| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Departments of Comparative Medicine and 2Neurology and Neurological Sciences, Stanford University, Stanford, California
Submitted 19 December 2007; accepted in final form 4 March 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
). Depth electrode recordings have shown that seizures initiate in the mesial temporal lobe (Spencer and Spencer 1994; Wennberg et al. 2002). The hippocampal dentate gyrus is a prominent structure of the mesial temporal lobe and has been proposed to normally suppress seizure activity by acting as a gate or low-pass filter (Collins et al. 1983; Heinemann et al. 1992). In patients and models of temporal lobe epilepsy, the dentate gyrus displays chronic abnormalities including altered expression of ligand- (Brooks-Kayal et al. 1998) and voltage-gated channels (Bender et al. 2003), loss of some GABAergic interneurons (de Lanerolle et al. 1989; Obenaus et al. 1993), and synaptic reorganization (Nadler et al. 1980; Represa et al. 1987; Sutula et al. 1989). These and other persistent changes within and outside the dentate gyrus may contribute to temporal lobe epileptogenesis. However, epilepsy is a dynamic disease, and it remains unclear whether and how persistent changes generate transient and spontaneous seizures (Lopes da Silva and Pijn 1999; Schwartzkroin 1997). Insight might be obtained by evaluating activity of individual dentate granule cells before and during spontaneous seizures.
During spontaneous seizures in alumina-induced neocortical epileptic foci, changes in unit activity have been reported (Ishijima 1972; Schmidt et al. 1976; Sypert and Ward 1967; Wyler 1982). Hippocampal unit activity has been recorded and analyzed in animal models of temporal lobe epilepsy while kindling (Shi et al. 2007) and during latent (Sanchez et al. 2006), interictal (Liu et al. 2003), and postictal periods (Zhou et al. 2007), but not during spontaneous seizures. In patients with temporal lobe epilepsy, unit recordings have been obtained from the mesial temporal lobe during spontaneous seizures (Babb and Crandall 1976; Babb et al. 1987; Verzeano et al. 1971). Those studies report changes in firing of some neurons around seizure onset. However, limitations of the clinical setting make it difficult to precisely determine electrode locations and identify types of neurons recorded, and there are few samples of single units available for quantitative analysis.
Although it is generally accepted that seizures are generated by pathophysiological neuronal activity, firing rates of individual neurons are rarely recorded as spontaneous seizures initiate. To address this, we evaluated a rat model of temporal lobe epilepsy. Single unit activity was isolated by tetrode recording and cluster cutting techniques. Individual neurons were classified as granule cells based on established electrophysiological criteria and anatomical verification of recording sites. Comparisons across seizures and animals were facilitated by aligning firing rate data to systematically determined electrographic onset of seizure activity recorded in dentate gyrus. We asked whether firing rates of granule cells change before or during spontaneous seizures.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Stanford University Institutional Animal Care and Use Committee. As described previously, 24- to 35-day-old, male Sprague-Dawley rats (Harlan, Indianapolis, IN) were treated with pilocarpine (380 mg/kg, ip) 20 min after atropine methyl bromide (5 mg/kg, ip) (Buckmaster 2004). Diazepam (10 mg/kg, ip) was administered 2 h after onset of status epilepticus and repeated as needed. Rats that did not develop status epilepticus received diazepam 2 h after pilocarpine and were used as controls. Beginning 
10 days after pilocarpine treatment, rats were video-monitored for 2 wk (40 h/wk) for motor seizures of grade 3 or greater on Racine's scale (Racine 1972). Rats that experienced status epilepticus (n = 4) were observed to have multiple spontaneous seizures (minimum = 14) before they were implanted for chronic recording, which began 51–86 days after pilocarpine treatment. Control rats (n = 2) were not observed to have spontaneous seizures and were recorded beginning 33–47 days after pilocarpine treatment.
To implant microdrives, rats were sedated with diazepam (10 mg/kg, ip), anesthetized with isoflurane (1.5%), placed in a stereotaxic frame, maintained on a heating pad with feedback control, and prepared for aseptic surgery. A concentric, bipolar stimulating electrode (SNEX-100, Rhodes Medical Instruments, Tujunga, CA) was implanted in the angular bundle (7.6 mm behind and 4.6 mm right of bregma and 
3 mm below brain surface) to aid tetrode localization into the granule cell layer. A microdrive (Ray Harlan, Wayland, MA) containing four independently moveable tetrodes was located dorsal to dentate gyrus (4.6 mm behind and 2.6 mm right of bregma) and affixed to the skull with cranioplastic cement and jeweler's screws. Each tetrode consisted of four PAC-coated nichrome wires, 15 µm in diameter (California Fine Wire, Grover Beach, CA), twisted together and fused. Following surgery, rats received lactated Ringer (10 ml, sc), antibiotic (enrofloxacin, 10 mg/kg, sc), and analgesic (buprenorphine, 0.05 mg/kg, sc). An image of a similar implanted microdrive is shown in Fig. 1 of Gothard et al. (1996), and the same model used in this study can be seen on-line (Headstage-16, http://neuralynx.com/SingleProd.asp?ProdID=51).
Tetrode recording
Beginning 4 days after surgery, tetrodes were advanced over multiple days through the CA1 pyramidal cell layer—as noted by multiple, single units and polarity reversals of sharp waves (Buzsaki et al. 1990)—and 750 µm further until encountering a second layer of units in the granule cell layer. Tetrode placement was facilitated by examining responses to perforant path stimulation and optimized to record multiple, single units. To enhance stability, tetrodes were not advanced on recording days until after recordings were obtained. Recordings were obtained from epileptic rats continuously before, during, and after spontaneous seizures, and in control rats while they rested quietly. Signals were buffered with a lightweight headstage (HS-16, Neuralynx, Tucson, AZ) attached to the microdrive, amplified (Lynx-8 amplifiers, Neuralynx), digitized (Cheetah Data Acquisition, Neuralynx), and saved to disk for off-line analysis. One channel of each tetrode was filtered (0.1–1,800 Hz) to record EEG, whereas, in parallel, all four channels were filtered (600–6,000 Hz) and sampled at 30.3 kHz to record action potentials.
Histology
After unit recording experiments were completed, rats were killed with urethane (2.0 g/kg, ip) and perfused at 30 ml/min through the ascending aorta for 2 min with 0.9% NaCl, 5 min with 0.37% sodium sulfide solution, 1 min with 0.9% NaCl, and 30 min with 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). The sodium sulfide solution contained 1.2% Na2S–9H2O and 1.0% NaH2PO4. Brains were removed and stored in fixative at 4°C at least overnight and were equilibrated in 30% sucrose in 0.1 M PB. Left hippocampi were straightened, frozen, and sectioned perpendicular to the septotemporal axis on a sliding microtome set at 40 µm. To determine whether mossy fiber sprouting developed in epileptic rats, a 1-in-6 series of sections from the left hippocampus was processed for Timm staining. Sections were mounted on slides, dried, and developed for 45 min in 120 ml of 50% gum arabic, 20 ml of 2 M citrate buffer, 60 ml of 0.5 M hydroquinone, and 1 ml of 19% silver nitrate. Blocks of tissue from the right cerebral hemisphere, which contained tracks of stimulating and recording electrodes, were sectioned coronally on a sliding microtome set at 40 µm. Continuous series of sections were stained with 0.25% thionin.
Identification of electrographic seizure onset in the dentate gyrus
In rodent models of temporal lobe epilepsy, the substrate for seizure generation is distributed over several structures, including the hippocampus, amygdala, piriform cortex, and entorhinal cortex (Bertram 1997). In this study, a systematic procedure was developed to identify the locally recorded electrographic seizure onset in the dentate gyrus at a resolution of seconds. First, EEG recordings (0.1–1,800 Hz) were evaluated subjectively to bracket a window of seizure onset by finding the latest time of normal EEG activity (latest-normal) and the earliest time of clear seizure activity (clear-seizure). Within the bracketed period, seizure onset was estimated "by eye" as the first appearance of persistent, rhythmic activity. This was done independently by each author, and the earlier of the two onsets used. Next, three objective measures of seizure onset were computed. The slow wave (Bragin et al. 2005) was identified by low-pass filtering EEG traces (cut-off, 1 Hz) and identifying the peak associated with the largest voltage change within the seizure onset window. Spectrograms were computed using discrete, prolate, spheroids (pmtm function, Matlab, Mathworks, Nawtick, MA), which produce estimates of spectral power optimized for discrete time windows (Percival and Walden 1993). Two frequency windows were examined. The peak of summed spectral power for low frequencies (20–200 Hz) looked for changes in the gamma frequency band. The peak of the summed spectral power for high frequencies (200–600 Hz) looked for changes in frequency ranges previously associated with seizure onset (Bragin et al. 2005). In both cases, summed spectral power was smoothed with a low-pass filter (cut-off, 1 Hz). The earliest of these four measurements—by eye, slow wave, 20–200 Hz power, or 200–600 Hz power—provided a conservative (i.e., as early as possible) estimate of dentate gyrus electrographic seizure onset time. In this study, seizure onset refers to this time point.
Cluster cutting
To avoid interference from possible seizure-related artifacts, cluster boundaries for each neuron were established using only interictal data obtained during the period preceding seizure onset. Epochs beginning 10 min before behavioral seizure onset and ending at seizure onset (i.e., not including data recorded during the seizure) were used to identify single units by grouping similar waveforms (Harris et al. 2000; McNaughton et al. 1983a) based on four features: peak amplitude, sum of the squared amplitude (i.e., energy), and the first two principal components. An automated, cluster identification program was used first (KlustaKwik-1.6, K. D. Harris et al., Rutgers University) followed by manual selection of final clusters (MClust-3.0.3, A. D. Redish et al., University of Minnesota). Cluster boundaries, which were identified from interictal data, were applied to all data obtained during that recording session, including data obtained during seizures. Action potentials of individual neurons were clearly separable during the early part of seizures, but became more difficult to identify as a seizure progressed. Several factors might have contributed to this observation including synchrony of unit activity with electrographic ictal spikes (Wyler 1982), synchrony with other units (Wyler et al. 1982), or increased background noise (Sypert and Ward 1967). Therefore we analyzed only the first minute following seizure onset.
When matching across sequential seizures, clusters isolated during an epoch before the first seizure were applied to data obtained before the second seizure and the best match identified by the greatest overlap of spikes with cluster boundaries. This process was repeated in reverse; i.e., by applying clusters isolated during an epoch before the second seizure to data obtained before the first. If two clusters chose one another as best matches, average waveforms from the first and second seizures were compared to determine whether they showed a high degree of similarity. If so, they were treated as describing the same neuron during the two seizures.
Statistical analysis
Perievent time histograms (PETHs) were computed by aligning data to seizure onsets and binning. Seizure onset occurred at the boundary between two bins. Mean duration between latest-normal EEG and clear-seizure activity represents the maximum uncertainty of seizure onset time. Bin widths of PETHs were set at 30 s, which is roughly two times that value.
Normality within groups was established using a Kolmogorov-Smirnov test. Bonferroni correction was applied to the repeated-measures t-test and ANOVAs. A P value of 0.05 was chosen as the cut-off for significance in all statistical tests. SE is reported in all cases.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All rats displayed a stimulating electrode track ending in the angular bundle (data not shown) and one or two tetrode tracks ending in the granule cell layer of the dentate gyrus (Fig. 1, A and B). Hilar neuron loss and aberrant mossy fiber sprouting were evident in all epileptic rats but not controls (Fig. 1).
|
In each epileptic rat, recordings were obtained 4–12 h/d until five spontaneous seizures were recorded in which no seizure activity was observed for the preceding hour. Both convulsive and nonconvulsive seizures were included (Table 1). The earliest of one subjective and three objective electrographic seizure onset markers was identified (Fig. 2). Each of the four seizure onset markers gave the earliest estimate in at least one case, and the by-eye estimate was earliest most often (Table 1). The average time between the earliest and latest seizure marker was <4 s and the maximum was <15 s. Thus variability in seizure onset markers was small relative to the bin size used to analyze unit data (30 s). Duration of the seizure onset window (between points of latest-normal EEG and clear-electrographic seizure activity) was 16.8 ± 1.4 s (range, 6.0–33.3 s). Latency between the beginning of the seizure onset window and the earliest seizure onset marker was 4.8 ± 0.6 s (range, 0.2–11.6 s).
|
|
10–20 s. This quiet period was associated with low-amplitude [200 µV root mean square (r.m.s.), compared with 400 µV during baseline periods], high-frequency (80–100 Hz) oscillations, similar to the low-voltage fast-type seizure onsets commonly observed in epileptic patients (Bragin et al. 2005). This was followed by a tonic phase that lasted another 20–30 s, containing a series of positive spikes occurring at 20–40 Hz. This tonic period was normally followed by a clonic phase that lasted 20–40 s consisting of low-frequency bursts (1–2 Hz) of high-frequency (30–50 Hz), large-amplitude positive spikes (3 mV or larger). Burst duration increased and frequency decreased toward the end of seizures. Seizures were followed by a relatively silent (<50 µV r.m.s.), postictal depression phase that lasted up to several minutes. This stereotyped pattern was observed across all levels of behavioral seizure severity, including nonconvulsive seizures. Electrographic seizures lasted 75.6 ± 4.0 s (range, 50.4–110.5 s). Behavioral seizure onsets consistently appeared to follow electrographic onsets. Seizure-related unit activity
Action potential waveforms diverged from background noise (20 µV r.m.s., 600–6,000 Hz) with an 80 µV amplitude, brief (width at half-maximum amplitude 70 µs) negative-going spike, and a subsequent positivity of longer duration (Fig. 2E). Before and during the earliest phases of seizures, waveform morphology of a given unit showed little variation (Fig. 2F). Following seizures, during postictal depression, firing frequencies were very low. However, low postictal firing rates are not likely attributable to changes in tetrode position, because within a few minutes waveforms returned.
Tetrodes recorded the same action potentials on four closely spaced wires, which aided identification and separation of waveforms generated by different neurons (McNaughton et al. 1983b). An example that shows part of the process of ascribing action potentials to individual neurons is shown in Fig. 3. For each recorded action potential, four features (peak amplitude, energy, and 1st 2 principal components) were computed for each of four tetrode channels, allowing all possible pairs of features to be viewed as scatter plots. Figure 3A shows two such scatter plots. Action potentials generated by the same neuron produce a cluster of nearby points that are separated by a combination of computer algorithm and investigator classification (see METHODS). Multiple features are evaluated, because they vary in capacity to discern individual neurons. For example, plotting energy of channel 2 versus 3 shows a distinct cluster of action potentials generated by the blue neuron in Fig. 3, whereas plotting the first versus second principal component of channel 1 more clearly separates action potentials generated by the red and black neurons.
|
; Ranck 1973). Mean firing rates of remaining individual units during the baseline period ranged from 0.019 to 1.83 Hz. Of the 123 units included, 109 had mean firing rates <1 Hz, and 14 had mean firing rates between 1 and 2 Hz. All displayed irregular firing as indicated by multimodal and/or skewed interspike interval distributions or very low firing rates. No differences were observed between control or epileptic rats with respect to mean firing rate, peak-to-valley spike width, or peak-to-valley spike amplitude (Fig. 4). Based on previously established criteria, including low baseline firing frequency (Jung and McNaughton 1993; Ranck 1973), all units were classified as putative granule cells. However, we cannot exclude the possibility that low-frequency firing interneurons may have been included.
|
|
Firing rates were normalized within each cell to its mean firing rate during the baseline period. Firing rates for preictal-increase (-decrease) patterns appeared to increase (decrease) from baseline abruptly and persist near the altered rate at least until seizure onset (Fig. 6). The timing of preictal shifts in firing rate ranged from 4.0 to 0.5 min before seizure onset. Firing rates of most ictal-increase cells increased markedly during the first 30 s of the seizure. A minority of ictal-increase cells delayed increased firing until the second 30 s of the seizure.
|
|
2 test). Persistent group membership exceeded that expected by chance. In addition, more cells than expected shifted between preictal-increase and ictal-increase groups. Examples of individual cells with preserved and different grouped membership across two sequential seizures are shown (Fig. 8, B and C).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Identification of electrographic seizure onset in dentate gyrus
Many techniques exist to aid spatial localization of seizure foci in patients (Fisch and Spehlmann 1999), but little emphasis has been placed on identifying reliable, temporal markers within EEG recordings around which seizure-related events may be aligned with a resolution of seconds (but see Meeren et al. 2002). In this study, sites of seizure onset are unclear because rodent models of temporal lobe epilepsy display variability in this regard (Bertram 1997), and we recorded only in dentate gyrus. However, by using four electrographic markers and choosing the earliest in each seizure, we conservatively estimated local seizure onset times within dentate gyrus. Results showed little effect of using multiple seizure onset markers, because variability in timing of seizure onset markers was small relative to the bin size of unit data. Realigning unit data to a single seizure marker (e.g., slow wave), instead of the earliest, did not significantly affect seizure-related unit firing patterns.
Seizure onset times were identified from low-frequency data (<2 kHz) and not from action potential data. Nevertheless, abrupt increases in firing rates across seizures and animals became apparent after unit activity was aligned by markers identified from low-frequency data, suggesting estimated onset times correlated with underlying ictal transition mechanisms reflected by sudden increases in unit firing.
Heterogeneous seizure-related activity patterns of granule cells
At least four different patterns of seizure-related unit activity were observed. What might account for the heterogeneity? One possibility is that recordings were obtained from different neuronal types—granule cells and various interneuron classes. This is unlikely because all units were recorded in the granule cell layer, displayed similar waveforms, and had low baseline firing frequencies (Jung and McNaughton 1993; Ranck 1973). What then might account for heterogeneity among granule cells? If groups of granule cells with similar seizure-related activity patterns were physically clustered, one might expect a tetrode, which has a limited recording range, to detect units with similar activity patterns. However, every possible combination of seizure-related activity groups was recorded simultaneously at least twice by individual tetrodes (Table 1), suggesting granule cells with different activity patterns were in close proximity and intermixed. In normal rats, granule cells display heterogeneous activity patterns in the form of place cells. In contrast to periods of quiet rest when many granule cells discharge at baseline rates, during exploration only a subset produce place-related responses, whereas others remain relatively silent (Jung and McNaughton 1993). This phenomenon is thought to be an emergent property of the synaptic connection matrix of granule cells. In one environment, place-related responses are reliably observed for a given granule cell, even though that cell is assumed to have anatomical and electrophysiological characteristics similar to its neighboring granule cells that are much less active in that environment. By analogy, heterogeneous seizure-related unit activity patterns might be an emergent network property among the population of similar granule cells. Actual granule cells, however, do show anatomical and functional variability. For example, dendritic structure varies according to soma location (Claiborne et al. 1990). Granule cells continue to be generated in adult animals (Kaplan and Hinds 1977), and new cells display distinct morphological (Zhao et al. 2006) and electrophysiological characteristics (Schmidt-Hieber et al. 2004). After epileptogenic treatments, neurogenesis increases (Parent et al. 1997), and subsets of granule cells develop basal dendrites (Spigelman et al. 1998) and form aberrant synapses (Represa et al. 1993). At this time, however, it is unclear which, if any, parameters correlate with seizure-related activity patterns.
Ictal-increase pattern
Previous studies found that some cells in the mesial temporal lobe increase firing at electrographic seizure onset in patients with temporal lobe epilepsy (Babb and Crandall 1976; Babb et al. 1987). Similarly, firing rates of ictal-increase cells rose abruptly when electrographic seizure activity was first detected in dentate gyrus of epileptic rats. During seizures, mean normalized firing rates of ictal-increase cells were >4 times higher than baseline levels. False-positive detections of action potentials are an unlikely explanation, because action potentials were attributed to individual neurons by clustering similar waveforms (McNaughton et al. 1983b). Clusters are boundaries in 16-dimensional feature space defined by the number of recording channels and waveform features. Using four channels per tetrode and four waveform features (peak amplitude, energy, and 1st and 2nd principal components) provides power to separate clusters and reduces probability of incorrectly assigning detected events (Henze et al. 2000). It is unlikely that artifacts or aberrant waveforms during seizures would fit clusters established from data obtained before seizures and mistakenly increase firing rates. During seizures, however, false negatives could have occurred, because of overlapping action potentials or changes in extracellular milieu that affect action potential waveforms. Therefore we may have underestimated the magnitude of increased firing during seizures.
Unchanged pattern
During spontaneous seizures in patients with temporal lobe epilepsy, firing frequencies change in only a fraction of neurons in an epileptic focus, whereas activity of some cells remains unchanged (Babb and Crandall 1976; Babb et al. 1987). Similarly, firing rates of 19% of granule cells recorded in epileptic rats did not change significantly before or during spontaneous seizures. This is remarkable considering that severe, electrographic seizure activity was recorded simultaneously from the same tetrodes. Five of seven cells that displayed unchanged firing patterns during a first seizure, subsequently showed preictal changes in firing rate during a second seizure (Fig. 8). Therefore neurons that display unchanged firing patterns do not seem to be damaged or disconnected from other neurons.
Preictal-decrease pattern
Many granule cells in epileptic rats showed a significant decrease in firing frequency before seizure onset. Analysis of multiunit activity recorded in amygdala during a spontaneous seizure in a patient suggested that activity of some units may decrease before seizure onset (Verzeano et al. 1971). One possible mechanism is that preictal-increase cells activate local inhibitory feedback circuits, which in turn inhibit preictal-decrease cells. Consistent with this hypothesis, mean normalized preictal-increase activity rose before preictal-decrease activity declined. Several preictal-decrease neurons displayed bimodal firing patterns, first increasing and then decreasing (Figs. 6 and 8C), suggesting competing factors of initial excitation followed by inhibition. Another possibility is that preictal-decreases are linked to behavioral state, not an impending seizure. When rats move or groom, granule cell firing rates usually decrease (Jung and McNaughton 1993). However, this is an unlikely explanation, because 21 preictal-decrease cells were recorded during seizures that began while rats were resting quietly and not grooming or moving (Table 1). It is unlikely that preictal-decrease activity is an artifact of missed action potentials, because many preictal-decrease units displayed normal firing rates during baseline periods and high firing rates after seizure onset, suggesting good recording conditions throughout the baseline, preictal, and seizure periods. Increased firing rates during seizures suggest that after seizure onset, possible sources of preictal inhibition weakened or were overcome by increased excitatory input.
It has been proposed that, in epileptic pilocarpine-treated rats with mossy fiber sprouting, granule cells are hyperinhibited and relatively quiet during spontaneous seizures (Harvey and Sloviter 2005). In contrast, granule cells display early and extensive Fos expression after spontaneous seizures in epileptic pilocarpine-treated mice (Peng and Houser 2005). In this study, 19% of recorded granule cells showed unchanged activity before and during seizures. Another 18% were preictal-decrease cells whose firing rate did not increase during seizures. However, a substantial fraction of recorded cells (63%) became active before and/or during spontaneous seizures. This percentage may underestimate the level of granule cell activity during seizures, because only the first minute after seizure onset was analyzed and some seizures lasted >1 min.
Preictal-increase pattern
The largest fraction of recorded granule cells (34%) displayed increased preictal activity. As described above, increased activity is not likely attributable to artifactual detections. Furthermore, many sources of electrographic artifacts can be excluded during the preictal period. Rats usually were resting quietly before seizure onset (Table 1), scratching and chewing were not observed during preictal periods included in our sample, and potential movement artifacts caused by convulsions were not a factor during preictal periods.
Significantly increased preictal activity begins 4 min before seizure onset recorded locally in dentate gyrus. Based on this finding, we propose the following hypotheses: First, although electrographic seizures may originate elsewhere, we believe they are unlikely to begin earlier than 4 min before electrographic onset measured in dentate gyrus. Second, in other brain regions, preictal changes in neuronal firing rate might occur before locally recorded electrographic seizure activity. Third, the gradual buildup of preictal activity may reflect reverberation and reinforcement across synaptically linked structures of the limbic system, rather than sequential, sudden "turning on" of one region followed by another. We have observed a similar, gradual buildup of preictal activity by CA3 pyramidal cells (unpublished observations). However, additional experiments are necessary to test these predictions.
The finding of increased preictal granule cell discharges may be novel, although analysis of multiunit activity recorded in amygdala during a spontaneous seizure in a patient suggested that some units may show increased preictal activity (Verzeano et al. 1971). It is tempting to speculate that preictal unit activity might be related to prodromal symptoms some patients consistently experience several minutes before seizure onset (Rajna et al. 1997). Quantitative analyses of EEG recordings detect significant preictal changes in patients with temporal lobe epilepsy (Iasemidis et al. 2005; Le Van Quyen et al. 2000, 2001; Lehnertz and Elger 1998; Litt et al. 2001; Martinerie et al. 1998; Navarro et al. 2005). The activity of preictal-increase cells may be related to some of these preictal EEG changes.
In this study, unit and EEG recordings were obtained only from dorsal dentate gyrus in one cerebral hemisphere. Sites of seizure onset may lie outside the recorded region. If other brain regions display seizure-related changes in firing rates before those observed in granule cells, seizure prediction horizons could be extended even further than suggested by this study. Whether they are within the initiating epileptic focus or not, preictal cells may provide opportunities for predicting seizures. Unpredictability of seizures puts patients at increased risk for accidents (Buck et al. 1997), causes anxiety, and degrades quality of life (Murray 1993). Although recent studies have shown the use of recording seizure-related events at frequencies 
500 Hz (Staba et al. 2004), our findings show additional, potentially useful information in even higher frequency ranges (6,000 Hz) that include action potential activity of individual neurons. Future studies will test whether preictal changes in firing rate are sensitive and specific predictors of spontaneous seizures.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: P. Buckmaster, Edwards Bldg. R321, 300 Pasteur Dr., Dept. of Comparative Medicine, Stanford Univ., Stanford, CA 94305-5342 (E-mail: psb{at}stanford.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Babb TL, Wilson CL, Isokawa-Akesson M. Firing patterns of human limbic neurons during stereoencephalography (SEEG) and clinical temporal lobe seizures. Electroencephalogr Clin Neurophysiol 66: 467–482, 1987.[CrossRef][Web of Science][Medline]
Bender RA, Soleymani SV, Brewster AL, Nguyen ST, Beck H, Mathern GW, Baram TZ. Enhanced expression of a specific hyperpolarization-activated cyclic nucleotide-gated cation channel (HCN) in surviving dentate gyrus granule cells of human and experimental epileptic hippocampus. J Neurosci 23: 6826–6836, 2003.
Bertram E. Functional anatomy of spontaneous seizures in a rat model of limbic epilepsy. Epilepsia 38: 95–105, 1997.[CrossRef][Web of Science][Medline]
Bragin A, Wilson CL, Fields T, Fried I, Engel J Jr. Analysis of seizure onset on the basis of wideband EEG recordings. Epilepsia 46: 59–63, 2005.[Web of Science][Medline]
Brooks-Kayal AR, Shumate MD, Jin H, Rikhter TY, Coulter DA. Selective changes in single cell GABA(A) receptor subunit expression and function in temporal lobe epilepsy. Nat Med 4: 1166–1172, 1998.[CrossRef][Web of Science][Medline]
Buck D, Baker GA, Jacoby A, Smith DF, Chadwick DW. Patients’ experiences of injury as a result of epilepsy. Epilepsia 38: 439–444, 1997.[CrossRef][Web of Science][Medline]
Buckmaster PS. Laboratory animal models of temporal lobe epilepsy. Comp Med 54: 473–485, 2004.[Web of Science][Medline]
Buzsaki G, Chen LS, Gage FH. Spatial organization of physiological activity in the hippocampal region: relevance to memory formation. Prog Brain Res 83: 257–268, 1990.[Web of Science][Medline]
Claiborne BJ, Amaral DG, Cowan WM. Quantitative, three-dimensional analysis of granule cell dendrites in the rat dentate gyrus. J Comp Neurol 302: 206–219, 1990.[CrossRef][Web of Science][Medline]
Collins RC, Tearse RG, Lothman EW. Functional anatomy of limbic seizures: focal discharges from medial entorhinal cortex in rat. Brain Res 280: 25–40, 1983.[CrossRef][Web of Science][Medline]
de Lanerolle NC, Kim JH, Robbins RJ, Spencer DD. Hippocampal interneuron loss and plasticity in human temporal lobe epilepsy. Brain Res 495: 387–395, 1989.[CrossRef][Web of Science][Medline]
Engel J Jr, Williamson PD, Wieser HG. Mesial temporal lobe epilepsy. In: Epilepsy: A Comprehensive Textbook, edited by Engel J Jr, Pedley TA. Philadelphia, PA: Lippincott-Raven, 1997, p. 2417–2426.
Fisch BJ, Spehlmann R. Fisch and Spehlmann's EEG Primer: Basic Principles of Digital and Analog EEG. New York: Elsevier, 1999.
Gothard KM, Skaggs WE, Moore KM, McNaughton BL. Binding of hippocampal CA1 neural activity to multiple reference frames in a landmark-based navigation task. J Neurosci 16: 823–835, 1996.
Harris KD, Henze DA, Csicsvari J, Hirase H, Buzsaki G. Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements. J Neurophysiol 84: 401–414, 2000.
Harvey BD, Sloviter RS. Hippocampal granule cell activity and c-Fos expression during spontaneous seizures in awake, chronically epileptic, pilocarpine-treated rats: implications for hippocampal epileptogenesis. J Comp Neurol 488: 442–463, 2005.[CrossRef][Web of Science][Medline]
Heinemann U, Beck H, Dreier JP, Ficker E, Stabel J, Zhang CL. The dentate gyrus as a regulated gate for the propagation of epileptiform activity. Epilepsy Res Suppl 7: 273–280, 1992.[Medline]
Henze DA, Borhegyi Z, Csicsvari J, Mamiya A, Harris KD, Buzsaki G. Intracellular features predicted by extracellular recordings in the hippocampus in vivo. J Neurophysiol 84: 390–400, 2000.
Iasemidis LD, Shiau DS, Pardalos PM, Chaovalitwongse W, Narayanan K, Prasad A, Tsakalis K, Carney PR, Sackellares JC. Long-term prospective on-line real-time seizure prediction. Clin Neurophysiol 116: 532–544, 2005.[CrossRef][Web of Science][Medline]
Ishijima B. Unitary analysis of epileptic activity in acute and chronic foci and related cortex of cat and monkey. Epilepsia 13: 561–581, 1972.[Medline]
Jung MW, McNaughton BL. Spatial selectivity of unit activity in the hippocampal granular layer. Hippocampus 3: 165–182, 1993.[CrossRef][Web of Science][Medline]
Kaplan MS, Hinds JW. Neurogenesis in the adult rat: electron microscopic analysis of light radioautographs. Science 197: 1092–1094, 1977.
Le Van Quyen M, Adam C, Martinerie J, Baulac M, Clemenceau S, Varela F. Spatio-temporal characterizations of non-linear changes in intracranial activities prior to human temporal lobe seizures. Eur J Neurosci 12: 2124–2134, 2000.[CrossRef][Web of Science][Medline]
Le Van Quyen M, Martinerie J, Navarro V, Boon P, D'Have M, Adam C, Renault B, Varela F, Baulac M. Anticipation of epileptic seizures from standard EEG recordings. Lancet 357: 183–188, 2001.[CrossRef][Web of Science][Medline]
Lehnertz K, Elger CE. Can epileptic seizures be predicted? Evidence from nonlinear time series analysis of brain activity. Phys Rev Lett 80: 5019–5022, 1998.[CrossRef][Web of Science]
Litt B, Esteller R, Echauz J, D'Alessandro M, Shor R, Henry T, Pennell P, Epstein C, Bakay R, Dichter M, Vachtsevanos G. Epileptic seizures may begin hours in advance of clinical onset: a report of five patients. Neuron 30: 51–64, 2001.[CrossRef][Web of Science][Medline]
Liu X, Muller RU, Huang LT, Kubie JL, Rotenberg A, Rivard B, Cilio MR, Holmes GL. Seizure-induced changes in place cell physiology: relationship to spatial memory. J Neurosci 23: 11505–11515, 2003.
Lopes da Silva FH, Pijn JP. Epilepsy as a dynamic disease of brain systems. Adv Neurol 81: 97–104, 1999.[Medline]
Martinerie J, Adam C, Le Van Quyen M, Baulac M, Clemenceau S, Renault B, Varela FJ. Epileptic seizures can be anticipated by non-linear analysis. Nat Med 4: 1173–1176, 1998.[CrossRef][Web of Science][Medline]
McNaughton BL, Barnes CA, O'Keefe J. The contributions of position, direction, and velocity to single unit activity in the hippocampus of freely-moving rats. Exp Brain Res 52: 41–49, 1983a.[Web of Science][Medline]
McNaughton BL, O'Keefe J, Barnes CA. The stereotrode: a new technique for simultaneous isolation of several single units in the central nervous system from multiple unit records. J Neurosci Methods 8: 391–397, 1983b.[CrossRef][Web of Science][Medline]
Meeren HK, Pijn JP, Van Luijtelaar EL, Coenen AM, Lopes da Silva FH. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci 22: 1480–1495, 2002.
Murray J. Coping with the uncertainty of uncontrolled epilepsy. Seizure 2: 167–178, 1993.[CrossRef][Web of Science][Medline]
Nadler JV, Perry BW, Cotman CW. Selective reinnervation of hippocampal area CA1 and the fascia dentata after destruction of CA3-CA4 afferents with kainic acid. Brain Res 182: 1–9, 1980.[CrossRef][Web of Science][Medline]
Navarro V, Martinerie J, Le Van Quyen M, Baulac M, Dubeau F, Gotman J. Seizure anticipation: do mathematical measures correlate with video-EEG evaluation? Epilepsia 46: 385–396, 2005.[CrossRef][Web of Science][Medline]
Obenaus A, Esclapez M, Houser CR. Loss of glutamate decarboxylase mRNA-containing neurons in the rat dentate gyrus following pilocarpine-induced seizures. J Neurosci 13: 4470–4485, 1993.[Abstract]
Parent JM, Yu TW, Leibowitz RT, Geschwind DH, Sloviter RS, Lowenstein DH. Dentate granule cell neurogenesis is increased by seizures and contributes to aberrant network reorganization in the adult rat hippocampus. J Neurosci 17: 3727–3738, 1997.
Percival DB, Walden AT. Spectral Analysis for Physical Applications. Cambridge, UK: Cambridge University Press, 1993.
Peng Z, Houser CR. Temporal patterns of fos expression in the dentate gyrus after spontaneous seizures in a mouse model of temporal lobe epilepsy. J Neurosci 25: 7210–7220, 2005.
Racine RJ. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr Clin Neurophysiol 32: 281–294, 1972.[CrossRef][Web of Science][Medline]
Rajna P, Clemens B, Csibri E, Dobos E, Geregely A, Gottschal M, Gyorgy I, Horvath A, Horvath F, Mezofi L, Velkey I, Veres J, Wagner E. Hungarian multicentre epidemiologic study of the warning and initial symptoms (prodrome, aura) of epileptic seizures. Seizure 6: 361–368, 1997.[CrossRef][Web of Science][Medline]
Ranck JB Jr. Studies on single neurons in dorsal hippocampal formation and septum in unrestrained rats. I. Behavioral correlates and firing repertoires. Exp Neurol 41: 461–531, 1973.[CrossRef][Medline]
Represa A, Jorquera I, Le Gal La Salle G, Ben-Ari Y. Epilepsy induced collateral sprouting of hippocampal mossy fibers: does it induce the development of ectopic synapses with granule cell dendrites? Hippocampus 3: 257–268, 1993.[CrossRef][Web of Science][Medline]
Represa A, Tremblay E, Ben-Ari Y. Kainate binding sites in the hippocampal mossy fibers: localization and plasticity. Neuroscience 20: 739–748, 1987.[CrossRef][Web of Science][Medline]
Sanchez JC, Mareci TH, Norman WM, Principe JC, Ditto WL, Carney PR. Evolving into epilepsy: multiscale electrophysiological analysis and imaging in an animal model. Exp Neurol 198: 31–47, 2006.[CrossRef][Web of Science][Medline]
Schmidt EM, Mutsuga N, McIntosh JS. Chronic recording of neurons in epileptogenic foci of monkey during seizures. Exp Neurol 52: 459–466, 1976.[CrossRef][Web of Science][Medline]
Schmidt-Hieber C, Jonas P, Bischofberger J. Enhanced synaptic plasticity in newly generated granule cells of the adult hippocampus. Nature 429: 184–187, 2004.[CrossRef][Medline]
Schwartzkroin PA. Origins of the epileptic state. Epilepsia 38: 853–858, 1997.[CrossRef][Web of Science][Medline]
Shi LH, Luo F, Woodward DJ, McIntyre DC, Chang JY. Temporal sequence of ictal discharges propagation in the corticolimbic basal ganglia system during amygdala kindled seizures in freely moving rats. Epilepsy Res 73: 85–97, 2007.[CrossRef][Web of Science][Medline]
Spencer SS, Spencer DD. Entorhinal-hippocampal interactions in medial temporal lobe epilepsy. Epilepsia 35: 721–727, 1994.[CrossRef][Web of Science][Medline]
Spigelman I, Yan XX, Obenaus A, Lee EY, Wasterlain CG, Ribak CE. Dentate granule cells form novel basal dendrites in a rat model of temporal lobe epilepsy. Neuroscience 86: 109–120, 1998.[CrossRef][Web of Science][Medline]
Staba RJ, Wilson CL, Bragin A, Jhung D, Fried I, Engel J Jr. High-frequency oscillations recorded in human medial temporal lobe during sleep. Ann Neurol 56: 108–115, 2004.[CrossRef][Web of Science][Medline]
Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol 26: 321–330, 1989.[CrossRef][Web of Science][Medline]
Sypert GW, Ward AA Jr. The hyperexcitable neuron: microelectrode studies of the chronic epileptic focus in the intact, awake monkey. Exp Neurol 19: 104–114, 1967.[CrossRef][Web of Science][Medline]
Verzeano M, Crandall PH, Dymond A. Neuronal activity of the amygdala in patients with psychomotor epilepsy. Neuropsychologia 9: 331–344, 1971.[CrossRef][Web of Science][Medline]
Wennberg R, Arruda F, Quesney LF, Olivier A. Preeminence of extrahippocampal structures in the generation of mesial temporal seizures: evidence from human depth electrode recordings. Epilepsia 43: 716–726, 2002.[CrossRef][Web of Science][Medline]
Wyler AR. Neuronal activity during seizures in monkeys. Exp Neurol 76: 574–585, 1982.[CrossRef][Web of Science][Medline]
Wyler AR, Ojemann GA, Ward AA Jr. Neurons in human epileptic cortex: correlation between unit and EEG activity. Ann Neurol 11: 301–308, 1982.[CrossRef][Web of Science][Medline]
Zhao C, Teng EM, Summers RG Jr, Ming GL, Gage FH. Distinct morphological stages of dentate granule neuron maturation in the adult mouse hippocampus. J Neurosci 26: 3–11, 2006.
Zhou JL, Shatskikh TN, Liu X, Holmes GL. Impaired single cell firing and long-term potentiation parallels memory impairment following recurrent seizures. Eur J Neurosci 25: 3667–3677, 2007.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  W. Zhang and P. S. Buckmaster Dysfunction of the Dentate Basket Cell Circuit in a Rat Model of Temporal Lobe Epilepsy J. Neurosci., June 17, 2009; 29(24): 7846 - 7856. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Queiroz, J. A. Gorter, F. H. Lopes da Silva, and W. J. Wadman Dynamics of Evoked Local Field Potentials in the Hippocampus of Epileptic Rats With Spontaneous Seizures J Neurophysiol, March 1, 2009; 101(3): 1588 - 1597. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Williams, A. M. White, S. Clark, D. J. Ferraro, W. Swiercz, K. J. Staley, and F. E. Dudek Development of Spontaneous Recurrent Seizures after Kainate-Induced Status Epilepticus J. Neurosci., February 18, 2009; 29(7): 2103 - 2112. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
