The direct actions of cannabidiol and 2-arachidonoyl glycerol at GABAA receptors.

Cannabidiol (CBD) is a major non-intoxicating component of cannabis and possesses anti-epileptic, anxiolytic and anti-hyperalgesic properties. The mechanism of action of CBD in producing such effects remains unclear. Despite evidence that some endogenous and synthetic cannabinoids interact with GABAA receptors, no-one has yet investigated the effects of CBD. Here we used two-electrode voltage clamp electrophysiology to compare the actions of CBD with those of the major central endocannabinoid, 2-arachidonoyl glycerol (2-AG) on human recombinant GABAA receptors (synaptic α1-6ßγ2 and extrasynaptic α4ß2δ) expressed on Xenopus oocytes. CBD and 2-AG were positive allosteric modulators at α1-6ßγ2 receptors, with low micromolar potencies. The maximal level of enhancement seen with either CBD or 2-AG were on α2-containing GABAA receptor subtypes, with approximately a 4-fold enhancement of the GABA EC5 evoked current, more than twice the potentiation seen with other α-subunit receptor combinations. Further we observed ß-subunit selectivity, whereby modulatory activity was higher at ß2/ß3 over ß1 subunits. The ß1-subunit homologous mutant ß2(V436T) substantially diminished the efficacy of both drugs to a third of that obtained with wild-type ß2 subunit combinations, but without changing potency. The potency of CBD increased and efficacy preserved in binary α1/α2ß2 receptors indicating that their effects do not involve the classic benzodiazepine site. Exploration of extrasynaptic α4ß2δ receptors revealed that both compounds enhanced GABA EC5 evoked currents at concentrations ranging from 0.01-1µM. Taken together these results reveal a mode of action of CBD on specifically configured GABAA receptors that may be relevant to the anticonvulsant and anxiolytic effects of the compound.

A pharmacological characterization of GABA, THIP and DS2 at binary α4ß3 and ß3δ receptors: GABA activates ß3δ receptors via the ß3(+)δ(-) interface.

There is growing evidence that GABA (γ-aminobutyric acid) can activate GABAA receptors (GABAARs) in the absence of an α subunit. In this study, we compared the pharmacology of homomeric and binary α4, ß3 or δ subunits with ternary α4ß3δ to identify subunit interfaces that contribute to the pharmacology of GABA, THIP, and DS2, and the antagonists, Zn(2+), gabazine and bicuculline. ß3δ receptors form functional GABA-gated channels when expressed in Xenopus oocytes with a pharmacology that differs to homomeric ß3, binary α4ß3 and ternary α4ß3δ receptors. GABA had similar potency at α4ß3 and ß3δ receptors (25µM and 26µM, respectively) but differed at α4ß3δ receptors where GABA exhibited a biphasic concentration-response (EC50 (1)=12.6nM; EC50 (2)=6.3µM). THIP activated ß3δ receptors (EC50=456µM) but was a more potent activator of α4ß3 (EC50=27µM) and α4ß3δ receptors (EC50 (1)=27.5nM; EC50 (2)=29.5µΜ), indicating that the α4 subunit significantly contribute to its potency. The δ-preferring modulator, DS2 had marginal or no effect at ß3δ and α4ß3 receptors, indicating a role for both the α4 and δ subunits for its potency. Gabazine inhibited GABA-elicited currents at ß3δ receptors whereas bicuculline activated these receptors. Mutational analysis verified that GABA binds to the ß3(+)δ(-) interface formed by the ß3 and δ subunits. In conclusion, evaluating agents against binary GABAARs such as ß3δ and α4ß3 receptors enables identification of interfaces that may contribute to the pharmacology of the more complex ternary α4ß3δ receptors.

A hydrophobic area of the GABA ρ1 receptor containing phenylalanine 124 influences both receptor activation and deactivation.

Experimental evidence suggests that GABA ρ1 receptors are potential therapeutic targets for the treatment of a range of neurological conditions, including anxiety and sleep disorders. Homology modelling of the GABA ρ1 extracellular N-terminal domain has revealed a novel hydrophobic area that extends beyond, but not including the GABA-binding site. Phenylalanine 124 (F124) is predicted to be involved in maintaining the structural integrity of the orthosteric-binding site. We have assessed the activity of a series of GABA ρ1 receptors that incorporate a mutation at F124. Wild-type and mutant human GABA ρ1 subunits were expressed in Xenopus laevis oocytes and AD293 cells, and the pharmacology and kinetic properties of the receptors were measured using electrophysiological analysis. Mutation of F124 had minimal effect on receptor pharmacology. However, the rate of deactivation was significantly increased compared to wild type. This study provides further information about the role of residues within a novel hydrophobic area of the GABA ρ1 receptor. This knowledge can help future studies into the design of potent and subtype-selective ligands with therapeutic value.

Regional Fos-expression induced by γ-hydroxybutyrate (GHB): comparison with γ-butyrolactone (GBL) and effects of co-administration of the GABAB antagonist SCH 50911 and putative GHB antagonist NCS-382.

γ-Hydroxybutyrate (GHB) has a complex array of neural actions that include effects on its own high-affinity GHB receptor, the release of neuroactive steroids, and agonist actions at GABAA and GABAB receptors. We previously reported partial overlap in the c-Fos expression patterns produced by GHB and the GABAB agonist, baclofen in rats. The present study extends these earlier findings by examining the extent to which GHB Fos expression and behavioral sedation are prevented by (2S)-(+)-5,5-dimethyl-2-morpholineacetic acid (SCH 50911), a GABAB antagonist, and NCS-382, a putative antagonist at the high-affinity GHB receptor. We also compare Fos expression caused by GHB and its precursor γ-butyrolactone (GBL), which is a pro-drug for GHB but lacks the high sodium content of the parent GHB molecule. Both GHB (1,000 mg/kg) and GBL (600 mg/kg) induced rapid sedation in rats that lasted over 90 min and caused similar Fos expression patterns, albeit with GBL causing greater activation of the nucleus accumbens (core and shell) and dentate gyrus (granular layer). Pretreatment with SCH 50911 (100mg/kg) partly reversed the sedative effects of GHB and significantly reduced GHB-induced Fos expression in only four regions: the tenia tecta, lateral habenula, dorsal raphe and laterodorsal tegmental nucleus. NCS-382 (50mg/kg) had no effect on GHB-induced sedation or Fos expression. When given alone, both NCS-382 and SCH 50911 increased Fos expression in the bed nucleus of the stria terminalis, central amygdala, parasubthalamic nucleus and nucleus of the solitary tract. SCH 50911 alone affected the Islands of Calleja and the medial, central and paraventricular thalamic nuclei. Overall, this study shows a surprising lack of reversal of GHB-induced Fos expression by two relevant antagonists, both of which have marked intrinsic actions. This may reflect the limited doses tested but also suggests that GHB Fos expression reflects mechanisms independent of GHB and GABAB receptors.

Mixed antagonistic effects of bilobalide at rho1 GABAC receptor.

Bilobalide was found to be a moderately potent antagonist with a weak use-dependent effect at recombinant human rho(1) GABA(C) receptors expressed in Xenopus oocytes using two-electrode voltage clamp methodology. Antagonism of bilobalide at homomeric rho(1) GABA(C) receptors appeared to be mixed. At low concentration, bilobalide (3 microM) caused a parallel right shift and surmountable GABA maximal response of the GABA dose-response curve characteristic of a competitive antagonist. At high concentrations, bilobalide (10-100 microM) caused nonparallel right shifts and reduced maximal GABA responses of GABA dose-response curves characteristic of a noncompetitive antagonist. The potency of bilobalide appears to be dependent on the concentrations of GABA and was more potent at lower GABA concentrations. The mechanism of action of bilobalide at rho(1) GABA(C) receptors appears to be similar to that of the chloride channel blocker picrotoxinin.

GABAC receptor ion channels.

1. The present review gives an overview of studies conducted on GABAC receptors over the past 10 years since the author started at the University of Sydney. It concentrates on the structure-activity relationship profiles of the receptor and how these studies were used to: (i) develop selective GABAC receptor ligands; and (ii) understand the impact of amino acid changes on GABAC receptor pharmacology and function. 2. Structure-activity relationship studies involving variations of both ligands and their receptor targets are vital to the discovery of drugs that interact selectively with particular native and mutant receptor subtypes. Such agents may be useful for treating anxiety, depression, epilepsy and memory related disorders, such as Alzheimer's disease.

The effects of cyclopentane and cyclopentene analogues of GABA at recombinant GABA(C) receptors.

The pharmacological effects of the enantiomers of cis-3-aminocyclopentanecarboxylic acids ((+)- and (-)-CACP), the enantiomers of trans-3-aminocyclopentanecarboxylic acids ((+)- and (-)-TACP), and the enantiomers of 4-aminocyclopent-1-ene-1-carboxylic acids ((+)- and (-)-4-ACPCA) were studied on human homomeric rho(1) and rho(2) GABA(C) receptors expressed in Xenopus oocytes using two-electrode voltage clamp methods. These compounds are conformationally restricted analogues of gamma-aminobutyric acid (GABA) held in a five-membered ring. (+)-TACP (EC(50) (rho(1))=2.7+/-0.2 microM; EC(50) (rho(2))=1.45+/-0.22 microM), (+)-CACP (EC(50) (rho(1))=26.1+/-1.1 microM; EC(50) (rho(2))=20.1+/-2.1 microM) and (-)-CACP (EC(50) (rho(1))=78.5+/-3.5 microM; EC(50) (rho(2))=63.8+/-23.3 microM) were moderately potent partial agonists at rho(1) and rho(2) GABA(C) receptors, while (-)-TACP (100 microM inhibited 56% and 62% of the current produced by 1 microM GABA at rho(1) and rho(2) receptors, respectively) was a weak partial agonist with low intrinsic activity at these receptors. In contrast, (+)-4-ACPCA (K(i) (rho(1))=6.0+/-0.1 microM; K(i) (rho(2))=4.7+/-0.3 microM) did not activate GABA(C) rho(1) and rho(2) receptors but potently inhibited the action of GABA at these receptors, while (-)-4-ACPCA had little effect as either an agonist or an antagonist. The affinity order at both GABA(C) rho(1) and rho(2) receptors was (+)-TACP>(+)-4-ACPCA >> (+)-CACP>(-)-CACP >> (-)-TACP >> (-)-4-ACPCA. This study shows that the cyclopentane and cyclopentene analogues of GABA affect GABA(C) receptors in a unique manner, defining a preferred stereochemical orientation of the amine and carboxylic acid groups when binding to GABA(C) receptors. This is exemplified by the partial agonist, (+)-TACP, and the antagonist, (+)-4-ACPCA.

1-Phenylpyrazolo[3,4-d]pyrimidines; structure-ac:tivity relationships for C6 substituents at A1 and A2A adenosine receptors.

Substitution of I-phenylpyrazolo[3,4-d]pyrimidines at C6 with N-alkyl-2-thiopropionamide groups has resulted in a series of 18 compounds which have been evaluated for binding at A1 and A2A adenosine receptors. Introduction of an N-ethyl group gave increased affinity at both A1 and A2A receptors for the amino compound 7b compared to the primary amide 7a. An additional hydrophobic pocket exists for substituents on the amide. This pocket allows an N-ethyl group for increased affinity at both A1 and A2A receptors, allows larger alkyl groups at A2A receptors but not at A1 receptors and there is an H-bond interaction requiring one H-bond donor. Molecular modeling studies have also enabled a proposal of the amino acid residues involved in ligand binding at both the A1 and A2A receptors.

(+)- and (-)-cis-2-aminomethylcyclopropanecarboxylic acids show opposite pharmacology at recombinant rho(1) and rho(2) GABA(C) receptors.

The effects of the enantiomers of (+/-)-CAMP and (+/-)-TAMP [(+/-)-cis- and (+/-)-trans-2-aminomethylcyclopropanecarboxylic acids, respectively], which are cyclopropane analogues of GABA, were tested on GABA(A) and GABA(C) receptors expressed in Xenopus laevis oocytes using two-electrode voltage clamp methods. (+)-CAMP was found to be a potent and full agonist at homooligomeric GABA(C) receptors (K:(D) approximately 40 microM: and I:(max) approximately 100% at rho(1); K:(D) approximately 17 microM: and I:(max) approximately 100% at rho(2)) but a very weak antagonist at alpha(1)beta(2)gamma(2L) GABA(A) receptors. In contrast, (-)-CAMP was a very weak antagonist at both alpha(1)beta(2)gamma(2L) GABA(A) receptors and homooligomeric GABA(C) receptors (IC(50) approximately 900 microM: at rho(1) and approximately 400 microM: at rho(2)). Furthermore, (+)-CAMP appears to be a superior agonist to the widely used GABA(C) receptor partial agonist cis-4-aminocrotonic acid (K:(D) approximately 74 microM: and I:(max) approximately 78% at rho(1); K:(D) approximately 70 microM: and I:(max) approximately 82% at rho(2)). (-)-TAMP was the most potent of the cyclopropane analogues on GABA(C) receptors (K:(D) approximately 9 microM: and I:(max) approximately 40% at rho(1); K:(D) approximately 3 microM: and I:(max) approximately 50-60% at rho(2)), but it was also a moderately potent GABA(A) receptor partial agonist (K:(D) approximately 50-60 microM: and I:(max) approximately 50% at alpha(1)beta(2)gamma(2L) GABA(A) receptors). (+)-TAMP was a less potent partial agonist at GABA(C) receptors (K:(D) approximately 60 microM: and I:(max) approximately 40% at rho(1); K:(D) approximately 30 microM: and I:(max) approximately 60% at rho(2)) and a weak partial agonist at alpha(1)beta(2)gamma(2L) GABA(A) receptors (K:(D) approximately 500 micro: and I:(max) approximately 50%). None of the isomers of (+/-)-CAMP and (+/-)-TAMP displayed any interaction with GABA transport at the concentrations tested. Molecular modeling based on the present results provided new insights into the chiral preferences for either agonism or antagonism at GABA(C) receptors.

Quantitative autoradiography of Na+-dependent [3H]L-aspartate binding to L-glutamate transporters in rat brain: structure-activity studies using L-trans-pyrrolidine-2,4-dicarboxylate (L-t-PDC) and 2-(carboxycyclopropyl)-glycine (CCG).

Sodium-dependent binding of [3H]L-aspartate was studied in thaw-mounted horizontal sections of fresh-frozen (i.e. not fixed) rat brain. After the incubation with [3H]L-aspartate, the sections were exposed against a 3H-sensitive film and the resulting autoradiograms were evaluated by quantitative densitometry. Effects of several inhibitors were examined and their potency expressed as IC50 and nH. Together with previously published data, the present study supports the view that [3H]L-aspartate binding to fresh-frozen sections of rat brain represents interaction of the radioligand with the substrate-binding sites on glutamate transporters. The most potent inhibitors were (2S,3S,4R)-2-(carboxycyclopropyl)-glycine (L-CCG III) and (2S,4R)-4-methylglutamate. In contrast, L-anti,endo-3,4-methanopyrrolidine dicarboxylate (L-a,e-MPDC) was about an order of magnitude less potent. Only subtle regional variations in the characteristics of inhibitors of [3H]L-aspartate binding were detected. It is not certain whether these differences reflect regional variations in the distribution of individual glutamate transporters or regional peculiarities in their pharmacological characteristics. In particular, (2S,4R)-4-methylglutamate, shown previously to differentiate between GLT-1 (principal glutamate transporter in the forebrain) and GLAST (expressed mainly in the cerebellum), did not strongly differentiate between the binding of [3H]L-aspartate in forebrain and cerebellum. Computer-assisted molecular modelling using selected glutamate analogues with restricted conformation (L-trans-pyrrolidine-2,4-dicarboxylate and four isomers of 2-(carboxycyclopropyl)-glycine: L- and D-CCG I, L-CCG III and L-CCG IV) identified at least one area of unfavourable steric interaction. We conclude that the quantitative autoradiographic studies using [3H]L-aspartate or other transporter-specific ligands, will be a useful tool to study the pharmacology of substrate binding sites on glutamate transporters in the mammalian brain in situ.

The 'ABC' of GABA receptors: a brief review.

1. In the mammalian central nervous system, GABA is the main inhibitory neurotransmitter. GABA is a highly flexible molecule and, thus, can exist in many low-energy conformations. Conformationally restricted analogues of GABA have been used to help identify three major GABA receptors, termed GABAA, GABAB and GABAC receptors. 2. GABAA and GABAC receptors are members of a super-family of transmitter-gated ion channels that include nicotinic acetylcholine, strychnine-sensitive glycine and 5HT3 receptors. GABAA receptors are hetero-oligomeric Cl- channels that are selectively blocked by the alkaloid bicuculline and modulated by steroids, barbiturates and benzodiazepines. To date, 16 human GABAA receptor cDNA have been cloned. 3. GABAB receptors are seven transmembrane receptors that are coupled to G-proteins and activate second messenger systems and Ca2+ and K+ ion channels. To date, three GABAB receptor proteins have been cloned and these resemble metabotropic glutamate receptors. GABAB receptors are hetero-oligomeric receptors made up of a mixture of a combination of the subunits. These receptors are selectively activated by (-)-baclofen and CCGP27492 and are blocked by phaclofen, the phosphonic acid analogue of baclofen. 4. In contrast, GABAC receptors represent a relatively simple form of transmitter-gated Cl- channel made up of a single type of protein subunit. Two human GABAC receptor cDNA have been cloned. These receptors are not blocked by bicuculline nor are they modulated by steroids, barbiturates or benzodiazepines. Instead, GABAC receptors are selectively activated by the conformationally restricted analogues of GABA in the folded conformation cis-4-aminocrotonic acid and (1s,2R)-2-(aminomethyl)-1-carboxycyclopropane. (1,2,5,6-Tetrahydropyridine-4-yl)methylphosphinic acid, a methylphosphinic acid analogue of GABA in a partially folded conformation, is a selective antagonist at GABAC receptors.

Aminomethyl-2,6-difluorophenols as a novel class of increased lipophilicity GABA(C) receptor antagonists.

3- and 4-(Aminomethyl)-2,6-difuorophenols were tested for activity against the three major classes of GABA receptors. 4-(Amninomethyl)-2,6difluorophenol was shown to be a competitive and somewhat selective antagonist at p1 GABA(C) receptors expressed in Xenopus oocytes (K(B) = 75.5 microM with a 95% Confidence Interval range of 75.2 microM to 75.8 microM). This is the first in a novel class of increased lipophilicity GABA(C) receptor antagonists with little activity at alpha1beta2gamma2 GABA(A) and GABA(B) receptors.

Stereoselective interaction of thiopentone enantiomers with the GABA(A) receptor.

1. As pharmacokinetic differences between the thiopentone enantiomers seem insufficient to explain the approximately 2 fold greater potency for CNS effects of (-)-S- over (+)-R-thiopentone, this study was performed to determine any enantioselectivity of thiopentone at the GABA(A) receptor, the primary receptor for barbiturate hypnotic effects. 2. Two electrode voltage clamp recording was performed on Xenopus laevis oocytes expressing human GABA(A) receptor subtype alpha1beta2gamma2 to determine relative differences in potentiation of the GABA response by rac-, (+)-R- and (-)-S-thiopentone, and rac-pentobarbitone. Changes in the cellular environment pH and in GABA concentrations were also evaluated. 3. With 3 microM GABA, the EC50 values were (-)-S-thiopentone (mean 26.0+/-s.e.mean 3.2 microM, n=9 cells) >rac-thiopentone (35.9+/-4.2 microM, n=6, P=0.1) >(+)-R-thiopentone (52.5+/-5.0 microM, n=8, P<0.02) >rac-pentobarbitone (97.0+/-11.2 microM, n=11, P<0.01). Adjustment of environment pH to 7.0 or 8.0 did not alter the EC50 values for (+)-R- or (-)-S-thiopentone. 4 Uninjected oocytes responded to >100 microM (-)-S- and R-thiopentone. This direct response was abolished by intracellular oocyte injection of 1,2-bis(2-aminophenoxy)ethane-N, N,N1,N1-tetraacetic acid (BAPTA), a Ca2+ chelating agent. With BAPTA, the EC50 values were (-)-S-thiopentone (20.6+/-3.2 microM, n=8) <(+)-R-thiopentone (36.2+/-3.2 microM, n=9, P<0.005). 5 (-)-S-thiopentone was found to be approximately 2 fold more potent than (+)-R-thiopentone in the potentiation of GABA at GABA(A) receptors expressed on Xenopus oocytes. This is consistent with the differences in potency for CNS depressant effects found in vivo.

GABA(C) receptor antagonists differentiate between human rho1 and rho2 receptors expressed in Xenopus oocytes.

The selective GABA(C) receptor antagonist, (1,2,5,6-tetrahydropyridin-4-yl)methylphosphinic acid (TPMPA), is eight times more potent against human recombinant p receptors than p2 receptors expressed in Xenopus oocytes. (3-Aminopropyl)methylphosphinic acid (CGP35024), the methylphosphinic acid analogue of GABA, and [(E)-3-aminopropen-1-yl]methylphosphinic acid (CGP44530), an open chain analogue of TPMPA, were five and four times, respectively, more potent as antagonists of p1 receptors than as antagonists of p2 receptors. Isoguvacine was a weak partial agonist at both p1 and p2 receptors with intrinsic activities (calculated as a percentage of the maximum whole cell current produced by a maximum dose of GABA) of 45 and 68%, respectively, of the maximum response produced by GABA. In agreement with other workers, it was found that imidazole-4-acetic acid was a partial agonist at both p1 and p2 receptors, showing higher intrinsic activity at p2 than at p1 receptors. The p1 receptor antagonist, trans-4-amino-2-methylbut-2-enoic acid (2-MeTACA), was a partial agonist at p2 receptors with an intrinsic activity of 34%. 2-MeTACA may be useful in differentiating between homo-oligomeric p1 and p2 receptors in native systems. These studies reveal significant differences in the antagonist profile of human recombinant p1 and p2 GABA(C) receptors.

Structural selectivity and molecular nature of L-glutamate transport in cultured human fibroblasts.

Uptake of L-[3H]glutamate by monolayers of fibroblasts cultured from human embryonic skin has been studied in the presence of several nonradioactive structural analogs of glutamate and aspartate. Results have suggested that the structural specificites of glutamate transporters in cultured human fibroblasts are similar to those of glutamate transporters in the mammalian brain. Only subtle differences have been detected: in the mammalian cerebral cortex, enantiomers of threo-3-hydroxyaspartate are almost equipotent as inhibitors of L-[3H]glutamate uptake while, in human fibroblasts, the D-isomer has been found to be an order of magnitude less potent than the corresponding L-isomer. Kinetic analysis of a model in which substrates are recognized by the glutamate transporter binding site(s) as both alpha- and beta-amino acids indicated that such a mechanism cannot explain the apparent negative cooperativity characterizing the effects of D- and L-aspartate. Molecular modeling has been used to estimate the optimum conformation of L-glutamate as it interacts with the transporter(s). Flow cytometry has indicated that all fibroblasts in culture express at least moderate levels of four glutamate transporters cloned from human brain. Small subpopulations (< 3%) of cells, however, were strongly labeled with antibodies against EAAT1 (GLAST) and EAAT2 (GLT-1) transporters. We conclude that these two transporters--known to be strongly expressed in brain tissue--can be principally responsible for the "high affinity" transport of glutamate also in nonneural cells.

Unsaturated phosphinic analogues of gamma-aminobutyric acid as GABA(C) receptor antagonists.

The phosphinic and methylphosphinic analogues of gamma-aminobutyric acid (GABA) are potent GABA(C) receptor antagonists but are even more potent as GABA(B) receptor agonists. Conformationally restricted unsaturated phosphinic and methylphosphinic analogues of GABA and some potent GABA(B) receptor phosphonoamino acid antagonists were tested on GABA(C) receptors in Xenopus oocytes expressing human retinal rho1 mRNA. 3-Aminopropyl-n-butyl-phosphinic acid (CGP36742), an orally active GABA(B) receptor antagonist, was found to be a moderately potent GABA(C) receptor antagonist (IC50 = 62 microM). The unsaturated methylphosphinic and phosphinic analogues of GABA were competitive antagonists of the GABA(C) receptors, the order of potency being [(E)-3-aminopropen-1-yl]methylphosphinic acid (CGP44530, IC50 = 5.53 microM) > [(E)-3-aminopropen-1-yl]phosphinic acid (CGP38593, IC50 = 7.68 microM) > [(Z)-3-aminopropen-1-yl]methylphosphinic acid (CGP70523, IC50 = 38.94 microM) > [(Z)-3-aminopropen-1-yl]phosphinic acid (CGP70522, IC50 > 100 microM). This order of potency differs from that reported for these compounds as GABA(B) receptor agonists, where the phosphinic acids are more potent than the corresponding methylphosphinic acids.

Contrasting modes of action of methylglutamate derivatives on the excitatory amino acid transporters, EAAT1 and EAAT2.

We have investigated the mechanism of action of a series of glutamate derivatives on the cloned excitatory amino acid transporters 1 and 2 (EAAT1 and EAAT2), expressed in Xenopus laevis oocytes. The compounds were tested as substrates and competitive blockers of the glutamate transporters. A number of compounds showed contrasting mechanisms of action on EAAT1 compared with EAAT2. In particular, (2S,4R)-4-methylglutamate and 4-methylene-glutamate were transported by EAAT1, with Km values of 54 microM and 391 microM, respectively, but potently blocked glutamate transport by EAAT2, with Kb values of 3.4 microM and 39 microM, respectively. Indeed, (2S,4R)-4-methylglutamate is the most potent blocker of EAAT2 yet described. (+/-)-Threo-3-methylglutamate also potently blocked glutamate transport by EAAT2 (Kb = 18 microM), but was inactive on EAAT1 as either a substrate or a blocker at concentrations up to 300 microM. In contrast to (2S,4R)-4-methylglutamate, L-threo-4-hydroxyglutamate was a substrate for both EAAT1 and EAAT2, with Km values of 61 microM and 48 microM, respectively. It seems that the chemical nature and also the orientation of the group at the 4-position of the carbon backbone of glutamate is crucial in determining the pharmacological activity. The conformations of these molecules have been modeled to understand the structural differences between, firstly, compounds that are blockers versus substrates of EAAT2 and, secondly, the pharmacological differences between EAAT1 and EAAT2.

Stimulation of [3H] GABA and beta-[3H] alanine release from rat brain slices by cis-4-aminocrotonic acid.

cis-4-Aminocrotonic acid (CACA; 100 microM), an analogue of GABA in a folded conformation, stimulated the passive release of [3H] GABA from slices of rat cerebellum, cerebral cortex, retina, and spinal cord and of beta-[3H]alanine from slices of cerebellum and spinal cord without influencing potassium-evoked release. In contrast, CACA (100 microM) did not stimulate the passive release of [3H]taurine from slices of cerebellum and spinal cord or of D-[3H]aspartate from slices of cerebellum and did not influence potassium-evoked release of [3H]-taurine from the cerebellum and spinal cord and D-[3H]-aspartate from the cerebellum. These results suggest that the effects of CACA on GABA and beta-alanine release are due to CACA acting as a substrate for a beta-alanine-sensitive GABA transport system, consistent with CACA inhibiting the uptake of beta-[3H]alanine into slices of rat cerebellum and cerebral cortex. The observed Ki for CACA against beta-[3H]alanine uptake in the cerebellum was 750 +/- 60 microM. CACA appears to be 10-fold weaker as a substrate for the transporter system than as an agonist for the GABAc receptor. The effects of CACA on GABA and beta-alanine release provide indirect evidence for a GABA transporter in cerebellum, cerebral cortex, retina, and spinal cord that transports GABA, beta-alanine, CACA, and nipecotic acid that has a similar pharmacological profile to that of the GABA transporter, GAT-3, cloned from rat CNS. The structural similarities of GABA, beta-alanine, CACA, and nipecotic acid are demonstrated by computer-aided molecular modeling, providing information on the possible conformations of these substances being transported by a common carrier protein.

1-Phenylpyrazolo[3,4-d]pyrimidines as adenosine antagonists: the effects of substituents at C4 and C6.

Forty-two 1-phenyl-pyrazolo[3,4-d]pyrimidines substituted at C6 with thioethers containing distal amide substituents and substituted at C4 with thiol, thiomethyl or amino were synthesized and tested for adenosine A1 and A2a receptor binding. Compared with a thiol at C4, both S-methylation and conversion to an amino resulted in increased affinity at both receptors with the C4 amino compounds having the highest affinity. The C-4 region of the receptor consists of an alkyl pocket containing a hydrogen-bonding site. The study established that for high affinity at both the A1 and A2a adenosine receptors the distal amide should be separated from the C6 thiol by only one carbon. In this study, 2'-(4-amino-1-phenylpyrazolo[3,4-d]pyrimidin-6-ylthio)-N-ethyl- ethanamide (4b) had the highest affinity at the A1 receptor with a Ki of 12.1 nM while 2'-(4-amino-1-phenylpyrazolo[3,4-d]pyrimidin-6-ylthio)ethanamid e (4a) had the highest affinity at the A2a receptor with a Ki of 44.9 nM.