Address correspondence and reprint requests to Dr. S.B. Churn at Department of Neurology, Virginia Commonwealth University, Richmond, VA 23298–0599, U.S.A. E-mail: SCHURN@HSC.VCU.EDU
Summary: Purpose: This study was conducted to characterize the early cellular changes in CaM kinase II activity that occur during the induction of status epilepticus (SE).
Methods: The pilocarpine model of SE was characterized both behaviorally and electrographically. At specific time points after the first discrete seizure, specific brain regions were isolated for biochemical study. Phosphate incorporation into a CaM kinase II–specific substrate, autocamtide III, was used to determine kinase activity.
Results: After the development of SE, the data show an immediate inhibition of both cortical and hippocampal CaM kinase II activity in homogenate, but a delayed inhibition in synaptic kinase activity. The maintenance of synaptic kinase activity was due to a translocation of CaM kinase II protein to the synapse. However, despite the translocation of functional kinase, CaM kinase II activity was not maintained, membrane potential was not restored, and the newly translocated CaM kinase II did not terminate the SE event. Unlike the homogenate samples, in the crude synaptoplasmic membrane (SPM) subcellular fractions, a positive correlation is found between the duration of SE and the inhibition of CaM kinase II activity in both the cortex and hippocampus.
Conclusions: The data support the hypothesis that alterations of CaM kinase II activity are involved in the early events of SE pathology.
Status epilepticus (SE) is a life-threatening emergency affecting >100,000 people annually in the United States (1–4) and is associated with a significant mortality rate (4–6). SE is commonly defined as continuous seizures lasting ≥30 min without recovery of consciousness (2). Recently, Lowenstein et al. (7,8) defined SE in adults and children older than 5 years as ≥5 min of continuous seizures or two or more discrete seizures between which incomplete recovery of consciousness occurs (7,8). This definition is justified by the fact that most tonic–clonic seizures cease within 5 min (9). Additionally, a loss of efficacy of front-line medications is found, including benzodiazepines (BZDs), as seizure duration increases (3). This decreased efficacy is observed both clinically and in experimental models. For this reason, we sought to elucidate the early cellular mechanisms that are involved in the loss of membrane-potential regulation and drug efficacy during SE.
Calcium/calmodulin-dependent protein kinase II (CaM kinase II) is a neuronally enriched protein that makes up the majority of the total protein in the postsynaptic density (10). CaM kinase II has been shown to modulate positively both glutamatergic (11–16) and γ-aminobutyric acid (GABA)ergic channels (17–19). For instance, an increase in CaM kinase II activity has been associated with induction of long-term potentiation (LTP) (16,20–26). Activation of CaM kinase II results in this increased phosphorylation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptors, which causes an increase in AMPA function, and is a critical step in the induction of LTP. Injection of CaM kinase II-α subunit also has been shown to modulate GABAergic channels. Activation of CaM kinase II has been shown to increase agonist-evoked inhibitory currents (17) and to increase both agonist (18) and allosteric modulator binding (19). Because CaM kinase II positively modulates both excitability and inhibitory synaptic receptor function, alteration of CaM kinase II activity would alter neuronal membrane excitability.
CaM kinase II has been well studied, and its activity is altered in many central nervous system pathologies (27). It has been shown that CaM kinase II is significantly inhibited after stroke (28). Additionally, multiple laboratories have demonstrated that epileptiform activity also causes an inhibition of CaM kinase II (29–31). One hour of SE has been shown to inhibit CaM kinase II (32–34) activity in both the short (35,36) and the long term (32) after a single SE episode. However, little research has looked at the early cellular changes that occur during SE induction, and no research has been done to characterize the temporal relation between the induction of SE and the inhibition of CaM kinase II that occurs.
If loss of CaM kinase II activity is involved in the early pathological events in SE, then it is important to determine at what time point during/immediately following the initiation of SE that the inhibition of CaM kinase II occurs. This study used EEG monitoring in the Pilo-SE model and behavioral observations to track the temporal profile of specific cellular changes occurring as the pathology is developing. Coincident with the development of SE, the data show an immediate inhibition of CaM kinase II activity in homogenate, but a delayed inhibition in synaptic activity. The data suggest that the inhibition of CaM kinase II activity is involved in the early cellular responses during the induction of SE.
MATERIALS AND METHODS
All materials were reagent grade and purchased from Sigma Chemical Company (St. Louis, MO, U.S.A.) and Fisher Scientific (Pittsburgh, PA, U.S.A.) unless otherwise stated. (γ-32P)adenosine triphosphate (ATP) was purchased from Perkin Elmer Life and Analytical Sciences (Boston, MA, U.S.A.). Adult male Wistar rats were purchased from Harlan Laboratories (Indianapolis, IN, U.S.A.). Autocamtide 3 was purchased from Calbiochem-Novabiochem Corp. (La Jolla, CA, U.S.A.).
Pilocarpine model of status epilepticus
All animal-use procedures were in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Virginia Commonwealth University Institutional Animal Care and Use Committee. Wistar rats were handled after arrival from Harlan Laboratories, for acclimation to handling before drug treatment. One week before the induction of SE, four surface electrodes were implanted into the skulls of rats under ketamine anesthesia, as described previously (36,37). Two frontal electrodes were implanted over frontal cortex [3.5 mm anterior to bregma, ±2.5 mm L/R (F1/F2)]. Two posterior electrodes were implanted over parietal cortex and hippocampus [2.0 mm posterior to bregma, ±2.5 mm L/R (P1/P2)]. A fifth electrode was fixed onto the surface of the skull as a ground. The electrodes were secured in place with dental acrylate, and the animals were allowed ≥5 days to recover from surgery before experiments were performed.
Induction of status epilepticus
Thirty minutes before the injection of pilocarpine, methylscopolamine, a muscarinic antagonist, was administered i.p. (1 mg/kg) to reduce adverse peripheral affects of the pilocarpine. Control and sham-surgery animals were hooked up to video-EEG machines (Viking IV; Nicolet, Madison, WI, U.S.A.), and baseline EEG recordings were obtained for ≥10 min after scopolamine injection. SE was induced in experimental animals by i.p. injection of 375 mg/kg pilocarpine HCl (Pilo), a muscarinic agonist. Behavioral and encephalographic activities were recorded throughout the procedure. Once initial seizure activity was observed, the time was noted, and rats were allowed to seize for specific amounts of time before the animals were processed. These time points consisted of 10, 15, 20, 30, 40, 50, 60, and 70 min after the first discrete seizure. Because this laboratory previously characterized SE onset as being ∼10 min after the first discrete seizure (36), these times approximate 0, 5, 10, 20, 30, 40, 50, and 60 min of SE, respectively. Electrographic data were analyzed for time to first seizure, time from first seizure to SE, average duration of SE for each animal, and percentage of time spent in each stage of SE for each animal. Spectral analysis of EEG activity was performed by using Insight II software (Persyst Corporation, Prescott, AZ, U.S.A.). After induction of SE, seizure activity was characterized according to the system of Handforth and Treiman (38,39). Seizure severity was determined based on the amount of time spent in each stage of SE. Behavioral seizures were assessed according to the scale of Racine (40), as described previously (36).
Brain region isolation
Brain region isolation was performed as previously described (32), with slight modifications (36). Specific brain regions (cortex and hippocampus) were dissected away on ice and immediately homogenized into an iced-cold buffer containing 50 mM Tris-HCl (pH 7.4), 7 mM EGTA, 5 mM EDTA, 320 mM sucrose, 1 mM dithiothreitol (DTT), and 0.3 mM phenylmethylsulfonyl fluoride (PMSF). The brain regions were homogenized by 10 up-and-down strokes with a Teflon pestle at 12,000 rpm (Fordham, Bethel, CT, U.S.A.). A portion of the sample was then aliquoted and frozen at −80°C until processed for biochemical analysis.
Isolation of subcellular fractions
Subcellular fractions were isolated by a differential centrifugation procedure (41) previously described (18,19). In brief, brain region homogenates were centrifuged at 5,000 g for 10 min (Beckman JA-17 rotor) to produce a crude nuclear pellet (P1). The supernatant from the spin (S1) was then centrifuged for 30 min at 18,000 g to produce a crude synaptoplasmic membrane/mitochondrial pellet (crude SPM), which was resuspended in 1/10 of the original volume of homogenization buffer to concentrate the fraction. All fractions were rapidly separated into aliquots and stored at −80°C for later use.
Substrate phosphorylation was performed as previously described with slight modifications (36,42). Brain region homogenates and crude SPM fractions were normalized for protein concentration by using the Bradford method with bovine serum albumin (BSA) as the standard (43). The CaM kinase II–dependent phosphorylation reaction solutions contained sample (5 μg), 10 mM Tris-HCl (pH 7.4), 7 μM (γ-32P) ATP, 10 mM MgCl2, 30 μM autocamtide 3 (AC-3), ±5 μM CaCl2, and ± 0.01 mg/ml calmodulin (CaM). Deionized water was added to bring the final volume of each tube to 100 μl. Basal reactions were assessed in the absence of calcium. ATP was added, the tubes were vortexed, and then placed into a water bath at 30°C. Tubes were allowed to equilibrate in a 30°C water bath before initiation of the reaction. For Mx reactions, Ca2+ was added after the first minute; the reaction progressed for 60 s, and was terminated by the addition of 20 mM EDTA. Then 20 μl of the reaction solution was spotted on P-81 filter paper in triplicate and washed 3 times in a phosphoric acid wash to remove any unbound phosphate. The P-81s were then washed in an acetone wash and were allowed to air dry completely. The filter papers were then placed into scintillation vials, 5 ml of cytoscint was added, and the reactions were counted by using a Beckman LS 2800 scintillation counter.
Immunodetection of CaM kinase II
Western analysis was performed to quantify the α-CaM kinase II subunit protein levels essentially as described previously (28,36,42). In brief, homogenate and crude synaptic cortical and hippocampal fractions were resolved on sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane by using the Trans-blot system (Bio-Rad, Hercules, CA, U.S.A.). The nitrocellulose membranes were then immersed for 1 h in blocking solution containing phosphate-buffered saline (PBS, pH 7.4), 0.05% (vol/vol) polyoxyethylene sorbitan monolaurate (Tween-20), 50 g/L Bio-Rad blotting-grade dry milk, and horse serum (Vector Laboratories Inc., Burlingame, CA, U.S.A. diluted 50 ml/10 ml). The nitrocellulose membrane was then incubated with primary antibody [clone SA-162, diluted 1:10,000, mouse monoclonal immunoglobulin G (IgG); Sigma Chemical Co.] in blocking solution for 1 h. Membranes were then washed 3 times in a wash solution containing PBS, Tween-20, and dry milk. The nitrocellulose membranes were then reacted with a secondary antibody in blocking solution for 45 min. The membrane was then washed 3 times and incubated with an avidin-biotinylated horseradish peroxidase complex for 30 min. Nitrocellulose was then washed 3 times in PBS for 10 min per wash. Blots were developed with a solution containing 10 ml PBS, 0.025% (vol/vol) H2O2, and 8 mg 4-chloro-1-napthol dissolved in 2 ml of methanol. Specific immunoreactive bands were quantified by computer-assisted densitometry with a BioRad GS-800 calibrated densitometer and the BioRad Quantity One software (Version 4.4.0), as described previously (32,36).
Data were analyzed by using GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA U.S.A.). Student's t tests were used for all single, parametric comparisons. For multiple comparisons, one-way analysis of variance (ANOVA) with a Tukey post hoc analysis with 95% confidence intervals was used. All data are presented as mean ± standard error of the mean (SEM).
Time-dependence of Pilo-induced SE
Electrographic and video recordings were used to characterize SE susceptibility, time to first seizure, number of discrete seizures, time from first seizure to SE, seizure severity once in SE, average duration of SE for each time point, percentage of time spent in each stage of SE for each time point, and death rate. Baseline recordings (Fig. 1A) were taken after scopolamine injection, and activity included low-amplitude, asynchronous spikes. Typical seizure progression was observed as reported previously (36). Baseline encephalographic activity developed into a sudden increase in amplitude and decrease in frequency. Spectral analysis of ictal activity during the first discrete seizure increased to a >10-fold increase in the delta component, which denoted the onset of the discrete seizure phase.
EEG activity was analyzed at specific time points after the first discrete seizure. These time points were 10, 15, 20, 30, 40, 50, 60, and 70 min after first discrete seizure, respectively termed 10-min, 15-min, 20-min, 30-min, 40-min, 50-min, 60-min, and 70-min animals. The 10-min animals typically experienced an average of 0–2 min of SE, and of that time, ∼50% was spent in the wax/wane (W/W) phase of SE. The fast/slow (F/S) phase of SE comprised 42% of the SE duration, whereas early continuous (EC) comprised the other 8%. In all experimental rats, EC was rarely observed during the first 10 min of SE. Two rats entered SE 5 min after the first discrete seizure, skewing the EEG data. The 15-min animals had similar EEG activity, spending ∼5 min in SE. Of the 5 min of SE, these rats typically spent ∼42% of their time in W/W, 42% of their time in F/S, and 16% of the time in EC. The 20-min animals, rats that experienced 10 min of SE, spent 23% of the time in W/W and 9% of the time in F/S. This time window was the first time EC occurred for the majority of the SE duration, accounting for the other 68% of the time. As the duration of SE increased, the rats tended to spend less time in W/W and more time in EC.
Similar to the 20-min animals, the 30-min rats spent 20% of the total time in SE in W/W, 3% in F/S, and 77% in EC. These rats spent ∼20 min in SE. At the 40-min time point, SE duration was roughly 30 min. Of the 30 min, the rats spent 14% in W/W, 37% in F/S, and only 49% in EC. In this experimental group, an observed reversion to F/S occurred. This observation of alternating back to previous stages has been observed in earlier studies in which rats typically exit and then reenter stages before going into the next phase of SE (36). The 50-min animals spent an average of 35 min in SE, although these data are skewed because of two animals spending ≥25 min in the discrete seizure phase and therefore only spending 25 min in SE, explaining the decrease in SE duration. These rats, however, spent 13% of the total SE duration in W/W, 35% of the time in F/S, 19% in EC, and 33% in fast spiking with pauses (FSWPs). This is the first observation of this later, more severe stage of SE; similar to previous reports that suggest that FSWPs occurs later in SE.
The two final time points in this study, 60- and 70-min animals, spent ∼50 and 60 min in SE, respectively. These animals spent ∼10% of SE in W/W (Fig. 1B), 40% in F/S (Fig. 1C), and the other 50% in EC (Fig. 1D) and FSWPs (Fig. 1E). The final phase of SE, late continuous (LC) (Fig. 1F), has an onset ∼ 70 min after the discrete seizure. This phase was observed in a few of the 70-min rats, but it made up a small portion of the total time in SE. As the rats seized for a longer period, it was noted that they spent more time in the later stages of SE, possibly suggesting that the longer the duration, the more severe the seizure (36).
SE-induced inhibition of CaM kinase II activity
SE-induced inhibition of CaM kinase II activity in cortical and hippocampal brain homogenate
Prolonged SE has been shown to cause an inhibition of CaM kinase II activity (32,33,36,44,45). This SE-induced inhibition has been shown for adolescent rats through adult rats (36). To determine whether the inhibition of CaM kinase II activity is an early event in SE, rats were allowed to seize, and at specific time points, brain regions were isolated, homogenized, and tested for CaM kinase II activity by using a previously described method of substrate phosphorylation reactions (36) (see Materials and Methods).
In control animals, the basal activity of cortical samples was ∼0.06 pmol PO4/μg protein (n = 25). No differences in basal activity were observed in cortical homogenates for any of the SE times studied when compared with control values. Under maximal conditions, cortical control homogenates had about a ninefold increase from basal levels to maximum levels, 0.56 pmol PO4/μg protein (n = 25; Fig. 2A). Although all samples examined displayed a significant increase in Ca2+-dependent activity when compared with basal levels (n = 25; p < 0.001, Student's t test), induction of SE resulted in an immediate inhibition of maximal CaM kinase II activity. In the 10-min animals, a significant inhibition (18%) in CaM kinase II activity was noted compared with that in controls (0.46 pmol PO4/μg protein, n = 9; p < 0.05, one-way ANOVA).
In 15-min animals, maximal CaM kinase II activity was ∼0.36 pmol PO4/μg protein (n = 4; p < 0.001). Similar results were observed in 20-, 30-, and 40-min animals with maximal levels of CaM kinase II activity approximately equal to 0.40 pmol PO4/μg protein (n = 15; p < 0.001), 0.36 pmol PO4/μg protein (n = 3; p < 0.001), and 0.41 pmol PO4/μg protein (n = 9; p < 0.001), respectively. The final time point in this study was the 70-min time point, which corresponds to previous studies in this laboratory (36). At this time point, maximal CaM kinase II activity was ∼0.38 pmol PO4/μg protein (n = 6; p < 0.001).
Hippocampal homogenate (Fig. 2B) displayed similar CaM kinase II activity as observed in cortical homogenates, as previously reported (36). Basal levels of control samples had an activity of ∼0.06 pmol PO4/μg protein (n = 15). Again, in all samples tested, basal reactions varied between 0.06 and 0.07 pmol PO4/μg protein (Fig. 2B). When maximally stimulated by using Ca2+ and calmodulin, control homogenates had an ∼11-fold increase from basal levels to maximum levels, 0.68 pmol PO4/μg protein (n = 15; p < 0.001, Student's t test). Under maximal stimulation, the 10-min animals displayed a statistically significant inhibition of CaM kinase II activity of ∼30% (0.4 pmol PO4/μg protein; n = 6; p < 0.05). Hippocampal homogenates decreased further at the 15-min time point with maximal activity of 0.3 pmol PO4/μg protein. At each of the following time points, 20, 30, 40, and 70 min, the CaM kinase II activity was between 0.3 and 0.4 pmol PO4/μg protein. Again, this is consistent with previous findings (36).
Initial electrographic characterization of seizure progression was performed to determine the temporal profile of SE-induced inhibition of CaM kinase II activity. The unexpected observation that significant inhibition of CaM kinase II activity occurred coincident with the induction of SE prevented the ability to correlate loss of enzyme activity with seizure progression during SE. Thus in both the cortex and hippocampus, loss of CaM kinase II activity reflects the onset of SE and is not a measure of seizure severity during SE.
Immunodetection of CaM kinase II activity in homogenate
Previously, no decrease in total protein levels of CaM kinase II was observed after 60 min of SE (35,36). Although unlikely, it is possible that the total protein was being degraded and replenished by 70 min. Therefore Western blot analysis was performed by using a previously characterized antibody (28,36,42) to determine protein levels in both cortical and hippocampal homogenate samples (Fig. 3A and B). Control, 10-, 15-, 20-, 30-, and 40-min time point samples were loaded. In cortical homogenate (Fig. 3A), the relative density for the control sample was 0.46. The 10-min time point relative density was 0.41, the 15-min time point was 0.47, the 20-min time point relative density was 0.47, and the 30- and 40-min time point relative densities were 0.49 and 0.48. For hippocampal samples (Fig. 3B), similar results were observed. Relative densities for control samples were 0.33, and all time points studied were between 0.27 and 0.35. None of these changes were significant when compared to each other or control values. This observation supports previous findings in this laboratory in which no statistically significant difference was observed in homogenate samples in either the cortex or hippocampus samples.
Subcellular SE-induced inhibition of CaM kinase II activity
SE-induced inhibition of CaM kinase II activity in cortical and hippocampal brain subcellular fractions
SE has been shown to cause an inhibition of CaM kinase II activity in both homogenate and subcellular fractions, mainly the crude SPM (46). The next phase of this study examined the subcellular activity of CaM kinase II. Previous studies have looked at 70 min of seizure activity, corresponding to ∼60 min of SE. This study looked at 10 min after the first discrete seizure, which has been previously characterized as the approximate onset of SE (36). In addition to the 10-min time point, the following time points were also studied in the crude SPM fraction: 15, 20, 30, 40, and 70 min after the first discrete seizure.
Basal levels of CaM kinase II activity in the control P2 samples were ∼0.05 pmol PO4/μg protein. This is similar to the values observed in homogenate samples. However, control maximal levels were significantly reduced when compared with control homogenate. Control maximal activity was ∼0.28 pmol PO4/μg protein, a decrease of ∼68% from homogenate. These values are similar to previous findings (46).
Unlike observations made in brain homogenates, no immediate decrease in total CaM kinase II activity was observed in the crude SPM fraction. In contrast, in the 10-min animals, crude cortical SPM CaM kinase II activity increased to 0.34 pmol PO4/μg protein, a statistically significant increase of 21% when compared with control animals (n = 7; p < 0.05, Students t test) (Fig. 4A). However, this is the only time point at which a significant increase in CaM kinase II activity was observed. In the 15-, 20-, and 30-min animals, CaM kinase II activity peaked at ∼0.30 pmol PO4/μg protein before decreasing to 0.28 pmol PO4/μg protein at 30 min. None of these changes was statistically significant when compared with control samples. However, crude SPM isolated from 70-min animals displayed a statistically significant inhibition of CaM kinase II activity with activity levels of 0.18 pmol PO4/μg protein (n = 4; p < 0.01, one-way ANOVA).
In crude SPM isolated from the hippocampus, similar results were observed (Fig. 4B). Control samples had a specific CaM kinase II activity of 0.27 pmol PO4/μg protein, findings that support previous work in this laboratory (46). Again, basal levels for all control and seizure animals were ∼0.05 pmol PO4/μg protein with slight, but not statistically significant, differences. A trend similar to that observed in the cortex was seen in the hippocampus, with an increase in activity at 10 min, 0.30 pmol PO4/μg protein, but then returning to control values at 15 min. The 15-min animals had a CaM kinase II activity level of 0.25 pmol PO4/μg protein, 20-min animals had an activity of 0.26 pmol PO4/μg protein, 30-min animals had activity of 0.23 pmol PO4/μg protein, and 40-min animals had CaM kinase II activity of 0.19 pmol PO4/μg protein, a statistically significant inhibition of 30% (n = 3; p < 0.01, one-way ANOVA) when compared with control. The 70-min animals had activity of 0.17 pmol PO4/μg protein (n = 4; p < 0.01, one-way ANOVA).
In the hippocampus, a definite trend in activity is seen. The hippocampal CaM kinase II activity peaks ∼10 min after the first seizure activity and then drops off below control levels. This finding correlates with another finding in this laboratory in which 70 min after the first discrete seizure, a 25% inhibition occurs in the crude SPM subcellular kinase activity (46).
Immunodetection of CaM kinase II activity in subcellular SPM
Western blots were performed to analyze the enzyme level of CaM kinase II in the crude SPM fractions after specific time points after the first discrete seizure. Control, 10-, 15-, 20-, 30-, and 40-min P2 samples were subjected to Western analysis to determine the amount of CaM kinase II protein in the sample. For cortical control samples, relative densities were similar to those seen in homogenate, approximately equal to 0.32 (Fig. 5A and C). The 10-min animals had an average relative density of 0.35 (n = 4), an increase of 16%. In animals that had 15 min of seizure activity, relative densities were ∼0.38 (n = 3). This is an increase of 38%, which is a statistically significant increase.
At all other times beyond 15 min, an increase of between 16 and 58% was noted, corresponding to relative densities of 0.39 and 0.43, respectively, each of which was statistically significant. The 20-min animals had a relative density of 0.43 (n = 5), the 30-min animals had a relative density of 0.36 (n = 5), and the 40-min animals had a relative density of 0.38 (n = 4). The jump in relative densities from control to SE samples suggests the possibility of a translocation of the kinase. This has been suggested previously, but the specific time point at which this event occurs was unknown (46).
In the hippocampal fractions, similar results were observed. Average hippocampal control relative densities were approximately equal to 0.30 (Fig. 5B and D). Relative densities for the 10-min samples was ∼0.34, an increase of 0.04 or a 13% increase in protein level. As the Western and graph indicate, each other time point has a relative density of 0.33 and 0.39, corresponding to a 10% to 40% increase when compared with control values. Concurrent to the cortex, the data suggest that a significant translocation of CaM kinase II protein to the SPM is found in the hippocampus, and this could be an explanation of why a delayed inhibition of kinase activity is seen.
SE results in a decrease in specific activity of CaM kinase II in crude SPM
Specific activity was determined by dividing the enzyme activity (pmol PO4) by relative CaM kinase II immunoreactivity (OD) (Fig. 6). The cortical data show an initial, but statistically insignificant increase in activity at 10 min in relation to the control sample. Control values were 0.26 pmol PO4/OD. At 15 min, a 30% decrease in specific activity (0.18 pmol PO4/OD; n = 5) was found, compared with control activity. The same 30% decrease in activity was observed at 20, 40, and 70 min. In the hippocampus, similar specific activity was observed for control samples. Unlike the cortex, a progressive decrease occurs through 70 min. This indicates that despite the translocation, an overall decrease is found in synaptic CaM kinase II activity with increased SE duration.
This study used electrographic mapping of seizure progression in the Pilo-SE model. By mapping seizure progression in the Pilo-SE model, we demonstrated an immediate inhibition of total CaM kinase II activity that was coincident with SE onset, despite no overall decrease in CaM kinase II protein levels. One possible mechanism to explain this observation is that a posttranslational modification occurred early in the course of SE that resulted in a rapid decrease in kinase activity. The results also show a delayed inhibition in synaptic CaM kinase II activity. The synaptic kinase activity was maintained by translocation of additional enzyme to the SPM. These observations suggest that the brain may be attempting to preserve synaptic kinase function; however, as SE duration increases, the attempt to maintain kinase activity fails. The data support the hypothesis that alterations of CaM kinase II activity are involved in the early events of SE pathology.
At least two scenarios exist whereby modulation of CaM kinase II activity is involved in the series of events of SE induction. One possibility for the initial increase in synaptic CaM kinase II activity is an extension of a pharmacologic response. On a synaptic level, high-frequency stimulation has been shown induce LTP, which is associated with an increase in synaptic CaM kinase II activity (25,26,47–49), autophosphorylation level (50), and translocation of additional enzyme to the synaptic region (51–53). One hypothesis for this is that the increase in CaM kinase II causes an enhancement of glutamate receptors (AMPA and NMDA), which has been shown to increase synaptic strength. This increases the function of the circuit by increasing the signal-to-noise ratio, relative to neighboring neuronal circuits.
The current study shows that at least two events associated with LTP also are observed in the Pilo-SE model: an initial increase in total synaptic CaM kinase II activity as well as a translocation. However, unlike LTP, the observed increase in SPM CaM kinase II occurs over the whole structures of both the cortex and the hippocampus, not just in specific synapses. This suggests that, unlike the increase in efficacy for specific synaptic complexes, an increase is found in whole-brain synaptic efficacy. That is, the two to three discrete seizures may be equivalent to whole-brain tetanic stimulation, which should result in an increase in whole-brain synaptic CaM kinase II. Instead of increasing the signal-to-noise ratio, as observed in LTP, an increase in whole-brain synaptic efficacy would increase overall brain responsiveness in the form of seizure activity. It also is possible that the increase in synaptic efficacy increases neuronal responsiveness enough to overwhelm the mechanisms that normally terminate seizure activity. Thus the translocation and increase in CaM kinase II activity may be involved in the mechanisms that induce SE in this model.
Support for this theory is that a transient increase takes place in the synaptic CaM kinase II activity at 10 min after the first discrete seizure as well as a translocation in the crude SPM. Both of these observations occur coincident with SE onset. Additionally, animals that have discrete seizures, but no SE, do not experience a translocation, nor do they have an inhibition of CaM kinase II activity after 1 h of SE (36). It has been noted and is currently being studied that rats that have only one discrete seizure typically do not develop SE, whereas rats that have two to three discrete seizures within 10 min of each other typically progress into SE. Therefore future studies will characterize CaM kinase II modulation in animals that display discrete seizures but do not develop SE. This would require the ability to segregate nonresponder animals accurately (animals that display discrete seizure activity but do not develop SE) from rats that develop SE. Preliminary data from our laboratory suggest that a measurable difference exists between the discrete seizure phase for these two groups that can predict which rats will enter SE (Kumar et al., unpublished data).
Another scenario is that the inhibition of CaM kinase II activity may be a surrogate marker for pathology in the Pilo-SE model. Supporting this theory is that, in the homogenate isolated from SE animals, an immediate inhibition of CaM kinase II activity is found. However, in rats that have discrete seizures, but do not develop SE, no inhibition occurs (36). Whereas a delayed inhibition of CaM kinase II activity was found in the crude SPM, an increase in calcium-independent activity was not observed. In many learning and memory models, autophosphorylation of the Thr286 residue in CaM kinase II results in significant calcium-independent activity (54–56). Because no increase in calcium-independent activity was observed in the present study, it is likely that other mechanisms are involved in the translocation of CaM kinase II to synaptic structures. For instance, the translocation may be an attempt by the neuron to maintain or regain synaptic function during SE. However, despite the translocation of functional kinase, CaM kinase II activity is not maintained, membrane potential is not restored, and the newly translocated CaM kinase II does not terminate the SE event.
Conversely, another possibility is that the translocation of CaM kinase II to the SPM is an epiphenomenon. It has been shown that cellular mechanisms are in place that transport CaM kinase II to the synapse in multiple models (57–60). In the crude SPM, increased protein levels maintain CaM kinase II activity toward an exogenously applied substrate. However, activity toward endogenous substrates must be determined to confirm whether a functional translocation of CaM kinase II exists in SE. It is possible that the kinase does not reach the appropriate endogenous targets. This could be due to the newly translocated CaM kinase II not being able to gain access to synaptic targets for the kinase because of the inhibited CaM kinase II blocking access to the endogenous substrate. To test this theory, endogenous substrates must be studied to determine CaM kinase II incorporation.
A more likely possibility is that all of these mechanisms are involved in the observed effects of CaM kinase II. During the discrete seizure phase, the initial translocation of the CaM kinase II is in response to a whole-brain LTP-like event. However, as SE progresses, even the SPM kinase is inhibited. At this point, loss of kinase activity may actually be a mechanism for induction of epilepsy, a process termed epileptogenesis. Once the inhibition of CaM kinase II activity is induced, it has been shown to be long-lasting (32). In other models of chronic kinase inhibition, significant seizure activity is observed. Butler et al. (61) showed significant seizure activity in a chronic knockdown mouse model. In addition, knockdown of CaM kinase II α-subunit protein expression also results in seizure activity in hippocampal neuronal cultures (44), possibly because of loss of GABAergic modulation (46). Thus inhibition of CaM kinase II activity has significant implications for regulation of neuronal excitability.
Because the inhibition in total CaM kinase II activity was observed within 10 min, it is not probable that the SE-induced inhibition is due to a change in transcription levels. Additionally, the inhibition was not due to a decrease in protein level. These two observations suggest that the inhibition of CaM kinase II is due to a posttranslational modification. Additionally, the inhibition of CaM kinase II is not likely due to autophosphorylation at the threonine305 residue because no change in phosphatase activity has been reported in SE. Support for this theory comes from studies in cerebral ischemia in which addition of exogenous phosphatases did not reverse the inhibition of CaM kinase II activity. In addition, SE has been shown to increase calcineurin activity (62). Although calcineurin does not directly act on CaM kinase II, increasing calcineurin activity should remove inhibition of phosphatase 1 by inhibitor 1. The overall result would be increased phosphatase activity, which should reduce Thr305 phosphorylation. Future studies will be directed at the cellular changes due to the inhibition of CaM Kinase II as SE progresses.
The findings in this article characterize the temporal aspects of both the electrographic and neurochemical changes during the induction of SE in the rat pilocarpine model. A significant inhibition of CaM kinase II activity was observed coincident with the induction of SE in both cortical and hippocampal homogenates. Thus CaM kinase II activity provides a neuronal marker for the aggressive treatment of SE (7,8). Future studies will be to directed at the cellular changes due to the inhibition of CaM kinase II as SE progresses.
Acknowledgment: We thank Jonathan Kurz for his valuable insight and assistance and Matt Ryan for technical assistance with EEG review. This work was supported by NIH grants P50-NS25630 and RO1-NS39970 to S.B.C.