Search

The presence of metabotropic glutamate receptors (mGluRs) of group II modulating glycine exocytosis from glycinergic nerve endings of mouse spinal cord was investigated. Purified synaptosomes were selectively prelabeled with [(3)H]glycine through the neuronal transporter GlyT2 and subsequently depolarized by superfusion with 12 mM KCl. The selective mGluR2/3 agonist LY379268 inhibited the K(+)-evoked overflow of [(3)H]glycine in a concentration-dependent manner (EC(50) about 0.2 nM). The effect of LY379268 was prevented by the selective mGluR2/3 antagonist LY341495 (IC(50) about 1 nM). N-acetylaspartylglutamate (NAAG) inhibited [(3)H]glycine overflow with extraordinary potency (EC(50) about 50 fmol). In contrast, glutamate was ineffective up to 0.1 nM, excluding that glutamate contamination of commercial NAAG samples is responsible for the reported activity of NAAG at mGluR3. LY341495 antagonized the NAAG inhibition of [(3)H]glycine release. The effect of a combination of maximally effective concentrations of LY379268 and NAAG exhibited no additivity. The non-hydrolysable NAAG analogue N-acetylaspartyl-β-linked glutamate (β-NAAG) antagonized NAAG and LY379268. In conclusion, our results show that glycinergic nerve endings in spinal cord are endowed with group II mGluRs mediating inhibition of glycine exocytosis. NAAG can activate these presynaptic receptors with extremely high affinity and with characteristics compatible with the reported mGluR3 pharmacology. This article is part of a Special Issue entitled 'Metabotropic Glutamate Receptors'.

Adenosine triphosphate (ATP) has been reported to enhance the release of glutamate by acting at P2X presynaptic receptors. Acetylcholine (ACh) can elicit glutamate release through presynaptic nicotinic cholinergic receptors (nAChRs) of the alpha7 subtype situated on glutamatergic axon terminals, provided that the terminal membrane is weakly depolarized. Considering that ATP and ACh are co-transmitters, we here investigate on the possibility that P2X and nAChRs co-exist and interact on the same glutamatergic nerve endings using purified rat neocortex synaptosomes in superfusion. ATP evoked Ca(2+)-dependent release of pre-accumulated D-[(3)H]aspartate ([(3)H]D-ASP) as well as of endogenous glutamate; (-)-nicotine, inactive on its own, potentiated the ATP-evoked release. The ATP analogue benzoylbenzoylATP (BzATP) behaved like ATP, but was approximately 30 times more potent; the potentiation of the BzATP-evoked release was blocked by methyllycaconitine or alpha-bungarotoxin. Adding inactive concentrations of (-)-nicotine, epibatidine or choline together with inactive concentrations of BzATP resulted in significant elevation of the [(3)H]D-ASP release mediated by alpha7 nAChRs. To conclude, P2X(7) receptors and alpha7 nAChRs seem to co-exist and interact on rat neocortex glutamatergic terminals; in particular, P2X(7) receptors exert a permissive role on the activation of alpha7 nAChRs, suggesting that ATP may not only evoke glutamate release on its own, but may also regulate the release of the amino acid elicited by ACh.

Postsynaptic glutamate AMPA receptors (AMPARs) can recycle between plasma membrane and intracellular pools. In contrast, trafficking of presynaptic AMPARs has not been investigated. AMPAR surface expression involves interactions between the GluR2 carboxy tail and various proteins including glutamate receptor-interacting protein (GRIP), AMPA receptor-binding protein (ABP), protein interacting with C kinase 1 (PICK1), N-ethyl-maleimide-sensitive fusion protein (NSF). Here, peptides known to selectively block the above interactions were entrapped into synaptosomes to study the effects on the AMPA-evoked release of [3H]noradrenaline ([3H]NA) and [3H]acetylcholine ([3H]ACh) from rat hippocampal and cortical synaptosomes, respectively. Internalization of pep2-SVKI to prevent GluR2-GRIP/ABP/PICK1 interactions potentiated the AMPA-evoked release of [3H]NA but left unmodified that of [3H]ACh. Similar potentiation was caused by pep2-AVKI, the blocker of GluR2-PICK1 interaction. Conversely, a decrease in the AMPA-evoked release of [3H]NA, but not of [3H]ACh, was caused by pep2m, a selective blocker of the GluR2-NSF interaction. In the presence of pep2-SVKI the presynaptic AMPARs on noradrenergic terminals lost sensitivity to cyclothiazide. AMPARs releasing [3H]ACh, but not those releasing [3H]NA, were sensitive to spermine, suggesting that they are GluR2-lacking AMPARs. To conclude: (i) release-regulating presynaptic AMPARs constitutively cycle in isolated nerve terminals; (ii) the process exhibits neuronal selectivity; (iii) AMPAR trafficking and desensitization may be interrelated.

The coexistence of nicotinic cholinergic receptors (nAChRs) and of N-methyl- d -aspartate (NMDA) receptors on the same noradrenergic axon terminals and the nAChR/NMDA receptor cross-talk were investigated by monitoring the release of noradrenaline (NA) evoked in superfused rat hippocampal synaptosomes by (−)-nicotine and NMDA alone or in combination. In medium containing a physiological concentration (1.2 mM) of Mg2+, the release of [3H]NA was very slightly increased by NMDA plus glycine, whereas it was significantly enhanced by (−)-nicotine. The (−)-nicotine/NMDA combination elicited supraadditive release which was totally abolished by the nAChR blocker mecamylamine and partly prevented by selectively blocking NMDA receptors. Supraadditive [3H]NA release was also observed by exposing synaptosomes to veratrine, but not to ionomycin. The supraadditive release elicited by the (−)-nicotine/NMDA or the veratrine/NMDA combination was sensitive to the protein kinase A/C inhibitor staurosporine and the selective protein kinase A inhibitor H89, but insensitive to the protein kinase C inhibitor Ro 31-8220. It is concluded that (i) release-modulating nAChRs and NMDA receptors coexist on hippocampal noradrenergic axon terminals; and (ii) nicotine permits NMDA receptor activation in the presence of Mg2+, possibly because the nicotine-induced influx of Na+ depolarizes the nerve ending membrane sufficiently to remove the Mg2+ block.

The release of [3H]D-aspartate ([3H]D-ASP) or [3H]GABA evoked by glycine and that of [3H]D-ASP or [3H]glycine evoked by GABA from spinal cord synaptosomes were studied in SOD1-G93A(+) mice, a transgenic model of amyotrophic lateral sclerosis, SOD1(+) mice and SOD1(-)/G93A(-) animals. Mutant mice were killed at advanced phase of pathology or during the presymptomatic period. In SOD1(-)/G93A(-) or SOD1(+) mice glycine evoked [(3)H]d-ASP and [(3)H]GABA release, while GABA caused [3H]D-ASP, but not [3H]glycine, release. The glycine-evoked release of [3H]D-ASP, but not that of [3H]GABA, and the GABA-evoked [3H]D-ASP release, but not that of [3H]glycine, were more pronounced in SOD1-G93A(+) than in SOD1(+) mice. Furthermore, these potentiations were already present in asymptomatic 30- to 40-day-old mice. Basal [3H]D-ASP release was also higher in SOD1-G93A(+) than SOD1(+) or SOD1(-)/G93A(-) mice. The release of endogenous glutamate and GABA was also enhanced in asymptomatic animals; the glycine-evoked release of endogenous glutamate, but not of endogenous GABA, was higher in SOD1-G93A(+) than in SOD1(+) animals. The effects of glycine and GABA were insensitive to receptor blockers, but sensitive to transporter inhibitors, indicating coexistence of glutamate and glycine transporters and of glutamate and GABA transporters on glutamate-releasing terminals. The glutamate release machinery seems excessively functional in SOD1-G93A(+) animals.

Rat cerebellar synaptosomes labeled with [ 3 H ] d -aspartate ( [ 3 H ] d -ASP) were exposed in superfusion to muscimol. The GABAA receptor agonist did not affect [ 3 H ] d -ASP basal release or the overflow provoked by 15 mM K+; muscimol potentiated the 35 mM K+-evoked overflow of [ 3 H ] d -ASP or endogenous glutamate. Membrane potential measured by Rhodamine 6G fluorescence was −65 mV under resting conditions and −32 mV in the presence of 35 mM K+. The membrane potential was not significantly affected by muscimol. The muscimol effect on the K+(35 mM)-evoked [ 3 H ] d -ASP overflow was not inhibited by omitting external Ca2+ or by entrapping BAPTA to chelate cytosolic Ca2+. Muscimol lost its ability to release glutamate following superfusion with d -aspartate to deplete cytosolic glutamate by heteroexchange suggesting that GABAA receptor activation elicits release of cytosolic glutamate. The non-transportable glutamate carrier blockers dihydrokainate or dl -TBOA did not reduce the muscimol potentiation. This was abolished by the anion channel blockers niflumic acid and NPPB. To conclude, when cerebellar parallel fiber terminals are sufficiently depolarized, activation of α6 GABAA receptors on these terminals mediates glutamate release in addition to that evoked by depolarization. This extra-release does not occur by exocytosis or transporter reversal but involves the opening of anion channels present on parallel fiber terminals.

Intrahippocampal perfusion of bicuculline (50 microM) in Mg2+-free medium caused elevation of extracellular cGMP and epileptic-like behaviour. Both effects were partially prevented by blocking NMDA receptors with MK-801 or Mg2+ ions. Similarly, the GABA(B) receptor antagonists CGP52432 (0.1-30 microM) and CGP35348 (0.3-1 mM) evoked increases of extracellular cGMP. CGP52432 also elicited behavioural responses ranging from wet dog shakes to convulsions. MK-801 or Mg2+ ions reduced the effects of CGP52432. Local application of muscimol (100-300 microM) or (-)baclofen (300 microM) caused inhibition of extracellular cGMP. Administration of the AMPA/kainate receptor antagonist NBQX (100 microM) caused cGMP elevation which was almost abolished by co-perfusion of muscimol and (-)baclofen. In the presence of physiological Mg2+, perfusion of AMPA (30 microM) failed to affect cGMP levels, although rats displayed wet dog shakes episodes. When AMPA was co-perfused with low concentrations of bicuculline or CGP52432, cGMP elevations were observed in 60% of the rats. Addition of both antagonists to AMPA resulted in 85% of rats displaying a cGMP response. To conclude: (a) extracellular hippocampal cGMP is controlled by inhibitory GABA(A) and GABA(B) receptors tonically activated through GABAergic interneurons receiving AMPA/kainate-mediated glutamatergic inputs; (b) the GABAergic receptors are not endogenously saturated and can be further stimulated by exogenous agonists; (c) blockade of the GABA-mediated inhibition causes increase of cGMP and epileptic-like behaviour, due largely to endogenous activation of NMDA receptors; (d) reproducible cGMP responses to AMPA can be observed when the inhibitory GABAergic inputs to the NO/guanylyl cyclase system are blocked, confirming the previously proposed existence of AMPA/kainate receptors able to increase the nucleotide synthesis.

Glutamate extracellular accumulation is an early event in brain ischemia triggering excitotoxic neuron damage. We have investigated how to control the glutamate efflux from human cerebrocortical slices superfused in conditions simulating an acute ischemic insult (oxygen and glucose deprivation). The efflux of previously accumulated [ 3 H]D-aspartate or endogenous glutamate increased starting 18 min after exposure to ischemia and returned almost to basal values in 6 min reperfusion with standard medium. Superfusion with Ca 2+ -free, EGTA (0.5 mM)-containing medium or with medium containing tetrodotoxin (TTX; 0.5 μM) inhibited the ischemia (24 min)-evoked [ 3 H]D-aspartate efflux by about 50% and 65%, respectively. The ischemia (24 or 36 min)-evoked efflux of [ 3 H]D-aspartate or endogenous glutamate was reduced at least 40% by the adenosine A 2A receptor antagonist SCH 58261 (1 μM); the compound was effective when added up to 15 min after exposure to ischemia. No effect of SCH 58261 on the ischemia-evoked [ 3 H]D-aspartate was found in Ca 2+ -free, EGTA-containing medium. To conclude, a significant component of the ischemia-evoked glutamate efflux in human cerebrocortical slices seems to occur by a vesicular-like mechanism. Endogenously released adenosine is likely to activate A 2A receptors that enhance vesicular-like glutamate release during ischemia; A 2A receptor antagonists would deserve consideration for their neuroprotective potential.

We have investigated the functional relationships between NMDA receptors and the NOS/sGC system in the rat pre-frontal cortex in vivo by microdialysis. cGMP basal levels were sensitive to NOS or sGC inhibitors (L-NARG or ODQ) or NO donors (SNAP) when enzymatic breakdown was blocked by the phosphodiesterase inhibitor IBMX, indicating that basal cGMP production derives, at least in part, from the NOS/sGC pathway activity and that the pre-frontal cortex possesses a very efficient degradation system for cGMP. The glutamate receptor agonist NMDA did not alter extracellular cGMP either in absence or presence of IBMX. cGMP was not augmented when NMDA was co-infused with the NOS substrate L-arginine, the glycine site agonist d-serine or the glutamate receptor agonist AMPA. Interestingly, the selective GABA(A) receptor antagonist bicuculline enhanced cGMP production, revealing that the cortical NOS/sGC system is tonically inhibited by endogenous GABA. However, in the presence of bicuculline, NMDA did not increase extracellular cGMP. In the presence of bicuculline, blockade of 5-HT1/2 receptors, known to inhibit the NMDA/NOS/sGC pathway, with the antagonist methiothepin did not unmask cGMP elevations by NMDA. Thus, it would seem that NMDA receptors do not regulate cortical NOS/sGC activity that, on the other hand, is modulated by endogenous GABA acting at GABA(A) receptors.

Adenosine released during cerebral ischemia is considered to act as a neuroprotectant, possibly through the inhibition of glutamate release. The involvement of A(1) and A(2A) receptors in the control of the rise of extracellular glutamate during ischemia was investigated by monitoring the effects of selective A(1) and A(2A) receptor antagonists on ischemia-evoked glutamate release in rat cerebrocortical slices.Slices were superfused with oxygen- and glucose-deprived medium and [(3)H]D-aspartate or endogenous glutamate was measured in the superfusate fractions. Withdrawal of Ca(2+) ions or addition of tetrodotoxin more than halved the ischemia-evoked efflux of [(3)H]D-aspartate or glutamate, compatible with a vesicular-like release. The glutamate transporter inhibitor DL-TBOA prevented the ischemia-evoked efflux of [(3)H]D-aspartate by about 40%, indicating a carrier-mediated efflux. The ischemia-evoked efflux of [(3)H]D-aspartate or glutamate was increased by the A(1) receptor antagonist DPCPX. The A(2A) antagonist SCH 58261 decreased [(3)H]D-aspartate or endogenous glutamate efflux (50 and 55% maximal inhibitions; EC(50): 14.9 and 7.6 nM, respectively); the drug was effective also if added during ischemia. No effect of either the A(1) or the A(2A) receptor antagonist was found on the ischemia-evoked efflux of [(3)H]D-aspartate in Ca(2+)-free medium. Our data suggest that adenosine released during cerebral ischemia can activate inhibitory A(1) and stimulatory A(2A) receptors that down- or up-regulate the vesicular-like component of glutamate release.

Previously, we have shown that presynaptic GABA(B) receptors regulating the release of various transmitters from CNS terminals can be differentially blocked by GABA(B) antagonists suggesting the existence of pharmacologically distinct GABA(B) receptor subtypes. We here examined the ability of CGP 36742 [(3-aminopropyl)n-butylphosphinic acid], a selective GABA(B) antagonist endowed with cognition enhancing activity, to block release-regulating GABA(B) receptors. In particular, CGP 36742 was tested against the inhibition of the depolarization-evoked release of GABA, glutamate, cholecystokinin and somatostatin produced by (-)baclofen in rat and human neocortex axon terminals. CGP 36742 potently antagonized (IC50 = 0.14 microM) the inhibition by (-)baclofen of somatostatin release from superfused rat neocortex synaptosomes. In contrast, the effects of (-)baclofen on GABA, glutamate and cholecystokinin release were insensitive to CGP 36742, at concentrations of up to 100 microM. In human neocortex synaptosomes CGP 36742 exhibited a pattern of selectivity identical to that in rat synaptosomes, although the antagonist was at least 10-fold less potent in human than in rat brain. CGP 36742 is the first compound displaying great selectivity for the GABA(B) presynaptic receptors regulating somatostatin release. Considering the proposed implication of the neuropeptide in cognitive processes, disinhibition of somatostatin release merits consideration as one of the mechanisms possibly involved in the behavioral activity of CGP 36742.

Granule cell axon terminals of rat cerebellum possess benzodiazepine-insensitive GABA A receptors mediating glutamate release. We have investigated the ability of benzodiazepines, ethanol and furosemide to modulate the function of these receptors in the cerebellum of alcohol-tolerant (AT) and alcohol-nontolerant (ANT) rats. AT and ANT synaptosomes, prelabeled with [ 3 H] d -aspartate, were superfused with GABA and various drugs during the K + -depolarization. GABA similarly enhanced [ 3 H] d -aspartate overflow in AT (EC 50 =1.7 μM) and ANT (EC 50 =3.9 μM) rats in a bicuculline-sensitive manner. Diazepam or zolpidem, at 0.1 μM, potentiated GABA at the GABA A receptor of ANT rats, but were ineffective at the AT receptor. Zolpidem acted with great potency (EC 50 =13.6 nM). Ethanol, added at 50 mM, potentiated GABA in ANT rats, but it was inactive at the GABA A receptor of the AT cerebellum. Furosemide significantly inhibited the effect of GABA in ANT, but not in AT synaptosomes. Our results show that one GABA A receptor (the receptor sited on granule cell terminals which mediates glutamate release) exhibits functional responses to diazepam and ethanol that differ between AT and ANT rats. However, the data with zolpidem and furosemide differ from previous results obtained with membranes of the granule cell layer suggesting that distinct GABA A receptor subtypes may exist on axon terminals versus soma/dendrites of granule cells.

The neurochemical and autonomic pharmacological profile of 1,2,3,4,10, 14b-hexahydro-2-methyl-pyrazino[2,1-a]pyrido[2,3-c]pyrido[2, 3-c] [2] benzazepine [+/-)Org 3770) and the related antidepressant drug, mianserin, have been compared. The uptake of [3H]noradrenaline ([3H]NA) in vitro was weakly affected by (+/-)Org 3770 (pKi = 5.6) in contrast to mianserin (pKi = 7.4). Both (+/-)Org 3770 and mianserin facilitated the release of [3H]NA in slices of cortex. The effects of NA mediated by alpha 2-adrenoceptors on the release of both [3H]NA or [3H]serotonin ([3H]5-HT) were antagonized by (+)Org 3770 with pKi values of 8.4 and 8.1, respectively. However, (-)Org 3770 only antagonized the effect of NA on the release of [3H]5-HT (pA2 = 7.7). The binding of [3H]rauwolscine to alpha 2-adrenoceptors was inhibited by (+/-)Org 3770 and mianserin with identical affinity (pKi = 7.0), whereas the binding of [3H]prazosin to alpha 1-adrenoceptors was less potently affected by (+/-)Org 3770 (pKi = 6.4) than by mianserin (pKi = 7.1). A similar difference was found for alpha 1- and alpha 2-adrenoceptors in vas deferens of the rat. The binding of [3H]mianserin to 5-HT2 receptors was less potently blocked by (+/-)Org 3770 (pKi = 8.1) than by mianserin (pKi = 9.4) while the binding of [3H]mepyramine to histamine-1 receptors was more potently affected by (+/-)Org 3770 (pKi = 9.3) than by mianserin (pKi = 8.75). The binding of [3H]quinuclidinylbenzilate to muscarinic cholinergic receptors was blocked equally by (+/-)Org 3770 (pKi = 6.1) and mianserin (pKi = 6.3). Similar data on tryptamine-D, histamine-1 and muscarinic cholinergic receptors in isolated organs were obtained. A prominent role for the blockade of alpha 2-adrenoceptors in the therapeutic effects of mianserin and (+/-)Org 3770 in depression is suggested, probably excluding a role of inhibition of the uptake of NA.

The effects of (−)-propranolol, (+)-propranolol and (±)cyanopindolol on the release of [ 3 H]5-hydroxytryptamine ([ 3 H]5-HT) were investigated in synaptosomes from the hippocampus of the rat, depolarized in superfusion with 15mM KCl. (−)-Propranolol, but not (+)-propranolol, inhibited in a concentration-dependent way the K + -evoked release of [ 3 H]5-HT. (±)-Cyanopindolol behaved similarly but was about 10 times more potent than (−)-propranolol. The inhibitory effects of (−)-propranolol and (±)-cyanopindolol were prevented by the autoreceptor antagonist methiothepin. Both β-adrenoceptor antagonists antagonized the inhibition by exogenous 5-HT of the K + -evoked release of [ 3 H]5-HT. The data suggest that some β-adrenoceptor antagonists may behave as mixed agonists-antagonists at the 5-HT autoreceptor. Synaptosomes in superfusion appear to be particularly suitable to study separately agonistic compared to antagonistic activity of compounds having a mixed agonist-antagonist profile.

In the present study the release of dopamine (DA) from a synaptosomal preparation obtained from the nerve endings of the tuberoinfundibular dopaminergic nerves has been analysed, utilizing a superfusion technique. The exposure of median eminence synaptosomes to increasing concentrations of K + ions produced a dose-dependent efflux of preaccumulated [ 3 H]DA. The removal of Ca 2+ ions from the superfusion medium completely prevented the high K + -induced release of [ 3 H]DA. Removal of Na + ions from the superfusion medium, a condition creating a favourable outward Na + gradient, produced a marked release of [ 3 H]DA. In order to establish whether this efflux of dopamine was a carrier-mediated process nomifensine, a dopamine transport inhibitor devoid of dopamine releasing effect in the median eminence, was added to the superfusion medium. Nomifensine largely counteracted the release of dopamine induced by the lack of Na + . d -Amphetamine produced a dose-related stimulation of [ 3 H]DA in synaptosomes from the median eminence. The possible existence of a direct control of dopamine release mediated by presynaptic receptors was tested by studying the effect of apomorphine on the efflux of preaccumulated [ 3 H]DA. Apomorphine was unable to modify both the spontaneous and the K + -evoked release of dopamine. The present results give support to the idea that dopamine released from the nerve endings of tuberoinfundibular neurones into the primary plexus acts as a neurotransmitter. The release of dopamine in these nerves does not seem to be regulated by autoreceptors.