Modulation of low calcium induced field bursts in the hippocampus by monoamines and cholinomimetics

Glutamate and dopamine receptors contribute to the lateral spread of epileptiform discharges in rat neocortical slices

PURPOSE

The effects of AMPA-type glutamate receptor as well as dopamine D1 and D2 receptors on the lateral propagation of epileptiform field potentials (EFP) were studied across adjacent areas of rat neocortical tissues.

METHODS

Epileptiform burst discharges were induced by superfusion of Mg(2+)-free artificial cerebrospinal fluid. Simultaneous field potential recordings of EFP were obtained from four microelectrodes placed 2-3 mm apart across coronal slices in the third layer of the neocortex. The effects of AMPA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), dopamine D1 receptor agonist SKF 81297, and dopamine D2 receptor agonist quinpirole on lateral propagation of burst discharges were investigated.

RESULTS

CNQX, applied focally between recording sites, blocked rapid propagation across treated areas and resulted in the emergence of spatially separate, independent pacemakers. Focal application of SKF 81297 between recording sites increased the repetition rate of EFP, but reduced the amplitude as well as the duration of epileptic discharges. However, addition of SKF 81297 to the bath medium abolished EFP. Both local and systemic applications of quinpirole irreversibly enhanced repetition rate of epileptiform burst discharges.

CONCLUSIONS

The results indicate the prerequisite of AMPA synaptic transmission for synchronized lateral propagation of Mg(2+)-free ACSF-induced epileptic activity and the modulatory effects of dopamine D1 and D2 receptors on both EFP initiation and propagation in epileptic tissues.

KCNQ/Kv7 channel regulation of hippocampal gamma-frequency firing in the absence of synaptic transmission

Synchronous neuronal firing can be induced in hippocampal slices in the absence of synaptic transmission by lowering extracellular Ca2+ and raising extracellular K+. However, the ionic mechanisms underlying this nonsynaptic synchronous firing are not well understood. In this study we have investigated the role of KCNQ/Kv7 channels in regulating this form of nonsynaptic bursting activity. Incubation of rat hippocampal slices in reduced (<0.2 mM) [Ca2+]o and increased (6.3 mM) [K+]o, blocked synaptic transmission, increased neuronal firing, and led to the development of spontaneous periodic nonsynaptic epileptiform activity. This activity was recorded extracellularly as large (4.7 +/- 1.9 mV) depolarizing envelopes with superimposed high-frequency synchronous population spikes. These intraburst population spikes initially occurred at a high frequency (about 120 Hz), which decayed throughout the burst stabilizing in the gamma-frequency band (30-80 Hz). Further increasing [K+]o resulted in an increase in the interburst frequency without altering the intraburst population spike frequency. Application of retigabine (10 microM), a Kv7 channel modulator, completely abolished the bursts, in an XE-991-sensitive manner. Furthermore, application of the Kv7 channel blockers, linopirdine (10 microM) or XE-991 (10 microM) alone, abolished the gamma frequency, but not the higher-frequency population spike firing observed during low Ca2+/high K+ bursts. These data suggest that Kv7 channels are likely to play a role in the regulation of synchronous population firing activity.

Dopamine Enhances Spatiotemporal Spread of Activity in Rat Prefrontal Cortex

Dopaminergic modulation of prefrontal cortex (PFC) is important for neuronal integration in this brain region known to be involved in cognition and working memory. Because of the complexity and heterogeneity of the effect of dopamine on synaptic transmission across layers of the neocortex, dopamine's net effect on local circuits in PFC is difficult to predict. We have combined whole cell patch-clamp recording and voltage-sensitive dye imaging to examine the effect of dopamine on the excitability of local excitatory circuits in rat PFC in vitro. Whole cell voltage-clamp recording from visually identified layer II/III pyramidal neurons in rat brain slices revealed that, in the presence of bicuculline (10 μM), bath-applied dopamine (30–60 μM) increased the amplitude of excitatory postsynaptic currents (EPSCs) evoked by weak intracortical stimulus. The effect was mimicked by the selective D1 receptor agonist SKF 81297 (1 μM). Increasing stimulation resulted in epileptiform discharges. SKF 81297 (1 μM) significantly lowered the threshold stimulus required for generating epileptiform discharges to 83% of control. In the imaging experiments, bath application of dopamine or SKF 81297 enhanced the spatiotemporal spread of activity in response to weak stimulation and previously subthreshold stimulation resulted in epileptiform activity that spread across the whole cortex. These effects could be blocked by the selective D1 receptor antagonist SCH 23390 (10 μM) but not by the D2 receptor antagonist eticlopride (5 μM). These results indicate that dopamine, by a D1 receptor–mediated mechanism, enhances spatiotemporal spread of synaptic activity and lowers the threshold for epileptiform activity in local excitatory circuits within PFC.

Changes in hippocampal histamine receptors across the hibernation cycle in ground squirrels

Hibernation is a physiological state characterized by a dramatic reduction in various functions, such as body temperature, heart rate, and metabolism. The hippocampus is thought to be important for regulation of the hibernation bout because it remains electrophysiologically active throughout this extremely depressed state. The question arises as to what neuronal influences act within the hippocampus during hibernation to sustain its activity. We hypothesized that histaminergic input might be an important contributor. Brain histamine is involved in functions relevant to hibernation, such as the regulation of diurnal rhythms, body temperature, and energy metabolism. Furthermore, we have previously shown that the histaminergic system appears to be activated during the hibernating state. In this study, we used receptor binding autoradiography, in situ hybridization, and GTP-gamma-S binding autoradiography to study changes in histamine receptors across the hibernation bout. We were able to demonstrate an increase in histamine H1 and H2 receptors in the hippocampus during hibernation, whereas the mRNA expression and receptor density of the inhibitory H3 receptor decreased. Histamine H3 receptors were shown to exhibit both histamine-activated and constitutive GTP-gamma-S-binding activity in the ground squirrel hippocampus, both of which decreased during hibernation, indicating a decrease in H3 receptor G-protein activation. Taken together, our results indicate that histamine may be involved in maintaining hibernation by sustaining hippocampal activity, possibly through H1 and H2 receptor activity and decreased inhibition by H3 receptors. The involvement of brain histamine, which is generally thought of as an arousal molecule, in maintaining a depressed state of the brain suggests a more general role for the amine in controlling arousal state.

Group I mGluRs modulate the pattern of non-synaptic epileptiform activity in the hippocampus

The hippocampus is well known for its susceptibility to epileptic seizures, in part because of its neuronal architecture that facilitates synchronization. Although synaptic networks are important for the genesis and spread of epileptiform activity, synchronization of neuronal activity can occur when action potential-dependent chemical synaptic transmission is absent. In particular, it is possible to induce epileptiform activity by perfusing hippocampal slices with a low-Ca(2+)/high-K(+) mediums. Using extracellular recording in area CA1 we have characterized the effects of metabotropic glutamate receptor (mGluR) activation on this non-synaptic bursting activity. Under control conditions, bursting occurred at intervals of 14-86 s with each burst comprising a long (up to 44 s) negative-going field potential of 2 to 13 mV superimposed upon which was sustained firing of population spikes. Activation of group I mGluRs by (S)-3,5-dihydroxyphenylglycine (DHPG) (25 microM) caused a dramatic increase in burst frequency (up to five-fold), which was accompanied by a decrease in the duration and amplitude of bursts. The selective mGluR(1) antagonist 2-methyl-4-carboxyphenylglycine (LY367385) and the selective mGluR(5) antagonist 2-methyl-6-(phenylethynyl)pyridine (MPEP) both restricted the increase in burst frequency induced by DHPG. However, only LY367385 inhibited the decrease in burst duration and amplitude. Combined application of both antagonists prevented all DHPG-induced changes in bursting activity. These data provide evidence for a role of both mGluR(1) and mGluR(5) subtypes in changing the frequency of non-synaptic bursting, with mGluR(1) alone causing alterations in burst duration and amplitude. These effects are likely to contribute to the group I mGluR-induced changes in synaptic epileptic activity that are already well documented.

Epileptiform discharge induced by 4-aminopyridine in magnesium-free medium in neocortical neurons: physiological and pharmacological characterization

An in vitro model of epileptiform activity was developed to study the role of excitatory and inhibitory neurotransmitters in the epileptogenesis. Intracellular recordings were obtained from rat neocortical slices exposed to 4-aminopyridine in a magnesium-free solution. Spontaneous epileptiform activity consisting of paroxysmal depolarization shifts with associated spontaneous depolarizing postsynaptic potentials were observed. The paroxysmal depolarization shifts were blocked either by D,L-2-amino-5-phosphonovalerate (50 microM), an N-methyl-D-aspartate receptor antagonist, or by 6-cyano-7-nitroquinoxaline-2.3-dione (10 microM), a non-N-methyl-D-aspartate receptor antagonist. These glutamate receptor antagonists also reduced the occurrence of spontaneous depolarizing postsynaptic potentials. Bicuculline methiodide, an antagonist of GABAA receptors, suppressed spontaneous depolarizing postsynaptic potentials, while it reduced the frequency of paroxysmal depolarization shifts and increased their duration. Hyperpolarization of the membrane potential by continuous current injection increased the frequency of paroxysmal depolarization shifts and reduced their duration, but it reduced the occurrence of spontaneous postsynaptic potentials. Paroxysmal depolarization shifts were blocked by tetrodotoxin (1 microM). The duration and the frequency of paroxysmal depolarization shift were reduced by dopamine (30-300 microM) in a dose-dependent manner. Our model suggests a different involvement of excitatory and inhibitory processes in the generation of epileptiform activity.

Suppression of drug-induced epileptiform discharges by cyclic AMP in rat hippocampus

The effect of cyclic adenosine 3',5'-monophosphate (cAMP) on epileptiform activity in rat hippocampal slices was investigated. Bath-applied cAMP reversibly decreased the frequency of extracellularly recorded discharges in the CA3 subfield induced by bethanechol- or theophylline-containing solutions. Because cAMP was presumed to be relatively membrane impermeant, we developed and tested the hypothesis that this cAMP-mediated effect occurred extracellularly through the catabolic conversion of cAMP to 5'-AMP and, in turn, to adenosine, a known inhibitory neuromodulator. Three predictions derived from this catabolic hypothesis were tested. First, blockers of the enzymes involved were predicted to antagonize this effect of cAMP. In contrast, the coapplication of a cAMP-phosphodiesterase inhibitor, 3-isobutyl-1-methylxanthine (IBMX), or a 5'-nucleotidase inhibitor, adenosine 5'-[alpha, beta-methylene] diphosphate (AMP-CP), enhanced the cAMP-induced suppressive effect. Second, the nonhydrolyzable cAMP analogs, dibutyryl- and 8-bromo-cAMP, were predicted to be ineffective. Low concentrations (5-40 microM) of these two derivatives, however, also suppressed bethanechol-induced discharges, while, at a higher concentration (100 microM), both analogs increased discharge frequencies. Third, enzymatic catabolism of adenosine was predicted to antagonize cAMP's effect, but coapplying adenosine deaminase (10 U/mL) did not diminish this action. Because these data did not support the catabolic hypothesis, other, as yet undefined, mechanisms must be responsible for the discharge-suppressant effect of cAMP.

The role of dopamine in epilepsy

The clinical benefits of dopamine agonists in the management of epilepsy can be traced back over a century, whilst the introduction of neuroleptics into psychiatry practice 40 years ago witnessed the emergence of fits as a side effect of dopamine receptor blockade. Epidemiologists noticed a reciprocal relationship between the supposed dopaminergic overactivity syndrome of schizophrenia and epilepsy, which came to be regarded as a dopamine underactivity condition. Early pharmacological studies of epilepsy employed nonselective drugs, that often did not permit dopamine's antiepileptic action to be clearly dissociated from that of other monoamines. Likewise, the biochemical search for genetic abnormalities in brain dopamine function, as predeterminants of spontaneous epilepsy, proved largely inconclusive. The discovery of multiple dopamine receptor families (D1 and D2), mediating opposing influences on neuronal excitability, heralded a new era of dopamine-epilepsy research. The traditional anticonvulsant action of dopamine was attributed to D2 receptor stimulation in the forebrain, while the advent of selective D1 agonists with proconvulsant properties revealed for the first time that dopamine could also lower the seizure threshold from the midbrain. Whilst there is no immediate prospect of developing D2 agonists or D1 antagonists as clinically useful antiepileptics, there is a growing awareness that seizures might be precipitated as a consequence of treating other neurological disorders with D2 antagonists (schizophrenia) or D1 agonists (parkinsonism).

Subcortical modulation of synaptic plasticity in the hippocampus

A two-stage model of memory trace formation in the hippocampus considers exploratory behavior associated with theta activity and consumatory behaviors or immobility associated with the occurrence of synchronous pyramidal cell discharges. Threshold, frequency and extent of CA3 bursts differ between differently performing strains of mice and are subject to subcortical modulation. Mechanisms for such actions are described: interference with potassium and other cation channels through second messengers and a direct action of histamine on the NMDA receptor channel. Furthermore long-lasting potentiations of synaptic transmission and excitability following brief exposures of hippocampal slices to histamine, noradrenaline, and serotonin are mediated by cyclic AMP and presumably protein kinase A.

Anticonvulsant effects of memantine and MK-801 in guinea pig hippocampal slices

The anticonvulsant properties of memantine (Mem) were compared to those of MK-801. Extracellular field recordings were obtained from area CA1 of guinea pig hippocampal slices in a total submersion chamber at 32 degrees C in normal oxygenated artificial cerebrospinal fluid (ACSF). Evoked responses were elicited by 0.07 Hz stimulation of the Schaffer collateral and commissural fibers. Bath perfusion of slices with Mg(2+)-free ACSF and N-methyl-D-aspartate (NMDA)-containing ACSF induced epileptiform afterdischarges following evoked responses. Pretreatment of slices by bath application of 100 microM Mem for 18-20 min prevented epileptiform afterdischarges under both convulsant conditions. Perfusion with 100 microM Mem alone for up to 50 min had no discernible effect on evoked responses. MK-801 was as effective at < or = 10 microM and required application for over 15 min to suppress afterdischarges completely. Both Mem and MK-801 suppressed epileptiform activity when applied after such activity was induced by NMDA or MG(2+)-free ACSF. The EC50 of Mem was 16.6 microM and that of MK-801 was 0.19 microM for blocking NMDA-induced evoked response suppression. Thus, in the guinea pig hippocampal slice preparation, Mem appeared to have anticonvulsant properties qualitatively similar to those of MK-801, but was 10-100 fold less potent.

Locus coeruleus noradrenaline system and focal penicillin hippocampal epilepsy: neurophysiological study

Previous experimental investigations have shown that several neuronal systems modulate the spontaneous and paroxysmal electric activity of the hippocampus. The locus coeruleus-noradrenaline (LC-NA) system exerts an inhibitory influence on several brain areas including the ipsilateral hippocampus. Selective destruction of the LC increases the susceptibility to epileptiform phenomena in different models of experimental epilepsy. Our experiments were conducted on 34 rats in which a steady epileptiform interictal activity of the hippocampus was obtained by means of intrahippocampal administration of penicillin. Electrical stimulation of LC caused a significant decrease of penicillin spiking of hippocampus. Stimulation sessions given 10-15 min after i.p. propranolol administration (2 mg/kg) failed to induce any significant modification in the hippocampal spiking frequency. Intrahippocampal injection of L-noradrenaline mimicked the inhibitory effect of LC electrical stimulation on hippocampus. Finally, intrahippocampal administration of isoproterenol HCl, a beta-adrenoceptor agonist, caused a significant decrease of hippocampal penicillin spiking; this effect was antagonised by i.p. propranolol administration. The experimental data show a modulating influence of the LC-NA system on penicillin focal hippocampal epilepsy that probably involves beta-adrenoceptors.

Potassium and calcium channel dependence of bursting in cultured neuronal networks

Increases in extracellular potassium concentrations reliably increase burst rates in cultured fetal murine spinal cord networks. This effect could be mimicked by either blocking voltage-gated potassium conductances or facilitating excitatory synaptic interactions, but not by blocking specific calcium-dependent potassium conductances or tonic depolarization. Spontaneous bursting in cultured networks is apparently dependent on potassium currents and intracellular calcium levels, but not on the pharmacologically characterized calcium-dependent potassium conductances.

Epileptic focus induced in rat by intrahippocampal cholera toxin: neuronal properties in vitro

Injecting 0.5-1.0 microgram of cholera toxin into rat hippocampus induces a chronic epileptic focus which generates interictal discharges and brief epileptic seizures intermittently over the following seven to 10 days. Here we examined the electrophysiological properties of hippocampal slices prepared from these rats three to four days after injection, at the height of the epileptic syndrome. These slices generated epileptic discharges in response to electrical stimulation of afferent pathways. In many cases epileptic discharges occurred spontaneously in the CA3 subregion; these usually lasted < 200 ms, but they could last < 0.6 s. Intracellular recordings from pyramidal layer cells revealed depolarization shifts synchronous with the epileptic field potentials. These depolarization shifts had slow onsets compared with those induced by blocking inhibition with bicuculline (depolarizations started a mean of 57 ms before, and reached 5.2 mV by, the onset of the cholera toxin epileptic field potential, compared with 12 ms and 3.6 mV respectively for 70 microM bicuculline methiodide). Extracellular unit recordings showed that the slow predepolarization seen in the cholera toxin focus was associated with an acceleration of the firing of other pyramidal layer neurons. The epileptic activity in this model cannot be attributed to the loss of synaptic inhibition, because inhibitory postsynaptic potentials could be evoked when the synchronous bursts were blocked by increasing [Ca2+]o from 2 to 8 mM. Observations of monosynaptic inhibitory postsynaptic currents isolated by application of 20 microM 6-cyano-7-nitroquinoxaline-2,3-dione, 50 microM DL-2-amino-5-phosphonovaleric acid and 100-200 microM 3-amino-2-(4-chlorophenyl)-2-hydroxy-propylsulphonic acid showed a small effect of the toxin only on the time course of the inhibitory postsynaptic current. On the other hand, there were significant changes in the intrinsic properties of individual neurons. The membrane potentials of cells in the cholera toxin focus did not differ from those in slices from rats injected with vehicle solution, but their input resistances were significantly increased. Unlike the other cellular changes in this model, the increase in input resistance was not seen in slices exposed acutely to 1 micrograms/ml cholera toxin for 30 min, suggesting there may be morphological changes in the chronic focus. Action potential accommodation and the slow afterhyperpolarization were depressed in both acute and chronic epileptic tissue, indicating impairments of Ca(2+)- and/or voltage-dependent K+ currents, and we conclude that these provide the most likely basis for cholera toxin epileptogenesis.

Effects of trifluoperazine on synaptically evoked potentials and membrane properties of CA1 pyramidal neurons of rat hippocampus in situ and in vitro

The effects of trifluoperazine (TFP), a phenothiazine antipsychotic, on hippocampal activity were studied in the CA1 subfield, both in situ and in slices. In the extracellular studies in situ and in vitro, both somatic population spikes and dendritic excitatory postsynaptic potentials (EPSP) fields were depressed reversibly by TFP, applied by microiontophoresis or in the bath (50-100 microM). Similar effects were also seen during iontophoretic applications of sphingosine in situ. Like TFP (at micromolar concentrations) sphingosine is a dual Ca2+/calmodulin-dependent kinase and protein kinase C (PKC) inhibitor. In intracellular recordings from slices, 50-100 microM TFP induced a slow depolarization and a decrease in input resistance (RN), probably through a gamma-aminobutyric acid (GABA)-mediated increase in Cl- conductance (GCl). TFP also reduced the slow afterhyperpolarization (AHP) as well as electrically evoked inhibitory postsynaptic potentials (IPSPs), but EPSPs were augmented in both amplitude and duration. When CA1 neurons were voltage clamped, TFP elicited a corresponding inward current (consistent with depolarization), increased the leak conductance, and enhanced excitatory synaptic currents; whereas inhibitory synaptic currents and high-threshold Ca2+ currents were reduced. In conclusion, these effects of TFP--which cannot be readily explained by its potent antidopamine action--are in keeping with other evidence that both Ca2+/calmodulin-dependent kinase and PKC can modulate GCl-conductance and high-threshold Ca(2+)-conductance, as well as inhibitory and excitatory postsynaptic currents.

Adenosine-mediated synaptic inhibition: partial blockade by barium does not prevent anti-epileptiform activity

Adenosine-induced inhibition of evoked postsynaptic potentials (PSPs) and epileptiform burst firing in the CA1 subfield of rat hippocampal slices was studied with intracellular recordings in vitro. Adenosine (50 microM) caused a membrane hyperpolarization which was abolished during superfusion with 2 mM Ba2+. The adenosine-induced inhibition of the PSPs was still evident, although the magnitude of the effect was significantly reduced. Adenosine also reduced Ba(2+)-induced burst firing, but less effectively than it did bursts evoked by TEA (5 mM). The results suggest that adenosine inhibits synaptic transmission and epileptiform activity by at least 2 mechanisms: a postsynaptic barium-sensitive increase in gK and a presynaptic effect independent of this adenosine-evoked outward potassium conductance.

Vasoactive intestinal polypeptide modulates neuronal excitability in hippocampal slices of the rat

Vasoactive intestinal polypeptide added at submicromolar concentrations to the perfusion fluid of rat hippocampal slices and slice cultures enhanced the excitability of CA1 and CA3 pyramidal cells in several ways. Specifically, cells were depolarized and the Ca(2+)- and cyclic AMP-dependent potassium conductance was blocked as demonstrated by reduction of the long-lasting afterhyperpolarization and the accommodation of firing. This was also found in tetrodotoxin-containing medium. In low Ca(2+)-high Mg2+ medium (in synaptic isolation) the firing rate was increased. Synaptic transmission was potentiated: extracellularly registered excitatory postsynaptic potentials and population spikes in response to stratum radiatum stimulation and intracellularly recorded excitatory postsynaptic potential-inhibitory postsynaptic potential sequences were enhanced. These results are in keeping with the known stimulation of adenylate cyclase by vasoactive intestinal polypeptide.

Postsynaptic inhibition by adenosine in hippocampal CA3 neurons: Co(2+)-sensitive activation of an inwardly rectifying K+ conductance

The properties of the current underlying the membrane hyperpolarization evoked by adenosine (50-100 microM) were investigated in hippocampal CA3 neurons in vitro using current-clamp and single-electrode voltage-clamp techniques. In voltage-clamp measurements, the adenosine-induced current (IAdo) was outward at rest and reversed at membrane potentials close to the equilibrium potential of K+ (EK), indicating that IAdo was carried by K+ ions. Determination of IAdo at several membrane potentials revealed a nonlinear current/voltage (I/V) relationship of the current displaying inward rectification in the hyperpolarizing direction. Similarly, adenosine increased the membrane slope conductance only at membrane potentials negative to rest, whereas the slope of the neuronal I/V curve remained unchanged when determined at potentials positive to rest. Since the electrophysiological properties of IAdo were very similar to those described for K+ conductances activated by other neuroactive substances like serotonin, opioid peptides and gamma-aminobutyric acid B receptor (GABAB) agonists, we conclude that IAdo belongs to a family of ligand-operated, inwardly rectifying K+ currents which apparently share a common mechanism to reduce postsynaptic excitability. As an additional feature, the postsynaptic adenosine response was reduced by bath application of Co2+ or Ni2+. The adenosine-induced membrane hyperpolarization was not affected by low-Ca2+ or low-Mg2+ solutions, nor by buffering of intracellular Ca2+, but a gradual decline of IAdo was observed following superfusion with Co2+ or Ni2+. In contrast, Mn2+ caused only a weak attenuation of the adenosine response.

Intracellular potassium activity in mammalian proximal tubule: effect of perturbations in transepithelial sodium transport

Intracellular potassium activity (alpha Ki) was measured in control conditions in mid-cortical rabbit proximal convoluted tubule using two methods: (i) by determination of the K+ equilibrium potential (EK) using Ba(2+)-induced variations in the basolateral membrane potential (VBL) during transepithelial current injections and (ii) with double-barrel K-selective microelectrodes. Using the first method, the mean VBL was -48.5 +/- 3.2 mV (n = 16) and the mean EK was -78.4 +/- 4.1 mV corresponding to alpha Ki of 68.7 mM. With K-selective microelectrodes, VBL was -36.6 +/- 1.1 mV (n = 19), EK was -64.0 +/- 1.1 mV and alpha Ki averaged 40.6 +/- 1.7 mM. While these last EK and VBL values are significantly lower than the corresponding values obtained with the first method (P less than 0.001 and P less than 0.01, respectively), the electrochemical driving force for K transport across the basolateral membrane (microK = VBL-EK) is not significantly different for both techniques (30.1 +/- 3.3 mV for the first technique and 27.6 +/- 1.8 mV for ion-selective electrodes). This suggests an adequate functioning of the selective barrel but an underestimation of VBL by the reference barrel of the double-barrel microelectrode. Such double-barrel microelectrodes were used to measure temporal changes in alpha Ki and microK in different experimental conditions where Na reabsorption rate (JNa) was reduced. alpha Ki was shown to increase by 12.2 +/- 2.7 (n = 5) and 14.1 +/- 4.4 mM (n = 5), respectively, when JNa was reduced by omitting in the luminal perfusate: (i) 5.5 mM glucose and 6 mM alanine and (ii) glucose, alanine, other Na-cotransported solutes and 110 mM Na. In terms of the electrochemical driving force for K exit across the basolateral membrane, microK, a decrease of 5.4 +/- 2.0 mV (P less than 0.05, n = 5) was measured when glucose and alanine were omitted in the luminal perfusate while microK remained unchanged when JNa was more severely reduced (mean change = -1.7 +/- 2.1 mV, NS, n = 5). In the latter case, this means that the electrochemical driving force for K efflux across the basolateral membrane has not changed while both the active influx through the Na-K pump and the passive efflux in steady state are certainly reduced. If the main pathway for K transport is through the basolateral K conductance, this implies that this conductance must have decreased in the same proportion as that of the reduction in the Na-K pump activity.

Inhibition of low calcium induced epileptiform discharges in the hippocampus by dopamine D1 receptor agonist, SKF 38393

The influence of dopamine D1 receptor agonist, SKF 38393 has been studied in vitro in the model of low calcium spontaneous epileptiform discharges. Application of SKF 38393 (3 microM) to the perfusing medium evoked a decrease in neuronal firing rate of hippocampal CA1 neurons. The effect of SKF 38393 was blocked by pretreatment with SCH 23390. It is concluded that simulation of hippocampal D1 dopamine receptors by SKF 38393 inhibits epilepsy-like events induced by low calcium concentration in the perfusing fluid.

Inhibition of N6-[3H]cyclohexyladenosine binding by carbamazepine

The mechanism of action of carbamazepine (CBZ) (Tegretol), despite widespread use in the management of partial and tonic-clonic seizures in adults, is not completely understood. In animals, adenosine and adenosine analogues have anticonvulsant effects that may be due to interactions with central A1 adenosine receptors. CBZ (at therapeutically relevant concentrations) inhibits the binding of agonists and antagonists to brain A1 adenosine receptors, but whether as an agonist/antagonist is not clear. The adenosine agonist, N6-[3H]cyclohexyladenosine ([3H]CHA), binds to membranes from rat cortex and hippocampus at two nanomolar binding sites or states. To clarify the actions of carbamazepine at the A1 adenosine receptor, its inhibitory actions were compared with those of known adenosine agonists and xanthine antagonists using 0.1 nM[3H]CHA, in which almost all binding is to the higher affinity state, or 10 nM [3H]CHA, in which there is a substantial contribution of binding from both states. The ratios of the IC50 values (concentration that inhibits specific binding by 50%) at 10 nM [3H]CHA to the IC50 values at 0.1 nM [3H]CHA were 18-31 for the agonists and 4-10 for the xanthine antagonists. CBZ had a ratio of 3. The inhibitory effects of GTP on [3H]CHA binding were less in the presence of the adenosine agonist, 2-chloroadenosine than were inhibitory effects in the presence of the xanthine antagonist theophylline or CBZ in both cortex and hippocampus. These in vitro studies indicate that CBZ is an antagonist at A1 adenosine receptors in cerebral cortical and hippocampal membranes from rat brain. Agonist activity at A1 adenosine receptors would have been compatible with the sedative anticonvulsant effects of CBZ, but these data do not support a role of the anticonvulsant action of carbamazepine on A1 adenosine receptors in cerebral cortex or hippocampus.

Combined electrochemical and electrophysiological studies of monoamine overflow in rat hippocampal slices

In vivo electrochemical measurements of chronoamperometric recordings from Nafion-coated electrodes were used to investigate monoamine overflow from selected regions of the rat hippocampal slice. Concurrent electrophysiological measurements of evoked CA1 pyramidal cell population spike responses were used to characterize changes in the electrical activity in the slices that occur during potassium-induced neurotransmitter overflow. Superfusion with elevated K+ (10-50 mM, 5 min) elicited consistent concentration-dependent increases in the electrochemical responses recorded from the dentate gyrus. At the onset of K+ perfusion, there was an initial increase in the population spike response, followed by electrical silence, which usually lasted 5-10 min following the return to normal medium, and required 20-30 min for complete recovery of the response. The potassium-induced electrochemical signal always increased following the decline in the electrophysiological response. Although the electrochemical signal usually returned to baseline much before the electrophysiological response (usually within 5 min), both signals remained refractory for some time. Cocaine pretreatment (10-50 microM) caused a dose-dependent augmentation of the electrochemical responses. Local pressure ejection of K+ via a micropipette elicited dose-dependent increases in the electrochemical signals that were of relativity brief duration as compared to superfusion with K+. Such potassium-evoked responses were highly localized, and were attenuated in amplitude in animals that had been previously treated with the selective noradrenergic neurotoxin, DSP-4. In addition to K+, local applications of methyl-amphetamine, tyramine and veratridine also elicited electrochemical signals, and the time courses of these responses were specific to the releasing agent that was used. Preliminary data obtained using high-speed electrochemical recordings of both oxidation and reduction current suggested that tyramine ejections evoked primarily norepinephrine overflow, while K+ evoked the overflow of both norepinephrine and serotonin. The present experiments demonstrate that simultaneous electrophysiological and electrochemical experiments can be used in an isolated preparation of brain such as the hippocampal slice to characterize the electrophysiological events that occur during stimulated transmitter release.

The brain histamine system in vitro

The electrophysiological properties of identified tuberomammillary histamine neurones were investigated in explant and slice preparations. The effects of histamine were studied on target neurones, mainly in the hippocampal slice. The results describe an important modulatory role of this diffusely projecting system.

Cholinergic modulation of epileptiform activity in the developing rat neocortex

The effects of carbachol on picrotoxin-induced epileptiform activity and membrane properties of neurons in the developing rat neocortex were examined in an in vitro slice preparation. Intracellular recordings were obtained in layer II-III neurons of slices prepared from rats 9-21 days of age. Epileptiform activity in 9- to 14-day-olds consisted of a sharply rising, sustained (10-30 s) membrane depolarization with superimposed action potentials. Bath application of carbachol (5-50 microM) raised the threshold for evoking epileptiform activity but, when such responses were evoked, their underlying depolarizations were increased in amplitude. Orthodromic stimulation in slices from 15- to 21-day-old animals evoked a prolonged epileptiform burst response that triggered an episode of spreading depression (SD). Carbachol reduced epileptiform responses and suppressed the occurrence of SD. It did not significantly affect the resting membrane potential or the height of the action potential but decreased the rheobase current needed to evoke an action potential and increased the input resistance. All effects of carbachol were antagonized by atropine (1 microM). These results indicate that carbachol has both pre- and postsynaptic effects in the developing neocortex and can significantly modulate neuronal excitability in the immature nervous system.

Synaptic and non-synaptic mechanisms underlying low calcium bursts in the in vitro hippocampal slice

1. The epileptiform activity generated by lowering extracellular [Ca++] was studied in the CA1 subfield of rat hippocampal slices maintained "in vitro" at 32 degrees C. Extracellular and intracellular recordings were performed with NaCl and KCl filled microelectrodes. 2. Synaptic potentials evoked by stimulation of the stratum radiatum and alveus were blocked upon perfusion with artificial cerebrospinal fluid (ACSF) containing 0.2 mM Ca++, 4 mM Mg++. Blockade of synaptic potentials was accompanied by the appearance of synchronous field bursts which either occurred spontaneously or could be induced by stimulation of the alveus. 3. Both spontaneous and stimulus-induced low Ca++ bursts recorded extracellularly in stratum pyramidale consisted of a negative potential shift with superimposed population spikes. This extracellular event was closely associated with intracellularly recorded action potentials rising from a prolonged depolarization shift. Steady hyperpolarization of the cell membrane potential decreased the amplitude of the depolarizing shift suggesting that synaptic conductance were not involved in the genesis of the low Ca++ burst. 4. Spontaneous depolarizing inhibitory potentials recorded in normal ACSF with KCl filled microelectrodes were reduced in size in low Ca++ ACSF. However, small amplitude potentials could still be observed at a time when low CA++ bursts were generated by hippocampal CA1 pyramidal neurons. 5. Bicuculline methiodide, an antagonist of gamma-aminobutyric acid (GABA), was capable of modifying the frequency of occurrence and the shape of synchronous field bursts. The effects evoked by bicuculline methiodide were, however, not observed when 81-100% of NaCl was replaced with Na-Methylsulphate.(ABSTRACT TRUNCATED AT 250 WORDS)

Modulation of neuronal excitability by endogenous adenosine in the absence of synaptic transmission

Rat hippocampal slices were superfused with low calcium, high magnesium medium. Reductions in flow rate were associated with a marked depression of antidromically elicited afterpotentials with little change in the initial antidromic population spike recorded from CA1 pyramidal neurons. The depression of the afterpotential at the lower flow rates was largely reversed by the adenosine antagonist, theophylline (100 microM), by adenosine deaminase (10 micrograms/ml) and was mimicked by the application of the adenosine reuptake blocker, dipyridamole (100 microM). Since synaptic transmission was blocked, it is concluded that sufficient endogenous adenosine exists in the absence of synaptic function to alter neuronal excitability.

Endogenous adenosine inhibits hippocampal CA1 neurones: further evidence from extra- and intracellular recording

Extracellular and intracellular recordings from CA1 pyramidal neurones of rats in vitro were used to study the effects of endogenous and exogenously applied adenosine. The adenosine receptor antagonist, caffeine, enhanced the intracellular recorded e.p.s.p.-i.p.s.p. sequence evoked by stimulation of the stratum radiatum which is antagonized by exogenous adenosine. The late, potassium dependent i.p.s.p. was not antagonized. The adenosine uptake inhibitor, nitrobenzylthioinosine (NBTI), mimicked the effects of exogenously applied adenosine. The effects of NBTI and of exogenously applied adenosine were antagonized by caffeine in the same manner. Exposure to adenosine deaminase enhanced the evoked field e.p.s.p. During this enhancement caffeines effects were significantly reduced. In low calcium high magnesium medium which abolishes synaptic activity, adenosine deaminase increased, NBTI decreased cell firing. We conclude that endogenous adenosine, release by a calcium independent mechanism, can exert an inhibitory tone on CA1 neurones in vitro. This is consistent with a role for adenosine as a mediator of negative feedback between the metabolic state and electrophysiological activity of nervous tissue.

Anticonvulsant action of the nucleoside transport inhibitor, soluflazine, on synaptic and non-synaptic epileptogenesis in the guinea-pig hippocampus

The effects of the nucleoside transport inhibitor, soluflazine, were examined on synaptic and non-synaptic epileptogenesis, and on paired-pulse facilitation and inhibition in the CA1 region of the guinea-pig hippocampal slice. In the model of synaptic epileptogenesis, excitation was enhanced by omitting Mg2+ from the artificial cerebrospinal fluid (ACSF). This procedure induced a second epileptogenic population spike (PS) after orthodromic stimulation, which was inhibited by soluflazine (IC50 value 1.2 x 10(-6) M). In the non-synaptic model of epileptogenesis spontaneous depolarizing 'burst' discharges were induced in CA1 by lowering the concentration of Ca2+ and increasing the concentration of K+ and Mg2+. The IC50 value of soluflazine was 6.0 x 10(-7) M for antagonizing 'burst' frequency and 7.5 x 10(-6) M for 'burst' amplitude, indicating a preferential effect on 'burst' initiation. After paired orthodromic stimuli to stratum radiatum, the amount of synaptic facilitation of PS amplitude was significantly increased by soluflazine. This was mainly due to a decrease in the size of the PS induced by the conditioning pulse. The amount of PS inhibition after antidromic/orthodromic stimulation was not significantly altered by soluflazine. With the exception of the failure of soluflazine to attenuate inhibition, the results obtained with soluflazine resemble those reported for adenosine. This strengthens the hypothesis that soluflazine increases the extracellular concentration of adenosine. Further, the results indicate that centrally active nucleoside transport inhibitors may be a new class of antiepileptic drug.

Pitfalls in the use of brain slices

In vitro brain slices are the preparation of choice for the detailed examination of local circuit properties in mammalian brain. However it is the investigator's responsibility to verify that the circuits under investigation are indeed confined within the boundaries of the functional region of the slice used. The medium in which the slice is maintained is under the full control of the investigator. This places the burden on the investigator to ensure that: (1) the properties of the medium are fully under control; (2) the effects of the medium on the slice are known; (3) the conditions under which the slice is being maintained bear some reasonable relation to those it enjoys (or endures) in vivo. Generalizations to in vivo conditions must be made with caution. If at all possible, similar studies (perhaps less extensive, due to the greater technical difficulties) should be done in vivo to provide a basis for comparison. Investigators using drugs should be aware of, and respect, the basic pharmacological principles cited in the text. In particular, the substantial freedom the investigator has in defining the extracellular medium should not be abused.

Effects of adrenal steroids on Na transport in the lower intestine (coprodeum) of the hen

The influence of adrenal steroids on sodium transport in hen coprodeum was investigated by electrophysiological methods. Laying hens were maintained on low-NaCl diet (LS), or on high-NaCl diet (HS). HS hens were pretreated with aldosterone (128 micrograms/kg) or dexamethasone (1 mg/kg) before experiment. A group of LS hens received spironolactone (70 or 160 mg/kg, for three days). The effects of these dietary and hormonal manipulations on the amiloride-sensitive part of the short-circuit current were examined. This part is in excellent agreement with the net Na flux, and therefore a direct electrical measurement for Na transport. After depolarizing the basolateral membrane potential with a high K concentration, the apical Na permeability and the intracellular Na activity were investigated by current-voltage relations for the different experimental conditions. Plasma aldosterone concentrations (PA) were low in HS hens, dexamethasone-treated HS hens and spironolactone-treated LS hens (less than 70 pM). In contrast LS hens and aldosterone-treated HS hens had a PA concentration of 596 +/- 70 and 583 +/- 172 pM, respectively. LS diet (chronic stimulation) had the largest stimulatory effect on Na transport and apical Na permeability. Hormone-treated animals had three- to fourfold lower values. Spironolactone supply in LS hens decreased Na transport and apical Na permeability about 50%. The results provide evidence that both mineralo- and gluco-corticoids stimulate Na transport in this tissue by increasing the apical Na permeability. Quantitative differences between acute and chronic stimulation reveal a secondary slower adaptation in apical membrane properties.

Non-synaptic modulation of repetitive firing by adenosine is antagonized by 4-aminopyridine in a rat hippocampal slice

In rat hippocampal slices which were superfused with low calcium (0.2 mM) medium, stimulation of the alvear fibers elicited an extracellularly recorded antidromic population spike in CA1 pyramidal neurons which was followed by 2-4 afterpotentials. Adenosine (20-40 microM) and the A1-adenosine agonist 1-phenylisopropyladenosine (L-PIA) blocked these afterpotentials without affecting the first spike. Addition of up to 5 mM tetraethylammonium to the superfused medium did not interfere with this adenosine action. But the addition of only 50 microM 4-aminopyridine (4-AP) antagonized almost completely the adenosine- or L-PIA-induced depression of antidromically evoked repetitive firing. It is concluded that functioning of 4-AP-sensitive potassium channels is a prerequisite for this 'antiepileptic' adenosine action. Since a similar pharmacological characteristic has been described for the A-current, it is likely that adenosine acts by turning on this particular potassium current.

Testing of prototype antiepileptics in hippocampal slices

The effects of six prototype anticonvulsant drugs, phenytoin, carbamazepine, midazolam, phenobarbital, ethosuximide and sodium valproate, were evaluated in two different experimental models of epileptiform activity using the in vitro slice preparation from the rat hippocampus. The relative potencies of the agents were determined: a) in the complete absence of synaptic transmission by recording spontaneous burst firing from the CA 1 pyramidal cell layer in a low calcium high magnesium solution and b) during blocked synaptic inhibition by observing the activity of each drug upon orthodromically evoked population spikes in penicillin containing medium. The rank order of potencies was a) in low Ca2+: carbamazepine, phenytoin, midazolam, phenobarbital, valproate, ethosuximide; b) in penicillin containing medium: midazolam, phenobarbital, carbamazepine, phenytoin, valproate, ethosuximide. These observations illustrate that the use of multiple paradigms is warranted when examining the mechanisms of action of new anticonvulsants.

Adenosine actions on CA1 pyramidal neurones in rat hippocampal slices

Intracellular recordings with a bridge amplifier of CA1 pyramidal neurones in vitro were employed to study the mechanisms of action of exogenously applied adenosine in the hippocampal slice preparation of the rat. Adenosine enhanced the calcium-dependent, long-duration after-hyperpolarization (a.h.p.) at least in part by a reduction in the rate of decay of the a.h.p. Both the reduced rate of decay and that of the control can be described with a single exponential. Antagonism of the calcium-dependent potassium current (and as a result, the a.h.p.) by bath application of CdCl2 or intracellular injection of EGTA (ethyleneglycolbis-(beta-aminoethyl ether)N,N'-tetraacetic acid) did not reduce the adenosine-evoked hyperpolarization or decrease in input resistance. Similarly, TEA (tetraethylammonium), which antagonizes both the voltage- and calcium-sensitive, delayed, outward rectification, had no effect on the adenosine-evoked changes in resting membrane properties. Adenosine did not affect the early, transient, outward rectification. During exposure to 4-aminopyridine (4-AP) in concentrations sufficient to antagonize this early rectification, the changes in resting membrane properties evoked by adenosine were unaffected. We conclude that the enhancement of the a.h.p. and accommodation by adenosine may be mediated by a change in the regulation of intracellular calcium. However, the mechanism responsible for the hyperpolarization and decrease in input resistance evoked by adenosine is both calcium and voltage insensitive. Thus, it appears distinct from that mediating the enhancement of the a.h.p. and accommodation.

Is adenosine an endogenous anticonvulsant?

The anticonvulsant properties of adenosine were tested pharmacologically on amygdala-kindled seizure activity in rats. The adenosine analogue 2-chloroadenosine and the adenosine uptake blocker papaverine both increased the latency to behavioral clonus as well as reduced the duration and severity of the clonic motor convulsion. Both drugs, however, failed to alter the postkindling afterdischarge (AD) threshold. Theophylline, an adenosine antagonist, had the opposite effects, prolonging the AD and motor seizure durations and facilitating partially kindled seizures, but again not altering the prekindling or postkindling AD thresholds of amygdala-elicited seizures. In contrast, carbamazepine raised AD thresholds, suggesting that it does not produce its anticonvulsant effects through adenosine systems. Since endogenous adenosine can impede seizure spread and seizure continuation, but does not affect seizure initiation from the amygdala, perhaps endogenous adenosine has the special property of being brought into play as an anticonvulsant only by the seizure itself.

Adenosine enhances afterhyperpolarization and accommodation in hippocampal pyramidal cells

Adenosine added to the perfusion fluid of rat hippocampal slices at 10 mumol X l-1 enhanced long lasting afterhyperpolarizations after single action potentials, bursts of action potentials or calcium spikes. Accommodation of firing during a depolarizing pulse was potentiated. An increase in calcium dependent potassium conductance is likely to mediate these effects. Adenosine at 50 mumol X l-1 induced a hyperpolarization accompanied by a reduction in input resistance. The hyperpolarization could be reversed at -85 mV. In TTX and TTX-barium treated slices the amplitude of the slow spike was decreased. This may result from a shunting of inward current in the dendrites due to an adenosine induced increase in potassium conductance. It is suggested that adenosine reduces pre- and postsynaptic excitatory signals principally by enhancing one or more potassium conductances. This effect is a powerful means for modulation of neuronal excitability and synaptic efficacy and can explain the antiepileptic activity of adenosine.

Low-calcium field burst discharges of CA1 pyramidal neurones in rat hippocampal slices

Incubation of rat hippocampal slices in solutions containing low Ca2+ and increased Mg2+ rapidly blocked synaptic responses and increased spontaneous firing of all the principal neurones. More remarkably, a rhythmic and synchronous bursting discharge developed, which was restricted to the CA1 population of pyramidal neurones. These 'field bursts' or 'spreading excitation' were rapidly abolished by restoring the Ca2+ to 2 mM, by increasing the Mg2+ to 6 mM or by decreasing K+ from 6 to 3 mM. The CA1 pyramidal cells depolarized after the change to the low-Ca2+ solution by about 10-20 mV. Individual field bursts were associated with a further depolarization of 10-12 mV surmounted by a burst of action potentials at about 20/s. This transient depolarization shift, recorded extracellularly as a negative field, could be attributed to the increase of [K+]o during the bursts, reaching 9-10 mM as measured by ion-sensitive electrodes. The bursts were followed by a hyperpolarization, seen extracellularly as a small soma-positive field, which was attributed to an electrogenic pump and/or a Ca2+-activated K+ conductance. Stimulation of the tightly packed pyramidal cell axons in the alveus elicited a train of population spikes, instead of the single spike normally seen, and could trigger a full field burst. Recordings of the alvear tract volley suggested that the repeated spikes arose within the pyramidal cells. Multiple recordings from CA1 revealed that field bursts usually, but by no means always, started near the caudal (subicular) end of the area. They spread through the cell layer at 0.04-0.12 m/s. The most rapid propagation was seen when the bursts had an abrupt onset; slower propagation (1-10 mm/s) occurred when the bursts started gradually, which generally was the case near the sites of burst initiation and termination. Usually the action potentials within each burst were synchronized into population spikes which spread across CA1 at 0.04-0.15 m/s. The site of initiation and the extent of the spread of these population spikes varied during each burst, as did their amplitude. The degree of spike synchronization was enhanced by various treatments expected to increase neuronal excitability. Measurements of transmembrane potential during the burst confirmed the role in the generation of population spikes of ephaptic or field interactions between the pyramidal cells. It is proposed that the increased firing of all neurones is due to the block of tonic inhibition, depression of after-hyperpolarization and to increased membrane excitability.(ABSTRACT TRUNCATED AT 400 WORDS)