Maintenance of GABA receptor function of small-diameter cockroach neurons by adenine nucleotides.

Small diameter (<20 microm) neurons from the sixth abdominal ganglion of the American cockroach, Periplaneta americana, were enzymatically isolated and responses to exogenously applied gamma-aminobutyric acid (GABA) were recorded using the whole-cell patch clamp technique. With a minimal intracellular medium, responses to repeated applications of GABA decreased to zero within a few minutes. The rate of rundown of GABA responses was decreased by the intracellular inclusion of the phosphatase inhibitors microcystin and okadaic acid, suggesting that phosphorylation is necessary for the maintenance of cockroach GABA receptor function. ATP (5 mM) prevented GABA response rundown. ADP (5 mM) also slowed GABA response rundown, but responses stabilized at a level about half that seen with ATP. In the presence of protein kinase A inhibitory peptide (PKI), ATP was only as efficacious as ADP in slowing rundown. PKI had no effect on the ability of ADP to slow rundown, suggesting that the beta-phosphate of ADP is not involved in PKA-dependent phosphorylation of the GABA receptor. These results suggest that in cockroach neurons, GABA receptor function is maintained intracellularly by adenine nucleotides, not only by phosphorylation, but also possibly by an interaction with a nucleotide recognition site unrelated to PKA-dependent phosphorylation.

Phenytoin delays ischemic depolarization, but cannot block its long-term consequences, in the rat hippocampal slice.

The anticonvulsant phenytoin has been reported to block anoxia-induced losses of synaptic activity in the rat hippocampal slice and experimental ischemia-induced losses of synaptic activity in the guinea pig hippocampal slice. We examined phenytoin in our rat hippocampal slice model of experimental ischemia (anoxia +2 mM D-glucose). In this model, ischemic depolarization (ID) occurs 4-5 min after the introduction of anoxic medium, and oxygen and D-glucose are restored 1 min after the onset of ID. In control slices, synaptic recovery is never observed following ID in 2 mM D-glucose. Phenytoin (30,100 and 300 microM), perfused for 20 min prior to, and for 10 min following anoxia, did not allow for synaptic recovery following ID. At the higher concentrations, however, it did increase the latency to ID. In addition, the presynaptic volley (PV), which normally disappears at the time of ID, was lost substantially earlier in the presence of phenytoin. These findings suggest that the anti-ischemic effects of phenytoin reported by others are due to delay of ID. This may suggest that phenytoin will be effective in preventing global ischemia-induced damage only when the ischemic insult is of short duration.

Assessment of long-term effects of transient anoxia on metabolic activity of rat hippocampal slices using triphenyltetrazolium chloride.

Hippocampal slices maintained in an oxygen-rich static interface chamber remained viable, as determined by the mitochondrial marker 2,3,5-triphenyltetrazolium chloride (TTC), for over 20 h in vitro. By contrast, slices exposed, after 1 h in vitro, to an anoxic environment for 25 min and then allowed to recover for 1-18 h, showed an initial slight decrease in TTC staining followed by a dramatic decrease at time points greater than 6.5 h after anoxia. These data are suggestive of delayed neuronal death. Furthermore, the decreases in TTC staining induced by anoxia could be prevented by conditions known to prevent cell death either in vitro or in vivo. For example, pretreatment of the slices with the N-methyl-D-aspartate antagonist 3-((RS)-2-carboxy-piperazin-4-yl)-propyl-1- phosphonic acid dose-dependently prevented the loss of TTC staining induced by 25 min anoxia. In addition, high-intensity TTC staining correlated with normal CA1 synaptic activity, even after more than 20 h in vitro, suggesting that TTC staining reflects functional neuronal activity. These data suggest that the use of TTC staining of in vitro hippocampal slices may represent a novel and convenient screen for anti-ischemic compounds.

Electrophysiological actions of delta opioids in CA1 of the rat hippocampal slice are mediated by one delta receptor subtype.

Various opioid agonists and antagonists were examined for their ability to alter extracellularly and intracellularly recorded CA1 pyramidal cell activity. All opioid agonists tested, with the exception of [D-ala2]deltorphin II, increased primary population spike amplitude. Of these active agonists, all except DPDPE and p-Cl-DPDPE produced secondary population spikes. DSLET and DAMGO, but not DPDPE, reduced the amplitude of the orthodromically stimulated IPSP. Naltrexone antagonized the actions of all agonists tested. The actions of DPDPE and p-Cl-DPDPE, but not those of DSLET, DAMGO or morphine, were antagonized by the delta antagonist naltrindole. Similarly, the delta antagonist ICI-174,864 blocked the actions of DPDPE, but not DSLET or DAMGO. Based on the inactivity of [D-ala2]deltorphin II and the lack of delta antagonist-sensitive actions of DSLET, the data suggest that the delta 1 subtype is the predominant delta subtype in the CA1 region of the hippocampus.

Polyamine spider toxins are potent un-competitive antagonists of rat cortex excitatory amino acid receptors.

We examined the effect of argiopine and argiopinine 3, low molecular weight polyamine venom components of the spider Argiope lobata, on rat cortical excitatory amino acid (EAA) receptors expressed in Xenopus oocytes. Responses to 100 microM N-methyl-D-aspartate (NMDA) with 10 microM glycine were blocked by both of the polyamine toxins in a dose-dependent manner. Both compounds had similar potencies against 100 microM kainate or 50 microM (S)-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (L-AMPA). Oscillatory responses to 2 microM quisqualate were unaffected by either polyamine toxin. Increasing concentrations of either NMDA, kainate or AMPA were unable to overcome the antagonism by either spider toxin. We were able to demonstrate a use-dependent phenomenon similar to that of phencyclidine; neither polyamine toxin affected the NMDA, kainate or AMPA response without the presence of the respective agonist.

Characterization of indole-2-carboxylate derivatives as antagonists of N-methyl-D-aspartate receptor activity at the associated glycine recognition site.

We have synthesized a series of indole-2-carboxylate derivatives and, with the use of radioligand binding, electrophysiological techniques and an in vivo transient bilateral carotid occlusion model of ischemic damage known to be sensitive to NMDA antagonists, have evaluated the indole-2-carboxylate derivatives ability to inhibit N-methyl-D-aspartate (NMDA) receptor activity through the associated glycine modulatory site. By using [3H]glycine to label this modulatory site, we found that the compounds with the highest affinity (Ki less than 1 microM) contained a chloro group at position C-6 and a polar, hydrogen-bond-accepting group at position C-3 of the indole ring. When these compounds were tested for their ability to modulate [3H]MK-801 [(+)-[3H]-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclophepten-5,10- imine maleate) binding, a functional assessment of NMDA receptor activation, binding was inhibited, indicative of NMDA receptor antagonist character. Schild regression analysis indicated that this antagonism was competitive with glycine. Next, several of these indole-2-carboxylate derivatives were analyzed electrophysiologically in rat cortex mRNA-injected Xenopus oocytes shown to express a functional NMDA receptor channel complex. These compounds inhibited NMDA receptor activity in a manner noncompetitive with NMDA. They also produced a parallel right-ward shift in the glycine dose response for potentiation of the NMDA responses in the oocytes and thus provided further evidence for a competitive interaction at the glycine site. Finally, in vivo transient bilateral carotid artery occlusion experiments revealed that these compounds were capable of reducing the damage typically associated with an ischemic insult in Mongolian gerbil hippocampal neurons.

Pharmacological characteristics of cyclic homologues of glycine at the N-methyl-D-aspartate receptor-associated glycine site.

In Xenopus oocytes, injected with mRNA from the brain of the rat, the characteristics of the cyclic homologues of glycine, ACPC, ACBC and cycloleucine have been examined. 1-Aminocyclopropane-1-carboxylate was a potent agonist at the NMDA-associated glycine site (EC50 = 0.09 +/- 0.02 microM) and exhibited characteristics consistent with a partial agonist. 1-Aminocyclobutane-1-carboxylate, in addition to its previously described antagonist properties, was found to possess agonist properties of low efficacy. Furthermore, ACBC did not completely block NMDA/glycine-induced currents, which is also consistent with partial agonist characteristics. In addition, small concentrations of glycine (less than 3 microM) did not alter the potency of ACBC, possibly suggesting that it is not simply a competitive glycine antagonist. Cycloleucine was a very weak glycine antagonist. These results suggest that as the size of the ring of cyclic homologues of glycine increases, there is a resulting transition from agonist to mixed agonist/antagonist to antagonist properties.

Cis-2,4-methanoglutamate is a potent and selective N-methyl-D-aspartate receptor agonist.

Cis- and trans-2,4-methanoglutamate were compared with L-glutamate as acidic amino acid ligands. Cis-2,4-methanoglutamate had a Ki of 0.052 microM against N-methyl-D-aspartate (NMDA)-specific L-[3H]glutamate binding compared with 0.050 microM for L-glutamate. Cis-2,4-methanoglutamate exhibited no significant affinity against [3H]kainate or [3H]alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate ([3H]AMPA) binding. Trans-2,4-methanoglutamate had no significant affinity for any of these sites. Cis-2,4-methanoglutamate increased [3H]N-1[2-thienyl]cyclohexyl-3,4-piperidine [( 3H]TCP) binding with EC50 of 0.35 +/- 0.14 microM. It produced an inward current in rat brain mRNA-injected Xenopus oocytes which was blocked by the NMDA antagonist, D-2-amino-7-phosphonoheptanoate (D-AP7). Cis-2,4-methanoglutamate (EC50 = 15.9 microM) was 100-fold more potent than L-glutamate (EC50 = 1,584 microM) in reducing the excitatory postsynaptic potential in CA1 of hippocampal slices. Cis-2,4-methanoglutamate is the most potent, selective NMDA agonist known.

NMDA receptor antagonists attenuate a portion of the penicillin-induced epileptiform burst.

In the CA1 region of the rat hippocampal slice, epileptiform activity was induced by the GABAA antagonist penicillin (PEN, 3.4 mM). The competitive N-methyl-D-aspartate (NMDA) receptor antagonists D-2-amino-7-phosphonoheptanoate (D-AP7) and D-2-amino-5-phosphonovolerate (D-AP5) attenuated extracellularly recorded evoked burst duration, the number of population spikes per evoked bursts and the frequency of spontaneously occurring bursts, but did not affect the sum of the population spike amplitudes or the evoked burst coastline measures due to increases in amplitude of the remaining secondary population spikes. Intracellular recordings showed that many of the secondary action potentials in the PEN burst were decreased in amplitude and broadened in duration, perhaps due to spike inactivation. D-AP7 allowed these secondary action potentials to increase in amplitude, which could explain the increases in secondary population spike amplitude seen extracellularly. Decrements in stimulus strength can mimic the effect of D-AP7 on PEN bursts. These data suggest that there is a portion of the PEN-induced epileptiform burst which is sensitive to NMDA antagonists.

D-cycloserine acts as a partial agonist at the glycine modulatory site of the NMDA receptor expressed in Xenopus oocytes.

In the Xenopus oocyte preparation, D-cycloserine (DCS) had the profile of a partial agonist at the glycine modulatory site of the NMDA receptor. Maximal NMDA responses in the presence of DCS were only 40-50% of those in the presence of glycine. Glycine and D-serine were agonists at this site. The actions of DCS were competitively antagonized by HA-966, a known glycine antagonist.

Epileptiform activity in vitro can produce long-term synaptic failure and persistent neuronal depolarization.

A large, extracellular negative DC shift, termed epileptic depolarization, could be elicited during zero magnesium-induced epileptic activity in the rat hippocampal slice. In 10 mM glucose medium, epileptic depolarization was elicited by high-frequency synaptic stimulation. During epileptic depolarization synaptic responses were abolished, but recovered in 10.4 +/- 2.1 min. In low glucose (2 mM) medium, epileptic depolarization either occurred spontaneously or could be elicited by high frequency synaptic stimulation, and no recovery of synaptic responses was observed for at least 30 min. This long-term synaptic failure was blocked by the competitive NMDA antagonists, 3-[+/-)-2-carboxypiperazin-4-yl)-propyl-1-phosphonate (CPP, 100 microM) and D-2-amino-7-phosphonoheptanoate (D-AP7, 100 microM) when added at the peak of epileptic depolarization, but not 5 min afterwards. Intracellular analysis showed that this extracellular DC shift was correlated with a membrane depolarization which approached 0 mV. With 10 mM glucose medium, the membrane potential returned to resting level in 6.3 +/- 1.9 min. In 2 mM glucose medium, neurons remained depolarized and no recovery was observed. This persistent depolarization could account for the loss of synaptic function recorded extracellularly. Application of 100 microM CPP blocked persistent depolarization and allowed for the recovery of the membrane potential. Epileptic depolarization was also observed during picrotoxin-induced epileptic activity. Both anoxic depolarization during experimental ischemia and epileptic depolarization can trigger long-term synaptic failure and persistent depolarization. Epileptic depolarization and anoxic depolarization may be triggers which can lead to neuronal failure in diseases associated with neuronal degeneration.

Glycine antagonist action of 1-aminocyclobutane-1-carboxylate (ACBC) in Xenopus oocytes injected with rat brain mRNA.

ACBC has been reported to have the binding profile of an antagonist at the glycine site of the NMDA receptor. In Xenopus oocytes injected with rat brain mRNA, we have confirmed the antagonist action of ACBC on NMDA responses. ACBC and HA-966, a known glycine antagonist, blocked NMDA responses in a non-competitive manner and blocked the potentiation of NMDA responses by glycine in a competitive manner. We conclude that ACBC blocks NMDA responses via a competitive interaction at the glycine modulatory site.