A selective binding site for 3H-NECA that is not an adenosine A2 receptor

Study of adenosine A2 receptors in membrane preparations from optic tectum of chicks

Binding properties of the subtypes of adenosine A2 receptors in membrane preparations and the effects of adenosine receptor ligands on cAMP accumulation in slices from the optic tectum of neonatal chicks have been investigated. [3H]2-[4-(2-p-carboxyethyl)phenylamino]-5'-N-ethylcarboxaminoadenosin e (CGS 21680), a selective ligand for adenosine A2a receptors, did not bind to optic tectal membranes, as observed with rat striatal membranes. CGS 21680 also did not induce cyclic AMP accumulation in optic tectum slices. However, 5'-N-ethylcarboxamidoadenosine (NECA), 2-chloro-adenosine or adenosine induced a 2.5- to 3-fold increase on cyclic AMP accumulation in this preparation. [3H]NECA binds to fresh non-washed-membranes obtained from optic tectum of chicks, displaying one population of binding sites, which can be displaced by NECA, 8-phenyltheophylline, 2-chloro-adenosine, but is not affected by CGS 21680. The estimated K(D) value was 400.90 +/- 80.50 nM and the Bmax was estimated to be 2.51 +/- 0.54 pmol/mg protein. Guanine nucleotides, which modulate G-proteins activity intracellularly, are also involved in the inhibition of glutamate responses by acting extracellularly. Moreover, we have previously reported that guanine nucleotides potentiate, while glutamate inhibits, adenosine-induced cyclic AMP accumulation in slices from optic tectum of chicks. However, the guanine nucleotides, GMP or GppNHp and the metabotropic glutamate receptors agonist, 1S,3R-ACPD did not alter the [3H]NECA binding observed in fresh non-washed-membranes. Therefore, the adenosine A2 receptor found in the optic tectum must be the adenosine A2b receptor which is available only in fresh membrane preparations, and its not modulated by guanine nucleotides or glutamate analogs.

Enhanced expression of G protein-coupled receptor kinase 2 selectively increases the sensitivity of A2A adenosine receptors to agonist-induced desensitization

1. G protein-coupled receptor kinases (GRKs) are thought to be important in mediating the agonist-induced phosphorylation and consequent desensitization of G protein-coupled receptor (GPCR) responses. We have previously shown that stable expression of a dominant negative mutant G protein-coupled receptor kinase 2 (GRK2) construct in NG108-15 mouse neuroblastoma x rat glioma cells suppresses the agonist-induced desensitization of A2A and A2B adenosine receptor-stimulated adenylyl cyclase activity (Mundell et al., 1997). To further determine the role of GRK2 in agonist-induced desensitization of these adenosine receptors, we stably overexpressed wild type GRK2 in NG108-15 cells. 2. In homogenates prepared from cells overexpressing GRK2, the acute stimulation of adenylyl cyclase by activation of A2A and A2B adenosine receptors was markedly reduced, but could be reversed by pretreating the cells with AD (adenosine deaminase), to remove extracellular adenosine from the medium. On the other hand, acute stimulation of adenylyl cyclase by secretin, iloprost, NaF and forskolin was the same in GRK2 overexpressing cells and plasmid-transfected control cells. 3. Cells overexpressing GRK2 were more sensitive to adenosine receptor agonist-induced desensitization than plasmid-transfected control cells. This effect was selective since the agonist sensitivity of desensitization for secretin and IP-prostanoid receptor-stimulated adenylyl cyclase activity was not affected by GRK2 overexpression. 4. These results further implicate GRK2 as the likely mechanism by which A2 adenosine receptors undergo short-term desensitization in NG108-15 cells, and indicate that even when overexpressed, GRK2 retains its substrate specificity for native receptors in intact cells. Furthermore, the susceptibility of GPCRs to desensitization appears to depend on the level of GRK expression, such that in cells that express high levels of GRK2, low agonist concentrations may be sufficient to trigger GRK-mediated desensitization.

Evidence for co-expression and desensitization of A2a and A2b adenosine receptors in NG108-15 cells

Using receptor-selective agonists and antagonists, the possible presence of both A2a and A2b adenosine receptor subtypes coupled to activation of adenylyl cyclase was investigated in NG108-15 neuroblastoma x glioma hybrid cells. The relatively non-selective adenosine receptor agonist 5'-(N-ethyl carboxamido)-adenosine (NECA; 1 nM-300 microM) produced a biphasic increase in adenylyl cyclase activity in cell homogenates, best fitted to two components with high (EC50 0.7 microM) and low (EC50 16.0 microM) potency, respectively. The selective adenosine A2a receptor agonist CGS-21680 (1 nM-300 microM) also produced a biphasic increase in adenylyl cyclase. The NECA-dependent increase in adenylyl cyclase activity was almost completely inhibited by the non-selective adenosine receptor antagonist xanthine amine congener (XAC; 30 microM), but only partially inhibited by the selective A2a adenosine antagonist 8-(3-chlorostyryl)caffeine (CSC; 1 microM). Experiments were also performed to investigate the time course of NECA-induced desensitization of putative A2a and A2b receptor responses. The A2a-response was quantified using 10 microM CGS-21680, whilst the A2b response was quantified using 100 microM NECA in the presence of 1 microM CSC. The t0.5 for desensitization for each subtype was found to be around 20 min. Neither activation (with dibutyryl cAMP; 1 mM) nor inhibition (with H-89; 10 microM) of cyclic AMP-dependent protein kinase altered the ability of NECA pretreatment to desensitize A2a or A2b receptor-activated adenylyl cyclase. However zinc (200 microM), an inhibitor of G-protein coupled receptor kinase 2 (GRK2), significantly reversed the agonist-induced desensitization of A2a and A2b receptor-activated adenylyl cyclase. These experiments suggest the co-existence of A2a and A2b receptors coupled in a stimulatory fashion to adenylyl cyclase in NG108-15 cells. Furthermore desensitization of A2a and A2b responses occurs at the same rate and may involve a G-protein-coupled receptor kinase.

Heterogeneous forms of adenotin-1 of different subcellular localization

The localization of the low-affinity adenosine binding protein adenotin-1 with respect to distribution in rat organs and subcellular compartments was investigated. Adenotin-1 was characterized by 5'-N-ethylcarboxamido[2,8-3H]adenosine ([3H]NECA) binding and Western blotting. Cytosolic as well as membrane fractions of all tissues contained adenotin-1. Highest levels of membrane-bound adenotin-1 were found in the liver (liver > kidney approximately spleen approximately lung > forebrain approximately cerebellum > fat heart - striated muscle), whereas highest levels of cytosolic adenotin-1 were detected in spleen, liver, lung and fat. Subcellular fractions from rat liver were prepared by differential and density gradient centrifugation. Like the homologous proteins endoplasmin or gp96, adenotin-1 is enriched in the endoplasmic reticulum. Cytosolic and membrane-bound adenotin-1 species are pharmacologically distinct, because in the liver particulate fraction adenotin-1 showed a more rapid binding kinetics, a twofold lower affinity for [3H]NECA (KD 227 nM vs. 105 nM) and a sevenfold higher affinity for 2-chloroadenosine than the cytosolic protein (Ki 1.48 microM vs. 9.25 microM). In rat liver cytosol, two different binding sites were found, which differed in [3H]NECA binding kinetics and displayed a hundredfold difference in their affinity for 2-chloro-5'-N-methylcarboxamidoadenosine (Ki 45.8 nM vs. 4.76 microM). The presence of adenotin-1 in subcellular fractions, as determined by radioligand binding, was confirmed by Western blotting. Adenotin-1 was detected as a 98-kDa band in all rat liver subcellular fractions, which agrees with the molecular mass determined for the purified protein. In the cytosol, a 65-kDa hand was labeled more intensely than the 98-kDa band. This additional band probably represents the pharmacologically distinct species of adenotin-1 found in the cytosol.

A dominant negative mutant of the G protein-coupled receptor kinase 2 selectively attenuates adenosine A2 receptor desensitization

G protein-coupled receptor kinases (GRKs) are thought to be important in mediating the agonist-induced phosphorylation and consequent desensitization of G protein-coupled receptor responses. NG108-15 mouse neuroblastoma X rat glioma cells express a wide range of G protein-coupled receptors and significant levels of GRK2. Therefore, to determine the role of GRK2 in agonist-induced desensitization of various G(s)-coupled receptors in NG108-15 cells, we stably transfected cells with a dominant negative mutant GRK2 construct (Lys220Arg). In homogenates prepared from cells overexpressing the dominant negative mutant GRK2, the acute stimulation of adenylyl cyclase by various receptor and nonreceptor agonists was the same as in control cells stably transfected with plasmid only. NG108-15 cells express both A2a and A2b adenosine receptors, which mediate activation of adenylyl cyclase, with both of these responses being subject to agonist-induced desensitization with a t1/2 of 15-20 min. In dominant negative mutant GRK2 cells, the rates of desensitization of A2a and A2b receptor-stimulated adenylyl cyclase were markedly slower than in plasmid transfected controls, with the latter being similar to wild-type cells. After a 20-min treatment with an adenosine agonist, the desensitization of A2a and A2b receptor-stimulated adenylyl cyclase in dominant negative mutant GRK2 cells was less than half that seen in plasmid transfected control cells. On the other hand, the agonist-induced desensitization of secretin and IP-prostanoid receptor-stimulated adenylyl cyclase was the same in dominant negative mutant GRK2 cells as in plasmid transfected control cells. These results indicate that in intact cells, GRK2 may mediate the desensitization of adenosine A2 receptors. Furthermore, there seems to be selectivity of GRK2 action between G(s)-coupled receptors because the agonist-induced desensitization of secretin and IP-prostanoid receptor-stimulated adenylyl cyclase was not affected by dominant negative mutant GRK2 overexpression.

Activation of nucleotide receptors inhibits high-threshold calcium currents in NG108-15 neuronal hybrid cells

A P2U (UTP-sensitive) nucleotide receptor has previously been cloned from NG108-15 neuroblastoma x glioma hybrid cells and it has been shown that activation of this receptor inhibits the M-type K+-current. We now report that UTP also inhibits Ca2+-currents in differentiated NG108-15 cells, but probably through a different nucleotide receptor. UTP (100 microM) inhibited the peak of the high-threshold current by 28.4 +/- 3.1% (n = 28) with no effect on the low-threshold current. Two components of high-threshold current were identified: one inhibited by 100 nM omega-conotoxin (CgTx) and one inhibited by 2 microM nifedipine and enhanced by 1 microM BAY K8644. UTP inhibited the former by 31.0 +/- 3.1%, with an IC50 of 2. 8 +/- 1.1 microM, and the latter 34.2 +/- 6.1% with an IC50 of 1.7 +/- 1.3 microM. Pertussis toxin pretreatment prevented inhibition of the CgTx-sensitive, nifedipine-resistant but not CgTx-resistant current. Inhibition was not prevented by intracellular BAPTA (20 mM) or cAMP (1mM). Effects of UTP on both currents were imitated by UDP, ATP, ADP, AP4A and ATPgammaS but weakly or not at all by 2-MeSATP, GTP, AMP-CPP or ITP. Since the receptors which inhibit Ca2+-currents are activated by ATP, it is suggested that they might mediate auto-inhibition of transmitter release by ATP if present on purinergic nerve terminals.

KEYWORDS

nucleotides, UTP, ATP, calcium currents, neuroblastoma cells

Regulation of the stress-like protein adenotin in PC 12 cells by ethanol exposure

The effects of chronic ethanol exposure on the stress-like protein adenotin were investigated using the radioligand [3H]-5'-N-ethylcarboxamidoadenosine ([3H]NECA). A 4-day exposure to 150 mM ethanol increased both the KD and the density of [3H]NECA binding sites. These changes were not due to residual ethanol as the acute addition of ethanol did not alter [3H]NECA binding. Chronic ethanol exposure of A126-1B2-1 cells, which are a mutant PC 12 cell line deficient in protein kinase A (PKA), increased the cellular density of adenotin, but did not affect the KD for the radioligand. Conversely, when PC 12 cells were exposed to 10 microM forskolin for either 2 or 4 days, the cellular density of adenotin was not altered, but the affinity of adenotin for [3H]NECA was reduced significantly. An increase in KD was not observed after a 1-hr exposure of PC 12 cells to forskolin, indicating that the reduction in affinity for the radioligand was not due simply to a PKA-mediated phosphorylation of adenotin. The present study demonstrated that chronic ethanol regulates adenotin through two different mechanisms. The ethanol-induced increase in the density of adenotin does not involve PKA, while the reduction in affinity ap pears to involve a cAMP-dependent mechanism.

Effects of acute and chronic ethanol on cyclic AMP accumulation in NG108-15 cells: differential dependence of changes on extracellular adenosine

1. This study investigated the effects of acute and chronic ethanol on basal, agonist- and forskolin-stimulated cyclic AMP formation in NG108-15 mouse neuroblastoma x rat glioma hybrid cells, and examined the role of changes in extracellular adenosine concentrations on the effects observed. 2. NG108-15 cells incubated acutely with ethanol (1-200 mM) displayed concentration-dependent increases in basal and iloprost-stimulated (300 nM; a prostanoid IP receptor agonist) cyclic AMP accumulation but a concentration-dependent decrease in forskolin-stimulated (10 microM) accumulation. 3. Cells treated chronically with ethanol (200 mM) for 48 h displayed increases over control in basal, iloprost- (0.001-10 microM) and forskolin (0.01-100 microM)-stimulated cyclic AMP formation. However, chronic ethanol did not affect [3H]-iloprost binding to cell membranes. 4. Inclusion of adenosine deaminase (ADA; 1 unit ml-1) during the incubation period to measure cyclic AMP accumulation completely abolished the increase in basal accumulation following chronic ethanol, but did not affect the increase in iloprost stimulation. On the other hand ADA partially reversed the increase in forskolin stimulation following chronic ethanol, but even in the presence of high concentrations of ADA (5 units ml-1) the forskolin stimulation remained elevated above control. 5. Cells treated chronically with the adenosine receptor agonist 5'-(N-ethylcarboxamido)-adenosine (NECA; 10 microM for 48 h) displayed a reduction in subsequent NECA- and forskolin-stimulated cyclic AMP accumulation, but iloprost stimulation was not affected. ADA included acutely during the incubation period to measure cyclic AMP accumulation abolished the reduction in forskolin but not NECA stimulation produced by the chronic NECA pretreatment. 6. We have previously noted that ethanol inhibits NG108-15 cell proliferation and alters cell morphology.To mimic this, cells were incubated in the absence of foetal calf serum for 48 h. Following this time, basal, iloprost- and forskolin-stimulated cyclic AMP formation was enhanced over that in cells grown in the presence of serum.7. These results indicate that chronic ethanol enhances cyclic AMP formation in intact NG108-15 cells by more than one mechanism: one involves increased extracellular adenosine concentrations and the other a change in the transduction system beyond the receptor, possibly involving the adenylyl cyclase enzyme. Furthermore the ethanol-induced changes in cyclic AMP accumulation may relate to alterations in NG108-15 cell growth and development.

Receptor availability defines the extent of agonist-mediated G-protein down-regulation in neuroblastoma x glioma hybrid cells transfected to express the beta 2-adrenoceptor

Sustained exposure of neuroblastoma x glioma hybrid, NG108-15, cells transfected to express the human beta 2-adrenoceptor (clone beta N22) to isoprenaline or iloprost (an agonist at the endogenously expressed IP prostanoid receptor) resulted in a substantial and selective down-regulation of the alpha subunit of the G-protein Gs. Treatment of these cells with the irreversible beta-adrenoceptor antagonist bromoacetyl alprenolol menthane diminished both the potency and the maximal ability of isoprenaline but not of iloprost to cause Gs alpha down-regulation. These results demonstrate that the extent of agonist-mediated Gs alpha down-regulation is dependent upon the availability of receptor to agonist.

Binding characteristics of the adenosine A2 receptor ligand [3H]CGS 21680 to human platelet membranes

The binding characteristics of the selective adenosine A2 agonist [3H]CGS 21680 ([3H]2-[p-(2-carboxyethyl)-phenethyl-amino]-5'-N- ethylcarboxamidoadenosine) were determined in human platelet membranes. Specific binding was saturable, reversible and dependent upon protein concentration. Saturation experiments revealed a single class of binding sites with Kd and Bmax values of 1.4 microM and 5.9 pmol/mg of protein, respectively. Adenosine receptor agonists and antagonists competed for the binding of [3H]CGS 21680 (50 or 200 nM) to human platelet membranes showing a rank order of potency consistent with that typically found for interactions at the adenosine A2 receptor. Adenylate cyclase stimulation and platelet aggregation inhibition induced by adenosine agonists exhibited a rank order of potency close to that observed in binding experiments. However, the adenosine A1 receptor agonists, R- and S-N6-(2-phenylisopropyl)adenosine, (R-PIA) and (S-PIA), N6-cyclohexyladenosine (CHA) and 2-chloro-N6-cyclopentyladenosine (CCPA), which stimulate adenylate cyclase and inhibit platelet aggregation in the low microM range, displaced [3H]CGS 21680 only in the high microM range. In conclusion, we have found that [3H]CGS 21680, which is widely used as a specific A2 agonist in binding studies on brain tissues, is not appropriate for the characterization of the human platelet adenosine A2 receptor.

Gs alpha-dependent and -independent desensitisation of prostanoid IP receptor-activated adenylyl cyclase in NG108-15 cells

NG108-15 mouse neuroblastoma x rat glioma cells were treated with the prostanoid IP receptor agonist iloprost (1 microM) and the time course of changes in the levels of prostanoid IP receptors, adenylyl cyclase activity, and the alpha-subunit of the stimulatory guanine nucleotide binding regulatory protein, Gs, were measured. Incubation of cells with iloprost produced a biphasic time course of desensitisation of prostanoid IP receptor-activated adenylyl cyclase. Parallel analysis of iloprost-induced loss of membrane Gs alpha, NaF-stimulated adenylyl cyclase and [3H]iloprost binding suggested only monophasic curves, with t0.5 values similar to the initial phase of desensitisation of iloprost-stimulated adenylyl cyclase activity. This suggests that the loss of receptor and Gs alpha occur at the same time and account for the initial period of desensitisation due to iloprost pretreatment. Pretreatment of NG108-15 cells with cholera toxin produced a near complete loss of membrane-associated Gs alpha, but the loss of [3H]iloprost binding due to iloprost treatment was not affected by pretreatment with cholera toxin, suggesting that prostanoid IP receptors can be down-regulated in the absence of any coupling to Gs. The second phase of desensitisation of iloprost-stimulated adenylyl cyclase activity, during which there was no further change in NaF-stimulated adenylyl cyclase or in the membrane levels of Gs alpha, was not due to protein kinase A activation, since elevating intracellular cyclic AMP levels with forskolin did not subsequently decrease iloprost-stimulated adenylyl cyclase activity or [3H]iloprost binding. These results demonstrate that iloprost pretreatment of NG108-15 cells induces two distinct phases of desensitisation; an initial desensitisation due to concurrent loss of prostanoid IP receptors and Gs alpha, and then a further desensitisation by an as yet uncharacterized mechanism during which there is no further loss of Gs alpha.

Purification and characterization of an adenotin-like adenosine binding protein from human platelets

A low-affinity adenosine binding protein (adenotin) was purified from human platelet membranes by a four-step procedure. Purification was achieved after extraction from human platelet membranes with 0.3% 3-[3-(cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). Further purification included Sepharose CL6B gel filtration, DEAE-Sepharose CL6B, and hydroxylapatite chromatography. The protein was purified 884-fold to homogeneity with a 25% yield of binding activity from the membranes. 5'-[8(n)-3H]-N-ethylcarboxamidoadenosine ([3H]NECA) binds to the purified protein with a KD of 155 (144-167) nmol/l and a Bmax of 1.85 +/- 0.10 nmol/mg of protein. Sodium dodecylsulfate polyacrylamide gel electrophoresis of purified protein revealed a single band at 98 kDa. The 2-chloro-substituted adenosine analogs 2-chloro-5'-N-methylcarboxamidoadenosine (CIMECA) and 2-chloro-5'-N-ethylcarboxamidoadenosine (CINECA) were identified as new high affinity ligands of the purified protein showing Ki values of 18 nmol/l and 28 nmol/l, respectively. The low-affinity adenosine binding protein showed a pharmacological profile as follows: CIMECA > 5'-N-ethylcarboxamidoadenosine (NECA) > 2-chloroadenosine (CIA) > 2-[4-(2-carboxyethyl)phenethylamino]-5'-N-ethylcarboxamidoadenosin e (CGS 21,680) > R-N6-phenylisopropyl-adenosine (R-PIA). Amino-terminal sequence analysis revealed homologies to endoplasmin, glucose regulated protein (GRP94), tumor rejection antigen precursor (GP96), and some stress related proteins.

Desensitization of adenylate cyclase responses following exposure to IP prostanoid receptor agonists. Homologous and heterologous desensitization exhibit the same time course

Pretreatment of NG108-15 cells with 0.03-25 microM prostaglandin E1 (PGE1) produced decreases in the maximal stimulation of adenylate cyclase activity produced by iloprost, N-ethylcarboxamidoadenosine and sodium fluoride. The rate of desensitization to all three agents was dependent on the concentration of PGE1 used, but at each concentration of PGE1 the rate of loss of responsiveness to each agent was the same, suggesting that the decreases in responsiveness may be mediated by a single process. Functional desensitization was accompanied by a decrease in the specific binding of [3H]iloprost, consistent with a 75-80% decrease in IP receptor number, with no change in the coupling of the remaining IP receptors to G protein. At each concentration of PGE1 used, the times taken for half maximal decreases in receptor number and functional responsiveness were similar, suggesting that IP receptor down-regulation is a relatively early event in desensitization. IP receptor down-regulation could be inhibited partially by 100 microM chloroquine, suggesting that lysosomal breakdown of receptors may be occurring.

Differential activation of adenosine receptors decreases N-type but potentiates P-type Ca2+ current in hippocampal CA3 neurons

Adenosine is released in the brain in significant quantities in response to increased cellular activity. Adenosine has been shown either to decrease synaptic transmission or to produce an excitatory response in hippocampal synapses, resulting in increased glutamate release. Previous reports have shown that adenosine or its analogs reduced Ca2+ current in dorsal root ganglion and hippocampal neurons. Here we show that the selective activation of adenosine receptor subtypes has different effects on Ca2+ channels from acutely isolated pyramidal neurons from the CA3 region of guinea pig hippocampus. Activation of A1 receptors inhibited primarily N-type Ca2+ current. In contrast, activation of A2b receptors resulted in significant potentiation of P-type but not N-type Ca2+ current. This potentiation could be inhibited by blocking the cAMP-dependent protein kinase. Because of the ubiquity of adenosine, the differential effects on Ca2+ channels of adenosine receptor subtype activation may have significant implications for neuronal excitability.

The adenosine receptors present on the plasma membrane of chromaffin cells are of the A2b subtype

The adenosine receptors in the plasma membrane of chromaffin cells from bovine adrenal medulla were characterized. The presence of A1 receptors was discounted owing to the absence of R-[3H]phenylisopropyladenosine (R-PIA) and [3H]8-cyclopentyl-1,3-dipropylxanthine ([3H]-DPCPX) binding. The binding of the specific A2a ligand CGS-21680 was low. In contrast, the binding of 5'-(N-[3H]-ethylcarboxamidoadenosine ([3H]NECA) was relatively high (1.7 pmol/mg of protein at a ligand concentration up to 90 nM). This binding did not correspond to non-adenosine receptor NECA binding sites because the specific [3H]-NECA binding was similar when unlabeled adenosine, NECA, or R-PIA was used to measure the nonspecific binding. The rank order of potency of different ligands for the displacement of specific [3H]NECA binding was DPCPX greater than NECA greater than chloroadenosine greater than R-PIA greater than theophylline = CGS-21680. These results indicate that the receptors present on the plasma membrane of chromaffin cells are exclusively of the A2b subtype.

Concurrent down-regulation of IP prostanoid receptors and the alpha-subunit of the stimulatory guanine-nucleotide-binding protein (Gs) during prolonged exposure of neuroblastoma x glioma cells to prostanoid agonists. Quantification and functional implications

Neuroblastoma x glioma hybrid NG108-15 cells express a high-affinity IP prostanoid receptor. Saturation binding analysis of this receptor, using [3H]prostaglandin E1 ([3H]PGE1) as ligand, indicated that it was present at some 1.5 pmol/mg of membrane protein and displayed a dissociation constant for this ligand of 30-40 nM. Prolonged exposure of these cells either to PGE1 or to iloprost, which is a stable analogue of prostacyclin, caused a 40-70% decrease in levels of the receptor. The remaining receptors were capable of interacting with the stimulatory G-protein (Gs) of the adenylate cyclase cascade, as saturation analysis of the binding of [3H]PGE1 indicated that they had a similar affinity for the 3H-labelled ligand, and because the specific binding of [3H]PGE1 to these receptors was still sensitive to the presence of poorly hydrolysed analogues of GTP. We have recently demonstrated that prolonged exposure of NG108-15 ells to PGE1 causes a cyclic AMP-independent loss of Gs alpha-subunit (Gs alpha) from these cells [McKenzie & Milligan (1990) J. Biol. Chem. 265, 17084-17093]. Steady-state concentration of the larger 45 kDa form of Gs alpha (which is the predominant form expressed in these cells) was assessed to be 9.6 pmol/mg of membrane protein, and treatment with iloprost decreased levels of this polypeptide to some 3.0 pmol/mg of protein. Time courses of iloprost-mediated down-regulation of the IP prostanoid receptor, loss of Gs alpha protein as assessed by immunoblotting and loss of Gs alpha activity as assessed by the reconstitution of NaF stimulation of adenylate cyclase activity to membranes of S49 cyc- cells by sodium cholate extracts of NG108-15 cells were identical, suggesting that the loss of the IP prostanoid receptor and G-protein occurred in parallel. Each of these effects was half-maximal between 2 and 3 h of exposure to the agonist. Stoichiometry of loss of Gs alpha and IP prostanoid receptor was unchanged by the percentage receptor occupancy, and quantification indicated the loss of some 7-10 mol of Gs alpha/mol of receptor. This is the first report to demonstrate the temporal concurrence of loss of Gs alpha and of a receptor which interacts with this G-protein. Chronic activation of the IP prostanoid receptor on these cells results in the development of a heterologous form of desensitization to agents which function to activate adenylate cyclase [Kelly, Keen, Nobbs & MacDermot (1990) Br. J. Pharmacol. 99, 306-316]. Agonist regulation of Gs alpha levels in these cells may contribute to this process.

Cyclic AMP produces desensitization of prostacyclin and adenosine A2 receptors in hybrid cell lines but does not affect Gs function

Prostacyclin and adenosine A2 receptors stimulate adenylate cyclase activity in the related somatic hybrid cell lines NG108-15 and NCB20. The role of cAMP in the desensitization of these receptors has been examined. Pretreatment for 17 h with forskolin or 8-bromo-cAMP had the same effect in both cell lines. There was no change in the response to sodium fluoride or forskolin, suggesting that the function of Gs and adenylate cyclase were unaffected by increased levels of cAMP. Receptor responses were affected however; the maximum response to N-ethylcarboxamidoadenosine (an A2 receptor agonist) was reduced by 30-40%, there was a small but consistent shift to the right of the dose-response curve for iloprost (a stable analogue of prostacyclin) and [3H]iloprost binding studies revealed a loss of prostacyclin receptors. However, the loss of receptor responsiveness was much smaller than that which occurs following pretreatment with prostacyclin or adenosine A2 receptor agonists (Keen et al. (1989) Biochem. Pharmacol. 38, 3827-3833; Kelly et al. (1990) Br. J. Pharmacol. 99, 309-316) suggesting that cAMP may not play a major role in agonist mediated desensitization.

Non-eicosanoid functions of essential fatty acids: regulation of adenosine-related functions in cultured neuroblastoma cells

Studies have demonstrated that augmenting the omega 6 polyunsaturated-fatty-acid (PUFA) content of N1E-115 neuroblastoma cells by media supplementation with linoleic acid results in greater than or equal to 2-fold increases in basal levels of intracellular cyclic AMP (cAMP). Data suggested some involvement of increased production of adenosine from endogenous metabolites; however, increases in adenosine were not related to increased activity of 5'-nucleotidase or decreased uptake of extracellular adenosine. PUFA-dependent elevations in basal cAMP were evident within 1 min of exposure to a phosphodiesterase inhibitor; this phenomenon did not appear to be due to PUFA-dependent changes in Ca2+ uptake or to increases in sensitivity of adenylate cyclase to Ca2+. Forskolin-stimulated cAMP formation was 3-fold higher in PUFA-enriched cells than in control cells, which suggested a direct effect on the functioning of the catalytic unit. Linoleic acid supplementation resulted in a 2-fold increase in the maximum amounts of cAMP produced in response to the stable adenosine analogue, 5'-N'ethylcarboxy-amidoadenosine (NECA). The altered stimulatory response did not involve eicosanoid formation, but may have been related to an increase in the number of stimulatory adenosine receptors, as judged by binding of [3H]NECA. These studies indicate that membrane PUFA modulate adenosine-related functions in neuroblastoma cells, and suggest that a complex series of mechanisms is involved in this regulation.

Characterization of the human liver vasopressin receptor. Profound differences between human and rat vasopressin-receptor-mediated responses suggest only a minor role for vasopressin in regulating human hepatic function

The [Arg8]vasopressin (AVP) receptor expressed by human hepatocytes was characterized, and compared with the rat hepatic V1a vasopressin receptor subtype. In addition to determining the pharmacological profile of the human receptor, the cellular responses to AVP were measured in human and rat hepatocytes by assaying glycogen phosphorylase alpha activity and DNA synthesis. Marked differences were observed between human and rat hepatocytes regarding vasopressin receptors and the intracellular consequences of stimulation by AVP. Data presented in this paper demonstrate the following, (i) Vasopressin V1a receptors are present in low abundance on human hepatocytes. (ii) Species differences exist between human and rat V1a receptors with respect to the affinity of some selective antagonists. (iii) AVP-stimulated glycogen phosphorylase a activation in human hepatocytes was approx. 5% of that observed in rat cells. (iv) In contrast with rat hepatocytes, DNA synthesis in human cells in culture was not stimulated by AVP. It is concluded that vasopressin plays only a minor role in the regulation of human hepatic function. Furthermore, conclusions drawn from observations made with AVP and its analogues on rat hepatic function cannot be directly extrapolated to the human situation.

Primate vascular responses to octimibate, a non-prostanoid agonist at the prostacyclin receptor

1. Octimibate is a potent inhibitor of human platelet aggregation, and appears to act (at least in part) through the prostacyclin receptor, as described in the preceding paper. Here, the vascular effects, both in vitro and in vivo, of octimibate have been compared to those of the stable prostacyclin (PGI2) mimetic, iloprost. Since octimibate shows extensive species variation and is potent at inhibiting platelet aggregation in primates, all of the experiments reported here have been carried out with primate tissue or in vivo in cynomolgus monkeys. 2. Activation of adenylyl cyclase in human lung membranes appears to involve stimulation of the vascular PGI2 receptor. Octimibate, as well as iloprost, stimulates adenylyl cyclase in this preparation. The EC50 values for iloprost and octimibate are 50 nM and 340 nM respectively. These values are similar to those seen with human platelet membranes. As with platelets, the maximal activation achievable with octimibate is 60% of that seen with iloprost. This result suggests that octimibate is a partial agonist for stimulation of adenylyl cyclase. 3. Iloprost (10-100 nM) relaxes human coronary and mesenteric artery precontracted with KCl, and also relaxes cynomolgus monkey aorta precontracted with phenylephrine. Octimibate appears to be a partial agonist for relaxation of human coronary artery precontracted with KCl; the intrinsic activity of octimibate (10 microM) is 0.15 compared to iloprost, and octimibate surmountably antagonizes the relaxant effects of iloprost with a Kp of 200 nM. Octimibate (up to 10 microM) evokes only weak relaxation of human mesenteric artery (precontracted with KCl) and cynomolgus monkey aorta (precontracted with phenylephrine). 4. The effects of iloprost and octimibate were compared in vivo in cynomolgus monkeys. In addition to inhibiting ex vivo platelet aggregation, both compounds cause hypotension with little effect on heart rate. The dose-response curves for inhibition of ex vivo platelet aggregation and a fall in mean arterial blood pressure were compared. The dose-separation (i.e., the relative differences in effective concentrations) for the two responses is similar with both iloprost and octimibate. 5. Since the pern; beral resistance vessels are intimately involved in regulation of systemic arterial blood pressure, the effects of both agents were tested on human peripheral resistance vessels (150-400pm diameter) in vitro. These vessels are relaxed by both iloprost and octimibate following precontraction with KCI. The IC50 value for iloprost is 44nM, and 1.7 microM octimibate evokes 50% of the maximal relaxation obtained with iloprost. Thus, the relative potencies of the two compounds in relaxing human subcutaneous resistance vessels are similar to their relative potencies in inhibiting platelet responses. This result correlates with the lack of platelet versus vascular selectivity seen with the in vivo monkey studies. 6. These results suggest that octimibate, a partial agonist at the prostacyclin receptor, is unable to discriminate between platelet and vascular prostacyclin receptors in primates.

Octimibate, a potent non-prostanoid inhibitor of platelet aggregation, acts via the prostacyclin receptor

1. Octimibate, 8-[(1,4,5-triphenyl-1H-imidazol-2-yl)oxy]octanoic acid, is reported to have antithrombotic properties. This is in addition to its antihyperlipidaemic effects which are due to inhibition of acylCoA:cholesterol acyltransferase (ACAT). The aim of this study was to investigate the mechanism of the antithrombotic effect of octimibate, and to determine whether the effects of octimibate are mediated through prostacyclin receptors. 2. In suspensions of washed (plasma-free) human platelets, octimibate is a potent inhibitor of aggregation; its IC50 is approx. 10 nM for inhibition of aggregation stimulated by several different agonists, including U46619 and ADP. The inhibitory effects of octimibate on aggregation are not competitive with the stimulatory agonist; the maximal response is suppressed but there is no obvious shift in potency of the agonist. In platelet-rich plasma, octimibate inhibits agonist-stimulated aggregation with an IC50 of approx. 200 nM. 3. Octimibate also inhibits agonist-stimulated rises in the cytosolic free calcium concentration, [Ca2+]i, in platelets. Both Ca2+ influx and release from intracellular stores are inhibited. The effects of octimibate on aggregation and [Ca2+]i are typical of agents that act via elevation of adenosine 3':5'-cyclic monophosphate (cyclic AMP). Similar effects are seen with forskolin, prostacyclin (PGl2) and iloprost (a stable PGl2 mimetic). 4. Octimibate increases cyclic AMP concentrations in platelets and increases the cyclic AMP-dependent protein kinase activity ratio. Octimibate stimulates adenylyl cyclase activity in human platelet membranes, with an EC50 of 200 nM. The maximal achievable activation of adenylyl cyclase by octimibate is 60% of that obtainable with iloprost. Octimibate has no effect on the cyclic GMP-inhibited phosphodiesterase (phosphodiesterase-ITI), which is the major cyclic AMP-degrading enzyme in human platelets.5. Octimibate inhibits, apparently competitively, the binding of [3H]-iloprost (a stable PGl2 mimetic) to platelet membranes; the estimated Ki is 150 nm. 6. The platelets of different species show considerable differences in the apparent potency of their inhibition of aggregation by octimibate; platelets from cynomolgus monkeys are 3 fold more sensitive than those from humans, while rat, cat and cow platelets are 50, 100, and 250 fold less sensitive than human platelets. The sensitivity of these different species to iloprost, however, varies over a range of only 10 fold with no obvious difference between primates and non-primates. 7. Octimibate appears to be a potent agonist (aggregation), or partial agonist (adenylyl cyclase), at prostacyclin receptors and is the first non-prostanoid agent of this type to be identified. The species differences in relative potency of octimibate and iloprost may reflect the existence of receptor subtypes.

Adenosine receptor subtypes in the brainstem mediate distinct cardiovascular response patterns

A limited occipital craniotomy was conducted on urethane-chloralose anesthetized, spontaneously breathing rats to expose the caudal medulla in the region of the obex. Microinjections of highly selective agonists for adenosine receptor subtypes were made into the medial region of the caudal nucleus tractus solitarius (NTS) at the level of the posterior portion of the area postrema. Cardiorespiratory parameters were subsequently recorded for a 60-min test period following microinjection of drug or vehicle solutions. The following selective adenosine receptor agonists were used: the A1 agonist, N6-cyclopentyladenosine (CPA), which is 480-fold selective for A1 receptors in rat brain binding assays, and the A2 agonist, 2-p-(2-carboxyethyl)phenethylamino-5'-N-ethylcarboxamidoadenosine (CGS 21680), which is 170-fold selective for A2 receptors in rat brain binding studies and over 1500-fold selective in functional assays. The results showed that distinct and converse cardiovascular response patterns were elicited by these selective agonists for adenosine receptor subtypes following microinjections into the caudal NTS. Specifically, CGS 21680 selectively elicited potent dose-related decreases in mean arterial blood pressure (ED50 = 0.064 nmol/kg) and dose-related decreases in pulse pressure (ED50= 0.058 nmol/kg). Conversely, CPA selectively elicited potent dose-related increases in mean arterial blood pressure (ED50 = 0.62 nmol/kg) and dose-related increases in pulse pressure (ED50 = 0.70 nmol/kg). Additionally, the overall agonist-mediated response patterns were dramatically different wherein the CGS agonist exhibited a considerably more rapid time course in eliciting its hypotensive responses whereas CPA exhibited a more delayed and substantially longer time course to exert its hypertensive responses. Additionally, these distinct and converse cardiovascular response patterns were further shown to be receptor-selective since the depressor responses elicited by the A2 receptor agonist, CGS 21680, and the pressor responses elicited by the A1 receptor agonist, CPA, were completely and selectively blocked, respectively, by the selective A2 receptor antagonist, CGS 15943A, and the selective A1 receptor antagonist, DPCPX. Taken together, these findings provide persuasive in vivo evidence showing that pharmacologic activation of adenosine receptor subtypes in the caudal NTS of rats elicits specific response patterns with selective and opposite actions on cardiorespiratory behavior. These data also indicate that separate physiologic responses are specifically mediated by A2 receptors in the intact nervous system and thereby lend additional support to the case for using in vivo models to assess the functional role of adenosine A2 receptors in brain function.

Binding of the adenosine A2 receptor ligand [3H]CGS 21680 to human and rat brain: evidence for multiple affinity sites

A new radiolabeled adenosine receptor agonist, 2-[p-(2-carboxyethyl)phenethylamino]-5'-N-ethylcarboxamidoadeno sin e (CGS 21680), apparently specific for high-affinity binding sites of the A2 subtype in rat brain, was used to identify and pharmacologically characterize adenosine receptors in human brain. The binding of [3H]CGS 21680, as determined by standard radioligand binding technique in the presence of exogenously added adenosine deaminase, reached equilibrium after 40 min at 25 degrees C. In saturation studies, a single class of high-affinity binding sites with values for KD of 22 +/- 0.5 nM and Bmax of 444 +/- 63 fmol/mg of protein were observed. Similar binding characteristics were observed regardless of whether rapid filtration or centrifugation was used to separate bound versus free ligand. Of the 14 brain regions examined, [3H]CGS 21680 binding was highest in putamen, followed by globus pallidus and caudate nucleus. The level of [3H]CGS 21680 binding in these areas of basal ganglia was identical to 5'-N-[3H]ethylcarboxamidoadenosine ([3H]NECA) binding in the presence of 50 nM N6-cyclopentyladenosine (CPA). The rank order of agonist potencies as determined by a series of competition experiments was NECA greater than or equal to CGS 21680 greater than 2-chloroadenosine greater than N6-(R)-phenylisopropyladenosine greater than N6-cyclohexyladenosine greater than N6-(S)-phenylisopropyladenosine. This potency order was the same for the binding of [3H]CGS 21680 to rat, and of [3H]NECA in the presence of 50 nM CPA to rat and human, brain membranes.(ABSTRACT TRUNCATED AT 250 WORDS)