Synthesis and physicochemical characterization of (6S)-5-formiminotetrahydrofolate; a reference standard for metabolomics.

The one-carbon carrier of the formate oxidation level derived from the interaction of tetrahydrofolate and formiminoglutamate, which has been tentatively identified as 5-formiminoltetrahydrofolate, has been prepared by chemical synthesis. Treatment of a solution of (6S)-tetrahydrofolate in aqueous base with excess ethyl formimidate in the presence of anti-oxidant under anaerobic conditions afforded a gummy solid which, based on mass spectral analysis, conformed to a monoformimino derivative of tetrahydrofolate. Further physicochemical characterization by validated methods strongly suggested that the product of chemical synthesis was identical to the enzymatically produced material and that it was, in fact, (6S)-5-formiminotetrahydrofolate. Conditions and handling methods toward maintaining the integrity of this highly sensitive compound were identified and are described, as is analytical methodology, useful for research studies using it.

Structure and receptor activity for classical cannabinoids.

In this decade, classical cannabinoid SAR has increasingly been interpreted in terms of ligand-receptor interaction. Membrane receptor binding affinity, inhibition of receptor coupled processes, functional in vitro assays that distinguish between agonists and antagonists, transfected receptor systems, and distinction among receptor subtypes have been applied to cannabinoid pharmacology studies. These studies have revealed information about the pivotal receptor interaction that mediates the neurochemical system characterized by the receptor. The positive correlation between binding affinity and behavioral effects validates ligand-receptor interaction as a focus for developing new cannabinoids with the potential for selective pharmacological effects. Added to these direct receptor studies are the computational methods that compare active and inactive ligands for steric, electronic, and conformational similarities that reveal an SAR for activity with a sophistication and a perspective not possible before computer methodology. Merging these studies with knowledge of the receptor sequence, consequences of mutations of the receptor, and a calculated structure of the receptor are evolving a physical picture of the interaction of cannabinoids with its receptor(s). This picture of a ligand-receptor interaction guides medicinal chemists in their interpretation of classical cannabinoid SAR and the design of new analogs for selective pharmacological activity and as probes of the cannabinoid receptor. It is from a receptor perspective that this review of the SAR of classical cannabinoids is presented.

Synthesis and pharmacological comparison of dimethylheptyl and pentyl analogs of anandamide.

(Dimethylheptyl)anandamide [(16,16-dimethyldocosa-cis-5,8,11,14-tetraenoyl)ethanolamine ] (17a) and its amide analogs were synthesized by Wittig coupling of a ylide derived from a fragment of arachidonic acid. These amides were compared to the endogenous cannabinoid receptor ligand arachidonylethanolamide (anandamide, 2a) and its amide analogs in pharmacological assays for potential enhancement of cannabimimetic activities. The receptor affinity to rat brain membranes of the dimethylheptyl (DMH) analogs increased by an order of magnitude in most comparisons to the corresponding anandamides in displacement assays versus the cannabinoid agonist [3H]CP 55,940 or antagonist [3H]SR141716A, for which rank order differences in affinity were observed. An order of magnitude enhancement of potency with comparable or higher efficacy in behavioral assays in the mouse tetrad of tests of cannabinoid activity was observed in 17a versus 2a. In contrast, no enhancement in potency for the pentyl to DMH side chain exchange was seen in the mouse vas deferens assay. The data indicate a structural equivalence between classical plant cannabinoids and 2a as well as different receptor-ligand interactions that characterize multiple receptor sites or binding modes.

Importance of the C-1 substituent in classical cannabinoids to CB2 receptor selectivity: synthesis and characterization of a series of O,2-propano-delta 8-tetrahydrocannabinol analogs.

The separation of the mood-altering effects of cannabinoids from their therapeutic effects has been long sought. Results reported here for a series of C-9 analogs of the cyclic ether O,2-propano-delta 8-tetrahydrocannabinol (O,2-propano-delta 8-THC) point to the C-1 position in classical cannabinoids as a position for which CB2 subtype selectivity occurs within the cannabinoid receptors. O,2-Propano-11-delta 8-THC, O,2-propano delta 9,11-THC, O,2-propano-9-oxo-11-nor-hexahydrocannabinol (O,2-propano-9-oxo-11-nor-HHC), and O,2-propano-9 alpha- and O,2-propano-9 beta-OH-11-nor-HHC were synthesized and evaluated in radioligand displacement assays for affinity at the CB1 and CB2 receptors and in the mouse vas deferens in vitro assay and the mouse tetrad in vivo assay for cannabinoid activity. Evaluation of binding affinity at the CB1 and CB2 receptors revealed that each compound possesses a modest increased affinity for the CB2 receptor. Analogs which contained an oxygen attached to C-9 (i.e., oxo and hydroxy derivatives) showed the highest affinity and selectivity for CB2 (for O,2-propano-9-oxo-11-nor-HHC, Ki(CB1) = 90 nM, Ki(CB2) = 23 nM, selectivity ratio 3.9; for O,2-propano-9 beta-OH-11-nor-HHC, Ki(CB1) = 26 nM, Ki(CB2) = 5.8 nM, selectivity ratio 4.5). Each compound was found to produce a dose-dependent inhibition of electrically-evoked contractions of the mouse isolated vas deferens when administered at submicromolar concentrations. This inhibition could readily be prevented by the selective CB1 cannabinoid receptor antagonist SR-141716A. The analogs exhibited unique in vivo profiles with O,2-propano-delta 9,11-THC exhibiting antinociception with reduced activity in three other in vivo measures and O,2-propano-9 beta-OH-HHC exhibiting lack of dose responsiveness in all measures. The CB2 selectivities in the O,2-propano analogs may be due to differences in solvation/desolvation that occur when the ligands enter the CB1 vs CB2 binding site. Alternatively, the CB2 selectivities may be a results of an amino acid change from a hydrogen bond-accepting residue in CB1 to a hydrogen bond-donating residue in CB2.

The bioactive conformation of aminoalkylindoles at the cannabinoid CB1 and CB2 receptors: insights gained from (E)- and (Z)-naphthylidene indenes.

The aminoalkylindoles (AAIs) are agonists at both the cannabinoid CB1 and CB2 receptors. To determine whether the s-trans or s-cis form of AAIs is their receptor-appropriate conformation, two pairs of rigid AAI analogues were studied. These rigid analogues are naphthylidene-substituted aminoalkylindenes that lack the carbonyl oxygen of the AAIs. Two pairs of (E)- and (Z)-naphthylidene indenes (C-2 H and C-2 Me) were considered. In each pair, the E geometric isomer is intended to mimic the s-trans form of the AAIs, while the Z geometric isomer is intended to mimic the s-cis form. Complete conformational analyses of two AAIs, pravadoline (2) and WIN-55, 212-2 (1), and of each indene were performed using the semiempirical method AM1. S-trans and s-cis conformations of 1 and 2 were identified. AM1 single-point energy calculations revealed that when 1 and each indene were overlayed at their corresponding indole/indene rings, the (E)- and (Z)-indenes were able to overlay naphthyl rings with the corresponding s-trans or s-cis conformer of 1 with an energy expense of 1.13/0.69 kcal/mol for the C-2 H (E/Z)-indenes and 0.82/0.74 kcal/mol for the C-2 Me (E/Z)-indenes. On the basis of the hypothesis that aromatic stacking is the predominant interaction of AAIs such as 1 at the CB receptors and on the demonstration that the C-2 H (E/Z)- and C-2 Me (E/Z)-indene isomers can mimic the positions of the aromatic systems in the s-trans and s-cis conformers of 1, the modeling results support the previously established use of indenes as rigid analogues of the AAIs. A synthesis of the naphthylidene indenes was developed using Horner-Wittig chemistry that afforded the Z isomer in the C-2 H series, which was not produced in significant amounts from an earlier reported indene/aldehyde condensation reaction. This approach was extended to the C-2 Me series as well. Photochemical interconversions in both the C-2 H and C-2 Me series were also successful in obtaining the less favored isomer. Thus, the photochemical process can be used to provide quantities of the minor isomers C-2 H/Z and C-2 Me/E. The CB1 and CB2 affinities as well as the activity of each compound in the twitch response of the guinea pig ileum (GPI) assay were assessed. The E isomer in each series was found to have the higher affinity for both the CB1 and CB2 receptors. In the rat brain membrane assay versus [3H]CP-55,940, the Ki's for the C-2 H/C-2 Me series were 2.72/2.89 nM (E isomer) and 148/1945 nM (Z isomer). In membrane assays versus [3H]SR141716A, a two-site model was indicated for the C-2 H/C-2 Me (E isomers) with Ki's of 10. 8/9.44 nM for the higher-affinity site and 611/602 nM for the lower-affinity site. For the Z isomers, a one-site model was indicated with Ki's of 928/2178 nM obtained for the C2 H/C-2 Me analogues, respectively. For the C-2 H/C-2 Me series, the CB2 Ki's obtained using a cloned cell line were 2.72/2.05 nM (E isomer) and 132/658 nM (Z isomer). In the GPI assay, the relative order of potency was C-2 H E > C-2 Me E > C-2 H Z > C-2 Me Z. The C-2 H E isomer was found to be equipotent with 1, while the C-2 Me Z isomer was inactive at concentrations up to 3.16 microM. Thus, results indicate that the E geometric isomer in each pair of analogues is the isomer with the higher CB1 and CB2 affinities and the higher pharmacological potency. Taken together, results reported here support the hypothesis that the s-trans conformation of AAIs such as 1 is the preferred conformation for interaction at both the CB1 and CB2 receptors and that aromatic stacking may be an important interaction for AAIs at these receptors.

Synthesis and in vivo studies of a selective ligand for the dopamine transporter: 3 beta-(4-[125I]iodophenyl) tropan-2 beta-carboxylic acid isopropyl ester ([125I]RTI-121).

A selective ligand for the dopamine transporter 3 beta-(4-iodophenyl)tropan-2 beta-carboxylic acid isopropyl ester (RTI-121) has been labeled with iodine-125 by electrophilic radioiododestannylation. The [125I]RTI-121 was obtained in good yield (86 +/- 7%, n = 3) with high radiochemical purity (> 99%) and specific radioactivity (1210-1950 mCi/mumol). After i.v. administration of [125I]RTI-121 to mice, the rank order of regional brain tissue radioactivity (striatum > olfactory tubercles > > cortex, hippocampus, thalamus, hypothalamus, cerebellum) was consistent with dopamine transporter labeling. Specific in vivo binding in striatum and olfactory tubercles was saturable, and was blocked by the dopamine transporter ligands GBR 12,909 and (+/-)-nomifensine. By contrast, binding was not reduced by paroxetine, a serotonin transporter inhibitor, or desipramine, a norepinephrine transporter inhibitor. A variety of additional drugs having high affinities for recognition sites other than the neuronal dopamine transporter also had no effect. The [125I]RTI-121 binding in striatum and olfactory tubercles was inhibited by d-amphetamine in dose-dependent fashion. Nonmetabolized radioligand represents 85% of the signal observed in extracts of whole mouse brain. Thus, [125I]RTI-121 is readily prepared, and is a useful tracer for dopamine transporter studies in vivo.

The design, synthesis and testing of desoxy-CBD: further evidence for a region of steric interference at the cannabinoid receptor.

Cannabidiol CBD, a non-psychoactive constituent of marihuana, has been reported to possess essentially no affinity for cannabinoid CB1 receptor binding sites in the brain. Our hypothesis concerning CBD's lack of affinity for the cannabinoid CB1 receptor is that CBD is not capable of clearing a region of steric interference at the CB1 receptor and thereby not able to bind to this receptor. We have previously characterized this region of steric interference at the CB1 receptor [P.H. Reggio, A.M. Panu, S. Miles J. Med. Chem. 36, 1761-1771 (1993)] in three dimensions using the Active Analog Approach. We report here a conformational analysis of CBD which, in turn, led to the design of a new analog, desoxy-CBD. Modeling results for desoxy-CBD predict that this compound is capable of clearing the region of steric interference by expending 3.64 kcal/mol of energy in contrast to the 12.39 kcal/mol expenditure required by CBD. Desoxy-CBD was synthesized by condensation of 3-pentylphenol with p-mentha-2,8-dien-1-ol mediated by DMF-dineopental acetal. Desoxy-CBD was found to behave as a partial agonist in the mouse vas deferens assay, an assay which is reported to detect the presence of cannabinoid receptors. The compound produced a concentration related inhibition of electrically-evoked contractions of the mouse vas deferens, possessing an IC50 of 30.9 nM in this assay. Taken together, these results support the hypothesis of the existence of a region of steric interference at the CB1 receptor. While the energy expenditure to clear this region was too high for the parent compound, CBD, the removal of the C6' hydroxyl of CBD produced a molecule (desoxy-CBD) able to clear this region and produce activity, albeit at a reduced level.

A rational search for the separation of psychoactivity and analgesia in cannabinoids.

The compound 9-beta-hydroxy-hexahydrocannabinol [(-)-9 beta-OH-HHC] was designed to fit a combined theoretical profile of an analgesic cannabinoid (equatorial alcohol at C-9, phenol at C-1 and a C-3 side chain) with reduced psychoactivity (axial C-9 substituent which protrudes into the alpha face). (-)-9 beta-OH-HHC was synthesized by the addition of methyl Grignard to 9-oxo-11-nor-HHC. Its alpha epimer was obtained by the regiospecific epoxide ring opening of 9 alpha, 10 alpha-epoxy-HHC acetate. (-)-9 beta-OH-HHC and (-)-9 alpha-OH-HHC were each evaluated in a battery of tests in mice and were found to be 10-25 times less potent than (-)-trans-delta 9-tetrahydrocannabinol (delta 9-THC) in all tests including the tail flick test for antinociception (analgesia). Molecular mechanics calculations [MMP2(85)] revealed that, in the global minimum energy conformation of (-)-9 beta-OH-HHC, the axial methyl at C-9 protrudes into the alpha face of the molecule, while the axial hydroxyl at C-9 in (-)-9 alpha-OH-HHC protrudes into this same face. These calculations also identified a higher energy carbocyclic ring (twist) conformer of each in which there is no protrusion of a C-9 substituent of the carbocyclic ring into the alpha face. The minimal activity of both compounds is attributed to these higher energy forms.(ABSTRACT TRUNCATED AT 250 WORDS)

High-affinity cannabinoid binding sites in brain membranes labeled with [3H]-5'-trimethylammonium delta 8-tetrahydrocannabinol.

The binding of [3H]-5'-trimethylammonium delta 8-tetrahydrocannabinol (THC) [( 3H]TMA) to rat neuronal membranes was studied. TMA is a positively charged analog of delta 8THC modified on the 5' carbon, a portion of the molecule not important for its psychoactivity. Unlabeled TMA inhibits field-stimulated contractions of the guinea-pig ileum (IC50 = 1 microM) in the same presynaptic manner as delta 9THC. [3H]TMA binds saturably and reversibly to brain membranes with high affinity (KD = 89 nM) to apparently one class of site (Hill coefficient, 1.1). Highest binding site density occurs in the brain, but several peripheral organs also display specific binding. Detergent solubilizes the sites without affecting their pharmacological properties. Molecular sieve chromatography reveals a bimodal peak of [3H]TMA binding activity of approximately 60,000 daltons apparent molecular weight. delta 9THC competitively inhibits [3H]TMA binding potently (Ki = 27 nM) and stereoselectively. For some cannabinoids potency in behavioral and physiological tests parallels their affinity for the [3H]TMA binding site. However, several nonpsychotropic cannabinoids are active at the binding site.

Radioimmunoassays for cannabinoids.

The simplicity, sensitivity, and specificity of radioimmunoassay have made it an attractive procedure for the analysis of delta-9-tetrahydrocannabinol (THC) in biological fluids or tissues. The presence of closely related compounds such as metabolites may interfere with radioimmunoassay results. Appropriate design of immunogens may diminish such interference. This work has been directed towards the use of the amyl side chain for linking cannabinoid compounds to proteins to form immunogens. Although the amyl side chain is metabolized to some extent, the metabolites are not quantitatively significant in most cases. 5'-Carboxy-delta-8-THC and 5'-carboxy-delta-9-THC were linked to bovine serum albumin. Immunization of rabbits with the resulting conjugates resulted in the formation of antisera with high selectivity for delta-9-THC vs. its carboxylic acid metabolite, 11-nor-9-carboxy-delta-9-THC. Delta-8-THC radioligands (4',5'-tritium and 5'-iodine-125) could be used with these antisera for analysis of delta-9-THC in plasma. Sensitivity with tritium-labeled material is about 2.5 ng/ml. 5'-Oxo-11-nor-9-carboxy-delta-8-THC was used to prepare an immunogen which led to the generation of an antiserum highly specific for 11-nor-9-carboxy-delta-9-THC. This antiserum and iodine-125-5'-iodo-11-nor-9-carboxy-delta-8-THC were used to develop a highly specific assay for 11-nor-9-carboxy-delta-9-THC in plasma.

Comparative receptor binding analyses of cannabinoid agonists and antagonists.

To further characterize neuronal cannabinoid receptors, we compared the ability of known and novel cannabinoid analogs to compete for receptor sites labeled with either [3H]SR141716A or [3H]CP-55,940. These efforts were also directed toward extending the structure-activity relationships for cannabinoid agonists and antagonists. A series of alternatively halogenated analogs of SR141716A were synthesized and tested in rat brain membrane binding assays along with the classical cannabinoids, Delta9-tetrahydrocannabinol, cannabinol, cannabidiol, the nonclassical cannabinoid CP-55,940, the aminoalkylindole WIN55212-2 and the endogenous fatty acid ethanolamide, anandamide. Saturation binding isotherms were performed with both radioligands, as were displacement studies, allowing an accurate comparison to be made between the binding of these various compounds. Competition studies demonstrated that all of the compounds were able to displace the binding of [3H]CP-55,940 with rank order potencies that agreed with previous studies. However, the rank order potencies of these compounds in competition studies with [3H]SR141716A differed significantly from those determined with [3H]CP-55,940. These results suggest that CP-55,940, WIN55212-2 and other agonists interact with cannabinoid binding sites within the brain which are distinguishable from the population of binding sites for SR141716A, its analogs and cannabidiol. Structural modification of SR141716A significantly altered the affinity of the compound and its relative ability to displace either [3H]CP-55,940 or [3H]SR141716A preferentially within the rat brain receptor membrane preparation.

Investigation of the role of the phenolic hydroxyl in cannabinoid activity.

Structure-activity relationship studies have suggested that the phenolic hydroxyl group is essential for the pharmacological activity of the cannabinoids. However, it remains to be established whether it is the hydrogen of the phenolic hydroxyl that is important (possibly because this hydrogen can participate in a hydrogen bonding interaction) or whether it is the oxygen of the phenolic hydroxyl that is important (possibly because one of the lone pairs of electrons in this oxygen can serve as a hydrogen bond acceptor). Two new etherified cannabinoids were prepared in which the phenolic hydroxyl oxygen is incorporated into a fourth ring. These new compounds were designed to test the importance both of the phenolic hydroxyl oxygen and of the orientation of its lone pairs of electrons for cannabinoid pharmacological activity. O,2-Propano-delta 8-tetrahydrocannabinol (0,2-Propano-delta 8-THC) was designed to mimic delta 9-THC in its phenol conformation I (C2-C1-O-H = 7 degrees). O,10-Methano-delta 9-tetrahydro-cannabinol (0,10-Methano-delta 9-THC) was designed to mimic delta 9-THC in its phenol conformation II (C2-C1-O-H = 167 degrees). Molecular mechanics calculations revealed that 1) there are two accessible minimum energy conformers for O,2-propano-delta 8-THC, which differ principally in the conformation of the new fourth ring, and 2) there are three accessible minimum energy conformers for O,10-methano-delta 9-THC, the first two of which differ mainly in the conformation of the new fourth ring, whereas the third possesses an alternate pyran ring conformation. Wave functions and molecular electrostatic potential (MEP) maps were calculated for each accessible conformer of O,2-propano-delta 8-THC and of O,10-methano-delta 9-THC. The resultant MEP maps compared well with the corresponding MEP maps generated for delta 9-THC in each of its two minimum energy conformations (two phenolic hydroxyl positions). These results imply that 1) O,2-propano-delta 8-THC should be capable of being recognized at a site that would recognize delta 9-THC in its phenol conformation 1 and 2) O,10-methano-delta 9-THC should be capable of being recognized at a site that would recognize delta 9-THC in its phenol conformation II. Pharmacological evaluation of the analogs revealed that O,10-methano-delta 9-THC was inactive in all mouse tests, as well as the rat drug discrimination model. O,2-Propano-delta 8-THC was similar to delta 8-THC in that it depressed rectal temperature and produced antinociception and ring immobility in mice.(ABSTRACT TRUNCATED AT 250 WORDS)

Selective labeling of the dopamine transporter by the high affinity ligand 3 beta-(4-[125I]iodophenyl)tropane-2 beta-carboxylic acid isopropyl ester.

The iodine-125 analog of the dopaminergically selective cocaine analog 3 beta-(4-iodophenyl)tropane-2 beta-carboxylic acid isopropyl ester (RT1-121) was evaluated as a probe for the dopamine transporter in rat striatum. Saturation and kinetic studies indicated that [125I]RTI-121 binds to both high and low affinity components. The Kd of the high affinity component was 0.14 +/- 0.01 nM (mean +/- standard error), whereas the low affinity component demonstrated an affinity of 1.59 +/- 0.09 nM. The corresponding numbers of striatal binding sites labeled by [125I]RTI-121 were 295 +/- 6 and 472 +/- 59 pmol/g of tissue (original wet weight), respectively. Intrastriatal injections of 6-hydroxydopamine eliminated > 90% of specific [125I]RTI-121 binding in the striatum. The pharmacological profile of specific [125I]RTI-121 binding in the rat striatum was consistent with that of the dopamine transporter. There was a strong (r = 0.98, p < 0.0001) correlation between the potencies of drugs that displaced specific [125I]RTI-121 binding and the potencies of these drugs to inhibit the uptake of [3H]dopamine. In contrast, no correlation was found for the potencies of drugs to inhibit the uptake of either [3H]norepinephrine or [3H]serotonin. Autoradiographs produced using [125I]RTI-121 demonstrated a distribution of label consistent with the distribution of dopaminergic neurons in rat brain. Because of its high affinity and high selectivity for the dopamine transporter, [125I]RTI-121 may be an extremely useful ligand for the dopamine transporter.

Mapping dopamine transporters in the human brain with novel selective cocaine analog [125I]RTI-121.

The novel cocaine analog RTI-121 [3 beta-(4-iodophenyl)tropane-2 beta-carboxylic acid isopropyl ester] was evaluated as a probe for the in vitro labeling and localization of the dopamine transporter in the human brain. Saturation binding experiments conducted in sucrose-phosphate buffer (10 mM sodium phosphate, pH 7.4, 0.32 M sucrose) revealed high- and low-affinity binding components with affinity values (KD) of 0.25 +/- 0.04 and 4.9 +/- 1.6 nM (mean +/- SE) and densities (Bmax) of 56.8 +/- 13.8 and 147.7 +/- 23.4 pmol/g tissue, respectively. In contrast, when saturation binding experiments were performed in phosphate-buffered saline (10 mM Na2HPO4, 1.8 mM KH2PO4, 136 mM NaCl, 2.8 mM KCl, 10 mM NaI, pH 7.4), a 9-fold decrease in the density of the low-affinity component was noted, suggesting that the low-affinity RTI-121 binding site is associated with the region of the transporter involved in the ionic dependence of substrate recognition and/or uptake. The rank order of potency for inhibition of [125I]RTI-121 binding to human caudate membranes demonstrates that the radioligand selectively labels the dopamine transporter (GBR 12909 > RTI-121 > mazindol > nomifensine > (-) cocaine > desipramine > citalopram). Autoradiographic mapping of [125I]RTI-121 revealed very high densities of cocaine recognition sites over areas known to be rich in dopaminergic innervation, including the caudate, putamen, and nucleus accumbens. Moderate densities were also observed over the substantia nigra and the ventral tegmental area. Low-to-background labeling of [125]RTI-121 was seen throughout the cerebral cortex, amygdaloid nuclei, globus pallidus, and thalamus. In comparison with the autoradiographic distribution of the cocaine analogs [3H]WIN 35,428 (or CFT) and [125I]RTI-55 (or beta-CIT), the labeling pattern for [125I]RTI-121 was more restricted. These studies demonstrate that [125I]RTI-121 labels dopamine-rich brain regions with greater selectivity than other currently available cocaine analogs, which makes it a potentially superior imaging probe for mapping the dopamine transporter in the human brain.