the phenyl ring and the ammonium group are similar in the two geometrical isomers. The cis isomer, however, has steric bulk out of the plane of the molecule and this appears to result in a loss of agonist efficacy. The addition of the N-phenethyl group to the 7-OH and 7,8-diOH cis compounds, however, appears to be sufficient to restore high affinity for dopaminergic receptors unlike previously synthesized cis compounds. These cis compounds, however, appear to be mixed agonist/antagonists or antagonists on functional assays of dopaminergic activity.

Absolute configurations and conformations of the opioid agonist and antagonist enantiomers of picenadol

The absolute configurations of the enantiomers of the opioid picenadol [cis-1,3-dimethyl-4-propyl-4-(3-hydroxyphenyl)piperidine; cis-3-methyl, 4-propyl] have been determined by an X-ray crystallographic study of the chloride salt of the (+)-enantiomer. The agonist (+)-enantiomer and the antagonist (-)-enantiomer were found to have the 3R,4R and 3S,4S absolute configurations, respectively. The conformational properties of the enantiomers were also examined with MM2-87 calculations. There was good agreement between the computed global minimum and the crystallographic structure with the phenyl ring approximately bisecting the piperidine ring by both methods. This orientation of the phenyl ring differs from that of related opioids such as the phenylmorphans, prodines, meperidine, and ketobemidone in which the phenyl ring tends to eclipse one edge of the piperidine ring. Because the phenyl ring bisects the piperidine ring in picenadol, there is little difference in the three-dimensional orientations of the phenyl rings of the two enantiomers when one superimposes the piperidine rings. The agonist (+)-enantiomer is ambiguous with respect to an opioid ligand model, which suggests that agonist activity requires a specific range of dihedral angles for the phenyl ring. While the global minimum of the agonist is not consistent with the model, a second conformer that is only 1.2 kcal/mol above the global minimum is consistent. An alternative explanation is that agonist or antagonist activity is solely due to the presence of the 3-methyl group on the different edges of the piperidine ring.(ABSTRACT TRUNCATED AT 250 WORDS)

Homology modeling of the dopamine D2 receptor and its testing by docking of agonists and tricyclic antagonists

We present the first model of dopamine D2 receptor transmembrane helices constructed directly from the bacteriorhodopsin (bR) coordinates derived from two-dimensional electron diffraction experiments. We have tested this model by its ability to accommodate rigid agonist and semirigid antagonist molecules which were docked into the putative binding pocket with stabilizing interactions. The model is consistent with structure-activity relationships of agonists and antagonists that interact with the receptor. It also illuminates data on a Na+ site for regulation of receptor function. The plausibility of the model is increased by its consistency with many mutagenesis studies on G protein-coupled receptors. Further, this model provides a basis to suggest testable molecular mechanisms for changes in the D2 conformational states for high- and low-affinity binding and signal transduction. Changes in the conformational state of the receptor are hypothesized to be due partly to movement of helix 7. In contrast to the model presented here, other published models were built using ideal helical structures or following the sense of the bacteriorhodopsin structure rather than the actual available coordinates. The presented model for the dopamine G protein-coupled receptor can be reconciled with the recent rhodopsin projection structure (Schertler, G. F. X.; Villa, C.; Henderson, R. Projection Structure of Rhodopsin.

Chiral resolution of 1,3-dimethyl-4-phenylpiperidine derivatives using high-performance liquid chromatography with a chiral stationary phase

A number of racemic 1,3-dimethyl-4-phenylpiperidines which serve as intermediates in the synthesis of opioid analgesics have been resolved on two commercially available high-performance liquid chromatography columns containing cellulose-based chiral stationary phases: Chiralcel OD and Chiralcel OJ. The resolution results were complementary between the two columns. Also, the polarity of substituents appears to play an important role on the ability of the Chiralcel OD column to resolve pairs of enantiomers.

Conformational analysis of the opioid phenylmorphan and its 9 alpha-methyl analogue in solution using high-resolution nuclear magnetic resonance spectroscopy

The solution conformations of the opioid phenylmorphan (5-m-hydroxyphenyl-2-methylmorphan) and its 9 alpha-methyl analogue were studied using one- and two-dimensional high resolution NMR techniques. The NMR spectra were analyzed by interpreting the phase-sensitive 2-D COSY and double quantum filtered COSY spectra, 1H-1H vicinal coupling constants, and nuclear Overhauser effects in the phase-sensitive 2-D NOESY spectra. The results show that, for both compounds, a chair-chair conformation of the cyclohexane and piperidine rings is exclusively preferred with some distortion of the rings from perfectly staggered chairs. For phenylmorphans, the phenyl ring is oriented to fit into the cleft formed by the cyclohexane and piperidine rings. Thus, for the (+)-enantiomer, the phenyl group assumes the same orientation with regard to the piperidine ring as morphine consistent with the morphine-like properties of the compound. For the 9 alpha-methyl analogue, the plane of the phenyl ring essentially bisects the piperidine ring to which it is attached and is outside of the required range of opioid agonists. This is consistent with the atypical properties of the two enantiomers. The NMR results are compared to the conformations of (-)-phenylmorphan and the (+)-9 alpha-methyl analogue in the crystal state and to the results of molecular mechanics (MM2) studies.

Biologically active conformers of phenothiazines and thioxanthenes. Further evidence for a ligand model of dopamine D2 receptor antagonists

Conformational analyses have been performed on several phenothiazine and thioxanthene dopamine antagonists using the MM2-87 program and parameter set. The compounds that were examined are thioridazine (2), methotrimeprazine (3), cis- and trans-chlorprothixene, and a piperidylidene derivative of chlorprothixene. In addition, (+)-2 and (-)-3 were determined by X-ray crystallography to have the R absolute configuration. The above compounds were superimposed onto loxapine, which was used as a template for the previously proposed dopamine D2 receptor ligand model. The conformational properties and receptor affinities of these compounds were found to be entirely consistent with the ligand model. For example, a conformer of (+)-R-2 that is consistent with the ligand model is lower in energy than a consistent conformer for (-)-S-2, which agrees with the higher D2 receptor affinity of the former. Similarly, in agreement with the much higher affinity of (-)-R-3 relative to (+)-S-3, only the former contains a low energy conformer consistent with the ligand model. The ligand model is also consistent with the greater potency of cis-thioxanthenes over the trans isomers. These results emphasize the importance of the correct orientation of the ammonium hydrogen for high affinity at the D2 receptor. The pharmacophore for D2 receptor ligands is compared with a recently proposed pharmacophore for D1 ligands.

Enantiomeric conformers of the opioid agonist ketobemidone HCl in the crystal state

A crystal of the potent opioid agonist ketobemidone [1-methyl-4-(3-hydroxyphenyl)-4-propionylipiperidine] HCl was analyzed by X-ray crystallography. The crystal was monoclinic, space group P2(1)/n with four molecules in the unit cell. In agreement with MM2 calculations (J. Med. Chem. 25:1127-1133, 1982), the crystal contains mirror image conformers in which the phenyl ring is equatorial to the piperidine ring. The conformers are enantiomers since they are not superimposable. One conformer is predicted to be responsible for the typical morphine-like activity of the compound since it closely matches the preferred conformer of the morphine-like (+)-phenylmorphan whereas the other conformer resembles the preferred conformers of (+)-beta-prodine and (-)-phenylmorphan which have atypical opioid properties and/or structure-activity relationships. The importance of considering the conformational enantiomers of a nonchiral receptor ligand in centrosymmetric crystal structures is emphasized.

Conformational preferences of the kappa-selective opioid agonist U50488. A combined molecular mechanics and nuclear magnetic resonance study

The conformational preferences of the kappa-selective opioid agonist U50488 have been studied using MM2-87 calculations and nuclear magnetic resonance (NMR) spectroscopy. The calculations were performed for the protonated form with a dielectric constant of 80 and the unprotonated form with dielectric constants of 1.5 and 80. A systematic search found 72 stable conformers with certain consistent conformational preferences for some of the important dihedral angles. The preferred conformers proved to be compact structures stabilized by intramolecular attractive van der Waals interactions, though at least some of these appear to be electrostatically unfavorable. The conformation of U50488 was also examined in aqueous solution using one-dimensional (1D) and two-dimensional (2D) high-resolution 1H NMR techniques such as the interpretation of 1H-1H vicinal coupling constants, 1D and 2D nuclear Overhauser effect (NOE) experiments, and 2D correlated spectroscopy (COSY) experiments. Five crystallographic conformations were examined as well. There was generally good agreement between all three methods of conformational analysis. There appeared to be a reasonable geometrical agreement between the relatively rigid kappa-agonist (-)-ketazocine and a gauche conformer of U50488. The proposed pharmacophore is also consistent with other kappa-selective analogs of U50488 including one in which the peptide bond is incorporated into a lactam ring. The low affinity of U50488 for mu-receptors was attributed to its cyclohexane ring which occupies space not present in the nonselective (-)-ketazocine.

Phenylmorphans and analogues: opioid receptor subtype selectivity and effect of conformation on activity

The morphine-like (+)-phenylmorphan, the atypical (-)-enantiomer, and some analogues have been tested in receptor binding assays selective for opioid mu 1, mu 2, delta, kappa 1, and kappa 3 receptors. The affinities of all of the compounds except one, including the atypical (-)-phenylmorphan, were greatest for mu 1 and mu 2 receptors. The only exception was the (+)-9 alpha-methyl analogue which had slightly greater affinity for the kappa 1 receptor. The selective receptor binding assays provide evidence that opioids in which the phenyl ring is constrained to be equatorial on the piperidine ring can have considerable affinity for mu receptors. In addition, dose-response curves were determined for (+)- and (-)-phenylmorphan using the mouse tail-flick assay with the (+)-enantiomer found to be about 7 times more potent. Pretreatment with the selective opioid antagonists beta-FNA (mu 1 and mu 2), naloxonazine (mu 1), nor-BNI (kappa 1), and naltrindole (delta) suggests that the antinociceptive activity of both enantiomers is mediated through mu receptors. The pretreatment with naloxonazine, which attenuated the antinociceptive effect, shows that both (+)- and (-)-phenylmorphan are mu 1 agonists while intrathecal administration shows that both are mu 2 agonists. Conformational energy calculations on the compounds were also performed using the MM2-87 program. Consistent with previous conformational results for the phenylmorphans (J. Med. Chem. 1984, 27, 1234-1237), the most potent antinociceptive compounds preferred a particular orientation of the phenyl ring.

Absolute configuration and conformation of the pure opioid antagonist (+)-2,9 alpha-dimethyl-5-(m-hydroxyphenyl)morphan

(+)-2,9 alpha-Dimethyl-5-(m-hydroxyphenyl)morphan is the only phenylmorphan analog whose affinity for opioid kappa-receptors is greater than its affinity for opioid mu-receptors. Pharmacologically, the compound is a pure opioid antagonist devoid of agonist activity in in vivo assays of antinociception. The absolute configuration of the compound has been determined to be (1R,5S,9R) from an X-ray crystallographic study of the chloride salt. Thus, the absolute configuration corresponds to that of the atypical opioid agonist (-)-phenylmorphan while the weak atypical agonist (-)-2,9 alpha-dimethyl-5-(m- hydroxyphenyl)morphan corresponds to the potent morphine-like (+)-phenylmorphan. The preferred orientations of the phenyl ring for the two stereoisomers were determined using the molecular mechanics program MM2-87 and found to vary from that of the two parent compounds. The atypical properties of the two 9 alpha-methyl analogs is consistent with an opioid ligand model which proposes that morphine-like properties require a particular range of phenyl orientations. There was good agreement between the structure obtained from X-ray crystallography and computed with the MM2-87 program.

Conformational properties of semirigid antipsychotic drugs: the pharmacophore for dopamine D-2 antagonist activity

Conformational energy calculations using the MM2-87 program have been performed on the tetracyclic spiro amines 1 (A23887) and 2 (A31472) which have previously been shown to have considerable affinity for dopamine D-2 receptors. These compounds are important for defining the pharmacophore for D-2 antagonist activity due to their limited conformational freedom. Possible foldings of the multicyclic structure were energy minimized and the barriers for inversion and for rotation of the ammonium group were computed. The conformational properties of 1 and 2 are consistent with a pharmacophore recently proposed by Liljefors and Bøgesø. The greater affinity of (S)-octoclothepin for D-2 receptors as compared with its enantiomer was attributed to the latter having an incorrect orientation of the ammonium hydrogen despite the correct folding of the tricyclic structure. Other D-2 antagonists with limited conformational freedom such as butaclamol, isobutaclamol, loxapine, clozapine, and resolved cyproheptadine analogues were also found to be consistent with the pharmacophore. In addition, 1, 2, and their enantiomers were tested on radioligand binding assays for dopamine D-1, dopamine D-2, noradrenergic alpha-1, serotonergic 5-HT2, muscarinic, and sigma receptors. 1 and 2 have greater affinities than their enantiomers in the D-1, D-2, alpha-1, and 5-HT2 assays though there was little difference between 2 and its enantiomer in the latter two assays. In the muscarinic assays, 2 and its enantiomer, which were approximately equipotent, had greater affinity than 1 and its enantiomer. None of the compounds had substantial affinity for sigma receptors. Since the same enantiomers of 1, 2, butaclamol, and the resolved cyproheptadine analogues also have greater affinities for D-1 receptors, the conformational requirements of D-1 ligands appear to be quite similar to those of D-2 ligands.

The development of computer simulations of the geometries and thermodynamics of biological molecules

The historical development of computer simulations of molecular geometry and thermodynamics (molecular mechanics, force field method) is outlined. The advantages and disadvantages of techniques such as energy minimization, molecular dynamics and free energy perturbation are discussed. An example is included that shows how energy minimization studies of dopamine D-2 antagonists have been used to develop an understanding of the three-dimensional pharmacophore necessary for this pharmacological activity.

Requirements for the activation of protein kinase C: comparison of the molecular geometries of phorbol and diacylglycerol

MM2 calculations have been performed on a number of derivatives of phorbol and diacylglycerol (DAG) to establish the molecular features required for the activation of protein kinase C by a detailed comparison of the molecular geometries in these two classes of compounds. For DAG, a dihedral angle of about -60 degrees appears to be required for the oxygens at C2 and C3 because that angle is fixed at this value in phorbols. There is good agreement between the computed Boltzmann distribution for the O1-C1-C2-O2 dihedral angle and NMR results for the same angle in phospholipids, as obtained by others. A conformer of DAG is identified with dihedral angles corresponding to those of beta-phorbols. This conformer, however, is 3.2 kcal/mol above the global minimum found for DAG. The molecular geometry of this conformer is consistent with that of a number of active and inactive rigid analogues of DAG. The preferred conformation in beta-phorbol diesters is found to be stabilized by an antiparallel stacking of the ester carbonyl groups. The lack of activity of alpha-phorbol esters appears to be due to differences in a portion of the molecule containing the five-membered/seven-membered rings, which are far from the DAG-like end of the phorbol molecule. It is proposed that some of the biological activities of phorbol diesters may be due to this portion of the beta-phorbol molecule, which might represent a second active region, distinct from that resembling DAG.

Conformational search in enkephalin analogues containing a disulfide bond

A systematic conformational search has been performed for the 14-membered ring in model compounds for disulfide-containing enkephalin analogues. The model compounds examined are [formula: see text], and the corresponding compounds with L-amino acids at the C-terminus. About 100 starting conformations were generated for each compound with the RNGCFM program and energy minimized with the AMBER program. Between 21 and 38 conformers within 3 kcal/mole of the apparent global minimum were found for each compound. There appeared to be fewer possible conformations of the disulfide-containing side chain than of the main chain. [formula: see text], whose parent compound is selective for opioid delta receptors, was found to prefer conformers with a positive dihedral angle of the disulfide bond, which is consistent with the previous proposal that delta-receptor selectivity may be associated with this conformational preference. Additional calculations were performed on the complete structure of [formula: see text] (DPDPE) with various possible conformations of the tyrosine and phenylalanine side chains. Conformational free energies and entropies were computed for these conformers from the molecular vibrations obtained from a normal mode analysis. As was found previously, conformers with low energies tended to have lower entropies, which resulted in a narrowing of the free energy differences between conformers. A conformer is identified that has the lowest energy hitherto found for DPDPE. It is suggested that DPDPE may be a useful compound for evaluating conformational search strategies because of its relatively small size and the number of conformers that have already been identified. Conformational energy calculations are also reported for naltrindole using the MM2(87) program. Naltrindole, which incorporates two aromatic 6-membered rings in a rigid structure, is a highly selective and potent opioid delta-receptor antagonist and may be an important clue regarding the biologically active conformer of DPDPE. Various conformers of DPDPE have been superimposed quantitatively onto the structure of naltrindole using the SUPER program and those conformers of DPDPE that are the best fit to naltrindole are reported.

Conformational analysis of enkephalin analogs containing a disulfide bond. Models for delta- and mu-receptor opioid agonists

Conformational analysis of the cyclic opioids H-Tyr-D-Pen-Gly-Phe-D-Pen-OH (DPDPE) and H-Tyr-D-Cys-Gly-Phe-D-Cys-OH (DCDCE) have been performed using the AMBER program. DPDPE is considerably more selective for delta-receptors than DCDCE. Using the RNGCFM program, a large number of ways were found to close the 14-membered disulfide-containing ring structure. However, intramolecular hydrogen bonds were only possible in gamma-turn and inverse gamma-turn conformations centered on the glycine residue which were associated with opposite chiralities of the disulfide bond. With the cyclic part of the molecules in either a gamma-turn or inverse gamma-turn, a systematic conformational analysis was performed on the tyrosine and phenylalanine sidechains. This showed that conformers with the tyrosine and phenylalanine phenyl rings in the vicinity of the disulfide bond were preferred due to attractive van der Waals forces. For DPDPE, however, this was only possible with a positive dihedral angle for the disulfide bond due to the presence of the beta-carbon methyls of Pen2. In contrast, these preferred conformers were possible with both chiralities of the disulfide bond in DCDCE. Conformational entropies and free energies were computed from the translational, rotational, and vibrational energy levels available to each conformer. The conformational entropies were found to vary significantly and to result in a re-ordering of the lowest energy minima. Based on these conformational differences in DPDPE and DCDCE and their differing pharmacological selectivities, tentative conformational preferences for delta- and mu-receptor opioid peptides are proposed.

Receptor affinity, neurochemistry and behavioral characteristics of the enantiomers of thioridazine: evidence for different stereoselectivities at D1 and D2 receptors in rat brain

The binding characteristics of the enantiomers of thioridazine were assessed in the brain of the rat using competitive radioreceptor assays with tritiated ligands selective for dopamine D1 (SCH-23390), D2 (spiperone), norepinephrine alpha-1 (prazosin) and muscarinic (quinuclinidinyl benzilate) receptors. (+)-Thioridazine was shown to have 2.7 and 4.5 times higher affinity than (-)-thioridazine for D2 and alpha-1 receptors, respectively. In contrast, (-)-thioridazine had 10 times higher affinity for the D1 receptor. Both enantiomers showed similar affinities for the muscarinic receptor. In a second experiment, thioridazine, dopamine, norepinephrine, serotonin and their metabolites were assayed in the brain of the rat after acute administration of the enantiomers of thioridazine and the assessment of catalepsy. (+)-Thioridazine was 4.1 times as potent as (-)-thioridazine in elevating the turnover of dopamine in the striatum, but neither enantiomer affected the other monoamines. The concentration of thioridazine and its metabolites in the brain, for a given dose, was similar for both enantiomers. (-)-Thioridazine induced slightly more catalepsy than (+)-thioridazine and appeared to be more toxic at large doses. While racemic thioridazine had an intermediate effect between that of its two enantiomers in the binding and neurochemical assays, it appeared to induce more catalepsy than either enantiomer, suggesting a synergistic effect in this behavioral assay. It was concluded that (+)- and (-)-thioridazine act as partially selective D2 and D1 antagonists, respectively. Therefore, clinical administration of only one enantiomer of thioridazine, rather than the currently prescribed racemate, may result in an improved therapeutic profile and so be worthy of further investigation.

Neuropharmacology and stereochemistry of dopamine receptor agonist and antagonist enantiomeric pairs

Neuropharmacological evaluation of the R and S isomers of 11-hydroxy-N-n-propylnoraporphine (11-OH-NPa) supports the impression that the 11-OH group in aporphines (analogous to the meta hydroxyl of dopamine, DA) is sufficient to confer high affinity and activity at DA receptors. As in the case of the catechol congeners, (R)-apomorphine (APO) and (R)-N-n-propylnorapomorphine (NPA), (R)-11-OH-NPa is a potent DA agonist while, like (S)-NPA, (S)-11-OH-NPa is a DA antagonist. Thus, (R) and (S)-11-OH-NPa are an additional pair of compounds in which one enantiomer is a DA agonist and the other an antagonist. Other analogous pairs are the enantiomers of 3-(3-hydroxyphenyl)-N-n-propylpiperidine (3-PPP), and cis-1-methyl-5-hydroxy-2-(di-n-propylamino)tetralin (5-OH-MDAT). All contain a meta hydroxyphenyl, an N-n-propyl, and a phenethylamine moiety which can be superimposed in a consistent way to discriminate the DA agonists from the antagonists, with the key feature in this discrimination being the direction of the ammonium hydrogen. An energy penalty must be incurred by 3-PPP to assume the required conformations and it may account for the relatively low potency of the 3-PPP enantiomers. This analysis supports the view that rigid analogs of flexible compounds when "frozen" in their biologically active conformation exhibit higher affinity interactions with the receptor.

A stereochemical and conformational model of dopaminergic agonist and antagonist activity: further evaluation

Conformational energy calculations using the Molecular Mechanics II (MM2) program have been performed on 2-aminotetrahydronaphthalene (ATN) and 2-aminoindan derivatives which are active or inactive at dopamine receptors. The results were used to test a stereochemical and conformational model previously proposed for dopaminergic activity. The conformer predicted to be optimal for agonist activity was found to have relatively low energy (less than 1.5 kcal/mol) for all of the agonists examined. The model successfully: (1) explained the relative activity or inactivity of compounds such as cis- and trans-1-methyl-5-hydroxyl ATN derivatives and the corresponding cis- and trans-octohydrobenzo[f]quinolines; (2) predicted the more potent antipode of 2-aminoindan dopaminergic agonists; and (3) explained the structure--activity peculiarities of 3-(3-hydroxyphenyl)-N-alkylpiperidines in which the potency is increased for (3S)-isomers and decreased for (3R)-isomers when the N-alkyl group is greater than propyl. Predictions of postsynaptic dopaminergic antagonism were also made for some of the compounds. In agreement with previous conclusions, the inactivity of ATN derivatives with a 2-methyl or 5-propyl group was attributed to steric interference at the receptor since those groups did not have a significant conformational effect on the receptor ligand.

Molecular geometries and steric energies of phorbol 10,11-diacetate and 1,2-diacetylglycerol molecules

Protein kinase C, an enzyme that is stimulated physiologically by diacylglycerol (DAG) and phospholipids in the presence of Ca2+, is involved in a novel cellular signaling system that is activated by the binding of appropriate agonists to certain classes of receptors. Phorbol esters are tumor promoters that can replace DAG in the activation of protein kinase C. Molecular similarities between the two compounds have been proposed to be responsible for the capacity to activate the enzyme. We have studied the molecular geometries and conformational energies of DAGAc and PDAc using the Molecular Mechanics II program and parameter set developed by Allinger and Yuh (1980). This was done to establish whether conformers of the two compounds are geometrically similar and which hydroxyl group of the phorbol molecule corresponds to the C3 hydroxyl of DAG which must be unsubstituted for activation of protein kinase C.

Geometrical correspondence between phenazocine and the enkephalins

Calculations have been performed on phenazocine using Allinger's MM2 (molecular mechanics II) program with full energy minimization. The N-phenethyl group was found to have considerable flexibility with a number of low-energy conformers. The best N-phenethyl axial conformer was 1.6 kcal/mol higher in energy than the best equatorial one. Calculations were also performed on the beta isomer of phenazocine with the result that the energy difference between the best equatorial and axial conformers rose to a substantial 4.6 kcal/mol. The hypothesis that opiate agonism requires an N substituent in the axial position does not appear to be consistent with the increased potency of beta isomers in which axial N substituents are thermodynamically more unstable. Comparisons have also been made between the low-energy conformers of phenazocine and those that have been observed or proposed for the enkephalins. One conformation of the tyrosine portion of the enkephalins that was observed by X-ray crystallography by Karle et al. was found to be a good fit to morphine-like opiates. The backbone conformer suggested by Gorin et al. was found to be the best fit to the two phenyl rings of phenazocine.

Preferred conformers for the pharmacologically typical and atypical antipodes of phenylmorphan opiates

The conformational preferences of phenylmorphan have been determined by the MM2 (Molecular Mechanics II) program using full energy minimization. Chair-chair conformations of the cyclohexane and piperidine rings were preferred by 2.6 kcal/mol or more. With the preferred chair-chair conformation, three stable orientations of the phenyl ring were found with relative energies of 0.0, 1.0, and 1.2 kcal/mol. The barrier to rotation of the phenyl ring was computed to be 4 kcal/mol. The preferred phenyl orientation for the (+)-antipode was similar to that of morphine using a previously postulated molecular model for opiate substrates. This is consistent with the typical morphine-like pharmacological properties of this antipode. The preferred phenyl orientation of the atypical (-)-antipode appears to be most similar to the phenyl orientation that is invariably preferred by more active prodine antipodes. The preferred conformer was similar to the one observed by X-ray crystallography.

Conformational properties of butaclamol and isobutaclamol. Regularities in the structures of semirigid neuroleptics

Conformational energy calculations have been performed on butaclamol and isobutaclamol using Allinger's MM2 (Molecular Mechanics II) program. Cis arrangements of rings D and E were found to be preferred by 1.4-1.9 kcal/mole for both compounds. Nevertheless, based on a molecular comparison with a number of semirigid neuroleptics, most notably loxapine and octoclothepin, it is suggested that trans arrangements are required for neuroleptic activity in the two compounds. However, trans conformer B of butaclamol, which was previously postulated as the biologically active form, was found to be 4.1 kcal/mole higher in energy, suggesting that it is less likely to play a significant pharmacological role. The biologically active forms are identified as trans conformer A for butaclamol and trans conformer B for isobutaclamol. Certain regularities in the structures of the semirigid neuroleptics are noted. It is also speculated that the cis conformers of protonated butaclamol may have unfavorable geometries for ion solvation, which would account for the anomalously low pKa measured for the compound. A similar explanation would also account for a trans conformer being found in the crystal structures of the bromide salts of butaclamol and dexaclamol.

Conformation-activity study of 4-phenylpiperidine analgesics

A conformational study of various 4-phenylpiperidine analgesics (the prodines, ketobemidone, meperidine, and 1,3,4-trimethyl-4-phenylpiperidines) has been performed with Allinger's Molecular Mechanics II (MM2) program. Phenyl equatorial conformations were found to be preferred for the prodines, ketobemidone, and meperidine. For ketobemidone and meperidine, however, phenyl axial conformations were computed to be only 0.7 and 0.6 kcal/mol higher in energy. It was suggested that phenyl axial conformers can explain the potency-enhancing effect of a phenyl m-hydroxy group in these two compounds. In contrast, phenyl axial conformers were computed to be relatively unfavorable for the prodines, being 1.9, 2.8, and 3.4 kcal/mol higher in energy for 3-demethyl-, alpha-, and beta-prodine, respectively. In addition, relative concentrations of an analgesic conformation can be related to the potencies of the three prodines. A phenyl axial conformer was computed to be preferred by 0.7 kcal/mol for the 3-demethyl compound of 1,3,4-trimethyl-4-phenylpiperidine, with phenyl equatorial conformers preferred by 1.3 and 3.3 kcal/mol for the alpha and beta compounds. Phenyl axial conformers were unexpectedly found to be especially destabilized by a 3-methyl group in the beta configuration due to the steric crowding of the three piperidine substituents. Detailed comparisons were made between the computed structures and those observed by X-ray crystallography.