|Home | About | Journals | Submit | Contact Us | Français|
The kindling model of temporal lobe epilepsy and the memory model of long-term potentiation (LTP) may have common underlying mechanisms. This is evident by the demonstration that certain signaling molecules play a key role in both. Recently, a brain specific isoform of protein kinase C (PKMζ) has been shown to play a significant role in both maintaining LTP and memory storage. We were interested in determining if this kinase had a crossover role in kindling-induced epileptogenesis. Using developing and adult rats we examined the role of PKMζ in kindling. In developing (P15) rats we determined the effect of PKMζ on retention of amygdala kindling and kindling rate by intra-amygdala administration of a selective PKMζ antagonist, ZIP (10 nmol). In adult rats we examined the effect of PKMζ inhibition, ZIP (10 nmol), on afterdischarge (AD) thresholds and kindling retention using rapid hippocampal kindling. Inhibition of PKMζ by the antagonist ZIP did not affect kindling rate or retention in developing rats. In addition there was also no observed effect on AD thresholds and kindling retention in adult rats. Our results show that, despite the similarities between kindling and LTP in their induction, there is dissociation in the role that PKMζ plays within the two in maintenance. This may be of importance in establishing a separation between the pathophysiological processes involved in sustaining kindling and the physiological mechanisms involved in maintaining LTP and memory storage.
Kindling is a model of complex partial seizures with secondary generalization that results from repetitive, high-frequency electrical stimulation of limbic brain regions, such as the hippocampus and amygdala [4, 7, 16, 17, 20]. This model has been widely used to study the basic mechanisms involved in the epileptogenesis of temporal lobe epilepsy (TLE) and has often been used to evaluate the efficacy of antiepileptic drugs . The response of the brain to kindling is that of progressively worsening behavioral seizures with increased duration afterdischarges (AD) recorded on the EEG [7, 10, 20]. This is the result of the recruitment of vast brain areas outside the initial seizure focus which become reinforced through daily stimulations that ultimately result in generalized seizures. The recruitment and reinforcement in these networks is reminiscent of the plasticity that can occur in long-term potentiation (LTP) [2, 22].
LTP is a form of synaptic strengthening that results from high-frequency stimulation in certain brain regions, most notably the hippocampus . Many authors have proposed that LTP is the underlying mechanism by which organisms learn and form memories [19, 26, 27]. Since LTP was first described by Bliss and colleagues in the 1970’s there has been much study of the mechanisms involved in LTP. Over the past 30 years there have been numerous molecules that have been implicated in the underlying synaptic strengthening that occurs with LTP; however, not all of these have been shown to influence memory , and those that do involve the initial induction phase of LTP and the early consolidation phase of memory formation. More recently a new molecule implicated in the persistent maintenance of LTP and the storage of memory has been described  [21, 25]. This molecule is a brain-specific isoform of the protein kinase C, protein kinase M zeta (PKMζ) [6, 21, 24]. Blockade of PKMζ by a selective pseudosubstrate inhibitor ZIP rapidly abolishes preestablished LTP both in hippocampal slices and in vivo and the storage of active place avoidance in the hippocampus and conditioned taste aversion memory in the insular cortex, in rats, for a period lasting up to one month. These results show that persistent protein kinase activity may play a crucial role in memory formation and maintenance for long periods of time .
Since LTP is considered to be the mechanism by which learning and memory are achieved, and is maintained by PKMζ, wewere interested in whether this kinase also plays a role in maintaining the synaptic alteration of kindling. In order to determine if PKMζ activity influenced kindling we used both developing and adult male rats. The use of both developing and adult animals would allow us to determine if there were any age-related differences in PKMζ activity and the consolidation of kindling. Furthermore, since developmental kindling is rapid (single day) it also would allow us to examine if PKMζ can affect kindling rate. Thus we used PKMζ inhibition, by intracerebral ZIP administration to the kindling site, and determined its effect on kindling retention (developing and adult rats), AD thresholds (adult rats) and kindling rate (developing rats).
Male rat pups from timed pregnant Sprague-Dawley rats (Taconic Farms Germantown, NY USA) were used in the developmental kindling studies. All animals were housed under standard environmental conditions with a constant ambient temperature of 23°C (60% humidity) and a 12:12 hr light:dark cycle with food and water available ad-libitum. On postnatal day 13 (P13, date of birth = P0) separate litters of animals were prepared for amygdala kindling. Under ketamine/xylazine anesthesia combination cannula/bipolar electrodes (25 gauge cannula equipped with stainless steel bipolar electrode, Plastics One Roanoke VA) were aimed at the left basolateral amygdala using stereotaxic techniques. The coordinates for the basolateral amygdala for P13 rats were: 3mm posterior to bregma, 3.8mm lateral to the midline, and 8mm ventral to the skull surface with the incisor bar was set at −2.5mm. Upon completion of surgery animals were placed back with their dams and allowed to recover until P15 when the kindling procedure was started.
The kindling stimulation parameters consisted of biphasic square wave pulses of 1 sec duration with an amplitude of 400 μA (peak to peak) delivered at 60 Hz . Afterdischarges were amplified, digitized and recorded using a PC. In order to study the effect of PKMζ inhibition on kindling retention we stimulated animals every 20 min until they achieved 5 consecutive stage 4 (bilateral forelimb clonus with rearing) or higher seizures . The animals were then placed back with their dams for 24hrs before the rekindling was initiated. Twenty four hours after the final seizure animals were stimulated to ensure that they were kindled i.e. displayed stage 4 or higher seizures. Animals were then left to recover for 1 hour. Rats were then treated with either ZIP or a scrambled ZIP control peptide dissolved in PBS. ZIP (10 nmol) was infused into the basolateral amygdala in a volume of 1 μl over 3 min: the control peptide was treated the same. This dose was selected based upon previous studies that found ZIP efficacious in the blockade of memory storage and LTP [19, 21, 24, 25]. Thirty minutes later all animals were stimulated again and seizure severity (behavioral) and duration (EEG) was ascertained. In addition animals were also tested at 60 minutes post treatment and again at 24 hours to determine if there were any time related effects of ZIP.
To examine the effect of PKMζ inhibition on kindling rate, a separate group of animals was used. In this experiment the stimulation parameters were the same but the kindling procedure was changed. On P15 animals were given a single AD producing stimulation. One hour later animals were treated with ZIP or the scrambled peptide and placed back in their home cages. Twenty four hours later animals restarted the kindling procedure and were followed to the same endpoint as in the previous experiment. Electrode placement was confirmed using posthoc histology.
Adult Sprague Dawley male rats (200 – 300g; approximately 1.5 – 3 months old) were equipped with combination cannula/bipolar electrode in the dorsal hippocampus (AP -3.8, ML -2.5, SI -2.8). After a 1.5 week recovery period, the animals underwent a rapid kindling protocol according to the procedures of Lothman & Williamson . This began with an AD measurement (stimulus: 10s train 1ms biphasic pulses at 20 Hz, 0.5ms each phase), and continued with four alternate days of 12 stimulations each at a current supratheshold to the initial threshold AD. Animals were considered kindled when they established and maintained Racine class 4 and 5 seizures in response to stimulations on the final day of kindling . After establishment of the kindled state animals were left to rest for a period of 1 week. On the day of infusion, AD thresholds were again measured for each animal, subsequently each animal received 10 nmol of ZIP in 1 μl PBS or scrambled ZIP slowly over 20 min while under light ketamine/xylazine anesthesia. On the day after infusion, AD thresholds were again measured in all animals. One hour later animals received 4 additional stimuli, separated by 1 hour, each at their respective kindling current. Behavioral and electrographic responses were ascertained throughout the kindling procedures. Electrode placement was also confirmed using posthoc histology.
Data are presented as means +/− standard error of the mean (SEM). For the effects of ZIP on seizure severity and duration in developing animals, post treatment means were compared to baseline (in this case rekindled severity and duration) using paired t-tests. For the effects of ZIP on kindling rate means were compared using a oneway ANOVA. For adult animals the severity of seizures after treatment was also compared to baseline using a paired t-test. Significance for all tests was set at p < 0.05.
After animals were fully kindled they were treated with ZIP or a scrambled control peptide and retested at 30 min, 60 min and 24 hr after treatment. Treatment with ZIP or the scrambled peptide did not affect kindling retention at any of the time points examined (figure 1A, B and C both groups n = 4) as all animals displayed similar seizure stages at all points tested. We also examined the effect of ZIP on AD duration recorded quantitatively on the EEG. Again we also did not find any significant changes in seizure duration due to ZIP at any time point tested, with only a minor trend towards a decrease in seizure duration at 24 hours (figure 1D, E, and F both groups n = 4).
In a separate group of animals we examined the effect of ZIP on kindling rate. We found that ZIP given prior to kindling did not effect the rate at which developing animals kindled (figure 2A and B, naïve animals n = 8, scrampled ZIP n = 5 and ZIP n = 5).
After rapid kindling, the AD thresholds of all animals were reduced. The day after ZIP injection, the AD thresholds of animals in both groups were slightly higher, with no difference seen between the experimental and control group (data not shown). All animals, regardless of injection status, responded to the stimuli with four Racine stage 4 or 5 generalized seizures, with the exception of one animal which had one stage 0 seizure followed by three generalized seizures. No obvious effect of ZIP was observed (figure 3, scrampled ZIP n = 2 and ZIP group n = 3).
In the current study we show that inhibition of PKMζ does not affect kindling or kindling retention in either developing or adult rats. This was shown by the inability of the selective inhibitor ZIP to cause any significant alteration in kindling rate, kindling retention, and AD thresholds. This finding is in contrast to the significant role that has been established for PKMζ in LTP and memory [21, 24, 25].
Both kindling and LTP are phenomena that can arise from high-frequency stimulation of discrete brain regions. At the core of each is a heavy reliance on activation of the glutamatergic system. Both rely on initial activation of NMDA receptors which results in increased intracellular Ca+2 that can influence a wide variety of intracellular signaling pathways [12, 16, 18, 28]. The lack of effect of PKMζ in this study is not likely due to the dose of ZIP that we used because this dose has been shown to effectively block memory storage for a long periods of time and after strong training [19, 25]. Thus we expected that if there were to be an effect on kindling that this dose would be sufficient. Since we did not even observe a trend towards any differences (other than seizure duration) it is unlikely that increasing the dose would have had much of an effect on the main variables of kindling retention and kindling rate.
This lack of effect is likely due to two factors. The first is the inability of ZIP to affect the local AD. The second factor is that kindling not only recruits networks outside the original focus by means of synaptic potentiation, but also significantly modifies them. In adult rats some of these modifications include synaptogenesis , neurogenesis , and neurodegeneration either in the focus , or in distal areas . While it is true that in vivo LTP can produce synaptogenesis (mossy fiber sprouting in the dentate gyrus) it is not accompanied by other seizure/kindling induced alterations, such as cell loss and neurogenesis . However, in the absence of pathological changes (such as in developing rats ) ZIP also had no effect. This suggests that there may be other alterations involved in the recruitment and refinement of neural networks in kindling that are dissociated from PKMζ activity and LTP.
Although this study is largely negative in nature, it is important to note that several compounds that affect LTP (for example NMDA antagonists) have shown efficacy in kindling. These may be important in the processes of the induction of epileptogenesis (without SE and spontaneous seizures) and are worth examining in the hunt for new antiepileptic therapies. Our results indicate, however, that the maintenance mechanisms that sustain epileptogenesis and behavioral memory are distinct. This gives hope that the molecular mechanisms that sustain pathophysiological hyperexcitability may be ameliorated without affecting the storage mechanism of long-term memories.
This work was supported by the National Institutes of Health Grants NS20253, NS058303 and the Heffer Family Medical Foundation (S.L.M.), NS049307, NS052519 and the Betsy and Jonathan Blattmachr family (H.B.), NS59074 (D.J.E.), MH53579 and MH57068 (T.C.S.). Dr. S.L. Moshé is the recipient of the Martin A. and Emily L. Fisher Fellowship in Neurology and Pediatrics. The authors would also like to acknowledge the assistance of H. Wong, P.K. Mansuripur, A. Soufer, and D.J. Ellens.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.