Chronic intrathecal infusion of phosphorothioate or phosphodiester antisense oligonucleotides against cytokine responsive gene-2/IP-10 in experimental allergic encephalomyelitis of lewis rat.

The role of CRG-2 protein (murine IP-10) in experimental allergic encephalomyelitis (EAE), an animal model of multiple sclerosis, was tested by blocking crg-2 mRNA synthesis using chronic infusion of antisense oligonucleotides into lumbar subarachnoid space by osmotic minipumps. After injection with myelin basic protein, rats developed hind limb paralysis and their spinal cords showed crg-2 mRNA expression and the presence of inflammation. When antisense phosphorothioate oligonucleotides to crg-2 (AS-PScrg2) or nonspecific oligonucleotides (AS-PSscram) were infused at rates from 0.18 to 5.4 nmol/hr, they caused hind limb paralysis that was clinically different from EAE and tissue necrosis of the spinal cord. Natural phosphodiester (AS-PO) oligonucleotides that were infused at rates from 1.8 to 5.4 nmol/hr had no toxic effects. AS-POcrg2 reduced the initial appearance of crg-2 mRNA and this may be responsible for the significant reduction in the EAE scores at the height of clinical disease. The antisense treatment did not alter inflammation of the spinal cord. While AS-PS oligonucleotides were unsuitable for intrathecal administration, AS-POcrg2 were not toxic and reduced paralysis due to EAE.

Muscarinic m3 receptors and dynamics of intracellular Ca2+ cerebellar granule neurons.

Activation of muscarinic receptors with carbachol has no effect on intracellular Ca2+ concentration in cerebellar granule cell cultures. Only after elevating intracellular Ca2+ concentrations using either 40 mM KCl or activating glutamatergic receptors was carbachol able to increase intracellular Ca2+. The response lasted about 10 s, and the median increase in intracellular Ca2+ with either 100 microM or 300 microM carbachol was about 85 nM. Carbachol at 30 microM elicited an increase in intracellular calcium that was half maximal. After a 16 min or 32 min delay following cell depolarization, responses to carbachol were only found in about 20% of the cells studied and the median increase in intracellular Ca2+ was about 14 nM. (-)-Quinuclidinylxanthene-9-carboxylate hemioxalate (QNX) at 10 nM (a muscarinic m3 antagonist), but not methoctramine at 5 microM (a muscarinic m2 antagonist), attenuated the action of 100 microM carbachol. This carbachol-mediated response was elicited in the absence of extracellular Ca2+ and in the presence of 10 microM dantrolene, but was blocked by 1 microM thapsigargin. Lastly, treatments of cerebellar granule cell cultures with carbachol (100 microM) for up to 12 h results in a desensitization of the carbachol-mediated release of stored Ca2+. Thus, carbachol acts on muscarinic m3 receptors to induce the release of Ca2+ from IP3-sensitive Ca2+ stores that must be filled by a prior period of high intracellular Ca2+.

Muscarinic m2 receptors in cerebellar granule cell cultures from rat: mechanism of short-term desensitization.

Cerebellar granule cell cultures of rat express only muscarinic m2 and m3 receptor subtypes and exhibit the pharmacological profile of muscarinic m2 receptors that couple to guanine nucleotide binding proteins to inhibit adenylyl cyclase. In vivo pretreatment with muscarinic agonists desensitizes the muscarinic m2 receptor with 50% complete desensitization within 15 to 20 min. After a 1-hr pretreatment with a maximal concentration of carbachol (short-term desensitization), m2 receptor responsiveness reappeared after a 1-hr treatment of cultures with atropine. However, after a 6-hr pretreatment with carbachol (long-term desensitization), m2 receptor responsiveness did not reappear after 1-hr treatment with atropine. Short-term desensitization was homologous for the m2 receptor because treatment of cultures with carbachol did not alter gamma-aminobutyric acidB receptor-mediated inhibition of adenylyl cyclase. Muscarinic m2 receptor desensitization was not mimicked by the addition of analogs of cyclic AMP, cyclic GMP or diacylglycerol to the cultures. The agonist-induced desensitization was not blocked by a cyclic AMP analog, 8-(4-chlorophenylthio)-cyclic AMP. Pretreatment with antisense oligodeoxynucleotides against the mRNA-encoding beta adrenergic receptor kinase attenuated the desensitization by carbachol (100 microM, 1 hr) of m2 receptors. Irreversible labeling of muscarinic m2 and m3 receptors with [3H]propylbenzilycholine mustard followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis produced a loss of the muscarinic m2 receptor (66-kDa protein), but not the muscarinic m3 receptor (92-kDa protein). We suspect that the short-term desensitization results from the phosphorylation of the muscarinic m2 receptor followed by loss of receptor from the plasma membrane.

Characterization of a decrease in muscarinic m2 mRNA in cerebellar granule cells by carbachol.

Studies involving carbachol (100 microM) treatment of cerebellar granule cells for 1, 3, 6, 9, 12 and 24 hr show a decrease in the mRNA encoding for the muscarinic m2 receptor. The response was transient, decreasing m2 mRNA by 25 to 50% in 6 and 9 hr, respectively. The data presented in this work were quantified by ribonuclease protection assay, using a [32P]-cRNA probe corresponding to nucleotide +1138 to 1650 of the rat m2 muscarinic receptor. Because cerebellar granule cells express muscarinic m2 and m3 receptors, we tested whether the carbachol-mediated decrease in m2 mRNA resulted from a homologous or heterologous activation of muscarinic receptors. At a 1 microM concentration, methoctramine specifically blocked the muscarinic m2 receptor and reversed carbachol's action. These data suggested that carbachol acts via a possible homologous activation of muscarinic m2 receptors. The half-life of the receptor mRNA measured in the presence of actinomycin D with and without carbachol were similar. Because carbachol treatments decrease the steady-state levels of m2 mRNA without changing the half-life of the message, we suggest that a carbachol treatment induces a decrease in the transcription of the gene for the muscarinic m2 receptor.

A specific antisense oligodeoxynucleotide to mRNAs encoding receptors with seven transmembrane spanning regions decreases muscarinic m2 and gamma-aminobutyric acidB receptors in rat cerebellar granule cells.

Many receptors that inhibit adenylyl cyclase belong to a common superfamily of receptors that couple to guanine nucleotide binding proteins. These receptors are thought to span the membrane seven times and to have a highly homologous amino acid sequence (LACADL) in the second membrane spanning region. The antisense oligodeoxynucleotide seven-transmembrane spanning receptor (7TMR), a 15-mer that binds to mRNA encoding this amino acid sequence, was added to cerebellar granule neuron cultures to decrease receptors of this superfamily. Intact antisense 7TMR was found to enter neurons. This 4- to 6-day treatment with antisense oligonucleotide decreased the total number of muscarinic receptor binding sites by about 40% and completely eliminated muscarinic m2 receptors that inhibit cyclic AMP formation. Antisense 7TMR treatment at 25 microM prevented the gamma-aminobutyric acid (GABA)B-mediated inhibition of cyclic AMP formation by about 40%. The treatment was effective in decreasing GABAA receptors only when the antisense oligonucleotide was given 1 day after plating the cells, and the receptor response assay was performed 6 days later. The half-maximal concentration of antisense 7TMR was approximately 5 microM in blocking GABAB receptors. Antisense 7TMR appeared to be specific because another antisense oligodeoxynucleotide sequence (15-mer) having four mismatches with 7TMR had no effect on either muscarinic m2 or GABAB receptor-mediated responses and did not affect the total number of muscarinic binding sites. These results are consistent with the view that antisense oligonucleotides decrease proteins in which the nucleotide sequence is known such as the muscaranic m2 receptor.(ABSTRACT TRUNCATED AT 250 WORDS)

Propylbenzilylcholine mustard has greater specificity for muscarinic m2 receptors than for m3 receptors present in cerebellar granule cell culture from rat.

Both muscarinic m2 receptors that inhibit adenylyl cyclase and m3 receptors that stimulate the hydrolysis of inositol phospholipids are expressed in cerebellar granule cells. In order to determine whether a reserve population of either of these receptors is present in this cell culture, the irreversible muscarinic receptor antagonist, propylbenzilylcholine mustard (PBCM), was used at different concentrations to bind various proportions of available muscarinic receptors. After pretreating the cell cultures with low concentrations of PBCM (< 1 nM), the muscarinic m2 receptor-mediated response decreased. Higher concentrations of PBCM (1-3 nM) were needed to reduce the muscarinic m3 receptor-mediated response. These results suggested that either a reserve population of muscarinic m3 receptors is present or that PBCM shows greater specificity for muscarinic m2 receptors. Because the muscarinic m2 receptor is a 66 kDa protein, whereas the muscarinic m3 receptor is a 92 kDa protein, these receptors can be separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis after being labeled with [3H]PBCM. PBCM appears to specifically bind the 66 kDa m2 receptor at concentrations lower than those required to bind to the 92 kDa m3 receptor. A linear correlation was found between the increased binding of [3H]PBCM to each receptor and the proportional loss of that receptor-mediated response. Thus, a reserve population of either muscarinic m2 or m3 receptors does not appear to exist in cerebellar granule cells. These studies also show that PBCM has greater affinity for the muscarinic m2 receptor than the muscarinic m3 receptor.

Proposed antagonists at GABAB receptors that inhibit adenylyl cyclase in cerebellar granule cell cultures of rat.

The effects of various proposed GABAB receptor antagonists on baclofen-mediated inhibition of adenylyl cyclase were studied in cultured cerebellar granule cells from rat. (+/-)-Baclofen maximally inhibited adenylyl cyclase by approximately 60% of the basal enzyme activity with an EC50 value of 10 microM. 3-Aminopropane sulfonic acid (3-APS) and 5-aminovaleric acid (5-AVA) produced similar responses to that seen with (+/-)-baclofen. Saclofen reversed the action of (+/-)-baclofen, 50 microM, with a half maximal inhibitory concentration (IC50) of about 1.0 mM. The most effective antagonist in blocking the action of (+/-)-baclofen was 3-aminopropyl-diethoxy-methyl-phosphonic acid (CGP 35,348). In the presence of (+/-)-baclofen, 50 microM, the IC50 for CGP 35,348 was 290 microM and its inhibitory constant (KA) was 180 microM. The agonist-like actions of 3-APS and 5-AVA were antagonized by CGP 35,348 suggesting that 3-APS and 5-AVA may act as weak agonists at the GABAB receptor that inhibits adenylyl cyclase. All antagonists tested, except the new compound CGP 35,348, have very low potencies at GABAB receptors that inhibit adenylyl cyclase, though these compounds have been quite effective at other GABAB receptor-mediated events. Thus, the GABAB receptor which inhibits adenylyl cyclase differs pharmacologically from other reported GABAB receptor/effector systems and supports the existence of multiple receptor subtypes.

Stimulation of high affinity gamma-aminobutyric acidB receptors potentiates the depolarization-induced increase of intraneuronal ionized calcium content in cerebellar granule neurons.

In the treatment of spasticity, the therapeutic cerebrospinal fluid levels of (+/-)-baclofen, a gamma-aminobutyric acid (GABA)B receptor agonist, are below 1 microM. However, the mechanism of the therapeutic action of (+/-)-baclofen remains unknown, because, for the most part, the action of (+/-)-baclofen on GABAB receptors requires micromolar concentrations. Using fura-2 fluorescence microscopy, intracellular ionized calcium was measured in cerebellar granule neurons. Stimulation of a high affinity GABAB receptor potentiated by 2-3-fold the rise in intracellular calcium observed after depolarization of the cell with a Krebs Ringer's buffered solution containing 40 mM K+. Both GABA (100 nM) and (+/-)-baclofen (10-100 nM) stimulated this high affinity receptor. The potentiation of the depolarization-induced rise in intracellular calcium by (+/-)-baclofen (100 nM) was completely blocked by the GABAB receptor antagonist CGP 35348 (200 microM). Also, the intracellular calcium response induced by the activation of high affinity GABAB receptors was prevented by dantrolene (10 microM). The cerebellar granule neurons contained calcium-induced calcium release (CICR) stores. Caffeine (3 mM) and ryanodine (100 microM) potentiated the depolarization-induced rise in intracellular calcium, and this response to both drugs was blocked by dantrolene (10 microM). Because dantrolene does not prevent the rise in intracellular calcium after cell depolarization (this calcium originated from the influx of extracellular calcium), (+/-)-baclofen acting via the high affinity GABAB receptor indirectly activates the CICR stores, allowing the influx of extracellular calcium to trigger the release of calcium from these dantrolene-sensitive CICR stores. Thus, this high affinity GABAB receptor might become activated during persistent depolarization caused by pathological states and could be a mechanism to be studied for the therapeutic action of (+/-)-baclofen in spasticity.

Role of central GABAB receptors in physiology and pathology.

There are two major classes of gamma-aminobutyric acid (GABA)-sensitive receptors: GABAA and GABAB. The GABAA receptor, the better known of the two GABA receptors, is a heterooligomeric complex that forms a chloride channel. Multiple subtypes of the GABAA receptor result from the composition of different subunits. In contrast to the GABAA receptor, the GABAB receptor protein has not been isolated and purified to homogeneity. Various effector systems, however, have been identified for the GABAB receptor using a limited GABAB-specific pharmacologic reportoire. In almost all cases, activated GABAB receptors employ a guanosine triphosphate-binding protein to transduce a signal intracellularly. There may be multiple subtypes of the GABAB receptor. Because the responses elicited by activation of GABAB receptors are small in terms of their intensity and are considered to be modulatory, the role these receptors play in the central nervous system (CNS) may not be very obvious. However, it is our view that in a finely tuned instrument such as the brain, treatment with neuromodulators (drugs that produce slight changes in brain neurochemistry) may be safer than most current drugs. Moreover, neuromodulators may have far greater potential as pharmacotherapeutic agents for CNS disorders. Thus, in this article, we will review the pharmacologic characteristics of the GABAB receptor, known physiologic roles that this receptor plays in the CNS, and the importance of this receptor in certain disease states.

gamma-Aminobutyric acid-B receptors inhibit glutamate release from cerebellar granule cells: consequences of inhibiting cyclic AMP formation and calcium influx.

In cerebellar granule cells, baclofen acted with micromolar concentrations at proposed gamma-aminobutyric acid-B receptors to inhibit the formation of cyclic AMP and depolarization-induced release of glutamate. Nanomolar concentrations of baclofen inhibited depolarization-induced influx of calcium. All three responses to baclofen were attenuated after pertussis toxin pretreatment of cell cultures. The inhibition of calcium influx and glutamate release were reversed by the cyclic AMP analog, 8-(4-chlorphenylthio)-cyclic AMP. The release of glutamate was dependent on the influx of extracellular calcium, which enters the cell through dihydropyridine-sensitive voltage-dependent calcium channels. Because the decrease in calcium influx by baclofen and nifedipine were additive, the baclofen-mediated decrease in calcium influx may not involve a dihydropyridine-sensitive calcium channel. These results show similarities between the baclofen-mediated inhibition of cyclic AMP formation and glutamate release, suggesting a relationship between these two events. The baclofen-mediated inhibition of calcium influx may not be related to baclofen's inhibition of glutamate release.

Baclofen inhibits with high affinity an L-type-like voltage-dependent calcium channel in cerebellar granule cell cultures.

In primary cultures of cerebellar granule cells, D,L baclofen (p-chlorophenyl-GABA) inhibited approximately 50% of the calcium-45 influx induced with cell depolarization. The half maximal effective concentration for baclofen was 4 nM. Basal calcium influx was not influenced by baclofen thus suggesting that its inhibitory action could be exerted via a voltage dependent calcium channel (VDCC). Whole-cell recordings by patch-clamp technique showed a calcium current that appeared to be similar to the reported L-type VDCC. Nanomolar concentrations of baclofen also inhibited this calcium current by about 60%. However, in order for baclofen to be active, it needed to be placed into the incubation buffer at least five minutes before patching a cell raising the possibility that baclofen may be acting to inhibit the VDCC via a second messenger system.

Specificity of methoctramine in blocking muscarinic receptors which inhibit adenylate cyclase in cerebellar granule cells.

In primary cultures of cerebellar granule cells, activation of muscarinic receptors stimulates both hydrolysis of phosphatidylinositol (PI) and inhibition of adenylate cyclase. The specificity of three muscarinic receptor antagonists, pirenzepine, methoctramine and (-)quinuclidinyl xanthene-9-carboxylate [(-)QNX], in blocking carbachol-stimulated hydrolysis of PI and inhibition of adenylate cyclase were determined. Pirenzepine was found to be nonspecific in blocking the carbachol-stimulated hydrolysis of PI and inhibition of adenylate cyclase, while methoctramine specifically antagonized carbachol-stimulated inhibition of adenylate cyclase with 600 times greater potency than carbachol-stimulated hydrolysis of PI. (-)Quinuclidinyl xanthene-9-carboxylate was approximately 20 times more potent in blocking the carbachol-stimulated hydrolysis of PI than inhibition of adenylate cyclase. Studies of the ability of these three antagonists to block the binding of [3H]quinuclidinyl benzilate [( 3H]QNB) to muscarinic sites on membranes from cerebellar granule cells, revealed that all three antagonists displayed binding characteristics, characteristic of two binding sites, possibly representing the two types of muscarinic receptors. However, the ratio of the affinities for each of the two binding sites was about ten for pirenzepine, 100 for methoctramine and 650 for (-)QNX. Thus, the specificity of these antagonists, in blocking the inhibition of adenylate cyclase and hydrolysis of PI did not correlate with their specificities obtained with the binding studies with [3H]QNB.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification of muscarinic receptor subtypes present in cerebellar granule cells: prevention of [3H]propylbenzilyl choline mustard binding with specific antagonists.

Subtypes of muscarinic receptors (possible m1 to m5) can be identified by their molecular size, specific effector systems and antagonist specificity. In membranes prepared from primary cultures of cerebellar granule cells, [3H]propylbenzilylcholine mustard [( 3H]PBCM) irreversibly binds to muscarinic receptive proteins, having two major molecular sizes, 92 and 66 kDa. With relatively short periods of incubation (approx. 30 min, 30 degrees C) of [3H]PBCM with atropine, a nonspecific competitive receptor antagonist, the irreversible labeling of these muscarinic proteins by [3H]PBCM could be prevented. Methoctramine, a specific competitive antagonist at muscarinic receptors coupled to inhibition of adenylate cyclase, protected most of the muscarinic receptors having a molecular size of 66 kDa from binding of [3H]PBCM. These 66 kDa receptive proteins are suggested to be muscarinic m2 and m4 subtypes. (-)Quinuclidinyl xanthene-9-carboxylate [(-)QNX], a somewhat specific competitive antagonist at muscarinic receptors coupled to hydrolysis of phosphatidylinositol, prevented the binding of [3H]PBCM to 92 kDa muscarinic receptive proteins and some 66 kDa muscarinic receptive proteins. The 92 kDa receptive proteins are suggested to be the muscarinic m3 subtype and the 66 kDa proteins could be either the m2 or m4 receptor subtype. Lastly, pirenzepine, a nonspecific antagonist at muscarinic receptors mediating inhibition of adenylate cyclase and hydrolysis of PI in these cultures, resembled (-)QNX in preventing binding of [3H]PBCM to the 92 kDa receptive proteins and some 66 kDa receptive proteins. The suggested subtypes of muscarinic receptors, specifically bound by pirenzepine should be the m3 (92 kDa) and the m4 (66 kDa) subtypes, since pirenzepine reportedly exhibits a low affinity for the muscarinic m2 subtype.(ABSTRACT TRUNCATED AT 250 WORDS)

Islet-activating protein inhibits the beta-adrenergic receptor facilitation elicited by gamma-aminobutyric acidB receptors.

gamma-Aminobutyric acidB (GABAB) receptor recognition sites that inhibit cyclic AMP formation, open potassium channels, and close calcium channels are coupled to these effector systems by guanine nucleotide binding proteins (G proteins). These G proteins are ADP-ribosylated by islet-activating protein (IAP), also known as pertussis toxin. This process prevents receptor coupling to these G proteins. In slices of cerebral cortex and hippocampus from rat, stimulation of GABAB receptors with baclofen, a receptor agonist, also potentiates the accumulation of cyclic AMP stimulated by beta-adrenergic agonists. It was unknown whether those GABAB receptors that potentiate the beta-adrenergic response were also sensitive to IAP. IAP was injected intracerebroventricularly into rats to ADP-ribosylate IAP-sensitive G proteins. Four days after the IAP injection, 38% and 52% of these G proteins from cerebral cortex and hippocampus, respectively, were ADP-ribosylated by the IAP injection. In slices of both structures prepared from IAP-treated rats, the GABAB receptor-mediated potentiation of the beta-adrenergic receptor response was attenuated. Thus, many GABAB receptor-mediated responses are coupled to IAP-sensitive G proteins.

Dihydropyridine binding sites regulate calcium influx through specific voltage-sensitive calcium channels in cerebellar granule cells.

In primary cultures of cerebellar granule cells, [3H]nitrendipine binds with high affinity to a single site (KD 1 nM and Bmax 20 fmol/mg protein). The 1,4-dihydropyridine (DHP) class of compounds such as nitrendipine, nifedipine, and BAY K 8644 displace [3H]nitrendipine binding at nanomolar concentrations. Verapamil partially inhibits whereas diltiazem slightly increases the [3H]nitrendipine binding. In these cells, the calcium influx that is induced by depolarization is very rapid and is blocked by micromolar concentrations of inorganic calcium blockers such as cadmium, cobalt, and manganese. The calcium influx resulting from cell depolarization is potentiated by BAY K 8644 and partially inhibited (approximately 40%) by nitrendipine and nifedipine. Other non-DHP voltage-sensitive calcium channel (VSCC) antagonists, such as verapamil and diltiazem, completely blocked the depolarization-induced calcium influx. This suggested that nitrendipine and nifedipine block only a certain population of VSCCs. In contrast, verapamil and diltiazem do not appear to be selective and block all of VSCCs. Perhaps some VSCCs can be allosterically modulated by the binding site for the DHPs, whereas verapamil and diltiazem may block completely the function of all VSCCs by occupying a site that differs from the DHP binding site.

Gamma aminobutyric acid B receptor-mediated inhibition of adenylate cyclase in cultured cerebellar granule cells: blockade by islet-activating protein.

In primary cultures of cerebellar granule cells, the gamma aminobutyric acid B (GABAB) receptor couples to an inhibitory mechanism of adenylate cyclase. The inhibition of adenylate cyclase can be observed either by the measurement of cellular cyclic AMP content or by in vitro measurement of adenylate cyclase from plasma membrane of these cerebellar granule cells. The GABAB receptors can be stimulated by GABA and the GABA analog, baclofen. This receptor-mediated inhibition of adenylate cyclase was blocked by the pertussis toxin, islet-activating protein. Furthermore, the authors show that islet-activating protein catalyzed the ADP ribosylation of the guanine nucleotide inhibitory unit (MW 41,000) in the cerebellar granule cells. In summary, the authors provide evidence supporting the presence of a GABAB receptor in the cerebellar granule cells and suggest that this receptor couples with the guanine nucleotide inhibitory unit to inhibit the formation of cyclic AMP.

Dihydropyridines change the uptake of calcium induced by depolarization into primary cultures of cerebellar granule cells.

The uptake of calcium (Ca++) into cerebellar granule cells in primary culture was increased by depolarizing the cells with either 60 mM KC1 or veratridine. Nitrendipine, at concentrations of 100 nM or greater, antagonized approximately 40 percent of the depolarization induced Ca++ uptake. The half maximal concentration of nitrendipine was 7nM. Furthermore, another dihydropyridine derivative, BAY K 8644 enhanced the uptake of Ca++ and in the presence of nitrendipine, this facilitation of Ca++ uptake was reduced. Thus, these data indicate the existence of voltage dependent Ca++ channels which are sensitive to dihydropyridines in primary cultures of cerebellar granule cells.

Co-localized adenosine A1 and gamma-aminobutyric acid B (GABAB) receptors of cerebellum may share a common adenylate cyclase catalytic unit.

In synaptosomal membranes from rat cerebellum, additive responses to adenylate cyclase activity are observed between the beta adrenergic receptors present on the Purkinje cells and the adenosine A-1 receptors or gamma-aminobutyric acid B (GABAB) receptors, which are both associated with the granule cells. In contrast, nonadditive responses are found with the activation of the adenosine A-1 and the GABAB receptors. Because both receptors are mainly associated with the same cell type, the nonadditive response indicates an interaction between the adenosine A-1, GABAB receptors and the adenylate cyclase. The present study suggests that the nonadditivity results from a limited number of adenylate cyclase catalytic units, which both receptor systems share. This conclusion was derived indirectly by showing that 1) a GABAB agonist did not affect the adenosine A-1 recognition site; 2) both receptors additively activated the high-affinity guanosine 5'-triphosphatase, which is believed to reflect the activation of the inhibitory guanine nucleotide unit; and 3) the nonadditivity was still observed after stimulation of adenylate cyclase activity with forskolin.

gamma-aminobutyric acid B receptors are negatively coupled to adenylate cyclase in brain, and in the cerebellum these receptors may be associated with granule cells.

Baclofen and gamma-aminobutyric acid (GABA) are shown to inhibit basal adenylate cyclase activity in brain of rat. The response is mediated through the GABAB receptor, and the rank order of potency for agonists is (-)-baclofen (EC50 = 4 microM) greater than GABA (EC50 = 17 microM) greater than muscimol greater than (+)-baclofen. GABAA agonists are not effective inhibitors of cyclase activity. The response is bicuculline-insensitive, and diazepam does not modify the GABA or (-)-baclofen inhibition of adenylate cyclase. Studies with neurologically mutant mice correlated a loss in GABAB receptor-mediated inhibition of cyclase with a loss in cerebellar granule cells. Thus, the GABAB receptor is negatively coupled to adenylate cyclase in various brain areas, and, in the cerebellum, data suggest a granule cell localization of this activity.