Kindling reduces sensitivity of CA3 hippocampal pyramidal cells to competitive NMDA receptor antagonists

alpha-Amino-3-hydroxy-5-methylisoxazole-4-propionate receptor autoradiography in mouse brain after single and repeated withdrawal from diazepam

Withdrawal from chronic treatment with benzodiazepines is associated with increased neuronal excitability leading to anxiety, aversive effects and increased seizure sensitivity. After repeated withdrawal experiences, seizure sensitivity increases while withdrawal-induced anxiety and aversion decrease. We used autoradiographical methods employing [(3)H]Ro48 8587, a selective ligand for glutamatergic alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors, to study withdrawal-induced changes in AMPA receptor binding in areas of the mouse brain postulated to be involved in these responses. Mice were given 21 days treatment with diazepam (15 mg/kg, s.c. in sesame oil) followed by withdrawal (single withdrawal) or three blocks of 7 days treatment interspersed with 3-day periods to allow washout of drug (repeated withdrawal). In keeping with heightened excitability in withdrawal from chronic diazepam treatment, the single withdrawal group showed, 72 h after their final dose of diazepam, increased [(3)H]Ro48 8587 binding in several brain areas associated with emotional responses or seizure activity, including hippocampal subfields, amygdalar and thalamic nuclei and motor cortex. In contrast, the repeated withdrawal group showed no changes in [(3)H]Ro48 8587 binding in any brain area studied. These observations are consistent with up-regulation of AMPA receptor-mediated transmission being important in withdrawal-induced anxiety and aversion but not in increased seizure sensitivity associated with repeated withdrawal. As changes in AMPA receptor subunit expression alter the functionality of the receptor, future studies will address this possibility.

Development of long-lasting potentiation effects in the dentate gyrus during pentylenetetrazol kindling

In the present study kindling was induced in rats by repeated intraperitoneal injection of pentylenetetrazol (PTZ) once every 48 h. The resulting seizure stages were registered after each PTZ application. The development of PTZ-induced kindling and the time course of possible potentiation effects in the dentate gyrus were examined. The efficacy of perforant pathway transmission to the granule cells was tested in every second kindling session by measuring the monosynaptic evoked field potentials recorded in the dentate gyrus following single test stimuli of the perforant pathway at different times after PTZ injection in freely moving animals. The data suggest that establishment of a PTZ kindling is associated with the development of long-lasting potentiation of the field potentials. After completion of kindling it was demonstrated that kindled rats also show a diminished learning performance. The relationship between the development of potentiation phenomena in hippocampal substructures and learning impairment is discussed.

Long-term reduction of benzodiazepine receptor density in the rat cerebellum by acute seizures and kindling and its recovery 6 months later by a pentylenetetrazole challenge

Seizures induced by an acute pentylenetetrazole (50 mg/kg) injection were accompanied by a long-term (at 1-48 h, but not on day 7) decrease in the density (B(max)) of [3H]-diazepam binding to benzodiazepine receptors in rat cerebellar cortex with no change in affinity (K(d)). Kindling for 24 days by daily administrations of pentylenetetrazole (20 mg/kg) led to the same decrease in benzodiazepine receptor density (at 1-48 h, but not on day 7) as that observed after a single dose of pentylenetetrazole (50 mg/kg). This suggests a common mechanism for both acute and kindling-induced seizures, dependent on the long-term receptor changes. The increased susceptibility to seizures persisted for 6 months after the termination of kindling, with BDZ receptor density in cerebellar cortex reduced almost by half. In age-matched controls, an acute dose of PTZ (30 mg/kg) induced seizures and decrease in both B(max) and K(d) of [3H]-diazepam binding. In kindled rats, at 6 months post-kindling, the same dose of PTZ (30 mg/kg) restored the benzodiazepine receptor density to the level found 6 months before, at the time of termination of kindling. Also, the severity of seizures was enhanced in the kindled rats. The results are discussed in terms of a balance of inhibitory and excitatory processes, in which the reduced BDZ receptor density at 6 months post-kindling may represent a compensatory reaction to outbalance some alterations in excitatory systems that have been reported to be induced by kindling.

Upregulation of NMDA receptors in hippocampus and cortex in the pentylenetetrazol-induced "kindling" model of epilepsy

"Kindling" is a phenomenon of epileptogenesis, which has been widely used as an experimental model of temporal lobe epilepsy. At the present work we investigated the contribution of NMDA receptors in the Pentylenetetrazol-induced "kindling" model in the mouse brain, by using quantitative autoradiography and the radioactive ligands [3H]MK801 and [3H]L-glutamate (NMDA-sensitive component). One week after establishment of kindling, a small but significant increase in [3H]MK801 as well as NMDA-sensitive [3H]glutamate binding was seen, being restricted to the molecular layer (ML) of the dentate gyrus (DG) and the CA3 region of the hippocampus. These binding augmentations persisted one month after establishment of kindling. A significant increase of NMDA receptor binding was also observed in the cortex-somatosensory and temporal one week after acquisition of the kindled state. The upregulation of NMDA receptors seen in DG and CA3 region of the hippocampus could be associated with the kindling process of this model especially with its maintenance phase, since it persists at long term, is area-specific and consistent with electrophysiological data. The increase of NMDA receptors seen in the cortex of the kindled animals could underlie the hyperexcitability detected by electrophysiological studies in this area.

Chronic epileptogenic cellular alterations in the limbic system after status epilepticus

Status epilepticus (SE) is associated with both acute and permanent pathological sequellae. One common long term consequence of SE is the subsequent development of a chronic epileptic condition, with seizures frequently originating from and involving the limbic system. Following SE, many studies have demonstrated selective loss of neurons within the hilar region of the dentate gyrus, CA1 and CA3 pyramidal neurons. Selective loss of distinct subpopulations of interneurons throughout the hippocampus is also frequently evident, although interneurons as a whole are selectively spared relative to principal cells. Accompanying this loss of neurons are circuit rearrangements, the most widely studied being the sprouting of dentate granule cell (DGC) axons back onto the inner molecular layer of the dentate gyrus, termed mossy fiber sprouting. Less studied are the receptor properties of the surviving neurons within the epileptic hippocampus following SE. DGCs in epileptic animals exhibit marked alterations in the functional and pharmacological properties of gamma-aminobutyric acid (GABA) receptors. DGCs have a significantly elevated density of GABA(A) receptors in chronically epileptic animals. In addition, the pharmacological properties of GABA(A) receptors in post-SE epileptic animals are quite different compared to controls. In particular, GABA(A) receptors in DGCs from epileptic animals show an enhanced sensitivity to blockade by zinc, and a markedly altered sensitivity to modulation by benzodiazepines. These pharmacological differences may be due to a decreased expression of alpha1 subunits of the GABA(A) receptor relative to other alpha subunits in DGCs of post-SE epileptic animals. These GABA(A) receptor alterations precede the onset of spontaneous seizures in post-SE DGCs, and so are temporally positioned to contribute to the process of epileptogenesis in the limbic system. The presence of zinc sensitive GABA receptors combined with the presence of zinc-containing "sprouted" mossy fiber terminals innervating the proximal dendrites of DGCs in the post-SE epileptic hippocampus prompted the development of the hypothesis that repetitive activation of the DG in the epileptic brain could result in the release of zine. This zinc in turn may diffuse to and block "epileptic" zinc-sensitive GABA(A) receptors in DGCs, leading to a catastrophic failure of inhibition and concomitant enhanced seizure propensity in the post-SE epileptic limbic system.

Measurement of NMDA receptor protein subunits in discrete hippocampal regions of kindled animals

Kindling refers to a phenomenon in which repeated application of initially subconvulsive electrical stimulations produces limbic and clonic motor seizures of progressively increasing severity. Once established, the increased excitability is lifelong. A diversity of studies demonstrate that kindling results in long lasting (28 days) alterations of the functional and pharmacologic properties of NMDA receptors, indicating that kindling may cause changes intrinsic to the NMDA receptor itself. Our previous studies disclosed no differences in NMDA receptor subunit gene or splice isoform mRNA expression between control and kindled animals 28 days after the last kindled seizure. Here, we extend those earlier studies by measuring levels of subunit protein for NMDAR1, NR2A, and NR2B in the hippocampus of control and kindled animals, 28 days after the last kindled seizure. We report that kindling does not effect long-lasting changes in the levels of NMDA receptor subunit protein. Together these findings support the idea that alterations in NMDA receptor protein expression do not contribute to the novel properties of NMDA receptors induced by kindling.

Brain thyrotropin-releasing hormone content varies through amygdaloid kindling development according to afterdischarge frequency and propagation

PURPOSE

Thyrotropin-releasing hormone (TRH), present in extra hypothalamic brain areas, has been proposed to have neuromodulatory functions and to be susceptible to change by electrical stimulation paradigms. We measured TRH concentrations of several brain areas during kindling development before its establishment and determined whether the changes detected in TRH levels were related to the behavioral stages of kindling, the number of stimulations required to reach these stages and, with the electrophysiological parameters characteristic of this paradigm (amygdaloid afterdischarge (AD) frequency, duration, and propagation).

METHODS

Male Wistar rats were implanted stereotaxically with indwelling bipolar electrodes in the basolateral nucleus of the amygdala and with two stainless-steel electrodes epidurally in frontal cortex. Amygdaloid kindling was induced by daily electrical stimulation; AD frequency and duration were recorded and analyzed throughout the development of kindling. TRH was extracted from several regions and quantified by radioimmunoassay (RIA).

RESULTS

Modifications in TRH concentrations were detected, depending on the region assayed, from stage II of kindling. A positive correlation was noted between the levels of TRH and the frequency and propagation of AD, but not with the number of stimulations. The rate of change in TRH concentration in relation to AD frequency or duration was highest in frontal cortex followed by hippocampus and amygdala.

CONCLUSIONS

A graded response was noted in the increase in TRH concentration dependent on the increase of AD frequency and propagation. The rate of response correlated with the region's epileptogenic susceptibility.

Pentylenetetrazol kindling changes the ability to induce potentiation phenomena in the hippocampal CA1 region

The present study describes changes of response enhancement of hippocampal field potentials in slices of kindled rats using different methods to induce long-lasting potentiation. Eight-week-old male Wistar rats were subjected to pentylenetetrazol (PTZ) kindling induced by intraperitoneal injection of 45 mg/kg once every 48 h until the occurrence of seizure stages 4-5. Eight to 12 days after the last kindling session, transverse hippocampus slices were prepared and maintained in an artificial medium. Evoked-field potentials were recorded in the CA1 region upon stimulation of the Schaffer collaterals. Potentiation was induced: 1. By moderate tetanic stimulation of the Schaffer collaterals, 2. by changing the perfusion medium to 0-magnesium for 30 min, and 3. by changing the medium to 4 mM Ca2+ for 7 min. In slices from kindled rats, long-term potentiation (LTP) after tetanic stimulation and increase of the evoked potential by 0-magnesium were significantly enhanced in comparison to slices from sham-kindled rats. However, Ca(2+)-induced LTP could not be induced in slices from kindled rats. The results support the assumption that PTZ kindling also induces lasting changes in the responsiveness of hippocampal structures, expressed as an enhanced ability to induce potentiation. An alteration of N-methyl-D-aspartate (NMDA) receptor-coupled processes can be assumed. The inability to induce Ca(2+)-induced LTP points to more complex effects of PTZ, perhaps also on nonNMDA coupled ionic channels.

Strain differences in pentylenetetrazol-kindling development and subsequent potentiation effects

Rats from two different strains, i.e. Wistar rats and Lister hooded rats, were investigated for their ability to acquire the kindling syndrome. After having received 13 kindling stimulations (injection of pentylenetetrazol), the animals were tested for subsequent alterations in induction and maintenance of hippocampal long-term potentiation (LTP) and, moreover in glutamate binding. It was found that rats from both strains did not differ in the response to the initial injection of pentylenetetrazol (PTZ) and the amplitude of the population spike. This suggests that some aspects of basic central excitability are equivalent. Wistar rats acquired the kindling syndrome rapidly whereas seizure outcome was poor in Lister rats. As regards hippocampal LTP, the population spike was only dramatically increased in Wistar rats after kindling completion. Glutamate binding was not altered in animals from the Lister strain. The results suggest that changes in glutamate binding and the increase in the population spike are characteristic consequences of kindling.

Altered expression of group I metabotropic glutamate receptors in the hippocampus of amygdala-kindled rats

Kindling is a well documented model of acquired focal epilepsy and synaptic plasticity in the nervous system. Previous biochemical studies have indicated an increase in mGluR-mediated phosphoinositide hydrolysis in the amygdala or hippocampus of fully kindled animals. In this study we have used in situ hybridisation techniques to examine the mRNA expression of group I metabotropic glutamate receptors (mGluR1 and mGluR5 both linked to phosphoinositide hydrolysis) in the hippocampus of amygdala-kindled animals sacrificed 24 h, 7 days or 28 days following the last electrically evoked stage 5 seizure, and in implanted non-stimulated control rats. Results indicate an initial up-regulation in mGluR1 mRNA (expressed as percentage of control) bilaterally in the DG (35-40%) and CA3 (16-48%), and unilaterally in CA4 (12%) in the 24 h post-kindled group. In kindled animals studied 7 days after the last seizure, these changes were either reduced or had returned to control levels. By 28 days mGluR1 mRNA levels had returned to control levels, with only a persistent increase in expression unilaterally in the DG (14%). In contrast, an initial down-regulation in mGluR5 mRNA was observed bilaterally in CA4 (-45 and -25%) and CA1 (-46 and -45%), and unilaterally in DG and CA3 (-27 and -42% respectively) 24 h after the last kindled seizure. In the 7 and 28 day kindled groups significant alterations in expression of mGluR5 mRNA were still apparent. These data show that the mRNAs for mGluR1 and mGluR5 are differentially regulated by kindling, indicating that the expression of each of these receptors is under independent regulatory control. These perturbations in mRNA expression may contribute to kindling epileptogenesis but are unlikely to account for the maintenance of the kindled state.

Regulation of alternative splicing of NMDAR1 in the kindling model

Kindling refers to a phenomenon in which repeated application of initially subconvulsive electrical stimulations produces limbic and clonic motor seizures of progressively increasing severity. Once established, the increased excitability is lifelong. Several lines of investigation suggest that the N-methyl-D-aspartate (NMDA) subtype of glutamate receptor participates in the expression of the increased neuronal excitability of the kindled brain. Many studies demonstrate that kindling results in altered NMDA receptor functional and pharmacological properties, indicating that kindling may cause changes intrinsic to the NMDA receptor itself. It is possible that altered expression of NMDA receptor subunit genes and splice isoforms of genes leads to subunit combinations resulting in the novel NMDA receptor properties identified in the hippocampus of kindled animals. To begin to address this possibility, we previously examined the hippocampal expression of known NMDA receptor genes and found no differences in expression between control and kindled animals either 24 h or 28 days after the last kindled seizure. Here, we extend that earlier study by examining the expression of NMDAR1 splice isoforms in the hippocampus of control and kindled animals. We report that kindling induces the transient reduction of specific splice isoforms of NMDAR1 containing the first of the carboxy-terminal splice cassettes (exon 21). We discuss the potential significance of this regulation in terms of its relevance to previous findings in the kindling model and possible effects on NMDA receptor function.

Potentiation effects in the dentate gyrus of pentylenetetrazol-kindled rats

The study examines changes in the function of perforant pathway dentate granule cell synapses after pentylenetetrazol (PTZ) kindling. Field potentials evoked in the dentate area by test stimuli to the perforant pathway were recorded in freely moving rats at different times after injection of PTZ. In fully kindled animals, but not in sham-kindled controls, subconvulsive test doses of PTZ induced long-lasting potentiation of the population spike. Also, potentiation was not induced in naive controls injected with equieffective doses of the convulsant. The slope function of the field EPSP was depressed 90-120 min after PTZ administration, in both kindled and control animals, indicating that this was an effect of acute-injected PTZ. Later on, only in kindled animals that showed seizure stages 4 or 5 did it increase in parallel with the population spike potentiation. Finally, when compared to controls the kindled animals showed a greater pop spike potentiation induced by moderate tetanization of the perforant pathway. The model offers the possibility of differentiating between acute effects of the convulsant drug and kindling-related changes in neuronal plasticity.

Alternative splicing at the C-terminal but not at the N-terminal domain of the NMDA receptor NR1 is altered in the kindled hippocampus

Several lines of evidence suggest that N-methyl-D-aspartate (NMDA) receptors significantly contribute to the development of kindling. In addition, a lasting enhancement of the NMDA receptor function has been suggested to play a significant role in the chronic hyperexcitability occurring in the hippocampus after kindling epileptogenesis. We have investigated whether hippocampal kindling induces changes in the NMDA receptor at the molecular level by assessing the expression of mRNAs of the different spliced variants at the N-terminal (exon 5) and C-terminal (exon 21) position of the NMDA receptor 1 (NR1) gene by means of the reverse transcription-polymerase chain reaction. Alternative splicing at exon 5 confers different sensitivity of the NMDA receptor to polyamines while exon 21 encodes a 37-amino acid insert containing the major phosphorylation sites for protein kinase C. One week after the acquisition of stage 5 of kindling in rats (generalized tonic-clonic seizures), the relative abundance of the two alternatively spliced forms at the C-terminal domain, respectively containing (+) or lacking (-) exon 21, was reversed compared to controls (implanted with electrodes but not stimulated) in the dorsal hippocampus ipsilateral and contralateral to the electrical stimulation. The exon 21+/exon 21- mRNA ratio for controls was 1.3 +/- 0.04 (mean +/- SE); for ipsilaterally kindled rats it was 0.64 +/- 0.05 (P < 0.05), and for contralaterally kindled rats it was 0.48 +/- 0.07 (P < 0.01). Similar bilateral effects were observed in the ventral hippocampus (temporal pole). No changes were found 4 weeks after stage 5 seizures and 1 week after the induction of a single afterdischarge. No significant alterations were induced by kindling in the relative abundance of the spliced variants containing or lacking exon 5. Our findings show selective changes in alternative splicing of the NR1 gene after repeated application of an epileptogenic stimulus. This may generate receptors with different functional properties, which may contribute to the increased sensitivity for the induction of generalized seizures during kindling.

Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: involvement of glucocorticoid secretion and excitatory amino acid receptors

Repeated restraint stress of rats for 21 days causes atrophy of apical dendrites of hippocampal CA3c pyramidal neurons. This effect is mimicked by daily corticosterone treatment for 21 days and is prevented y the anti-epileptic drug, phenytoin, known to interfere with excitatory amino acid release and action. The present study was designed to investigate the involvement of endogenous corticosterone secretion and excitatory amino acid receptors in the stress-induced hippocampal dendritic atrophy. Treatment of chronically stressed rats with the steroid synthesis blocker cyanoketone prevented stress-induced dendritic atrophy. Cyanoketone-treated animals showed an impaired corticosterone secretion in response to the stressor, while basal levels were maintained. Besides the involvement of endogenous corticosterone secretion, N-methyl-D-aspartate receptors also play a role, since the competitive receptor antagonist, CGP 43487, blocked stress-induced dendritic atrophy. In contrast, NBQX, a competitive inhibitor of the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptors, was ineffective at a dose that blocks ischemic damage. These results indicate that the reversible atrophy induced by 21 days of daily restraint stress requires corticosterone secretion and that excitatory mechanisms involving N-methyl-D-aspartate receptors play a major role in driving the atrophy.

N-methyl-D-aspartate and kainate receptor gene expression in hippocampal pyramidal and granular neurons in the kindling model of epileptogenesis

To investigate the changes underlying kindling epileptogenesis in the rat hippocampus, the levels of the messenger RNAs encoding for the subunits of the N-methyl-D-aspartate-receptor (1, 2A-D) and the kainate-receptor (1, -2, GluR-5, -6, -7) were determined in hippocampal principal neurons using in situ hybridization techniques and semi-quantitative analysis of the autoradiograms. Schaffer collateral-commissural pathway kindled rats were investigated at three different stages of kindling acquisition, always 24 h after the last stimulation. Furthermore, fully kindled rats were studied at long-term (28 days) after termination of kindling stimulations. NR1 messenger RNA levels were slightly decreased in CA1 area of fully kindled animals. In the fascia dentata region, a minor increase of NR2A and NR2B transcripts was found at all stages of kindling acquisition. KA-2 messenger RNA was enhanced in all hippocampal subfields during kindling development. However, none of these changes persisted at long-term after the last seizure and only the low-abundant GluR-7 expression was slightly depressed in the fascia dentata. From our observations we conclude that it is unlikely that alterations in N-methyl-Daspartate or kainate receptor gene expression play an important role in kindling acquisition or maintenance.

Hippocampal loss of N-methyl-D-aspartate receptor subunit 1 mRNA in chronic temporal lobe epilepsy

The hippocampal distribution of mRNA for the N-methyl-D-aspartate (NMDA) receptor subunit 1 (NR1) was examined by non-radioactive in situ hybridization in 21 archival formalin-fixed and paraffin-embedded surgical specimens from patients with pharmacoresistant chronic epilepsy and in normal control specimens obtained at autopsy. Using the digoxigenin-labeling procedure, ribonucleotide probes were found to be significantly more sensitive than synthetic oligonucleotide probes. In normal autopsy specimens and in surgical specimens without Ammon's horn sclerosis there was intense NR1 expression in a great majority of the dentate gyrus granular cells. Many neurons in the hippocampal pyramidal cell layer also revealed a strong signal intensity. The strata oriens and moleculare of Ammon's horn and the molecular layer of the dentate gyrus contained only few labeled neurons. In the subiculum and entorhinal cortex most neurons throughout various layers were positive. In hippocampal specimens of patients with chronic epilepsy there was a loss of NR1-positive cells that was closely related to the overall neuronal loss in the respective specimen and to Ammon's horn sclerosis. These data suggest that the loss of NR1 expression is a secondary phenomenon rather than an event that is relevant for the pathogenesis of epileptic seizures.

Mapping of cerebral blood flow changes during audiogenic seizures in Wistar rats: effect of kindling

The quantitative autoradiographic [14C]iodoantipyrine technique was applied to the measurement of rates of local cerebral blood flow (LCBF) during audiogenic seizures in Wistar AS rats belonging to a genetic strain selected at the Centre de Neurochimie (Strasbourg, France) for their sensitivity to sound. Seizures were elicited in native rats never exposed to sound (single audiogenic seizures) or in rats previously exposed to 10-40 seizure-inducing sound stimulations until generalization of the seizure to forebrain areas (referred to as "kindled animals"). During single audiogenic seizures, rates of LCBF increased over control values in all areas but the genu of the corpus callosum. The highest increases in LCBF (180-388%) were recorded in the inferior and superior colliculus, reticular formation, monoaminergic cell groupings, especially the substantia nigra, posterior vegetative nuclei, and many thalamic and hypothalamic regions. The lowest increases were seen in forebrain limbic regions and cortical areas. In kindled animals, LCBF rates increased over control levels in 67 areas of the 75 studied. LCBF increases were generally of a lower amplitude in kindled than in naive rats. Differences between the two groups of seizing rats were located mostly in brain-stem regions, mainly the inferior colliculus, reticular formation, substantia nigra, and posterior vegetative nuclei. Conversely, rates of LCBF were similar in forebrain areas of naive and kindled animals. In conclusion, the present data show that there is a good correlation between the structures known to be involved in the expression of audiogenic seizures (inferior colliculus, reticular formation, substantia nigra mainly) and the large increase in LCBF during single audiogenic seizures, while rates of LCBF increase to a lesser extent in forebrain areas not involved in this type of seizures. The circulatory adaptation to kindled seizures is rather a decreased response in brain-stem regions and no change in the forebrain, although the kindling process induces a generalization of the seizure from brain-stem to anterior regions.