SEARCH

SEARCH BY CITATION

Keywords:

  • Mesial temporal lobe epilepsy;
  • Kainate rats;
  • Neuronal mechanisms;
  • Fast ripples

Abstract

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

Summary:  Many different types of epileptic seizures and epileptic syndromes exist. The process of epileptogenesis and the progressive nature of epilepsy, however, can most easily be investigated in the acquired epilepsies, in which a brain insult presumably gives rise to changes in neuronal systems that ultimately become capable of generating spontaneous ictal events. Invasive in vivo and in vitro research can be carried out in patients with acquired epileptogenic lesions in the course of epilepsy surgery; however, such studies are possible only for those epileptic conditions that can be treated surgically, and can be used only to examine an end stage of the epileptogenic process. Consequently, experimental animal models of human epileptic conditions are still required to study mechanisms by which specific cerebral insults initiate the epileptogenic process and the progression of an epileptic disturbance. Most current parallel human/animal invasive research has been focused on temporal lobe epilepsy, and particularly that form associated with hippocampal sclerosis, the most common human epileptogenic lesion. Studies indicate that epileptogenesis in this condition is initiated by specific types of cell loss and neuronal reorganization, which results not only in enhanced excitation, but also in enhanced inhibition, predisposing to hypersynchronization. Even within this single, well-studied epileptic disorder, evidence is found for more than one type of ictal onset, and individual seizures can demonstrate a transition from one ictal mechanism to another. Recent in vivo and in vitro parallel, reiterative investigations in patients with mesial temporal lobe epilepsy, and in rats with intrahippocampal kainate-induced hippocampal seizures, have revealed the presence of interictal epileptiform events, termed “fast ripples,” which appear to be unique in tissue capable of generating spontaneous seizures. Pursuit of the fundamental mechanisms underlying these abnormalities should elucidate the neurobiologic basis of epileptogenicity in this disorder. Furthermore, if these events are markers for epileptogenicity, they may have clinical value for diagnosis and pharmacologic, as well as surgical, treatment. Further research is needed to determine if these observations are relevant to other types of epilepsies.

From clinical observations, we know that many different types of epileptic seizures and many epileptic syndromes do not share a common pathogenesis (1,2). Advances in neurodiagnostics and genetics have helped us in recent years to identify and describe more accurately a wide variety of epileptic disorders, some of which are benign and others of which are severe. The benign syndromes are, for the most part, age related and inherited, whereas the severe syndromes are mostly acquired (3). The fact that an acquired cerebral insult can induce epileptogenic alterations in neuronal structure and function, which over time develops a capacity to generate spontaneous seizures, indicates that this is a progressive process. It is reasonable to assume, therefore, that this progression does not necessarily stop when spontaneous seizures appear. Clinical evidence supports a conclusion that some forms of epilepsy are progressive, not only with respect to worsening of the epileptic condition, but also with respect to worsening of nonepileptic cerebral functions, including disturbances in interictal behavior, and epileptic encephalopathies manifest as developmental delay and severe cognitive disturbances (4).

Because different types of human epilepsy reflect different pathophysiologic disturbances, information obtained from one epileptic condition is not necessarily generalizable to others. Consequently, clinical research intended to elucidate the fundamental mechanisms of human epilepsy must focus on fairly pure populations of patients with well-defined epileptic syndromes and epileptic seizures. Patients with benign forms of epilepsy do not require extensive neurodiagnostic testing or therapeutic surgical interventions; thus clinical research into basic mechanisms of these disorders must be carried out noninvasively. Although techniques for carrying out noninvasive investigations in humans are rapidly advancing, particularly those involving functional neuroimaging, such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI), as well as magnetoencephalography (MEG), invasive research on basic mechanisms of human epilepsy is limited to patients with refractory seizures who are candidates for surgical treatment. Fortunately, the most common form of human epilepsy, mesial temporal lobe epilepsy, which is also one of the most intractable to pharmacotherapy, is a surgically remediable syndrome (5,6). Consequently, large numbers of patients with relatively stereotyped pathophysiologic and anatomic substrates for their spontaneous seizures are readily available for invasive research on fundamental neuronal processes. The remainder of this article concerns basic research on patients with mesial temporal lobe epilepsy in the setting of an epilepsy surgery program (7,8).

RESEARCH METHOD

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

Invasive research into fundamental mechanisms of human mesial temporal lobe epilepsy makes use of the fact that some patients referred for surgical treatment require in vivo intracerebral recordings, and that most subsequently undergo surgical resections, making epileptogenic tissue available for in vitro investigations (9). In addition to in vivo macroelectrode EEG recording from multiple limbic and neocortical sites made possible by stereotactic depth electrode placement, these standard electrodes can be equipped with microwires and microdialysis probes, which make it possible to examine local circuitry of the limbic system by using techniques of stimulation, single-cell, and field-potential recording, and to measure neurotransmitter release and other chemical changes in the extracellular fluid (10) (Fig. 1). The practice of performing a standardized en bloc temporal resection as surgical treatment for mesial temporal lobe epilepsy (11) permits large identifiable specimens of epileptogenic brain to be used for in vitro electrophysiologic studies and provides tissue for microanatomic and molecular biologic analyses.

image

Figure 1. Diagram of flexible probe used for concomitant recording of EEG and single-unit activity and for cerebral microdialysis. Low magnification (lower) shows the platinum contacts for EEG recordings. Upper: Magnification of the distal part of the probe, showing inflow of artificial cerebrospinal fluid and outflow of the dialysate, as well as the cuprophan membrane through which substances in the extracellular fluid migrate along a concentration gradient into the probe. At the distal end is the membrane of the microdialysis probe and four platinum–iridium microwires used for single-unit recording. (Adapted from ref. 10, with permission.)

Download figure to PowerPoint

Practical limitations

Invasive procedures on patients with epilepsy offer extraordinary research opportunities, but are also associated with many limitations that confound experimental design and data interpretation (12). The most important of these are the following.

  • 1
    It is not usually possible to obtain normal control data because the brains of patients with medically refractory mesial temporal lobe epilepsy are diffusely abnormal, with wide areas of potential epileptogenicity, which frequently include contralateral mesial temporal structures.
  • 2
    Each patient is unique, and even with a syndrome as well defined as mesial temporal lobe epilepsy, considerable variability from patient to patient can make it difficult to average or summarize data across subjects.
  • 3
    Most epilepsy surgery centers involved in basic research can examine only one to two patients per month, and sample size may not be sufficient to obtain statistically significant results.
  • 4
    Patients with medically refractory seizures who are candidates for invasive diagnostic testing and surgical treatment represent the end stage of the epileptogenic process; thus this investigative approach provides little or no insight into the neuronal mechanisms underlying the development of epileptogenesis itself.

Strategies to overcome limitations

For some in vitro studies, absence of control data from human epileptogenic brain can be offset by use of autopsy material or surgical specimens from patients who do not have epilepsy. For the most part, however, invasive investigations, particularly in vivo studies, accommodate to lack of control tissue by incorporating multiple correlative analyses, for instance, comparing microanatomic and molecular biologic data from areas that generate specific types of epileptogenic electrophysiologic events, such as interictal spikes, with areas that do not, or examining the electrophysiologic and molecular biologic characteristics of tissue that show specific microanatomic changes, such as mossy fiber sprouting, with comparable areas that do not.

Interpatient variability cannot be completely eliminated, but problems due to differences among patients can be reduced by the use of a standardized presurgical evaluation protocol that includes standardized depth electrode placements, a standardized therapeutic resection, and standardized collection of clinical data for correlation with results of more basic investigations.

Many important research questions can be answered with qualitative rather than quantitative data, so that statistical analysis requiring large populations of patients is not always necessary. When sample size is an issue, however, the use of common standardized clinical and research protocols at several centers can permit pooling of data. Thus, collaborative multicenter projects can greatly enhance the statistical power of results obtained from basic research on patients.

Finally, the only way to study invasively the progressive nature of epilepsy, observe the epileptogenic process from the beginning, or carry out other research objectives that would require approaches either impractical or unethical in patients, is to develop accurate animal models of human disorders. Parallel reiterative human and animal studies are necessary to identify various component parts of human epilepsy that can be adequately reproduced in animals, to validate findings in laboratory animals by determining whether they exist in patients, and to use this paradigm to extend our ability to manipulate experimental conditions in a manner that is relevant to human epilepsy.

OBSERVATIONS FROM INVASIVE BASIC RESEARCH

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

An important early conclusion from invasive studies in patients is that no single “focus” of epileptogenicity exists; the brain is diffusely abnormal in many ways (9,13). Intracranial recordings reveal that extensive areas of ipsilateral and contralateral brain tissue are capable of generating spontaneous interictal spikes, and that the spatial distribution of this irritative zone does not necessarily predict where seizures originate. Widespread nonepileptogenic dysfunction also is apparent from the distribution of EEG slowing, hypometabolism, and hypoperfusion observed with PET and single-photon emission computed tomography (SPECT), as well as deficits revealed by neuropsychological testing, such as material-specific memory disturbances. These regions of dysfunction also do not necessarily colocalize with the site of seizure generation.

Even within a single, relatively stereotyped, epileptic syndrome such as mesial temporal lobe epilepsy, and indeed in a single patient, more than one type of seizure often occurs, because of more than one pathophysiologic mechanism and anatomic substrate (14–16) (Fig. 2). Furthermore, within a single epileptic seizure, usually an evolution of the ictal process occurs such that multiple fundamental neuronal mechanisms become sequentially involved (Fig. 2). Consequently, no final common path gives rise to all ictal events in mesial temporal lobe epilepsy.

image

Figure 2. Examples of depth-recorded mesial temporal ictal onsets. A: Low-voltage fast seizure onset. Seizure begins with regional decrease in amplitude and increase in frequency of left mesial temporal sites and spreads to the contralateral mesial temporal lobe after 20 s. Chain montage, with most distal electrode tip of each electrode referred to an adjacent electrode tip. Time calibration between vertical lines, 1.0 s; amplitude, 200 μV. B: Hypersynchronous seizure onset with focal slow spiking that remained on the left side for 130 s before spreading to the contralateral mesial temporal lobe. Spread to the other side occurred after development of a high-frequency low-amplitude discharge. An interval of 46 s was removed between the upper and lower traces. R, right; L, left; A, amygdala; AP, anterior hippocampus; MP, middle parahippocampal gyrus; MG, middle hippocampus; PG, posterior parahippocampal gyrus. Calibrations: time, 5.0 s; amplitude, 400 μV. (From ref. 14, with permission.)

Download figure to PowerPoint

Interictal and ictal epileptiform discharges do not simply reflect enhanced excitation and/or decreased inhibition (17,18). Excitation may be enhanced, but often inhibition is enhanced as well. Increased interictal inhibitory tone can be demonstrated in the epileptogenic region by a variety of methods, one of which is paired-pulse stimulation. In patients with mesial temporal lobe epilepsy (19), as in kindled rats (20), paired-pulse facilitation can be seen with stimulation of association pathways, such as the Schaffer collaterals in the epileptic hippocampus, but paired-pulse suppression more commonly occurs with stimulation of the perforant path, demonstrating that enhanced interictal inhibition selectively involves specific local circuits (Fig. 3).

image

Figure 3. Mean differences in paired-pulse excitability plotted for perforant path and intrinsic associational pathways of the hippocampus based on fPSP amplitude measurement. A: Mean suppression for the epileptogenic perforant path was profound up to 100 ms, whereas the nonepileptogenic perforant path showed significantly less suppression, particularly at 50 ms. B: Mean excitability of nonepileptogenic intrinsic associational hippocampal pathways was not significantly different from that found in the nonepileptogenic perforant path (shown in A), whereas the mean epileptogenic intrinsic hippocampal S2 response was significantly facilitated at ISI 50 ms. Although the epileptogenic associational pathways showed greater excitability at all interstimulus intervals than either the nonepileptogenic intrinsic hippocampal pathway or the epileptogenic perforant pathway, these differences were not significant, except at 50 ms (p < 0.01). (From ref. 19, with permission.)

Download figure to PowerPoint

Enhanced inhibition of the epileptogenic region in mesial temporal lobe epilepsy may be a protective homeostatic response necessary to maintain the interictal state; however, evidence shows that enhanced inhibitory mechanisms also may be epileptogenic by promoting hypersynchronization (21) (Fig. 4). Some mesial temporal ictal EEG onsets are characterized by a build-up of low-voltage fast activity, consistent with a disinhibitory mechanism, but, more commonly, mesial temporal ictal EEG onsets are hypersynchronous, at times resembling the hypersynchrony underlying absence seizures (14) (Fig. 5). This EEG pattern suggests a strong role for enhanced inhibition in some forms of hippocampal seizure generation (17).

image

Figure 4. Unit histograms from human epileptic hippocampus show weak firing suppression (A) in neurons that were not firing synchronously as determined by cross-correlation histograms (C), and strong firing suppression (B) in neurons that were firing synchronously (D). This provides indirect evidence for recurrent inhibitory circuits as a mechanism of hypersynchronization. (From ref. 21, with permission.)

Download figure to PowerPoint

image

Figure 5. Forty continuous seconds of an EEG recorded from depth, sphenoidal, and scalp electrodes during a simple partial seizure of the right temporal lobe. Ictal onset consists of an increase in interictal spike discharges, maximal at the right anterior hippocampal electrode (left portion of the upper panel). After 8–9 s, these spikes become regular, eventually developing into a 3-Hz spike-and-wave pattern involving all derivations from the right mesial temporal lobe. Note that no low-voltage fast activity is seen, either initially, or at any part of the ictal episode. The patient reached for the call button at the arrow, at which point regular slow activity also is seen in the left anterior hippocampus and in the right sphenoidal electrode. The patient then indicated an aura, consisting of a sensation of fear in her stomach. Depth-electrode locations indicated as in Fig. 1. Superficial contacts from anterior (A), mid (M), and posterior (P) depth electrodes recorded from cortex of middle temporal gyrus (MTG). Calibration, 1 s. (From ref. 41, with permission.)

Download figure to PowerPoint

Cell loss and neuronal reorganization appear to provide the anatomic substrate for epileptogenesis in mesial temporal lobe epilepsy with hippocampal sclerosis. However, the roles of aberrant excitatory and inhibitory connections in the dentate gyrus and hippocampus proper in mediating the process of epileptogenesis, as well as ictal generation, remain to be elucidated (22). Similar EEG and behavioral seizures can be seen in patients with mesial temporal lobe epilepsy due to discrete structural lesions in the absence of hippocampal sclerosis (5). Consequently, no evidence demonstrates that specific epilepsy-related morphologic changes, such as mossy fiber sprouting, are either necessary or sufficient for the development of mesial temporal lobe epilepsy (23).

It follows from the preceding discussion that different pathophysiologic mechanisms of epileptogenesis and ictal generation exist within, and between, various affected mesial temporal structures. Disinhibition and hypersynchronization are distinctly different common mechanisms of ictal generation, and one mechanism can evolve into the other. Therefore, in vitro research that limits investigation to a particular point in time, or a small region of tissue, will not provide sufficient information to permit an understanding of the dynamic processes of epileptogenesis and ictal generation. In vivo studies also are necessary if we are to begin to comprehend spatiotemporal relations between the many pathophysiologic changes underlying diverse ictal events that characterize mesial temporal lobe epilepsy. Furthermore, the process of epileptogenesis, and the progressive nature of epilepsy, will never be completely understood from current research strategies in the epilepsy surgery setting. For the present, this will require parallel research with appropriate experimental animal models that can be studied over long periods. In the future, however, the structural and temporal resolution of noninvasive functional diagnostic techniques, such as fMRI and MEG, may sufficiently improve to permit their application to these basic research questions.

THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

A number of approaches can be used to develop lesions in rat hippocampus that resemble hippocampal sclerosis and that produce spontaneous limbic seizures. Among these are systemic (24) and intrahippocampal kainate injection (25), pilocarpine injection (26), and massed limbic stimulation (27). Our studies using intrahippocampal kainate injection demonstrate the development of unilateral hippocampal cell loss and mossy fiber sprouting similar to that observed in patients with mesial temporal lobe epilepsy, with the exception that the pyramidal cell loss (as in the other models) is more marked in CA1 in patients and in CA3 in the kainate rat (22,25). Nevertheless, this lesion in the kainate rat is associated with epileptogenesis and the appearance, after several months, of two types of spontaneous epileptic seizures (28) (Fig. 6). As in the human, one type is characterized by the build-up of low-voltage fast activity on EEG, which is associated with ictal motor manifestations, and the other begins with hypersynchronous EEG discharges, without motor manifestations. Although it is unlikely that any experimental animal model will precisely mimic human mesial temporal lobe epilepsy, as noted previously, it is important to identify component parts of this condition that can be reproduced in the laboratory, and then to validate new insights gained from animal experimentation by demonstrating their existence in patients. The remainder of this article provides an example of this type of parallel reiterative human–animal research by using the kainate rat.

image

Figure 6. Examples of two different types of seizure onsets in kainic acid–treated rats with recurrent spontaneous seizures. A: Low-voltage fast onset. B: Synchronous onset. 1, records from hippocampus; 2, records from entorhinal cortex.

Download figure to PowerPoint

Parallel studies using the intrahippocampal kainite rat and patients with mesial temporal lobe epilepsy

In vivo microelectrode recordings from normal rats reveal very fast (100–200 Hz) oscillations in the CA1 area of the hippocampus, termed ripples, which are believed to represent inhibitory postsynaptic potentials that serve to synchronize large areas of this structure (29,30). Similar electrophysiologic investigations were carried out in kainate rats with spontaneous seizures, and in patients with mesial temporal lobe epilepsy, in an effort to determine whether these normal oscillations were altered (31,32). No abnormalities in frequency, distribution, or associated unit activity of normal ripple activity were found in the epileptic rats or patients; however, a unique, even faster (250–500 Hz) oscillation was identified, which also was present in patients (Fig. 7). In the rat, these oscillations, termed fast ripples, differed from normal ripples in that they were most prominent in dentate gyrus and entorhinal cortex, whereas normal ripples never occur in dentate gyrus. Ten percent of principal cells fired in synchrony with these oscillations (Fig. 8). It has been suggested that fast ripples represent population spikes, rather than inhibitory postsynaptic potentials. Most important, fast ripples tended to occur as part of interictal spikes and could be found only in the kainate-lesioned regions capable of generating spontaneous seizures. Fast ripples do not occur, for instance, contralateral to a kainate lesion, or in mesial temporal structures of kindled rats, which do not generate spontaneous seizures. Microelectrode recordings from human mesial temporal structures also demonstrated that interictal spikes with fast ripples occurred only in areas generating spontaneous seizures (31,32) (Fig. 9).

image

Figure 7. Low- and high-frequency ripples in human entorhinal cortex (EC). A: Power spectrum of electrical activity recorded from microelectrode 2. Recording bandwidth, 0.1–10,000 Hz. Note peaks at 96 and 284 Hz. B–D: Examples of the unit activity, ripples, and fast ripples recorded from the same file with two electrodes within EC. E–G: Averages of events (number is indicated in parentheses). Because of similarities of amplitudes, the events were selected into different files by visual estimation. Single-unit activity was recorded only from microelectrode 2; note that ripples are in phase on both electrodes, and the fast ripples are out of phase. (From ref. 31, with permission.)

Download figure to PowerPoint

image

Figure 8. Fast ripples in the chronic epileptic rat brain. A: Spontaneous fast ripples recorded in the rat's dentate gyrus (DG) and entorhinal cortex (EC) 8 months after unilateral intrahippocampal kainate injection. The recordings were performed in the right dorsal hippocampus and at symmetric points in the right and left EC and ventral hippocampi. Notice that 300-Hz fast ripples occur in the right ventral DG and ipsilateral EC. B: The unit activity of presumed pyramidal neurons (bottom) discharge in a phase-locked fashion with the negative wave of the fast ripple. Three superimposed events are shown. This neuron revealed rare (fewer than one per 10 s) burst discharges with 1.5- to 2.0-ms interspike intervals. The fast ripples were recorded with 100- to 800-Hz bandpass filter; unit activity was recorded with 0.5- to 5.0-kHz bandpass filter. (From ref. 32, with permission.)

Download figure to PowerPoint

image

Figure 9. A: Example of kainate rat EEG with interictal spike with superimposed fast ripples. LEC and REC, left and right entorhinal cortex; LdHip and RdHip, left and right dorsal hippocampus; LpHip and RpHip, left and right posterior hippocampus. B: Interictal spike with superimposed fast ripples in the entorhinal cortex of a patient with mesial temporal lobe epilepsy. Arrows, extension of electrical activity indicated in the box. LAH, left anterior hippocampus; ROF, right orbitofrontal cortex; LEC and REC, left and right entorhinal cortex (two microelectrodes in each side).

Download figure to PowerPoint

Recordings from limbic structures of kainate rats and patients with mesial temporal lobe epilepsy revealed several characteristic EEG epileptiform interictal events (31,32) (Fig. 10). Interictal spikes, usually followed by slow waves, are nonspecific transients that can be recorded from wide areas ipsilateral to the site of ictal onset, as well as contralaterally. Sharp waves associated with gamma frequency (30–50 Hz) oscillations also are seen in epileptic limbic structures of both rats and humans, and have the same nonspecific spatial distribution as interictal spikes. These gamma oscillations resemble events induced by 0 Mg++ in rat hippocampal slices, which appear to depend on γ-aminobutyric acid (GABA)A mechanisms (33). Interictal spikes with fast ripples, or sharp waves with fast ripples followed by gamma oscillations (fast ripple tail gamma complex), appear to be predominantly associated with those regions capable of generating spontaneous seizures.

image

Figure 10. Examples of interictal events in a patient with mesial temporal lobe epilepsy (A) and kainate-treated rat (B–D). A1: Tail gamma oscillation. A2: Fast ripple. B1–3: Interictal spikes. C: Fast ripple. D1: Tail gamma. D2: Fast ripple–tail gamma complex. The numbers within dashed boxes show the frequency of oscillations indicated by arrows. (Modified from refs. 28 and 38, with permission.)

Download figure to PowerPoint

The view that fast ripples reflect pathologic processes responsible for epileptogenesis, as opposed to nonspecific changes related to injury, is supported by observations that spontaneous ictal events in the kainate rat typically involve fast ripples (28). Seizures that begin with low-voltage fast EEG activity are typically initiated by a spike associated with fast ripples, which is then followed by continuous gamma activity (Fig. 11). Hypersynchronous ictal onsets, conversely, consist of recurrent interictal spike-like events that either are associated with fast ripples initially or rapidly include fast ripples (Fig. 12). These evolve into more typical fast ripple tail gamma complexes over the course of the hypersynchronous ictal event. As in the human, hypersynchronous ictal onsets can evolve into low-voltage fast EEG activity, but in the kainate rat, low-voltage fast ictal onsets also can evolve into hypersynchronous discharges associated with fast ripple tail gamma complexes.

image

Figure 11. A: Low-voltage fast ictal onset with initial 5-Hz waves in a kainate rat, which gradually increase in amplitude, transforming into spike-and-wave and spike-burst patterns. This last pattern is more obvious in RpHip; however, it also is visible in REC, RdHip, and LpHip. B: Expanded examples of the seizure onset (1), spike wave (2) and spike burst (3) patterns in the RpHip are indicated by dashed boxes. Note that the seizure onset is a single spike with fast ripples. (Modified from ref. 28, with permission.)

Download figure to PowerPoint

image

Figure 12. A: Example of a hypersynchronous ictal discharge originating in the lesioned hippocampus of a kainate rat. B: Expanded parts of the seizure are indicated by dashed boxes. 1: Fast ripples superimposed on positive waves at the beginning of the seizure. 2: An additional wave appears during development of the seizure (double arrowhead). 3: Fast-ripple tail gamma complex. All are recorded only from right posterior hippocampus. (Modified from ref. 28, with permission.)

Download figure to PowerPoint

SYNTHESIS AND CONCLUSIONS

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

Synchronously firing bursting neurons have been considered a hallmark of many types of experimental epilepsy, but it has been difficult to demonstrate a consistent pattern of synchronously bursting neurons in long-term experimental models of mesial temporal lobe epilepsy, or in patients with mesial temporal lobe epilepsy (34–36). Unlike the experimental penicillin focus in which 90–100% of units participate in synchronous bursting during epileptiform events (37), it has been estimated that fewer than 10% of neurons in these long-term experimental and clinical conditions fire abnormally in bursts (35,36). Consequently, the probability of finding synchronously bursting neurons by using single-unit recording is likely to be too low to identify or characterize epileptogenic tissue with this technique. Conversely, a small number of synchronously bursting neurons can produce field potentials that should be easily recorded from the epileptogenic region by using wide-band microelectrode recordings. Fast ripples appear to be field potentials of synchronously bursting neurons and may represent a much more robust way to identify regions where small clusters of epileptogenic neurons exist (38). These electrophysiologic events, therefore, could constitute a marker for the epileptogenic region. If this should be the case, fast ripples could be used to localize the epileptogenic region for surgical resection. Furthermore, if fast ripples reliably represent epileptogenicity, they might also be used to monitor the effectiveness of antiepileptic drug (AED) treatment in individual patients, or to predict the development of epilepsy after potentially epileptogenic insults. The application of fast-ripple recordings to these clinical situations would be much more effective if they could be recorded noninvasively. It is theoretically possible that MEG could detect fast ripples. If interictal spikes with fast ripples represent a different pathophysiologic substrate than interictal spikes without fast ripples, these two EEG transients could have different metabolic signatures that would permit distinction between them with fMRI spike mapping (39).

If fast ripples reflect the same aspects of epileptogenicity in patients with mesial temporal lobe epilepsy, and in experimental animal models, this EEG phenomenon also might provide a cost-effective means to screen for potential AEDs. Mesial temporal lobe epilepsy might be among the most refractory epileptic disorders because virtually all current potential AEDs are tested against animal models of generalized tonic–clonic seizures and absence seizures. It is likely that many compounds that could have been clinically effective against limbic seizures were discarded during animal screening because they did not have anticonvulsant or antiabsence properties. The addition of a cost-effective laboratory model for screening AEDs against limbic seizures, therefore, would be of major clinical value.

A primary objective of basic research in patients is to identify fundamental neuronal mechanisms of epileptogenesis that will ultimately provide insights into new approaches to treatment and prevention. If fast ripples reflect unique mechanisms responsible for epileptogenesis, then experiments must be designed to elucidate what these electrographic oscillations mean, at the systems, cellular, and subcellular levels. Our working hypothesis is that fast ripples reflect the unique neuronal events responsible for initiating epileptic seizures, and that these occur in our model, and perhaps in human mesial temporal lobe epilepsy, predominantly in reverberating circuits between entorhinal cortex and dentate gyrus. The relatively restricted recurrent hypersynchronous discharges that characterize one type of ictal onset might reflect events in entorhinal–dentate circuits that are regularly interrupted by seizure-suppressing mechanisms that constitute the dentate gate (40). The wide distribution of tail gamma, conversely, may occur with maximal dentate activation (40), when breakdown of the dentate gate permits propagation of ictal activity into hippocampus, and from there to the rest of the brain. It is intriguing that this latter process might be mediated by GABAA mechanisms (33). Clinically, and in the rat, localized hypersynchronous ictal EEG events have no behavioral correlate, although an aura may exist in patients, whereas propagation of low-voltage fast ictal EEG activity is associated with complex partial seizures and motor symptoms in both patients and rats. In the rat, where we know that the lesion is in the hippocampus, the single hypersynchronous event that begins the low-voltage fast ictal onset may reflect initiation of the seizure in the dentate or entorhinal cortex, but in patients, these seizures usually are associated with extrahippocampal lesions and may bypass the dentate altogether (14). These concepts are currently being pursued, first in the animal laboratory, but it will be possible to validate many, if not most, of the insights gained from these studies because of the potential to carry out in vivo and in vitro invasive research on patients in the epilepsy surgery setting.

Acknowledgments

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES

Acknowledgment:  Original research reported by the author was supported in part by grants NS-02808, NS-15654, NS-33310, and GM-24839 from the National Institutes of Health, and Contract DE-AC03-76-SF00012 from the Department of Energy.

REFERENCES

  1. Top of page
  2. Abstract
  3. RESEARCH METHOD
  4. OBSERVATIONS FROM INVASIVE BASIC RESEARCH
  5. THE IMPORTANCE OF EXPERIMENTAL ANIMAL MODELS
  6. SYNTHESIS AND CONCLUSIONS
  7. Acknowledgments
  8. REFERENCES
  • 1
    Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981;22: 489501.
  • 2
    Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30: 38999.
  • 3
    EngelJJr, PedleyTA, eds. Epilepsy: a comprehensive textbook. Vols. 1–3. Philadelphia : Lippincott-Raven, 1997.
  • 4
    Heinemann U, Engel J Jr, Meldrum BS, et al. The progressive nature of epilepsy. Epilepsy research (supplement 12). Amsterdam : Elsevier, 1996.
  • 5
    Engel J Jr, Williamson PD, Wieser HG. Mesial temporal lobe epilepsy. In: EngelJJr, PedleyTA, eds. Epilepsy: a comprehensive textbook. Philadelphia : Lippincott-Raven, 1997: 241726.
  • 6
    Engel J Jr. Current concepts: surgery for seizures. N Engl J Med 1996;334: 64752.
  • 7
    Engel J Jr, Babb TL, Crandall PH. Surgical treatment of epilepsy: opportunities for research into basic mechanisms of human brain function. Acta Neurochir 1989;46(suppl):38.
  • 8
    Schwartzkroin PA. Basic research in the setting of an epilepsy surgery center. In: EngelJJr, ed. Surgical treatment of the epilepsies. 2nd ed. New York : Raven Press, 1993: 75573.
  • 9
    Lüders HO, Engel J Jr, Munari C. General principles. In: EngelJJr, ed. Surgical treatment of the epilepsies, 2nd ed. New York : Raven Press, 1993: 13753.
  • 10
    Fried I, Wilson CL, Maidment NT, et al. Cerebral microdialysis combined with single neuron and EEG recording in neurosurgical patients. J Neurosurg 1999;91: 697705.
  • 11
    Spencer DD, Spencer SS, Mattson RH, et al. Access to the posterior medial temporal lobe structure in surgical treatment of temporal lobe epilepsy. Neurosurgery 1984;15: 66771.
  • 12
    Engel J Jr. Research on the human brain in an epilepsy surgery setting. Epilepsy Res 1998;32: 111.
  • 13
    Engel J Jr. New concepts of the epileptic focus. In: WieserHG, SpeckmannEG, EngelJJr, eds. The epileptic focus. London : John Libbey Eurotext , 1987: 8394.
  • 14
    Velasco AL, Wilson CL, Babb TL, et al. Functional and anatomic correlates of two frequently observed temporal lobe seizure-onset patterns. Neural Plasticity 2000;7: 4963.
  • 15
    Schiller Y, Cascino G, Busacker N, et al. Characterization and comparison of local onset seizure and remote propagated electrographic seizures recorded with intracranial electrodes. Epilepsia 1998;39: 3808.
  • 16
    Spencer SS, Guimares P, Katz A, et al. Morphological patterns of seizures recorded intracranially. Epilepsia 1992;33: 53745.
  • 17
    Engel J Jr. Excitation and inhibition in epilepsy. Can J Neurol Sci 1996;23: 16774.
  • 18
    Nusser Z, Hájos N, Somogyi P, et al. Increased number of synaptic GABAA receptors underlies potentiation at hippocampal inhibitory synapses. Nature 1998;395: 1727.
  • 19
    Wilson CL, Khan SU, Engel J Jr, et al. Paired pulse suppression and facilitation in human epileptogenic hippocampal formation. Epilepsy Res 1998;31: 21130.
  • 20
    Kamphuis W, Gorter JA, Wadman WJ, et al. Hippocampal kindling leads to different changes in paired-pulse depression of local evoked field potentials in CA1 area and in fascia dentata. Neurosci Lett 1992;141: 1015.
  • 21
    Isokawa-Akesson M, Wilson CL, Babb TL. Prolonged inhibition in synchronously firing human hippocampal neurons. Epilepsy Res 1989;3: 23647.
  • 22
    Babb TL, Pretorius JK, Mello LE, et al. Synaptic reorganizations in epileptic human and rat kainate hippocampus may contribute to feedback and feedforward excitation. In: EngelJJr, WasterlainC, CavalheiroEA, et al., eds. Molecular neurobiology of epilepsy (Epilepsy Res suppl 9). Amsterdam : Elsevier, 1992: 193203.
  • 23
    Longo B, Mello LEAM. Blockade of pilocarpine- or kainate-induced mossy fiber sprouting by cycloheximide does not prevent subsequent epileptogenesis in rats. Neurosci Lett 1997;226: 1636.
  • 24
    Cavalheiro EA, Riche DA, Le Gal La Salle G. Long-term effects of intrahippocampal kainic acid injection in rats: a method for inducing spontaneous recurrent seizures. EEG Clin Neurophysiol 1982;53: 5819.
  • 25
    Mathern GW, Cifuentes F, Leite JP, et al. Hippocampal EEG excitability and chronic spontaneous seizures are associated with aberrant synaptic reorganization in the rat kainate model. Electroencephalogr Clin Neurophysiol 1993;87: 32639.
  • 26
    Turski L, Ikonomidou C, Turski WA, et al. Cholinergic mechanisms and epileptogenesis: the seizures induced by pilocarpine: a novel model of intractable epilepsy. Synapse 1989;3: 15471.
  • 27
    Lothman EW, Bertram EH, Kapur J, et al. Recurrent spontaneous hippocampal seizures in the rat as a chronic sequela to limbic status epilepticus. Epilepsy Res 1990;6: 1109.
  • 28
    Bragin A, Engel J Jr, Wilson CL, et al. Electrophysiologic analysis of a chronic seizure model after unilateral hippocampal KA injection. Epilepsia 1999;40: 121021.
  • 29
    Buzsáki G, Horvath Z, Urioste R, et al. High frequency network oscillation in the hippocampus. Science 1992;256: 10257.
  • 30
    Ylinen A, Bragin A, Nádasdy Z, et al. Sharp wave-associated high-frequency oscillation (200 Hz) in the intact hippocampus: network and intracellular mechanisms. J Neurosci 1995;15: 3046.
  • 31
    Bragin A, Engel J Jr, Wilson CL, et al. High frequency oscillations in human brain. Hippocampus 1999;9: 13742.
  • 32
    Bragin A, Engel J Jr, Wilson CL, et al. Hippocampal and entorhinal cortex high frequency oscillations (100-500 Hz) in kainic acid-treated rats with chronic seizures and human epileptic brain. Epilepsia 1999;40: 12737.
  • 33
    Köhling R, Vreugdenhil M, Bracci E, et al. Ictal epileptiform activity is facilitated by hippocampal GABAA receptor-mediated oscillations. J Neurosci 2000;20: 68209.
  • 34
    Babb TL, Carr E, Crandall PH. Analysis of extracellular firing patterns of deep temporal lobe structures in man. EEG Clin Neurophysiol 1973;34: 24757.
  • 35
    Colder BW, Frysinger RC, Wilson CL, et al. Decreased neuronal burst discharge near site of seizure onset in epileptic human temporal lobes. Epilepsia 1996;37: 11321.
  • 36
    Colder BW, Wilson CL, Frysinger RC, et al. Neuronal synchrony in relation to burst discharge in epileptic human temporal lobes. J Neurophysiol 1996;75: 2496508.
  • 37
    Matsumoto H, Ajmone-Marsan C. Cortical cellular phenomena in experimental epilepsy: ictal manifestations. Exp Neurol 1964;9: 30526.
  • 38
    Bragin A, Wilson CL, Engel J Jr. Chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters: a hypothesis. Epilepsy Res 2000;41: S14452.
  • 39
    Krakow K, Woermann FG, Symms MR, et al. EEG-triggered functional MRI of interictal epileptiform activity in patients with partial seizures. Brain 1999;122: 167988.
  • 40
    Lothman EW, Bertram EH III, Stringer JL. Functional anatomy of hippocampal seizures. Progr Neurobiol 1991;37: 182.
  • 41
    Engel J Jr. Seizures and epilepsy. Philadelphia : FA Davis, 1989.