Epilepsy Research
Volume 98, Issue 1 , Pages 25-34, January 2012

Phenotypic differences between fast and slow methionine sulfoximine-inbred mice: Seizures, anxiety, and glutamine synthetase

  • Arnaud Boissonnet

      Affiliations

    • Laboratoire de Neurobiologie, Rue de Chartres, Université d’Orléans, 45067 Orléans CEDEX 2, France
    • Glycodiag, 520 rue de Champteloup, 45520 Chevilly, France
    • ,
  • Tobias Hévor

      Affiliations

    • Laboratoire de Neurobiologie, Rue de Chartres, Université d’Orléans, 45067 Orléans CEDEX 2, France
    • ,
  • Jean-François Cloix

      Affiliations

    • Laboratoire de Neurobiologie, Rue de Chartres, Université d’Orléans, 45067 Orléans CEDEX 2, France
    • Corresponding Author InformationCorresponding author. Tel.: +33 0 238494979; fax: +33 0 238417244.

Received 9 June 2011; received in revised form 4 August 2011; accepted 18 August 2011. published online 03 November 2011.

Article Outline

Summary 

Seizures induced by the convulsant methionine sulfoximine (MSO) resemble human “grand mal” epilepsy, and brain glutamine synthetase is inhibited. We recently selected two inbred lines of mice: sensitive to MSO (MSO-Fast) and resistant (MSO-Slow). In the present study, the selection pressure was increased and consanguinity established. To gain insight into the mechanisms of epileptogenesis, we studied the behaviour of MSO-Fast and MSO-Slow mice based on their responses to various convulsants and anticonvulsants, and also the kinetics of glutamine synthetase. The results show that increasing the number of generations of sib-crossings resulted in an increase in the differences between MSO-Fast and MSO-Slow mice. The dose–response curve of MSO-dependent seizures demonstrated that the MSO-Slow mice were highly insensitive to MSO-dependent seizures compared with MSO-Fast inbred mice that were highly sensitivity. The MSO-Slow were resistant to convulsions induced by various convulsants having different mechanisms of action, whereas those in the MSO-Fast line were more sensitive to kainic acid-induced seizures. These data, in addition to the effects of anticonvulsant, strongly suggest that glutamatergic pathways are most likely involved in MSO-dependent seizures, rather than GABAergic ones. This hypothesis is corroborated by the glutamine synthetase activity, which is more elevated in the MSO-Slow line. Behaviour tests showed that MSO-Slow were less anxious than MSO-Fast. Collectively, these results showed that glutamatergic pathways could be involved in the epileptogenic action of MSO, which may be related to the glutamate/glutamine cycle in the brain.

Keywords: Epilepsy, Convulsants, Behaviour, Open-field, Glutamate/glutamine cycle, Glutamatergic pathways

 

Back to Article Outline

Introduction 

According to the World Health Organisation 1% of the worldwide population develop an epileptic syndrome. About 70% of the patients are treated using drugs, whereas the remainder are pharmacoresistant and undergo surgery whenever possible (Andrade and Minassian, 2007, Meldrum, 2007, Scharfman, 2007, Schmidt, 2009). The lack of an etiological treatment of epilepsy is due to poor knowledge on the basic mechanisms of seizure genesis, i.e., a sudden and temporary synchronization of neural activities (McNamara et al., 2006, Stafstrom, 2006, Scharfman, 2007, Reid et al., 2009). One difficulty in the study of epileptogenesis is how to find models which precisely mimic the human disease. Different animal models have been developed, corresponding essentially to spontaneous or genetic epilepsy, or to physical- or chemical-induced seizures. Among these latter models, one was described many years ago where its clinical pattern resembled the most striking form of human epilepsy (Wolfe and Elliot, 1962). This model depended upon methionine sulfoximine (MSO), which is known as a powerful inhibitor of glutamine synthetase (GS) (Griffith and Meister, 1978). Nevertheless many chemoconvulsants induce tonic and clonic seizures with similarity to human pathologies. MSO generates epileptiform seizures in a large variety of animals, and was shown in earlier studies to increase brain glycogen content (Folbergrova et al., 1969). Recently, an increase in brain glycogen was associated with human temporal lobe epilepsy (Dalsgaard et al., 2007). On account of the high resemblance of MSO models to human epilepsy, and because of the recent metabolic data on the condition, we became interested in looking for basic mechanisms that generate epileptiform activities in an MSO model, with the aim to better understand epileptogenesis. Our approach was based on selecting mice that respond differently to the administration of MSO, and on comparing the metabolism and the behaviour of selected inbred lines in order to find out abnormalities which can trigger seizures. In preliminary investigations, based on the first few stages of inbreeding, we observed that MSO separated types of mice that develop seizures minutes after MSO injection (MSO-Fast) from another type that developed seizures latter after administration of the same drug (MSO-Slow) (Cloix et al., 2010b). Moreover, metabolic differences were observed between these two types of mice, such as brain glycogen level and neurotransmitter contents (Cloix et al., 2010a). In the present study we aimed to increase the selection pressure to emphasise the difference between the two types of mice, after varying the doses and then making further crosses within lines in order to enhance the visible effects which could help to explain the seizures genesis.

Glutamate and GABA are two essential neurotransmitters acting as activator and inhibitor of brain activities, respectively. The neurotransmitters were described as being involved in the control of seizures in both patients and in animal models of epilepsies (Mohler, 2006, Kanner, 2008, Moult, 2009, Schmidt, 2009, Mares et al., 2010). Moreover, brain glutamate content and glutamate receptors, particularly metabotropic glutamate receptors (mGluR), are linked to the control of various physiological and physiopathological pathways, such as those involved in behaviour, mood and psychiatric disorders (Moult, 2009, Wieronska and Pilc, 2009, Krystal et al., 2010, Mares et al., 2010). As a relationship between epilepsy and psychiatric disorders has been hypothesized (Kanner, 2008, LaFrance et al., 2008, Hixson and Kirsch, 2009, Kanner, 2009), the present investigation is principally devoted to study the responses to various convulsants and anticonvulsants, and the behaviour of MSO-Fast and MSO-Slow mice in relation to GS activity.

Back to Article Outline

Experimental procedures 

Animals 

MSO-Slow and MSO-Fast mice were selected according to a previously described procedure based upon the crossing of eight parental strains: ABP/Le, A/J, BALB/c, C3H/HeJ, C57BL/6J, CBA/H, DBA/2J, and SWXL-4 (Cloix et al., 2010b). Thereafter, 3 additional MSO-challenges were performed in order to increase the selection pressure: MSO-Slow mice were challenged with 100mg/kg MSO, while MSO-Fast ones were challenged using 50mg/kg MSO. As previously described (Cloix et al., 2010b) two groups of 16 breeding pairs were established after each MSO challenges in such a way that no brother sister mating was allowed. The MSO-Fast group comprised mating pairs of 16 females and 16 males having the shortest seizure latencies; while MSO-Slow group comprised mating pairs of 16 females and 16 males having the longest or no seizure latencies. This was followed by 7–9 inbreeding generations of brother–sister crosses to generate mice that were 76.9±8.5 days old at the time of the experiments. All protocols were approved by the local ethical committee with the agreement number CREEA CL2007-023, and were in accordance with the European Community Council Directive of 24 November 1986 (86/609/ECC).

Chemicals and seizure latency 

Methionine sulfoximine (MSO), pentylenetetrazole (PTZ, an GABA antagonist), kainic acid (KA, an glutamatergic agonist), pilocarpine (PC, an cholinergic agonist), valproic acid (VPA, considered as an GABA agonist as it increased GABA concentration through various mode of action) and MK-801 (an NMDA antagonist) were obtained from Sigma–Aldrich (Saint Quentin Fallavier, France). The chemicals were dissolved in 0.9% NaCl and the pH was adjusted to 7.4 for KA and VPA only. They were administered intraperitoneally, and the latencies to seizures were determined as the duration between injection and seizure onset, as previously described (Cloix et al., 2010b). A maximal latency score was given to mice that did not seize during the observation period: PTZ, 180s, KA, 120min, PC, 120min, MSO, 600min. Doses of various convulsants and anticonvulsants were as follows: MSO, various doses as indicated hereafter; PTZ, 75mg/kg; KA, 25mg/kg, PC, 300mg/kg, VPA, 250mg/kg; MK-801, 1mg/kg. The ED50 for MSO were calculated using GraphPad Prism software with a non-linear fitting of data, normalised response and variable slope, and simply corresponded to the concentration required to provoke a response halfway between the baseline and maximum responses. In this analysis, only the generalised convulsions were considered. Twenty mice were used (10 females and 10 males) for the study of latency to generalised seizures, and 40 mice (20 females, 20 males) for the observation of seizure stages. The various stages (I–V) of MSO-dependent seizures were also determined accordingly to Racine's scale (Racine, 1972). In brief, the various stages and the main characteristics of each stage are as follows. Stage I: ataxia, akinesia, facial myoclony, sniffing. Stage II: head clony, head shaking. Stage III: myoclony of anterior members. Stage IV: myoclony of posterior members, tail whipping. Stage V: generalised seizures.

Glutamine synthetase activity 

Mice were sacrificed by decapitation, and heads were immediately frozen in liquid nitrogen and conserved at −80°C. Dorsal cortices were dissected out from the brain at −20°C and pulverised with a liquid nitrogen frozen homogeniser, and then kept at −80°C until used. Five animals, 3 females and 2 males were used to determine kinetic properties of GS from both inbred strains. Their cortices were pulverised and the powder was processed, as previously described (Blin et al., 2002) except homogenisation of 50mg of cortices were performed in 5vol of in 100mM imidazol buffer pH 7.2 using Ultra-Turax. Triplicate determination of glutamine synthetase was realized using 50μl of supernatant in a total volume of 300μl of incubation buffer (50mM l-glutamate, 20mM ATP, 10mM β-mercaptoethanol, 40mM MgCl2, 100mM hydroxylamine, 100mM imidazol, pH 7.2). The reaction was developed for 30min at 37°C and stopped by addition of 800μl of 0.37M FeCl3 solution in 0.67M HCl, 0.2M TCA. The mixture was centrifuged at 4°C for 5min at 10,000×g and the absorbance of the resulting supernatant was read at 535nm and compared to a standard curve of authentic l-glutamylhydroxamate. The activity of GS was therefore expressed as mM of l-glutamylhydroxamate/min/mg of protein, the latter was determined using the Lowry procedure (Lowry et al., 1951). Maximal enzyme activity and substrate affinity were determined by increasing the amount of substrate, i.e., glutamate from 0.2 to 50mM; while the inhibition constant was calculated for MSO, from 1.5μM to 25mM, at a glutamate concentration of 50mM. Values were determined using GraphPad Prism software. Vmax and Km were determined using plot of Lineweaver–Burk. Ki for MSO was calculated using log [MSO] as function of activity with a variable slope.

Behaviour 

We tested 40 animals, 20 females and 20 males. The 8-arms radial-maze test for spatial learning capacity was performed using a previously described method (Blin et al., 2002). Open-field analysis to determine anxiety was performed according to a slight modification of previously described protocols (Belzung and Griebel, 2001, Crusio, 2001, Einat et al., 2003). In brief, it corresponded to a dark plastic homemade open-field with 9 square-cases of 15cm each, separated by a 2cm blue line. A dark plastic wall of 30cm height surrounded the 9 cases. Mice were positioned in the centre square and observed for 5min. Between each mouse the open-field was wiped with a paper cloth soaked with 70% ethanol. The following values were recorded by the same experimenter: time before the first movement, corresponding to the time needed to pass to the next square; number of visited squares, which corresponds to line crossed by at least the animal's shoulders; the number of total rearing; and the number of defecation pellets. These values are considered as an index of animal anxiety (Belzung and Griebel, 2001, Crusio, 2001, Einat et al., 2003).

Electroencephalogram (EEG) recording was performed as follows. Under isoflurane anesthesia (1.5%), mice were implanted with three monopolar surface electrodes placed in the cranial bone. Two electrodes were set bilaterally over the parietal cortex and a ground electrode was placed over the frontal cortex. Electrodes were made of a tungsten wire (diameter 250μm) soldered to a male connector (Wire-pro, Farnell, Villefranche sur Saône, France). They were inserted in the skull so that only the tip (0.5mm) protruded onto tissue. The electrodes were glued to the skull with cyanoacrylate and dental cement. EEG was recorded on freely moving mice, after a recovery period of 7–10 days. EEG was monitored for a several hours lasting period of 8h using a Powerlab 26T model ML856 ADInstruments with a LabChart v7 sofware (Oxford, UK). Mice behaviour was noted down during the same period of time.

Statistical analyses 

Data are expressed as mean±standard error of the mean (SEM) for the numbers of animals indicated. The results of the latency toward MSO, various convulsants and anticonvulsants, and GS activity were analysed using the Student t-test. The results for the induction of various stages of convulsion and of radial-maze were evaluated by repeated measures of Analysis of Variance (ANOVA), with STRAIN (MSO-Slow vs MSO-Fast) and SEX as between-subjects factors, and DAYS as within-subject factor. Strain and sex comparisons for individual days and convulsion were made by means of two-way ANOVAs. As some animals submitted to radial-maze did not always take the food reward the first time they entered an arm on days 1 and 2, only the data from days 3–5 were used for the statistical evaluation. The results of open-field were evaluated by repeated measures of Analysis of Variance (ANOVA), with STRAIN (MSO-Slow vs MSO-Fast) and SEX as between-subject factors. Data was considered significant when the p value was less than 0.05%.

Back to Article Outline

Results 

Results from Fig. 1 demonstrate that the MSO-Slow mice are much less sensitive to MSO-dependant seizures than the MSO-Fast ones. Indeed, the calculated ED50 of MSO for the percentage of convulsion induced in the MSO-Fast line was 41.8±1.6mg/kg, while the computation was not possible for MSO-Slow line (Fig. 1A). According to the percentage of mice that convulsed as a function of MSO dose, the dose–response curve for the MSO-Slow line was almost flat, with a maximal value of 30% of convulsion for a 350mg/kg MSO dose. Conversely, the curve obtained with MSO-Fast inbred mice was steeper, and reached 100% of convulsion for a 75mg/kg MSO dose. Latency toward various MSO doses (Fig. 1B) in the MSO-Slow line was significantly higher than that of MSO-Fast ones. The latency to 75mg/kg MSO dose in MSO-Slow was above 600min, and was 277.6±64.5min for MSO-Fast line. In addition (Fig. 1C), the time needed to reach the different five stages of the Racine's scale was also significantly higher in MSO-Slow than in MSO-Fast ones, the data being obtained with 350mg/kg MSO for MSO-Slow and 75mg/kg MSO for MSO-Fast line.

  • View full-size image.
  • Figure 1. 

    Dose–response curve (A), latency as a function of MSO doses (B) and times necessary for inbred mice to reach various stages of MSO-dependent convulsion (C). Mice were administered with various doses of MSO and the development of seizures was observed. Ten males and 10 females were used, MSO-Slow (triangle, broken line) and MSO-Fast mice (square, full line). To observe stages I–V of seizures MSO was administered to MSO-Slow (n=40, 20 males, 20 females) and MSO-Fast (n=40, 20 males, 20 females) mice at a dose of 350 and 75mg/kg, respectively (C). Values are expressed as mean±SEM, and significance between two inbred mice was assumed using Student t-test as *p<0.05, **p<0.01, ***p<0.001. Statistical significance between the two inbred mice for stages I–V of MSO-dependent seizure is estimated using repeated ANOVA test: ***p<0.001.

EEG (Fig. 2) was recorded for 8h after MSO administration to MSO-Fast (75mg/kg, Fig. 2A) and MSO-Slow (350mg/kg, Fig. 2B). In these conditions, basal EEG recorded during pre-convulsive period was not different between the two strains of mice, and identical to that of before MSO injection. However, EEG recorded during the convulsive period showed a very high and long epileptic-activity in MSO-Fast mice as compared to MSO-Slow ones, that displayed a very low activity but higher than basal EEG. Moreover, this EEG was recorded while the MSO-Slow mice showed several typical behaviours of mice under MSO with no generalised convulsions. This result strongly suggested that MSO-Slow mice were resistant to MSO dependent-seizures while MSO-Fast were sensitive.

  • View full-size image.
  • Figure 2. 

    Typical electroencephalograms (EEG) of inbreed mice to 75mg/kg of methionine sulfoximine (MSO) and 350mg/kg for MSO-Fast and MSO-Slow mice, respectively. EEG from the convulsive period (upper panels A and B) was shown and 2h-EEG after MSO were displayed (lower panels A and B) as the control period. The EEG recording is one out of two performed on two different MSO-Fast mice (A) and two different MSO-Slow mice (B). Surface electrodes were implanted over the cortex through the skull under isoflurane anesthesia, and EEG was recorded 6–8 days later.

The two lines were challenged with various convulsants displaying different mechanisms of action (Fig. 3). No significant differences in terms of latencies and percentage of convulsion induced by PC and PTZ were observed, but the administration of KA induced a lower percentage of generalised convulsions (Fig. 3B) in MSO-Slow (40%) than in MSO-Fast (85%) mice. The latency toward KA-dependent seizures was significantly higher in MSO-Slow mice as compared to MSO-Fast ones (Fig. 3A), data that is consistent with previous studies (Cloix et al., 2010b). As a whole, these data indicate that only KA might discriminate between MSO-Slow and MSO-Fast lines. Among the convulsant tested, KA strongly suggested that glutamatergic pathway(s) might be involved in MSO-dependent seizures in these two lines of mice.

  • View full-size image.
  • Figure 3. 

    Latency (A) and percent of convulsion (B) toward various convulsants of MSO-Slow (full bars) and MSO-Fast (open bars) inbred mice. Latencies toward PC (pilocarpine, 300mg/kg) and KA (kainic acid, 25mg/kg) are given in minutes, while latency to PTZ (pentylenetetrazole, 75mg/kg) is given in seconds. We used 10 males and 10 females of the two inbred mice. Values are expressed as mean±SEM, and significance between two inbred mice was estimated using Student t-test: *p<0.05; ***p<0.001.

Fig. 4 shows the effects of various anticonvulsants on seizures induced by MSO in the MSO-Slow (350mg/kg) and MSO-Fast (75mg/kg) lines. The anticonvulsant effect induced by VPA was non-significant in both strains (Fig. 4A), while the anticonvulsant effect of MK-801 was detectable, and significant only in MSO-Fast strain (Fig. 4A). These data were obtained when the chemicals were administered 30min before MSO. In addition, special attention was given to the effect of MK-801, as it induced a very short episode of convulsions at the dose used to antagonise seizures (Velisek et al., 1991). We therefore questioned the interaction of MSO to antagonise the MK-801 effects administered 120min after a convulsive dose of MSO (Fig. 4B). In this condition, MSO also diminished the latency toward MK-801-dependent seizures only in the MSO-Slow strain (58.15±3.52min vs 47.95±16.87min, p=0.0378).

  • View full-size image.
  • Figure 4. 

    Effect of two anticonvulsants on MSO-dependent seizures in MSO-Slow and MSO-Fast mice. We used 10 males and 10 females of MSO-Slow inbred mice (full bars), and 10 males and 10 females of MSO-Fast inbred mice (open bars). MSO was administered at a dose of 350 and 75mg/kg to MSO-Slow and MSO-Fast inbred mice, respectively. Latencies to generalised convulsion are given in minutes, and values are expressed as mean±SEM. Statistical differences were assumed using Student t-test: *p<0.05, **p<0.01, ***p<0.001. (A) Valproic acid (VPA) was administered (ip, 250mg/kg) 30min before MSO (VPA+MSO); MK-801 (ip, 1mg/kg) was administered to mice 30min before MSO (MK-801+MSO). (B) Comparison of latencies to seizures induced by MK-801 administered either alone (MK-801) or 120min after MSO (MSO+MK-801).

The effect of one of the glutamate level-regulating enzymes, GS, was determined in both lines (Fig. 5). As shown in Fig. 5A, kinetic characteristics determined for both lines demonstrated an obvious difference between the two types of mice, as the enzyme activity was significantly higher in MSO-Slow than that in MSO-Fast line. Indeed, the Vmax value of GS extracted from brain cortices is significantly higher than that of MSO-Fast brain cortices (Fig. 5A). This data suggested either an intrinsically higher GS activity, or a higher level of enzyme expression in the MSO-Slow lines compared to MSO-Fast ones. The Km value for glutamate is significantly different between both lines of inbred mice (Fig. 5B). Similarly, the Ki for MSO is also not significantly different between the two strains (MSO-Slow: 0.54±0.24mM MSO; MSO-Fast: 0.43±0.30mM MSO; p=0.665; Fig. 5C), suggesting that the sensitivity of GS toward MSO might not contribute to the MSO-dependent seizures in both inbred lines.

  • View full-size image.
  • Figure 5. 

    Brain cortical glutamine synthetase activity in MSO-Slow and MSO-Fast mice. We used 5 females and 4 males of MSO-Slow inbred mice (full bars, full triangles), and 5 females and 4 males of MSO-Fast inbred mice (open bars, full squares). Values are expressed as mean±SEM and statistical difference was estimated using Student t-test: ***p<0.001, **p<0.01. (A) Glutamine synthetase activity expressed as nmol of glutamate hydroxamate produced per hour per mg of proteins, and is determined as Vmax by increasing glutamic acid concentration. Vmax is calculated using the linear part of the curve. (B) Km determination for glutamic acid using various concentrations. Both values in A and B were calculated using plot of Lineweaver–Burk, i.e., 1/V as a function of 1/S. (C) Determination of Ki for MSO using increasing MSO concentrations, and calculated using log [MSO] as function of activity with a variable slope.

Fig. 6 shows the data obtained with the open-field test using the two lines. There was no statistical difference for the 1st movement, whichever strain or sex effect was considered (STRAIN: F1,80=0.005, p=0.944; SEX: F1,80=0.407, p=0.525; STRAIN×SEX: F2,160=2.658, p=0.107). Conversely, in terms of anxiety, revealed by the numbers of squares crossed, rearing, defecation and time spent in the centre case, a statistical significant difference existed between MSO-Slow and MSO-Fast lines, the latter being more anxious (squared crossed, STRAIN: F1,80=26.335, p<0.0001; SEX: F1,80=2.371, p=0.065; STRAIN×SEX: F2,160=0.281, p=0.598; rearing, STRAIN: F1,80=9.812, p=0.003; SEX: F1,80=0.616, p=0.542; STRAIN×SEX: F2,160=0.024, p=0.878; defecation, STRAIN: F1,80=4.043, p=0.048; SEX: F1,80=4.619, p=0.006; STRAIN×SEX: F2,160=0.394, p=0.745). Moreover, the percent of returns to the centre case of the open field was significantly higher in MSO-Slow strains than in MSO-Fast ones. In addition, a significant sex effect was observed in terms of defecation in MSO-Slow line only (F1,80=9.239, p=0.003; data not shown). These results suggest that the MSO-Fast mice are more anxious, and probably consequently less active, than the MSO-Slow ones.

  • View full-size image.
  • Figure 6. 

    Data obtained using the open-field test. Mice from the 7–9 inbreeding procedure, analysed at 10–12 weeks of age, were submitted to the open-field test. Data are expressed as mean±SEM of 40 mice of both sexes corresponding to 20 mice of each sex. First movement, represents the times spend in the centre of open-field before first mice movement and are expressed in second. Total squares visited represent the numbers of lines crossed. Full bars: MSO-Slow inbred mice. Open bars: MSO-Fast inbred mice. Statistical differences were assumed using repeated ANOVA test: *p<0.05, **p<0.01, ***p<0.001.

In addition, data obtained in terms of spatial memory learning, using the radial-maze in the two inbred lines indicate that no statistical difference exists between the learning capacities of the two lines (data not shown). Indeed, the numbers of new entries and activity index increased in the same manner in both lines, whereas the number of errors, and defecation decreased in the same manner in both lines. The only significant difference was observed between males and females in terms of new entries in the MSO-Fast mice only (STRAIN: F1,80=0.330, p=0.567; SEX: F1,80=1.988, p=0.163; STRAIN×SEX: F2,160=4.584, p=0.036).

Back to Article Outline

Discussion 

The present data strongly suggest that in terms of convulsant and anticonvulsant effects, MSO-Slow inbred mice are distinctly different from the MSO-Fast ones, the latter being more responsive. It seems that the glutamatergic pathways could be one of the targets utilised by MSO to generate seizures in these inbred mice, and that these pathways are less involved in the MSO-Slow than in MSO-Fast lines. We may also assume that the inbreeding process in itself does not delete, but rather improves, the characteristics observed during the selection (Cloix et al., 2010a, Cloix et al., 2010b); and that such characteristics are involved in the observed differences between MSO-Slow and MSO-Fast lines. Moreover, the two strains also differ in their brain cortical GS activity, which is higher in MSO-Slow than in MSO-Fast. In addition, the MSO-Slow mice are less anxious than the MSO-Fast ones, and no difference between the two strains was observed in terms of spatial learning capacities.

Preliminary experiments were done using mice obtained during our selection procedure, and we observed different latencies after the administration of MSO to MSO-Slow and MSO-Fast lines (Cloix et al., 2010a). The augmentation of the selection pressure, the increase in the number of crossing from 6 to 9, and progress of inbreeding by 7–9 brother/sister matings, visibly expanded the differences between the two lines and are not deleterious to their characteristics. The ED50 of latency toward MSO increased from 42min in MSO-Fast mice to a noncomputable value in the MSO-Slow. Indeed, its calculation was not possible because of the flatness of the doses–latencies curve in the MSO-Slow line. This observation suggests that the additional selection, and further generations of inbreeding, enhanced the differences between the two lines. Moreover, the MSO dose necessary to provoke seizures in the MSO-Slow mice was at least 3- to 4-fold higher than that observed at the level of 6th generation of the selection process (Cloix et al., 2010b). Conversely, no change was observed in the MSO dose necessary to induce seizures between the 6th generation of the selection procedure (Cloix et al., 2010b) and the present value obtained after 3 additional MSO-challenges and inbreeding generations for the MSO-Fast line. Such a dissimilarity is confirmed by the differences observed in EEG of MSO-Fast mice compared to MSO-Slow ones. The better sensitivity of the MSO-Fast line to the analogue of glutamate, KA, as compared to PC and PTZ, supposes an involvement of glutamatergic pathways in the action of the convulsant MSO which was used for selection. The present data, corresponding to a different latency toward a commonly used dose of KA strongly suggests that such a response reflects differences in sensitivity toward KA of the MSO-Slow mice compared with MSO-Fast ones. The analyses of KA-dependent latency showed that MSO-Slow mice are more resistant to KA-induced seizures than MSO-Fast ones, which is in agreement with a lower involvement of glutamatergic pathways in MSO-Slow lines. Such an idea is reinforced by the anticonvulsant effect of the MK-801, which blocked selectively NMDA glutamatergic receptors, while less anticonvulsant action was observed with VPA, even if VPA was rapidly metabolised by mice and no effect of VPA in either line of mice was obtained. Indeed, the present data, in addition to those previously obtained with the two inbred lines (Cloix et al., 2010a, Cloix et al., 2010b), suggests less involvement of GABAergic pathways as compared to glutamatergic ones in the mechanisms of MSO-dependent seizures, with no exclusion of the putative involvement of such GABAergic pathways. This observation is corroborated by the higher level of GS activity in MSO-Slow compared to the MSO-Fast ones, with no significant effect on inhibition constant for MSO. This data supposes a higher conversion of glutamate into glutamine leading to a putative, yet to be demonstrated, lower level of glutamate in MSO-Slow line. The fate of brain glutamate under the influence of MSO, which is utilised to select the mice, is unclear. Nevertheless, the powerful inhibition of GS by MSO was thought to trigger the seizures, as glutamate was supposed to increase as a consequence of this enzyme inhibition. However, several data disclaim this eventuality, as a decrease in brain glutamate level was observed after the administration of convulsive doses of MSO (Engelsen and Fonnum, 1985, Fonnum and Paulsen, 1990, Somers and Beckstead, 1990) or no variation (Somers and Beckstead, 1990), while a deficiency of hippocampal GS, induced by MSO, causes seizures in rats (Eid et al., 2008). In addition, in rats with PTZ-induced repetitive epileptic seizures the regions showing a strong glial heat shock response, correspond with reduced GS-activity and GS-nitration, which together are clear indicators of a nitrosative stress response (Bidmon et al., 2008). As GS is confined to astrocytes (Martinez-Hernandez et al., 1977), Phelps (1975) suggested that glutamate might be massively drawn on to build the glycogen molecule, since glycogen accumulates to tremendous levels in the brain under the effect of MSO (Folbergrova et al., 1969, Hevor, 1994, Cloix and Hevor, 2009, Cloix et al., 2010a). For these reasons, the actual role of glutamatergic system in the epileptogeny generated by MSO remains unclear. Moreover, MSO is considered a powerful inhibitor of GS, as a consequence it should be considered that NH4+ might increase. This could lead to a difference between MSO-Fast and MSO-Slow strains in terms of swelling of astrocytes after MSO administration. In addition, it might be possible that a difference between both strains exists in blood–brain barrier sensibility toward MSO leading to a differential blood–brain barrier leaking. If such differences exist between both strains, they might contribute to the mechanisms of MSO-dependant seizures in both strains.

The relationship between brain glutamatergic pathways and epilepsy has already been reported in both man and animal models of epilepsy (Armijo et al., 2005, Andrade and Minassian, 2007, Sierra-Paredes and Sierra-Marcuno, 2007, Vincent and Mulle, 2009). Mood and psychiatric disorders in patients have been described as consequences of status epilepticus (Kanner, 2008, Kanner, 2009, Mula and Monaco, 2009). In addition, the population of epileptic patients is more prone to commit suicide (Kanner, 2009). In our experiments the behaviours of the two inbred lines of mice are visibly different, and the main data indicate that MSO-Fast are more anxious than MSO-Slow. Many reports underline the involvement of glutamatergic system in learning capacity and in anxiety (Mohler, 2006, Hashimoto et al., 2007, Venault and Chapouthier, 2007, Kanner, 2009, Duvoisin et al., 2010). One example is the direct binding of glutamate to cell membranes: glutamate binding to hippocampal membrane fractions of the fully kindled rats was significantly higher when compared to controls, the fully kindled ones having poorer learning performance, whereas binding of not fully kindled rats did not differ from that of controls (Rossler et al., 2000). In our experiments, the difference in GS activity may account for the difference in anxiety behaviour between the two lines due to the pivotal role of the latter enzyme in the glutamate/glutamine pathway. Indeed, even though GS is confined to astrocytes (Martinez-Hernandez et al., 1977), the latter provide glutamine to neurons which transform it into glutamate and GABA as neurotransmitters. Astrocytes withdraw glutamate released by glutamatergic neurons in the synaptic cleft; thus the magnitude of GS activity may influence glutamate/glutamine cycle in the brain. As this enzyme activity is statistically different between the two inbred strains of mice, it is highly probable that this impacted differently on the behaviour of each line. Moreover, we reported a higher level of serotonin in the cortex of MSO-Fast compared to MSO-Slow mice (Cloix et al., 2010a) which could contribute to the anxiety developed by MSO-Fast mice. Previous data (Chapouthier et al., 1998, Guillot et al., 1999, Rossler et al., 2000, Venault et al., 2006, Venault and Chapouthier, 2007, Kanner, 2008, Mula and Monaco, 2009), in addition to the present work, underlines a parallelism between an impairment of behaviour and seizure genesis. It remains to be established what is the basis of the actual relationship between these two physiological alterations.

Back to Article Outline

Acknowledgments 

This work was supported by the Ministère de l’Education Nationale, de la Recherche et de l’Enseignement Supérieur, by the Centre National de la Recherche Scientifique (C.N.R.S.), by the Région Centre, and the Conseil Général du Loiret. Arnaud Boissonnet received a grant from the Conseil Général du Loiret. This is in partial fulfilment of the requirements for the Ph.D. degree at the University of Orleans (A.B.). Many thanks to Ph. Moreau for his helpful technical assistance. Authors disclose no conflict of interests.

Back to Article Outline

References 

  1. Andrade DM, Minassian BA. Genetics of epilepsies. Expert Rev. Neurother. 2007;7:727–734
  2. Armijo JA, Shushtarian M, Valdizan EM, Cuadrado A, de las Cuevas I, Adin J. Ion channels and epilepsy. Curr. Pharm. Des. 2005;11:1975–2003
  3. Belzung C, Griebel G. Measuring normal and pathological anxiety-like behaviour in mice: a review. Behav. Brain Res. 2001;125:141–149
  4. Bidmon HJ, Gorg B, Palomero-Gallagher N, Schleicher A, Haussinger D, Speckmann EJ, et al. Glutamine synthetase becomes nitrated and its activity is reduced during repetitive seizure activity in the pentylentetrazole model of epilepsy. Epilepsia. 2008;49:1733–1748
  5. Blin M, Crusio WE, Hevor T, Cloix J-F. Chronic inhibition of glutamine synthetase is not associated with impairment of learning and memory in mice. Brain Res. Bull. 2002;57:11–15
  6. Chapouthier G, Launay JM, Venault P, Breton C, Roubertoux PL, Crusio WE. Genetic selection of mouse lines differing in sensitivity to a benzodiazepine receptor inverse agonist. Brain Res. 1998;787:85–90
  7. Cloix JF, Hevor T. Epilepsy, regulation of brain energy metabolism and neurotransmission. Curr. Med. Chem. 2009;16:841–853
  8. Cloix JF, Tahi Z, Boissonnet A, Hevor T. Brain glycogen and neurotransmitter levels in fast and slow methionine sulfoximine-selected mice. Exp. Neurol. 2010;225:38–47
  9. Cloix JF, Tahi Z, Martin B, Hevor T. Selection of two lines of mice based on latency to onset of methionine sulfoximine seizures. Epilepsia. 2010;51:118–128
  10. Crusio WE. Genetic dissection of mouse exploratory behaviour. Behav. Brain Res. 2001;125:127–132
  11. Dalsgaard MK, Madsen FF, Secher NH, Laursen H, Quistorff B. High glycogen levels in the hippocampus of patients with epilepsy. J. Cereb. Blood Flow Metab. 2007;27:1137–1141
  12. Duvoisin RM, Pfankuch T, Wilson JM, Grabell J, Chhajlani V, Brown DG, et al. Acute pharmacological modulation of mGluR8 reduces measures of anxiety. Behav. Brain Res. 2010;212:168–173
  13. Eid T, Ghosh A, Wang Y, Beckstrom H, Zaveri HP, Lee TS, et al. Recurrent seizures and brain pathology after inhibition of glutamine synthetase in the hippocampus in rats. Brain. 2008;131:2061–2070
  14. Einat H, Manji HK, Gould TD, Du J, Chen G. Possible involvement of the ERK signaling cascade in bipolar disorder: behavioral leads from the study of mutant mice. Drug News Perspect. 2003;16:453–463
  15. Engelsen B, Fonnum F. The effect of methionine sulfoximine, an inhibitor of glutamine synthetase, on the levels of amino acids in the intact and decorticated rat neostriatum. Brain Res. 1985;338:165–168
  16. Folbergrova J, Passonneau JV, Lowry OH, Schulz DW. Glycogen, ammonia and related metabolites in the brain during seizures evoked by methionine sulphoximine. J. Neurochem. 1969;16:191–203
  17. Fonnum F, Paulsen RE. Comparison of transmitter amino acid levels in rat globus pallidus and neostriatum during hypoglycemia or after treatment with methionine sulfoximine or gamma-vinyl gamma-aminobutyric acid. J. Neurochem. 1990;54:1253–1257
  18. Griffith OW, Meister A. Differential inhibition of glutamine and gamma-glutamylcysteine synthetases by alpha-alkyl analogs of methionine sulfoximine that induce convulsions. J. Biol. Chem. 1978;253:2333–2338
  19. Guillot PV, Sluyter F, Crusio WE, Chapouthier G. Mice selected for differences in sensitivity to a benzodiazepine receptor inverse agonist vary in intermale aggression. Neurogenetics. 1999;2:171–175
  20. Hashimoto K, Sawa A, Iyo M. Increased levels of glutamate in brains from patients with mood disorders. Biol. Psychiatry. 2007;62:1310–1316
  21. Hevor TK. Some aspects of carbohydrate metabolism in the brain. Biochimie. 1994;76:111–120
  22. Hixson JD, Kirsch HE. The effects of epilepsy and its treatments on affect and emotion. Neurocase. 2009;15:206–216
  23. Kanner AM. The use of psychotropic drugs in epilepsy: what every neurologist should know. Semin. Neurol. 2008;28:379–388
  24. Kanner AM. Psychiatric issues in epilepsy: the complex relation of mood, anxiety disorders, and epilepsy. Epilepsy Behav. 2009;15:83–87
  25. Krystal JH, Mathew SJ, D'Souza DC, Garakani A, Gunduz-Bruce H, Charney DS. Potential psychiatric applications of metabotropic glutamate receptor agonists and antagonists. CNS Drugs. 2010;24:669–693
  26. LaFrance WC, Kanner AM, Hermann B. Psychiatric comorbidities in epilepsy. Int. Rev. Neurobiol. 2008;83:347–383
  27. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 1951;193:265–275
  28. Mares P, Mikulecka A, Ticha K, Lojkova-Janeckova D, Kubova H. Metabotropic glutamate receptors as a target for anticonvulsant and anxiolytic action in immature rats. Epilepsia. 2010;51(Suppl. 3):24–26
  29. Martinez-Hernandez A, Bell KP, Norenberg MD. Glutamine synthetase: glial localization in brain. Science. 1977;195:1356–1358
  30. McNamara JO, Huang YZ, Leonard AS. Molecular signaling mechanisms underlying epileptogenesis. Sci. STKE. 2006;2006:re12
  31. Meldrum B. Status epilepticus: the past and the future. Epilepsia. 2007;48(Suppl. 8):33–34
  32. Mohler H. GABAA receptors in central nervous system disease: anxiety, epilepsy, and insomnia. J. Recept. Signal Transduct. Res. 2006;26:731–740
  33. Moult PR. Neuronal glutamate and GABAA receptor function in health and disease. Biochem. Soc. Trans. 2009;37:1317–1322
  34. Mula M, Monaco F. Antiepileptic drugs and psychopathology of epilepsy: an update. Epileptic Disord. 2009;11:1–9
  35. Phelps CH. An ultrastructural study of methionine sulphoximine-induced glycogen accumulation in astrocytes of the mouse cerebral cortex. J. Neurocytol. 1975;4:479–490
  36. Racine RJ. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr. Clin. Neurophysiol. 1972;32:281–294
  37. Reid CA, Berkovic SF, Petrou S. Mechanisms of human inherited epilepsies. Prog. Neurobiol. 2009;87:41–57
  38. Rossler AS, Schroder H, Dodd RH, Chapouthier G, Grecksch G. Benzodiazepine receptor inverse agonist-induced kindling of rats alters learning and glutamate binding. Pharmacol. Biochem. Behav. 2000;67:169–175
  39. Scharfman HE. The neurobiology of epilepsy. Curr. Neurol. Neurosci. Rep. 2007;7:348–354
  40. Schmidt D. Drug treatment of epilepsy: options and limitations. Epilepsy Behav. 2009;15:56–65
  41. Sierra-Paredes G, Sierra-Marcuno G. Extrasynaptic GABA and glutamate receptors in epilepsy. CNS Neurol. Disord. Drug Targets. 2007;6:288–300
  42. Somers DL, Beckstead RM. Chronic methionine sulfoximine administration reduces synaptosomal aspartate and glutamate in rat striatum. Neurosci. Lett. 1990;115:335–340
  43. Stafstrom CE. Epilepsy: a review of selected clinical syndromes and advances in basic science. J. Cereb. Blood Flow Metab. 2006;26:983–1004
  44. Velisek L, Veresova S, Pobisova H, Mares P. Excitatory amino acid antagonists and pentylenetetrazol-induced seizures during ontogenesis. II. The effects of MK-801. Psychopharmacology (Berl). 1991;104:510–514
  45. Venault P, Beracochea D, Valleau M, Joubert C, Chapouthier G. Mouse lines selected for difference in sensitivity to beta-CCM also differ in memory processes. Behav. Brain Res. 2006;173:282–287
  46. Venault P, Chapouthier G. Plasticity and anxiety. Neural Plast. 2007;2007:75617
  47. Vincent P, Mulle C. Kainate receptors in epilepsy and excitotoxicity. Neuroscience. 2009;158:309–323
  48. Wieronska JM, Pilc A. Metabotropic glutamate receptors in the tripartite synapse as a target for new psychotropic drugs. Neurochem. Int. 2009;55:85–97
  49. Wolfe LS, Elliot KAC. Chemical studies in regulation to convulsive conditions. In:  Elliot KAC,  Page IH,  Quastel JH editor. Neurochemistry. 2nd ed.. USA: Springfield; 1962;p. 694–697

PII: S0920-1211(11)00236-1

doi:10.1016/j.eplepsyres.2011.08.012

Epilepsy Research
Volume 98, Issue 1 , Pages 25-34, January 2012

Article Tools

* Image Usage & Resolution