Propentofylline: a nucleoside transport inhibitor with neuroprotective effects in cerebral ischemia

1. Adenosine is an endogenous neuroprotective agent; stimulation of A1 receptors decreases excitatory amino acid neurotransmission and stimulation of A2 receptors inhibits platelet and neutrophil activation and promotes vasodilation. 2. Post-ischemic administration of propentofylline (HWA 285) reduces neuronal damage in gerbils and improves glucose metabolism in all regions of brain in acute stroke patients. 3. Propentofylline inhibits the transport of adenosine into cultured cells and increases extracellular adenosine concentrations in ischemic brain. Thus, enhanced stimulation of adenosine receptors may account for some of the neuroprotective effects of this compound. 4. Propentofylline inhibits free radical production by cultivated microglia cells, stimulates nerve growth factor production and inhibits cAMP-phosphodiesterase activity. These effects may also be important for neuroprotection.

Modulation of nerve and glial function by adenosine--role in the development of ischemic damage

Adenosine is released during brain ischemia and provides neuroprotection by actions on nerve and glial cells. Activation of the adenosine A1 receptor enhances the K+ and Cl- conductance in neurons, leading to membrane hyperpolarization and postsynaptic reduction of neuronal Ca2+ influx through voltage- and NMDA receptor-dependent channels. In addition adenosine A1 receptor activation decreases excitatory amino acid release, possibly via inhibition of N- and P-type Ca2+ channels. The A1 and A2 receptors, coupled to Gi/G(o) and Gs proteins respectively, often co-exist and interact with the phospholipase C-dependent activation of the protein kinase C and the adenylyl cyclase. Activation of the A1 receptor may mimic metabotropic receptor stimulation in activating intracellular Ca2+ mobilization and PKC. A2 receptor mediated cAMP formation is depressed by high intracellular Ca2+ but enhanced by PKC activation. By modulating these metabolic signaling events, adenosine may influence acute cell functions, gene transcription and sustained changes of nerve and glial cells relevant for the development of ischemic damage. The neuroprotective adenosine effect seems to be amplified by treatment with propentofylline, which enhances adenosine release, influences the balance between A1 and A2 receptor mediated actions, depresses the free radical formation in activated microglia and influences astrocyte reactions.

Role of NMDA, AMPA and kainate receptors in mediating glutamate- and 4-AP-induced dopamine and acetylcholine release from rat striatal slices

Striatal slices, preincubated with [3H]dopamine and [14C]choline, were superfused continuously and subjected to electrical field stimulation (3 Hz) and perfused with amino acid analogues or 4-amino pyridine (4-AP). The released radioactivity was used to monitor release of the neurotransmitters dopamine (DA) and acetylcholine (ACh). Glutamate, NMDA (in the absence of Mg2+), AMPA, kainic acid, domoate and 4-AP all induced DA and ACh release in a concentration-dependent manner. The DA and ACh release induced by NMDA (15 microM) and glutamate (1 mM) was essentially abolished by Mg2+ (1.15 mM), whereas release induced by AMPA (100 microM), kainic acid (100 microM) or 4-AP (30 microM) was not reduced. Tetrodotoxin (1 microM) essentially abolished the effects of NMDA, markedly reduced the effects of glutamate, AMPA and 4-AP, whereas the effect of kainic acid was only modestly affected. MK-801 (30 nM) reduced NMDA-induced DA release by some 70% and ACh release by 30%. MK-801 reduced 4-AP-induced DA release by 40% but not ACh release. CNQX in a concentration (10 microM) that scarcely affected NMDA-induced ACh release, but blocked that induced by AMPA, kainic acid or domoate, reduced the ACh release induced by 4-AP. In summary, DA and ACh release from rat striatum can be stimulated by activation of NMDA and non-NMDA glutamate receptors, and this mechanism is activated by the potassium channel blocker 4-AP.

Carbachol-induced phosphatidylcholine hydrolysis and choline efflux in rat submandibular gland involves phospholipase D activation and is modulated by protein kinase C and calcium

The role of calcium and protein kinase C activation in carbachol-induced choline efflux from submandibular glands was investigated. The participation of phospholipase D in this signal transduction pathway was demonstrated by the formation of [14C]phosphatidylethanol in [14C]lysophosphatidylcholine-labelled submandibular gland cells treated with carbachol or noradrenaline in the presence of ethanol. Chelation of the intracellular calcium with BAPTA/AM reduced the carbachol stimulated outflow of [3H]choline. The calcium ionophore A23187 in a high concentration (10 microM) increased the basal [3H]choline outflow, but decreased the carbachol-induced outflow. Removal of the extracellular calcium enhanced the carbachol-stimulated outflow, which returned to control when calcium was re-added to the medium. Activation of protein kinase C by phorbol-12,13-dibutyrate (100 nM) or 1-oleyl-2-acetyl-sn-glycerol (20 microM) was without effect per se, but enhanced the carbachol-mediated outflow of [3H]choline. Phorbol-12,13-dibutyrate in combination with 1 microM A23187 induced a small efflux of [3H]choline. A 2 h treatment with phorbol-12,13-dibutyrate (1 microM), causing down-regulation of protein kinase C, significantly decreased the carbachol-stimulated [3H]choline outflow. In conclusion, elevation of intracellular calcium levels and protein kinase C activation are of importance for the carbachol-stimulated outflow of [3H]choline. Inflow of calcium, if anything, reduces the carbachol-stimulated outflow of [3H]choline. Since phosphatidylcholine breakdown generates diacylglycerol and this could lead to activation of protein kinase C, activation of this signal transduction pathway may be important for the protein content of the saliva and for the known trophic effects of parasympathetic innervation.

Adenosine A2B receptor signalling is altered by stimulation of bradykinin or interleukin receptors in astroglioma cells

The human astroglioma cell D384 possesses adenosine A2B receptors coupled to the formation of cyclic AMP. These cells also possess bradykinin B2 receptors coupled to phospholipase C and consequent increases in intracellular calcium and protein kinase C. Interleukin 1 beta causes an increase in c-fos, AP-1 transcriptional activity and an increased expression of several genes including NGF, but the initial signalling events are unknown. Bradykinin causes a rapid decrease in A2B receptor mediated cAMP formation, via a mechanism that involves calcium, but not cGMP, and appears to depend upon a direct decrease in adenylyl cyclase. Il-1 beta causes a slowly developing (18-24 h) increase in A2B receptor signalling. The results indicate that adenosine effects in glial cells, believed to be important in neuroprotection, are modified in the short and long-term by inflammatory mediators.

Evidence for functionally important adenosine A2a receptors in the rat hippocampus

Adenosine A2a receptors are not confined to dopamine-rich areas of the brain, since thermocycling analysis shows that adenosine A2a receptor mRNA is expressed also in the hippocampus (CA1, CA3 and dentate gyrus) and cerebral cortex. The expression of A2a mRNA in three main areas of the hippocampus was confirmed by in situ hybridization; A2a mRNA expression was mainly localized in the pyramidal and granular cells, the same hippocampal regions that showed adenosine A1 receptor mRNA expression. Receptor autoradiographic studies with [3H]CGS 21680 (30 nM), a selective adenosine A2a receptor agonist, showed specific binding sites in the hippocampus. The density of [3H]CGS 21680 binding was greatest in the stratum radiatum of the CA1 area, followed by the stratum oriens of the cornu Ammonis, stratum radiatum of the CA3 are and supra-granular layer of the dentate gyrus. This anatomical distribution of [3H]CGS 21680 binding was similar to the pattern of [3H]CHA binding in the hippocampus. Electrophysiological studies in the Schaffer fibers/CA1 pyramids showed that upon activation of the A2a receptors with CGS 21680 (10 nM) the ability of the adenosine A1 receptor agonist, CPA, to inhibit neuronal activity was significantly attenuated. These results show functionally important co-expression and co-localization of adenosine A2a and A1 receptors in the hippocampus. The results also suggest that adenosine A2a receptor-mediated neuromodulation is not confined to the basal ganglia, but is more widespread throughout the nervous system.

Changes in adenosine receptors in the neonatal rat brain following hypoxic ischemia

We used quantitative in situ hybridization and receptor autoradiography to study changes in adenosine receptors following hypoxia-ischemia (H-I) in the neonatal rat brain. Seven-day-old rat pups were subjected to a unilateral ligation of the common carotid artery followed by a 2 h 15 min hypoxic period (7.7% O2 in N2). Adenosine A1 receptor mRNA in cortex and several parts of hippocampus, and A2a mRNA was decreased in the ligated hemisphere 0 h, 1 h and 2 h following hypoxia. The binding of the A1 receptor selective antagonist [3H]8-cyclopentyl-1,3-dipropylxanthine (DPCPX) in the presence or in the absence of GTP decreased immediately after the hypoxic period in both hemispheres and returned thereafter gradually towards control. These results show that there are rapid changes in A1 receptor number on both sides of the brain, and of adenosine A1 and A2a receptor mRNA in the hemisphere that would later develop infarction. Decreases in adenosine receptors may worsen H-I brain damage and have consequences for the use of adenosine directed therapy.

Differences in the regional and cellular localization of c-fos messenger RNA induced by amphetamine, cocaine and caffeine in the rat

Male rats were treated i.p. with either 5 mg/kg amphetamine, 3 and 30 mg/kg cocaine or 100 mg/kg caffeine and killed after 30 min. Brains were sectioned and processed for radioactive in situ hybridization histochemistry for the labelling of either c-fos, enkephalin, substance P, neurokinin B, choline acetyltransferase, somatostatin or adenosine A2A receptor messenger RNA. The distribution of c-fos messenger RNA was investigated both at the regional level using film autoradiography, and at the cellular level using emulsion autoradiography. All drug treatments except 3 mg/kg cocaine induced an increased level of c-fos messenger RNA in cells that had a neuron-like morphology. The cells that contained the c-fos messenger RNA were identified by making pairs of 5-microns sections in which one section was processed for c-fos messenger RNA and the other was processed for one of the other messenger RNA species. After amphetamine treatment, only some 10% of the cells in the striatum were labelled, and to a variable extent. Instead there was prominent labelling of a band in the cortex that runs parallel to the cortical surface. There was also a moderate degree of labelling in the nucleus accumbens. c-fos-positive cells were substance P-positive and negative for enkephalin or A2A receptor messenger RNA. Cocaine (30 mg/kg) induced a modest labelling in the caudate-putamen, as well as in the accumbens. With cocaine treatment (30 mg/kg), about 30% of striatal neuron-like cells were c-fos labelled. Most c-fos-positive cells were substance P-positive, but none of the c-fos-positive cells were enkephalin-positive or A2A-receptor-positive. Cocaine (3 mg/kg) had no significant effect on c-fos. Caffeine gave rise to a strong hybridization signal in the caudate-putamen, particularly the dorsolateral part. No other region examined differed significantly from control. With caffeine treatment, about 73% of neuron-like cells were c-fos labelled in the lateral striatum, but labelling was much less pronounced in the medial part or in the accumbens. c-fos-labelled cells were found in enkephalin-positive and enkephalin-negative, substance P-positive and substance P-negative, neurokinin B-positive and neurokinin B-negative groups. No choline acetyltransferase-positive or somatostatin-positive cells were found that were also c-fos-positive with any of the treatments. We conclude that each of the different CNS stimulant drugs induces a highly specific pattern of c-fos messenger RNA.(ABSTRACT TRUNCATED AT 400 WORDS)

Expression of protein kinase C isoforms in smooth muscle cells in various states of differentiation

We analysed the expression of protein kinase C (PKC) isoforms in smooth muscle cells (SMC) in various states of differentiation, using Western blots and thermocycle amplification of mRNA. SMC isolated from intact tissues express three isoforms of PKC, namely alpha, delta, and zeta. Following cell culture, SMC additionally express mRNA for PKC-epsilon, but significant amounts of the corresponding protein were not detected. Transformed SMC, such as the cell line DDT1 MF-2, express both mRNA and protein for PKC-epsilon, lack the delta isoenzyme, whilst maintaining the expression of the alpha and zeta isoforms. Thus PKC-delta and epsilon isoenzyme expression appears to vary with the state of differentiation of these cells, with PKC-epsilon expression increasing as the cells become proliferative.

Propentofylline and other adenosine transport inhibitors increase the efflux of adenosine following electrical or metabolic stimulation of rat hippocampal slices

Propentofylline is a novel neuroprotective agent that has been shown to act as an adenosine transport inhibitor as well as an adenosine receptor antagonist. In the present series of experiments we have compared the effects of propentofylline with those of known adenosine transport inhibitors and receptor antagonists on the formation of adenosine in rat hippocampal slices. The ATP stores were labeled by incubating the slices with [3H]adenine. The total 3H overflow and the overflow of endogenous and 3H-labeled adenosine, inosine, and hypoxanthine were measured. Adenosine release, secondary to ATP breakdown, was induced both by hypoxia/hypoglycemia and by electrical field stimulation. Propentofylline (20-500 microM) increased the release of endogenous and radiolabeled adenosine, without increasing the total release of purines. Thus, the drug altered the pattern of released purines, i.e., increasing adenosine and decreasing inosine and hypoxanthine. This pattern, which was observed when purine release was induced both by electrical field stimulation and by hypoxia/hypoglycemia, was shared by the nucleoside transport inhibitor dipyridamole (1 microM) and by mioflazine (1 microM) and nitrobenzylthioinosine (1 microM). By contrast, other xanthines, including theophylline (100 microM) and 8-cyclopentyltheophylline (10 microM), enprofylline (100 microM), or torbafylline (300 microM), if anything, increased the total release of purines without alterations of the pattern of release. These results indicate that nucleoside transport inhibitors can decrease the release of purines from cells and at the same time increase the concentration of extracellular adenosine, possibly by preventing its uptake and subsequent metabolism. This change in purine metabolism may be beneficial with regard to cell damage after ischemia. The results also indicate that propentofylline behaves in such a potentially beneficial manner.

Release of adenosine and other purines from hippocampal slices stimulated electrically or by hypoxia/hypoglycemia. Effect of chlormethiazole

The ATP stores of rat hippocampal slices were labelled with [3H]-adenine. The efflux of endogenous and [3H]-purines was stimulated electrically or by combined hypoxia/hypoglycemia. The efflux of endogenous adenosine, inosine and hypoxanthine was also increased. Chlormethiazole (30-300 microM) decreased the [3H]-purine release evoked by hypoxia/hypoglycemia, but not that evoked by electrical field stimulation. The outflow of endogenous purines was not significantly altered by chlormethiazole, and there was no significant change in the relative proportions of endogenous purines released. Chlormethiazole (10-500 microM) did not block the uptake of adenosine by human erythrocytes. Chlormethiazole does not appear to have any primary effect on purine metabolism, but decreases the breakdown of ATP induced by a reduced supply of metabolites in hippocampal slices.

Synergistic effects between protein kinase C and cAMP on activator protein-1 activity and differentiation of PC-12 pheochromocytoma cells

In rat pheochromocytoma cells (PC-12) cells, we have studied the effect of protein kinase C (PKC) and cAMP on the activity of the nuclear transcription factor activator protein-1 (AP-1) and on differentiation of the cells into sympathetic nerve-like phenotype. By using mobility gel-shift assays, we found that both PKC and cAMP activation led to an increase in the binding of AP-1 to its consensus nucleotide sequence (TRE). When the PKC and cAMP pathways were activated simultaneously, a clear-cut synergistic effect was seen on the binding of AP-1 to TRE. Both PKC and cAMP activation were furthermore able to increase the AP-1 transcriptional activity in PC-12 cells transiently transfected with TRE-expressing plasmids. In agreement with the mobility gel-shift results, simultaneous activation of PKC and cAMP synergistically increased the AP-1 transcriptional activity. We next analyzed the effect of PKC and cAMP stimulation on differentiation and proliferation of PC-12 cells. Whereas PKC activation had no effect on the morphology of PC-12 cells, elevation of the intracellular cAMP level resulted in a marked increase in the number of neurite-bearing cells. This effect was paralleled by a strong inhibition of PC-12 cell proliferation. Interestingly, when PKC and cAMP activation were combined, the differentiation was further pronounced and growth further inhibited. These results show that both PKC and cAMP increase the AP-1 activity in PC-12 cells, and that these effects are synergistic. Moreover, we show that cAMP induces differentiation and inhibits growth of PC-12 cells, and that PKC activation acts synergistically with cAMP on these effects. The possible role of AP-1 in PC-12 cell differentiation is discussed.

Effects of adenosine A1 and A2 receptor activation on electrically evoked dopamine and acetylcholine release from rat striatal slices

The authors investigated how electrically evoked [3H]dopamine (DA) and [14C]acetylcholine (ACh) release in the rat striatum is affected, directly or indirectly, by adenosine A1 and A2A receptor activation. Striatal slices, preincubated with [3H]DA and [14C] choline, were superfused continuously and subjected to electrical field stimulation. The electrically evoked release of both [3H] and [14C] was tetrodotoxin sensitive and calcium dependent. The release of both labels was augmented by haloperidol and sulpiride but only the [14C]ACh release was increased by 7-chloro-2,3,4,5-tetrahydro-3-methyl-5-phenyl-1H-3-benzazepine-7-ol (each at 1 microM). The A1 receptor-selective agonist cyclohexyladenosine and the A2A receptor-selective agonist 2-[4-(2-carboxymethyl)-phenethylamino]-5'-N-ethylcarboxamido adenosine (CGS 21680) both caused a concentration-dependent (range, 0.01-1 microM) inhibition of electrically evoked DA and ACh release and with a surprisingly similar potency. The potency of both compounds, and particularly of cyclohexyladenosine, increased when they were allowed to equilibrate for a longer period with the slices. The effect of both compounds was blocked by the A1-selective antagonist 1,3-dipropyl-8-cyclopentyl xanthine (0.1 microM). CGS 21680 never caused any stimulation of DA or ACh release. The ability of the D2 agonist quinpirole to inhibit ACh release was reduced by CGS 21680 in the presence or absence of 0.1 microM 1,3-dipropyl-8-cyclopentyl xanthine. CGS 21680 did not alter the D2 receptor inhibition of DA release. In summary, adenosine analogs inhibit DA and ACh release from rat striatum by a receptor that may be different from those hitherto characterized.(ABSTRACT TRUNCATED AT 250 WORDS)

The binding of the adenosine A2 receptor selective agonist [3H]CGS 21680 to rat cortex differs from its binding to rat striatum

The binding of the reportedly A2A selective agonist CGS 21680 (2-[p-(2-carboxyethyl)phenylethylamino]-5'N-ethylcarboxamidoadenos ine) to cortex and striatum was examined in parallel using quantitative receptor autoradiography. [3H]CGS 21680 bound to a single site in rat striatum with KD 2.3 nM and Bmax 320 fmol/mg grey matter. In addition [3H]CGS 21680 bound to a single site in the cerebral cortex with KD 47 nM and Bmax 100 fmol/mg grey matter. In cat cortex [3H]CGS 21680 (2 nM) binding was strong and particularly evident in the most superficial layers. The potency order for inhibition of 2 nM [3H]CGS 21680 binding to rat striatum was NECA (5'-N-ethylcarboxamidoadenosine; IC50 9.0 nM) > 2-CADO (2-chloroadenosine; 87 nM) > R-PIA (N6-(R)-phenylisopropyladenosine; 110 nM). The potency order for inhibition of 2 nM [3H]CGS 21680 binding to rat cortex was NECA (3.0 nM) > 2-CADO (14 nM) > or = R-PIA (16 nM). Gpp(NH)p (5'-guanylyl imidodiphosphate) inhibited [3H]CGS 21680 binding to both cortex and striatum, but more potently in cortex (IC50 100 nM vs. 470 nM). The present results show that there is a cortical binding site for [3H]CGS 21680 which appears to be different from the the striatal A2A receptor, the A2B receptor and the A1 receptor.

Breakdown of membrane choline-phospholipids induced by endogenous and exogenous muscarinic agonist is potentiated by VIP in rat submandibular gland

The outflow of tritium from rat submandibular gland fragments, pre-labelled with [3H]choline, following electrical or pharmacological stimulation was studied. Electrical stimulation of the tissue increased the outflow of tritium in a frequency dependent manner. Atropine treatment decreased the electrically-induced release, indicating that the outflow did not reflect acetylcholine from nerve endings, but was largely brought about by postsynaptic receptors. In agreement with this hypothesis, treatment with noradrenaline or carbachol induced a dose dependent increase in tritium outflow from the gland fragments which could be blocked by prazosin or atropine, respectively. Moreover, analysis of the tissue-associated tritium revealed an incorporation primarily in the lipid fraction of the tissue (almost 80%), of which about 90% was in phosphatidylcholine, indicating that this was the source of the tritium outflow. Pre-incubation with vasoactive intestinal polypeptide (VIP), which coexists with acetylcholine in the parasympathetic neurons innervating the submandibular gland, increased the carbachol-induced tritium overflow significantly. The effect of VIP could be imitated by the adenylyl cyclase stimulator forskolin, which increased the carbachol-stimulated tritium efflux in a dose dependent manner. Taken together, our results suggests that muscarinic- and alpha 1-receptor agonists may activate a phospholipase coupled to phosphatidylcholine hydrolysis in the rat submandibular gland. Endogenous acetylcholine released from parasympathetic nerve endings appear to activate this mechanism. Furthermore, VIP treatment, and the concomitant cAMP-accumulation, potentiates the acetylcholine induced phosphatidylcholine hydrolysis, demonstrating a new type of interaction between the classical transmitter acetylcholine and the co-stored neuropeptide VIP.

Intracellular formation and release of adenosine from rat hippocampal slices evoked by electrical stimulation or energy depletion

In this study, the basal and evoked release of [3H]- and endogenous adenosine, inosine and hypoxanthine from rat hippocampal slices, labelled with [3H]adenine, was investigated. Evoked release was brought about by either electrical stimulation or energy depletion. The aim was to determine whether adenosine is formed intracellularly, and released as adenosine or extracellularly, from sequential extracellular hydrolysis of released ATP. All measurements were made in the presence of 5 microM erythro-9-(2-hydroxy-3-nonyl) adenosine (EHNA) to inhibit the enzyme adenosine deaminase. It was found that electrical field stimulation (5 min) increased the release of endogenous adenosine from hippocampal slices 10-fold and increased the proportion of [3H]-label associated with adenosine from approx 7% of the total released to 13% after the first stimulation and 20% after the second stimulation. Removal of oxygen and glucose from the superfusion medium (energy depletion) increased the release rate of endogenous adenosine 16-fold and increased the proportion of [3H]-label associated with [3H]adenosine from approx 10% of the total released to 50%. In order to prevent extracellular formation of adenosine, experiments were carried out in the presence of 50 microM alpha, beta-methylene ADP (AOPCP), an inhibitor of ecto-5'-nucleotidase. AOPCP was found to be without effect on either the basal or evoked release of adenosine. In contrast, L-homocysteine thiolactone (0.1-1.0 mM) which was used to "trap" intracellular adenosine reduced both the basal and evoked release of adenosine by 70-85%. This effect of L-homocysteine thiolactone also occurred in the presence of adenosine uptake inhibitors. It is concluded from these results that adenosine is formed predominantly intracellularly in hippocampal slices and is released as adenosine as a result of either tissue depolarisation or energy depletion. Furthermore, the finding that during energy depletion there is a proportionally greater release of adenosine than other ATP breakdown products, such as inosine and hypoxanthine, indicates that energy depletion is both a potent and selective stimulus for adenosine formation and release.

Long-term caffeine treatment leads to a decreased susceptibility to NMDA-induced clonic seizures in mice without changes in adenosine A1 receptor number

The effects of long-term caffeine treatment on N-methyl-D-aspartate (NMDA)-induced seizures in mice were studied. Caffeine was added (0.3 g/l) to drinking water for 14 days and the mice ingested 60-70 mg/kg/day. During the treatment, the plasma concentrations of methylxanthines (caffeine, theophylline and/or paraxanthine, theobromine) were measured. NMDA (150 mg/kg i.p.) was administered to control mice and to mice during and after the caffeine administration. A1 adenosine receptor density in the gyrus dentatus of hippocampus, measured by quantitative receptor autoradiography with [3H]cyclohexyl adenosine as the ligand, was not significantly altered after long-term caffeine treatment. NMDA-induced clonic seizures, wet dog shakes and mortality were significantly reduced at the end of long-term caffeine treatment but returned towards control at 1 and 2 days after withdrawal. At the end of caffeine treatment, tonic seizures were also absent. These results show that long-term treatment with caffeine in a dose that gives plasma levels of 6-10 microM decreases the effects of NMDA on e.g. seizure susceptibility, and that this effect cannot be ascribed to changes of A1 adenosine receptor density.

Bradykinin inhibits cyclic AMP accumulation in D384-human astrocytoma cells via a calcium-dependent inhibition of adenylyl cyclase

Bradykinin causes a concentration-dependent, transient rise in intracellular Ca2+ and a sustained inhibition of forskolin-, dopamine- and 5'-N-ethyl-carboxamidoadenosine (NECA)-stimulated cAMP accumulation in D384 astrocytoma cells. Chelation of intracellular calcium abolished bradykinin's inhibitory effect on cAMP accumulation. Chelating extracellular Ca2+ did not block the initial, but eliminated the sustained inhibition of cAMP accumulation. Increasing Ca2+ influx by calcium ionophore A23187 caused a concentration-dependent inhibition of stimulated cAMP accumulation. A hydroquinone derivative 2,5-di(tert-butyl)-1,4-benzohydroquinone (tBuBHQ), which inhibits microsomal Ca2+ sequestration, did not mimic the effect of bradykinin, although it increased [Ca2+]i even more than A23187 did. The inhibitory effect of bradykinin was not mediated by Ca2+/CaM-dependent stimulation of phosphodiesterase (PDE). Forskolin-stimulated adenylyl cyclase activity was inhibited by Ca2+ (10(-7) to 10(-3) M), both in ethyleneglycol-bis-(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) washed and native D384 plasma membranes. This effect was not altered by calmodulin (CaM) or CaM-antagonists. Bradykinin treatment, which attenuates cAMP accumulation in intact cells, did not do so in plasma membranes. These findings suggest that bradykinin-induced inhibition of cAMP formation in D384 cells requires mobilization of [Ca2+]i and subsequent entry of Ca2+ which directly interacts with a component of the adenylyl cyclase system.

Effect of long term caffeine treatment on A1 and A2 adenosine receptor binding and on mRNA levels in rat brain

The effect of long-term oral treatment with caffeine on A1 and A2 receptors in the rat brain was studied. Caffeine was added to the drinking water and the animals were sacrificed after a 12 day treatment period. The plasma caffeine concentration was close to 100 microM. A1 receptors were studied using quantitative autoradiography with [3H]cyclohexyladenosine (CHA). Caffeine treatment increased the number of A1 receptors in the CA3 subfield of the hippocampus from 337 to 393 fmol/mg with no change in KD (0.692 vs. 0.675 nM). A1 mRNA was measured using Northern blots and quantitative in situ hybridization. There was no increase in A1 mRNA. A2a receptors, located in dopamine rich regions of the rat brain, were studied with quantitative autoradiography using [3H]CGS 21680 as the ligand, and the A2a mRNA was determined using quantitative in situ hybridization. Caffeine treatment produced no significant change in either receptor number or mRNA, even though the apparent Bmax tended to increase from 322 +/- 8 to 352 +/- 8 fmol/mg. The results show that treatment with caffeine in a dose that causes tolerance to several effects of caffeine and increases some effects of adenosine analogues increases the number of A1 receptors without any change in A1 mRNA, suggesting that the adaptive changes are at a post-translational level. There were no significant changes in A2 receptors indicating that the two types are regulated differently and/or that the amount of endogenous agonist is sufficient to regulate A1, but not A2 receptors.