SEARCH

SEARCH BY CITATION

Keywords:

  • Absence epilepsy;
  • Differential mRNA display;
  • Hippocampus;
  • Cerebellum;
  • Gene expression;
  • Thalamocortical loop;
  • Computer modeling

Abstract

  1. Top of page
  2. Abstract
  3. COMPUTER MODELING
  4. DIFFERENTIAL GENE EXPRESSION
  5. REFERENCES

Summary:  Purpose: We present results obtained by computer modeling of the thalamic network and differential gene expression analysis in a rat strain with absence epilepsy, the genetic absence epilepsy rat from Strasbourg (GAERS).

Methods: (a) Computer modeling used equations from the Hodgking–Huxley model with a circuit of 13 reticular thalamic (nRt) and 39 thalamocortical (TC) neurons; (b) gene-expression analysis using differential mRNA display (DD), in situ hybridization, Northern blotting, and competitive reverse transcriptase–polymerase chain reaction (RT-PCR).

Results: (a) Computer modeling showed an increased network synchrony in the thalamic circuit as the value of conductance of low-voltage activated calcium channel (LVACC) is increased. (b) Using differential mRNA display, a 40% upregulation of the H-ferritin mRNA in the hippocampus was demonstrated. Looking for some candidate genes of the VACC family, no difference was found in the α1G mRNA expression between GAERS and control animals, whereas a decreased expression of the α1E subunit was observed in the cerebellum and the brainstem of the GAERS. This phenomenon was not observed in young animals when the epileptic phenotype is not expressed.

Conclusions: The use of computer modeling appeared to be an efficient way to evaluate the impacts of electrophysiologic findings in vivo from single cells on an entire circuit. No clear single gene defect was revealed so far in GAERS. More information could arise from linkage analysis. However, some brain structures like hippocampus or cerebellum classically not known to be involved in the production of absence spike-and-wave discharges could in fact participate in the development of this phenotype.

In experimental conditions, the genetic absence epilepsy rats from Strasbourg (GAERS) resemble humans with absence epilepsy (1). These animals exhibit absence-like seizures occurring spontaneously, as abrupt behavioral arrests, staring, and clonic twitching of the vibrissae, associated with high-amplitude spike-and-wave discharges (SWDs) at 7–9 Hz on the EEG. As in humans, (a) the illness is inherited, (b) the onset of seizures is age dependent, (c) seizure occurrence is maximal during phases of quiet wakefulness, (d) SWDs are suppressed by drugs effective against childhood absence epilepsy, and (e) SWDs were demonstrated to involve the lateral thalamic nuclei.

The hypothesis that the mechanisms of the thalamocortical synchronization could be implicated in the generation of spontaneous SWDs has been extensively investigated (2). At the molecular level, electrophysiologic studies have showed that a rhythmic sequence of γ-aminobutyric acid type B (GABAB) inhibitory postsynaptic potential (IPSP) and low-threshold calcium potential could be involved in the intrinsic pacemaker oscillations by which low-frequency oscillation of the membrane potential and associated burst firing can be produced by thalamocortical neurons (3,4). Many studies have therefore analyzed the properties of the GABAB receptor and T-type calcium current in the GAERS. Recently, however, some of the same researchers have cast some doubts on the exact mechanism of these oscillations during SWDs (5). Concerning GABAB, the results were contradictory, as some authors found no differences in density and affinity for this receptor in GAERS versus Wistar rats (6), whereas others find a twofold increase in affinity (7). Conversely, a selective increase in low-threshold calcium current conductance was recorded in the reticular thalamic nucleus of the GAERS (8), but this difference is detected considerably earlier (P11–P18) than the full development of SWDs (P30–P40).

In the present article we review, in light of more recent data from the current literature, results obtained on this model in our laboratory using (a) computer modeling to evaluate the impact of electrophysiologic findings from single units on the thalamic circuit; (b) differential gene expression analysis to see if the candidate genes or others are altered in this model. The details of these works have been previously published elesewhere (9–11).

COMPUTER MODELING

  1. Top of page
  2. Abstract
  3. COMPUTER MODELING
  4. DIFFERENTIAL GENE EXPRESSION
  5. REFERENCES

Introduction

The study by Tsakaridou et al. (8) found in comparing the voltage-dependent calcium channels of both normal and epileptic strains that the amplitude of the low-threshold calcium current (IT) of the GAERS rat was higher. Whereas the ITs current was found augmented in isolated neurons of the nRt, this low-threshold current was found not to be significantly different in isolated neurons of the thalamocortical (TC) neurons of both rats. No significant differences were found in the high threshold Ca2+ current, IL, between the nonepileptic and GAERS rat. This increase of the ITs, according to more detailed tests, could reflect either a higher number of T-type Ca2+ channels or an increase in single-channel conductance.

We have investigated the effect of an isolated increase of of ITs with the use of a computer model of thalamic network. This was done by examining the effects of increasing the maximum conductance of the low threshold calcium current (gTs max) in the nRt neurons of the model (9).

Methods

In brief, the simulated network consisted of 13 nRt and 39 TC neurons in a columnar arrangement. Each column consisted of one nRt neuron reciprocally connected to three TC neurons. Each nRt neuron also received input from the nRt and TC neurons in the column immediately adjacent to it, and the TC neurons received input from the nRt neurons of the adjacent column. The physiology of the neurons was constructed using the Hodgkin–Huxley formalism. Most of the parameters and equations necessary to describe the ionic currents of the model can be obtained from previous computer models of the thalamic network and have been described in detail. The intrinsic currents for the model neurons included the low-threshold calcium current, fast Na+, and fast K+ currents for both the TC and nRt neurons. The TC neuron had in addition an Ih (hyperpolarization activated) current, whereas the nRt neuron had a Ca2+-mediated K+ current. The synaptic input included GABAA inhibition for all the thalamic neurons, GABAB inhibition for the TC neurons, and AMPA synapses for the nRt neurons.

Results and discussion

By progressively increasing the maximal conductance of the low-threshold calcium channel in the nRt neurons from 1.7 to 2.7 mS/cm2, the initial stimulus being in each case a pulse in the nRt layer of the model, we found a progressive increase in the number of cycles of synchronized activity by the TC neurons (Fig. 1). The increase in the value of gTs max was found to increase the period of oscillations in the nRt neurons. In contrast, the low-threshold spike of the nRt neuron remain relatively unchanged by the increase of gTs max.

image

Figure 1. Number of cycles of synchronous activity in the network for each value of conductance of the T channel expressed in mS/cm 2 . The number of cycles of synchronous activity was computed as described in the text and references.

Download figure to PowerPoint

These results suggested that the enhanced synchrony in the network was primarily due to a phase shift in the firing of the nRt neurons with the respect to the TC neurons. Future questions regarding this subject include whether the change in ITs is due to changes in the subunits constituting the channel itself or whether it is due to changes in one of the eventual regulators of the channel such as kinases or phosphatases, for example.

DIFFERENTIAL GENE EXPRESSION

  1. Top of page
  2. Abstract
  3. COMPUTER MODELING
  4. DIFFERENTIAL GENE EXPRESSION
  5. REFERENCES

Introduction

Theoretically, the search for gene defects associated with such a rat strain can be done using linkage analysis or by searching for abnormally expressed transcripts in the brain. The first approach is currently investigated in another laboratory. Therefore, we first compared normal and GAERS strains using differential mRNA display and then explored some candidate genes to detect eventual modifications in gene expression in the brain of the GAERS.

The results obtained by electrophysiologic studies and computer models prompted us to first investigate the expression of the genes encoding α1 subunits of calcium channels with low thresholds of activation. We started exploring the first cloned α1 subunit, which at that time was thought to be a T-type channel and named α1E(10). Because now this subunit is known to support the R-type current, we also explored the first cloned α subunit encoding a true T-type current (i.e., α1G). We compared the data obtained in adulthood with those observed in young animals at an age where the epileptic phenotype is not yet expressed.

Methods

For more details, refer to the previously published articles (10,11).

Animals

Experiments were performed on male nonepileptic Wistar and GAERS rats. All adult animals (6–8 months old) were implanted with four contact electrodes, and 3 days after surgery, they underwent EEG recording in the freely moving state for 60–120 min to assess the presence or the absence of SWDs in the GAERS and Wistar rats, respectively. They were killed after a postoperation period of 5 days.

Young 25-day-old rats were from parents that were first EEG recorded. Because of the absence of SWDs in the young GAERS, some rats of the litter were kept until age 2 months, and then EEG recorded to assess the presence of SWDs.

Differential mRNA display

Total RNA was prepared from freshly isolated brain structures according to the acid guanidinium thiocyanate-phenol-chloroform single-step method (12). The detection of the various brain mRNAs was achieved using DD, mainly as described by Liang and Pardee (13).

In situ hybridization

In situ hybridization was done using [33P]dATP tailed oligonucleotides and standard hybridization conditions. Specificity controls are RNAse treatment and hybridization with a 100-fold excess of unlabelled oligonucleotide. The H-ferritin and α1G probes hybridize to specific regions of the corresponding mRNA showing ≤40–50% identity with the other sequences published in Genbank. The α1E probe is commercially available from Biognostik. For quantification, optical density measurements of the film were converted into Ci/g of tissue equivalent according to the standard calibration curve obtained using autoradiographic [14C] microscales.

Northern blot and dot blot

Nucleic acids were transferred onto positively charged nylon membranes. Hybridization was performed according to standard protocols. Results were analyzed using different amounts of material and after different exposure durations to ensure that all the signals compared on autoradiographic films were in the linearity range of detection. Differences in the amounts of RNA loaded were normalized according to GAPDH signals.

Competitive RT-PCR

Oligonucleotide primers for RT-PCR were designed according to the published rat α1E and α1G calcium channel subunits and GAPDH sequences. Total RNA (5 μg/reaction) was reversed transcribed into cDNA using random hexanucleotide primer and MMLV. For PCR MIMIC, a fixed amount of the resulting cDNA was amplified in the presence of decreasing amounts (twofold serial dilution) of an internal nonhomologous standard called MIMIC. These were prepared using the Clontech PCR MIMIC Construction Kit. For quantification, the resulting Polaroid negative was scanned on a HP Desk scan, and the signal was analyzed with Gel-Pro Analyser 3.0 software.

Results

DD allowed the identification of H-ferritin as an upregulated gene in GAERS

To detect possible abnormal gene expression associated with absence phenotype in GAERS, we decided to identify the differentially expressed genes in the brains of these animals. DD was performed in duplicate on total RNA isolated from brain hemisphere (i.e., without brainstem and cerebellum) of adult rats. Using all the primer sets (i.e., 80), 34 differentially expressed amplimers or cDNA fragments were isolated. As can be seen in Table 1, 24 of these have been analyzed, among which 10 exhibited more than one single sequence, for a total of 44 genes to be sequenced and checked by dot blotting. The majority of these sequences corresponds to unknown genes or EST. The known genes that show differences in expression in the GAERS brain include various proteins, none of them being directly involved in the control of membrane excitability (Table 1).

Table 1.  List of differentially expressed cerebral amplimers found by the differential mRNA display between normal and GAERS strains
AmplimerExpressionaSequencesbSize (bp)Clone identitycAC numberProbesdDot blote
  • In columns 2 and 3, characters are in both when the DD and dot-blot experiment showed the same variation of expression.

  • NA, not analyzed by dot-blot.

  • a

     A, augmented; D, diminished; GAERS versus normal.

  • b

     Indicates the number of different sequences per amplimer bands.

  • c

     (p) means identity on part of the sequence.

  • d

     Indicates if one or both strands of the amplimer have been tested by dot-blot.

  • e

     Dot-blot indicates an approximation in percent of the variation, GAERS versus normal.

 4A1426H-ferritinNM 012848BothNA
 5–6D1107NoneBoth79
 7–9A1500Capping protein (p)cBC 002053Both101
 8A3a381NoneBoth90
   b397NoneBoth92
   c257NoneBoth103
   d398None NA
10–11A1168RPL19X 82202Both89
12–13D1167NoneBoth75
14A1145NoneBoth109
15A3a158EST (p)BF 286810One sideNA
   b213H-ferritinNM 012848One sideNA
   c124Zinc-finger RIZU17837One sideNA
16A3a150EST (p)AW 26576One sideNA
   b198ESTAW 270765One sideNA
   c62NoneOne sideNA
17A1119NoneBoth93
18D4a194ESTBG 518751Both58
   c199ESTBB 473171Both54
   d195ESTBF 388190Both87
   e199NoneBoth86
19D1159NoneBoth96
20D1145ESTBF 398544Both76
21A3a135NoneNoneNA
   b55NoneNoneNA
   c179NoneNoneNA
22A3a88NoneOne sideNA
   b152NoneOne sideNA
   c141NoneBothNA
23A6a149Nectin-likeAF 195662One sideNA
   b187NoneBoth103
   c198Dematin (p)AF 079846Both106
   d198γ-actinX 52815Both89
   e198NoneBoth102
   f195Mss4L 10336Both98
24A4a156ESTAI 600125Both101
   b159NoneBothNA
   c155ESTAA 894007BothNA
   d159NoneBothNA
25–26D2a327CH1XM 010735Both32
   b292Proteasome p112 (p)NM 031978Both105
27A3a213NebuletteXM 016290Both105
   b216ERK-1M 61177Both111
   c266KinesineX 61435Both144

As an example, we summarize here results obtained with our first candidate. Using primer set T12VG/OPA18, one fragment appears (Table 1: DD4) differentially expressed in GAERS. This fragment shows 100% identity with the 3′ end of the rat H-ferritin mRNA. To confirm the putative increased expression of the H-ferritin mRNA, a Northern blot experiment was performed on new samples. The cDNA fragment hybridized to an appropriate size transcript of 0.9 kb in both strains. Quantification of the signal by densitometry indicated a 40% increase in GAERS as compared with the Wistar rat.

Total RNA was then isolated from cortex, subcortical structures, brainstem, and cerebellum of adult (6–8 months) and young (25 days) GAERS and Wistar rat brains and were analyzed using H-ferritin and GAPDH riboprobes. For a same amount of GAPDH between the two strains, only the subcortical structures from the adult GAERS brain showed an enhanced expression of H-ferritin mRNA. In the young GAERS, at an age at which no SWDs could be recorded, there is no upregulation in any structure.

Finally, analysis by in situ hybridization was performed. As expected, the H-ferritin mRNA expression was clearly observed in all adult brain structures, but in various degrees. Myelinated fascicular tracts such as corpus callosum, external capsule, anterior commissure, internal capsule, piriform cortex, lateral olfactory tract, fimbria, Ammon's horn, and the gyrus dentatus exhibited a high degree of expression. Conversely, the cellular layers of neuronal perikarya of archicortical origins such as limbic structures, as well as the fascicular layers and Purkinje cells of the cerebellum, also displayed a clear signal.

Quantification showed that the mean overexpression in the GAERS was detected in limbic hippocampal structures: the Ammon's horn, CA1 by 22%, CA2 21%, CA3 35%, and the gyrus dentatus by 35%. In agreement with the dot-blot experiment, no increase in mRNA level was recorded in the cortex or the cerebellum. Moreover, no statistically significant difference was observed in the thalamus of the GAERS versus Wistar rats.

The same experiment was performed on fully amygdala-kindled rats. However, when compared with control rats, no statistically significant difference was observed in the hippocampus of these animals.

Using competitive RT-PCR to investigate expression of some genes of the VACC family in the GAERS brain

Taking GAPDH as control, the competitive RT-PCR experiment indicates that the α1E mRNA level decreases by 50% in the GAERS cerebellum. A slight decrease also was seen in the brainstem of the epileptic strain, but no difference appears in the cortex and the subcortical structures of these rats (Fig. 2). To know if this decreasing expression could be associated with the absence phenotype, we performed the same experiments on RNA extracts from 25-day-old rat brains, at an age at which no SWDs could be recorded on the EEG. Because of the absence of differential expression in the cortex and in the subcortical structures of the adult rats, we have not analyzed these areas in the young rats. No significant altered level of α1E mRNA could be measured either in the cerebellum or in the brainstem. When identical competitive reactions were performed for α1G calcium channel subunit, the same amount of mRNA was found in adult GAERS and in Wistar rats. Because of this result, no experiment on young rats was carried out.

image

Figure 2. Relative expression of α 1E messenger RNA (mRNA) in the genetic absence epilepsy rat of Strasbourg (GAERS) versus Wistar rats at different ages as determined by competitive reverse transcriptase–polymerase chain reaction. Cx, cortex; Scx, subcortical structures (basal ganglia, thalamus); Cer, cerebellum; Bs, brainstem. For each structure, the GAPDH signal of GAERS was normalized to that of the Wistar rat. Vertical bars indicate ± SEM, and *, data significant at least at p[M1] < 0.05.

Download figure to PowerPoint

As expected from the MIMIC experiments, no altered expression of the α1E mRNA could be seen in the cortex or the subcortical structures such as the hippocampus and the thalamus using in situ hybridization. In the adult GAERS cerebellum, the mean decrease occurs in the white matter, but the granular cells layer shows a significant decrease of the α1E expression. The brainstem presents a decreased signal in the GAERS.

In the adult, the α1G subunit mRNA level does not exhibit any difference in any areas of the brain between the both strains. This is in agreement with the MIMIC experiment.

Discussion

Differential gene expression in GAERS brain

In the present study, we used DD to identify the eventual modifications of gene expression associated with an appropriate animal model of idiopathic absence epilepsy (i.e., the GAERS). Our results are not complete. We are currently running new sequences on genbank in order to obtain new information from our unknown genes that appear differentially expressed in the GAERS brain. The genes detected so far and presented in Table 1 showed only slight quantitative differences. It can be argued that we were totally unable to detect the absence of a single mRNA, as would be expected if the phenotype were due to one or multiple gene defects. This may be due to either a lack of sensitivity of the technique or to the fact that this strain result from several discrete gene abnormalities.

Recent work using microarray technology, comparing the differential gene expression in the nRT nucleus of GAERS, reached similar conclusions (14).

H-ferritin in GAERS brain

Of the various bands that differ in intensity between the epileptic and the normal Wistar rat brain, one was identified as the H-ferritin mRNA. This amplimer was shown to be overexpressed, and this result was confirmed by Northern blot. Moreover, the mRNA level is increased only when the epileptic phenotype is fully expressed. With in situ hybridization, it appears that the increased expression was localized only in the hippocampus, whereas no significant modifications were observed in thalamic nuclei, the structure known to be involved in the production of the absence phenotype.

What could be the significance of a selective increase in H-ferritin expression in hippocampal structures of a rat model of absence epilepsy? Ferritin is a multimeric iron-storage protein composed of 24 subunits forming a hollow shell capable of storing ≤4,500 atoms of metal. It is composed of different ratios of two different types of subunits termed H (heavy) and L (light) chains. The L-rich ferritin predominates in tissues specialized for long-term storage such as liver and spleen, whereas the H-rich form is found mainly in tissue with high iron utilization and low storage, such as the heart or the brain. From various studies, it appears that H-ferritin has two main functions. First, it provides the iron needed for the oxidative metabolism or neurotransmitter synthesis, because some enzymes need iron as cofactor. Second, it acts as a cytoprotectant against oxidant-mediated injury as it restricts the availability of intracellular free iron to participate in the Fenton reaction. More recent data show that several conditions can modify the transcriptional expression of the H-ferritin mRNA: factors controlling cell growth and differentiation like tumor necrosis factor, thyrotropin, insulin, and interleukin-1, as well as conditions like muscle denervation or oligodendrocytic adhesion. An increased H-ferritin gene expression also was obseved during growth arrest in rabbit vascular smooth muscle cells, and this was accounted for by an increase in both the transcription of this gene as well as by posttrancriptional events. Interestingly, in this model, cyclic adenosine monophosphate (cAMP) promotes the transcriptional event. In summary, the activation of the H-ferritin gene can be achieved in any situation in which cells need to grow and differentiate using growth factors, second messengers, and probably nuclear transcriptional factors.

However, our results could not be related to an increase in metabolic activity, as in the adult GAERS, the LCMRglc is increased by 20–50% in almost all the brain structures (15). Moreover, in the young GAERS, when no SWDs could be recorded, we saw no increase in H-ferritin mRNA expression, whereas the LCMRglc is increased in some specific brain structures including the hippocampus (16).

If in the GAERS hippocampus, no SWDs could be recorded, discharges were sometimes present and supposed to be associated with other types of epilepsy (1). In two models of absence epilepsy, the Stargazer and Tottering mice, mossy fiber sprouting was observed (17,18). Because cell death and axon growth occur during mossy fiber sprouting, this phenomenon could be related to an increase in ferritin mRNA expression. For this reason, we studied the H-ferritin mRNA expression in fully amygdala-kindled rats producing limbic epilepsy and extensive mossy fiber sprouting. With in situ hybridization, no difference in H-ferritin mRNA expression was recorded in the fully kindled rats. Therefore, it seems unlikely that our observation in the GAERS is related to abnormal discharges in the hippocampus and/or to mossy fiber sprouting.

It has been largely confirmed that the SWDs of absence epilepsy involve thalamic nuclei (see introduction). In the case of H-ferritin, no significant modifications were observed in any thalamic nucleus. Our results increase the list of observations implicating eventual modifications in limbic structures associated with the true absence epileptic phenotype. In the GAERS, a decreased expression of the subunit of the GABAA receptor in CA1 was reported (19). In the WAG/Rij rat, an upregulation of the mRNA-encoding prodynorphin was observed only in the hippocampus (20). Because in these models, no SWDs are recorded in the hippocampus, these modifications in gene expression may represent part of a protective mechanism and a response of the seizure-prone hippocampus preventing the spread of excessive TC activity. However, the relation with absence epilepsy remains to be demonstrated.

LVACC of the T type as candidate gene in GAERS?

The first principal finding of our investigations conducted in a rat model of absence epilepsy is that no clear modifications of the level of the mRNA encoding α1E and α1G are observed in the forebrain structures thought to play a key role in producing SWDs. More precisely, we did not observe a high degree of expression of α1E and no expression of α1G in nRt. T-type current in the nRt is probably linked to the presence of other recently identified α subunits: α1H and α1I (see ref. 10 for an exhaustive discussion). These have been now investigated by others, showing a small increase of the α1H mRNA level in the nRt of the GAERS (21).

Second, data obtained with both quantitative RT-PCR and in situ hybridization demonstrate a decrease of α1E transcript in the cerebellum and more precisely in the cortical and subcortical layers in the GAERS. This phenomenon is not present in young animals, at an age at which the epileptic phenotype is not yet expressed, suggesting its contribution in producing SWDs. To our knowledge, this would constitute the first report of a putative involvement of α1E subunit in an epileptic phenotype. This is also the first describing a decreased expression of a calcium-channel subunit in the cerebellum of an absence model showing no ataxia. The data presented here also differ from those obtained in the lethargic mice, in which an increased α1G mRNA level was shown in the cortex, the thalamus, and the cerebellum, whereas the α1E mRNA level remains unchanged (22). Thus, different modifications of gene expression could be associated with absence epilepsies.

Although no SWDs are recorded in the cerebellum of GAERS, our data reopen the question of the involvement of the cerebellum in controlling the activity of the TC loop responsible for the production of SWDs. Such a mechanism could have specific neuroanatomic circuits because a major input from the cerebellum projects directly into the ventrobasal complex of the thalamus (10). Most of the thalamofugal efferents originate from the dentate and interpositus nuclei of the deep cerebellar nuclei and project directly onto TC cells via the superior cerebellar peduncle. In the generalized penicillin epilepsy in the cat, surface cerebellar stimulation at high frequencies led to a decreased neuronal activity in the neocortex, whereas single-pulse shocks resulted in a short latency activation (23). More interestingly in the WAG/Rij rat, another model of generalized absence epilepsy, high-voltage spindles recorded from the sensimotor neocortex were correlated with single- or multiple-unit activity in the cerebellar cortex and deep cerebellar nuclei. These findings suggest that a rhythmic output from the cerebellum may contribute to the maintenance of generalized petit mal seizures (24). Ascending projections from the brainstem also are involved in the control of wakefulness and attention, and as a consequence, in occurrence of absence seizures. Altered activity in the brainstem would therefore potentially facilitate occurrence of corticothalamic SWDs (25).

Our results also can be analyzed in light of what is known at the molecular level in other models of absence epilepsy. In the lethargic mice, a mutation in the calcium channel β4 subunit results in a decrease of the N-type current in both the forebrain and the cerebellum, but surprisingly, at the protein level, the expression of the α1B calcium-channel subunit is selectively decreased in the cerebellum (26). This came probably from the facts that β subunits (a) modulate the electrophysiologic properties but also the membrane incorporation or stability of the α1 subunit (27), and (b) in some brain structures, other β subunits can replace β4. In that view, it will be interesting to analyze the expression of the calcium-channel β subunits in the GAERS brain.

Acknowledgment: This study was supported by grants from the FNRS (CC 1.5.081.98 and FRSM 3.4539.99), the University of Liège (Credits speciaux), and the CEE (Contract CHRXCT93-0247). We thank Dr. Matagne from UCB-Belgium for giving us amygdala-kindled rats, and Dr. J. Connor, for the H-chain ferritin antibodies.

REFERENCES

  1. Top of page
  2. Abstract
  3. COMPUTER MODELING
  4. DIFFERENTIAL GENE EXPRESSION
  5. REFERENCES
  • 1
    Marescaux C, Vergnes M, Depaulis A. Genetic absence epilepsy in rats from Strasbourg. J Neural Transm 1992;suppl 35:3769.
  • 2
    Avoli M, Gloor P. Role of the thalamus in generalized penicillin epilepsy: observations on decorticated cats. Exp Neurol 1982;77:386402.
  • 3
    Gloor P, Fariello G. Generalized epilepsy: some of its cellular mechanisms differ from those of focal epilepsy. Trends Neurosci 1988;11:638.
  • 4
    Crunelli V, Leresche N. A role for GABAB receptors in excitation and inhibition of thalamocortical cells. Trends Neurosci 1991;14:1621.
  • 5
    Charpier S, Leresche N, Deniau JM, et al. On the putative contribution of GABA(B) receptors to the electrical events occurring during spontaneous spike and wave discharges. Neuropharmacology 1999;38:1699706.DOI: 10.1016/s0028-3908(99)00139-2
  • 6
    Knight AR, Bowery NG. GABA receptors in rats with spontaneous generalized nonconvulsive epilepsy. J Neural Transm 1992;suppl 35:18996.
  • 7
    Mathivet P, Bernasconi R, Reymann N, et al. Different binding characteristics of cortical GABAB receptors in GAERS compared to nonepileptic control rats using agonist and antagonist radioligands. Pharmacol Rev Commun 1996;8:2037.
  • 8
    Tsakiridou E, Bertollini L, De Curtis M, et al. Selective increase in T-type calcium conductance of reticular thalamic neurons in a rat model of absence epilepsy. J Neurosci 1995;15:31107.
  • 9
    Thomas E, Grisar T. Increased synchrony with increase of low-threshold calcium conductance in a model thalamic network: a phase-shift mechanism. Neural Comput 2000;12:155371.
  • 10
    De Borman B, Lakaye B, Minet A, et al. Expression of mRNA encoding α1E and α1G subunit in the brain of a rat model of absence epilepsy. Neuroreport 1999;10:56974.
  • 11
    Lakaye B, De Borman B, Minet A, et al. Increased expression of mRNA endoding ferrtin heavy chain in brain structures of a rat model of absence epilepsy. Exp Neurol 2000;162:11220.
  • 12
    Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:1569.DOI: 10.1006/abio.1987.9999
  • 13
    Liang P, Pardee A. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 1992;257:96771.
  • 14
    Holter JL, Crunelli V, Leresche N, et al. Gene expression profile in the GAERS model of absence epilepsy. Soc Neurosci Abstr 2000;26:1178.
  • 15
    Nehlig A, Vergnes M, Marescaux C, et al. Cerebral energy metabolism in rats with genetic absence epilepsy is not correlated with the pharmacological increase or suppression of spike-wave discharge. Brain Res 1993;618:18.
  • 16
    Nehlig A, Vergnes M, Boyet S, et al. Metabolic activity is increased in discrete brain regions before the occurrence of spike-and-wave discharges in weanling rats with genetic absence epilepsy. Dev Brain Res 1998;108:6975.
  • 17
    Qiao X, Noebels JL. Developmental analysis of hippocampal mossy fiber outgrowth in a mutant mouse with inherited spike-wave seizures. J Neurosci 1993;13:462235.
  • 18
    Stanfield BB. Excessive intra and supragranular mossy fibers in the dentate gyrus of tottering (tg/tg) mice. Brain Res 1989;480:2949.
  • 19
    Snead OC III, Depaulis A, Banerjee PK, et al. The GABAA receptor complex in experimental absence seizures in rat: an autoradiographic study. Neurosci Lett 1992;140:912.
  • 20
    Lason W, Przewlocka B, Van Luijtelaar G, et al. Proenkephalin and prodynorphin mRNA level in brain of rats with absence epilepsy. Neuropeptides 1994;27:3437.
  • 21
    Talley EM, Solorzano G, Depaulis A, et al. Low-voltage-activated calcium channel subunit expression in a genetic model of absence epilepsy in the rat. Mol Brain Res 2000;75:15965.
  • 22
    Barun S, Lin FH, Hosford DA. Expression of voltage-dependent calcium channel (VDCC) subunits α1A, β1, β2 and β3 in lethargic (LH/LH) mouse model of absence seizures. Soc Neurosc Abstr 1998;24:1824.
  • 23
    Wray DV, Hablitz JJ. Neuronal firing patterns during generalized penicillin epilepsy in the awake cat. Brain Res Bull 1977;2:31721.
  • 24
    Kandel A, Buzsaki G. Cerebellar neuronal activity correlates with spike and wave EEG patterns in rat. Epilepsy Res 1993;16:19.
  • 25
    Danober L, Deransart C, Depaulis A. Pathophysiological mechanisms of genetic absence epilepsy in the rat. Prog Neurobiol 1998;55:2757.
  • 26
    Mc Enery MW, Copeland TD, Vance CL. Altered expression and assembly of N-type calcium channel alpha 1B and beta subunits in epileptic lethargic (lh/lh) mouse. J Biol Chem 1998;273:214358.
  • 27
    Walker D, De Waard M. Subunit interaction sites in voltage-dependent Ca++ channels : role in channel function. Trends Neurosci 1998;21:14854.