The mechanisms that cause acquired temporal lobe epilepsy are unknown. Suspected mechanisms include neuron loss, synaptic reorganization, and granule cell dispersion, but determining which abnormalities mediate epileptogenesis has been problematic because the most frequently used chemoconvulsant-based animal models exhibit extreme variability and minimal evidence of hippocampal epileptogenesis. Continuous monitoring of behavior and granule cell layer activity in awake rats after hippocampal injury caused by stimulation-induced status epilepticus has now shown that granule cells generate spontaneous field depolarizations, population spikes, and epileptiform discharges in the first days post-injury, prior to each generalized behavioral seizure. Thus, injury-associated hippocampal epileptogenesis is coincident with initial neuron loss, not delayed secondary processes. We hypothesize that neuron loss in the entorhinal cortex disrupts the functional separation of Layer II “grid cells,” causing abnormal synchronous discharges that invade the dentate gyrus. This, in turn, produces population spikes and epileptiform discharges in granule cells disinhibited by injury-induced hilar neuron loss. Long delays between injury and generalized behavioral seizures, when they occur, may primarily involve a “kindling” process in which initially focal (subclinical) discharges gradually increase in duration and cause clinical seizures. Neuroprotection in the immediate post-injury period, and prolonged anti-kindling therapy, might be the most effective anti-epileptic strategy.

WHY FOCUS ON TEMPORAL LOBE EPILEPSY AND THE DENTATE GYRUS?

There are many forms of epilepsy, and many brain regions that can generate spontaneous seizures, so why focus specifically on acquired temporal lobe epilepsy and the dentate gyrus in particular? There are several reasons. Of all the epilepsies, temporal lobe epilepsy is perhaps the most amenable to study because seizures often arise from the hippocampal formation, which is a brain region with a highly organized structure that facilitates the assessment of often subtle structural changes. In addition, the temporal lobe is removed surgically to treat refractory epilepsy, and is therefore often available for study.⁶ From an experimental perspective, the similarity of hippocampal structure from rodents to humans, and the fact that epilepsy can be produced in rodents by the same insults that cause epilepsy in humans, makes it feasible to relate the results of experimental studies to the human neurological disorder.

The early discovery that spontaneous epileptic seizures appeared to arise from a damaged and shrunken temporal lobe⁷ focused attention on surviving hippocampal neurons as likely sources of seizure activity. The likely role of the dentate gyrus as a primary source of seizures was supported by the observation that dentate granule cells are consistently among the few surviving hippocampal neuron populations,^8,9 and because the loss of neurons in the hippocampal “endfolium” was the single common pathology shared by all epilepsy patient brains that exhibited any detectable hippocampal cell loss.⁹ The anatomical realization that the neurons of the vulnerable endfolium, previously thought to be part of the pyramidal cell layer,¹⁰ were instead synaptically connected with the dentate gyrus,^11–13 raised the possibility that the loss of dentate hilar neurons⁹ might be associated with granule cell pathophysiology.¹⁴

Additional reasons for focusing on the dentate gyrus relate to the observation that several conspicuous secondary structural abnormalities develop after injury. In this chapter, we address selective neuron loss,¹⁴ synaptic reorganization,¹⁵ granule cell dispersion,¹⁶ and the timing of their development in relation to the time of onset of spontaneous granule cell epileptiform discharges and behavioral seizures. Clearly, whether a particular pathology precedes or follows the onset of spontaneous seizures is directly relevant to whether that pathology can be causally related to the epileptogenic process.

EPILEPTOGENIC NEURON LOSS AND IMMEDIATE GRANULE CELL HYPEREXCITABILITY; IS NEURON LOSS SUFFICIENT TO CAUSE EPILEPSY?

Do dentate granule cells become hyperexcitable and also spontaneously “epileptic” after hippocampal injury? And if so, precisely how soon after hippocampal injury do granule cells begin to generate spontaneous epileptiform discharges? A series of experimental studies in vivo has consistently demonstrated that granule cell hyperexcitability is caused immediately by prolonged seizures or head trauma, and is closely associated with extensive hilar neuron loss.^14,17–20 In our most detailed study, we found that whenever unilateral perforant pathway stimulation produced bilateral hilar neuron loss, then and only then was bilateral granule cell hyperexcitability evident.¹⁸ Conversely, whenever granule cell hyperexcitability was restricted to the ipsilateral dentate gyrus, only extensive ipsilateral hilar neuron loss was evident. These different responses to identical stimulation controlled for the effects of seizure activity per se because although unilateral stimulation always produced bilateral seizure discharges, contralateral granule cell hyperexcitability was uniquely associated with contralateral hilar neuron loss.¹⁸

We attributed the immediate granule cell hyperexcitability that we observed after prolonged perforant pathway stimulation to the extensive loss of the two main hilar neuron populations. These are the peptide-containing inhibitory interneurons that innervate the dentate outer molecular layer, and the excitatory hilar mossy cells that innervate the dentate inner molecular layer.^14,18 Loss of hilar inhibitory interneurons that project to the distal dendritic region innervated by the entorhinal cortex was hypothesized to produce a direct disinhibitory effect on granule cells.^14,18 The loss of excitatory hilar mossy cells was also hypothesized to cause granule cell hyperexcitability, but to do so indirectly as a result of decreased excitation of surviving inhibitory basket cells,^14,18 which mossy cells normally innervate and excite.^21–24

The hypothesized direct disinhibitory effect of hilar inhibitory interneuron loss^14,18 was supported by several subsequent studies.^25,26 However, the observation that extensive loss of hilar peptide-containing interneurons following ischemic insult²⁷ has minimal effect on granule cell excitability²⁸ argues against a prominent role for hilar interneuron loss in the specific injury-associated granule cell hyperexcitability that reliably follows prolonged perforant pathway stimulation.^14,18 Conversely, selective and extensive (~90%) mossy cell loss in a conditional mossy cell-knockout mouse caused immediate granule cell hyperexcitability,²⁹ supporting the view that extensive mossy cell loss is the primary cause of the granule cell hyperexcitability observed following kainate-, pilocarpine-, or stimulation-induced status epilepticus.^14,18,30,31

The main obstacle for our hypothesis - that seizure-induced neuron loss alone is a sufficient and primary epileptogenic mechanism-^18,32 was stated succinctly by Wasterlain and colleagues, who noted that, “... this hypothesis does not tackle the problem of the “silent period” between the initial injury and the development of spontaneous seizures. Because mossy cells are injured at the time of the original seizures and disappear within hours to days, it is hard to understand why spontaneous seizures are delayed by weeks to months, unless the loss of inhibition is permissive but not sufficient for epileptogenesis, and its role is only to permit further changes which, in turn, produce chronic epilepsy.”³³

The widely-accepted notion that status epilepticus-induced brain damage in animals is reliably followed by a silent, seizure-free “pre-epileptic” state lasting for weeks to months,^33–35 and that spontaneous epileptic seizures only begin because synaptic reorganization or other secondary processes that develop during the “silent period” have finally matured,^15,33,34 is anecdotal and based mainly on occasional behavioral observation,³⁶ which inevitably risks missing early behavioral seizures and all subclinical seizures.³⁷ The belief in a seizure-free, post-injury “latent period” ³⁵ has nonetheless formed the logical basis for focusing on delayed secondary processes as likely epileptogenic mechanisms, rather than on the effects of initial neuron loss and other immediate events. However, recent studies that involved continuous monitoring of chemoconvulsant-treated rats have now consistently shown that spontaneous behavioral seizures begin almost immediately after status epilepticus-induced injury.^31,38–41 Continuous video and electrographic monitoring of electrically stimulated rats, in which residual chemoconvulsant can play no role in any early seizures that occur, have now also shown that spontaneous granule cell-onset population spikes and epileptiform discharges, as well as behavioral seizures, also begin within the first days post- status epilepticus.⁴¹ Thus, epileptogenesis after status epilepticus is coincident with initial neuron loss (Figure 1), which occurs over a period of several days,⁴² indicating that neuron loss could well be the primary epileptogenic mechanism.⁴¹

Figure 1

Correlation between spontaneous granule cell layer activity and behavioral seizure expression in an epileptic rat 2–8 days after 3 hours of perforant pathway stimulation-induced convulsive status epilepticus (SE). Two focal (subclinical) seizures (more...)

“Hyperexcitable” is Not Synonymous with “Spontaneously Epileptic”; Does Hippocampal Epileptogenesis Require a Second Pathology?

Although extensive and specific mossy cell loss causes immediate granule cell hyperexcitability, this hyperexcitability alone is apparently not enough to cause spontaneous granule cell epileptiform discharges or spontaneous behavioral seizures.²⁹ This is not surprising because granule cell hyperexcitability to afferent excitation does not require that granule cells become a spontaneously discharging population. It is therefore notable that there is a second temporal lobe pathology common to all status epilepticus models, in the entorhinal cortex (Figure 2).^41,43,44

Figure 2

Acute Fluoro-Jade B (FJB) staining showing neurodegeneration 4 days after 3 hours of perforant pathway stimulation-induced SE. A: FJB fluorescence. B,C: Gray-scale, inverted image of the same horizontal brain section. Note selective degeneration of neurons (more...)

One possible explanation for the immediate development in awake animals of both granule cell hyperexcitability to afferent excitation and spontaneous granule cell-onset seizures, comes from the analysis of an animal model in which extensive hilar neuron loss, entorhinal cortex damage, and extra-temporal injury are consistently produced by electrical stimulation-induced status epilepticus.⁴¹ In this model, as in animals subjected to kainate-^20,30 or pilocarpine-induced status epilepticus,³¹ or experimental head trauma,¹⁹ extensive hilar neuron loss is reliably associated with immediate granule cell hyperexcitability to afferent excitation.⁴¹ In the case of perforant pathway-stimulated rats, in which granule cell activity and behavior were monitored continuously, spontaneous granule cell epileptiform discharges began almost immediately and preceded each spontaneous behavioral seizure (Figure 1).⁴¹ Importantly, clinical behavioral seizures always and only occurred when granule cells generated prolonged, negative-going population spikes (Figure 1). Clearly, these spontaneous granule cell discharges cannot be the result of recurrent excitatory connections among granule cells¹⁵ because the hyperexcitability, spontaneous epileptiform discharges, and behavioral seizures preceded the formation of recurrent axonal connections.⁴¹ We suggest that although granule cells become immediately hyperexcitable^{14,18–20,30} as a direct result of extensive hilar mossy cell loss,^18,29 granule cells do not generate synchronous epileptiform discharges until they receive abnormal excitatory input from the disinhibited Layer II neurons of the injured entorhinal cortex.^43,46 The appearance in the granule cell layers of spontaneous, large-amplitude dendritic field depolarizations virtually identical to those evoked by perforant pathway stimulation⁴¹ (Figure 1) suggests that spontaneous granule cell discharges may involve at least two principal pathologies; 1) mossy cell loss that disinhibits granule cells,^29,32 and 2) entorhinal cortex damage (Figure 2), which causes abnormal entorhinal Layer II neuron activity that invades the dentate gyrus.⁴¹ Future experiments will need to determine whether synchronous epileptiform activity in the entorhinal cortex precedes granule cell epileptiform discharges, or whether asynchronous entorhinal activity precedes synchronous granule cell discharges. Regardless, whether spontaneous granule cell seizure discharges cause behavioral seizures immediately (Figure 1), or result in a latent period before clinical seizures first occur, may be a measure of the extent of injury in the entorhinal cortex, in the hilus, and in the downstream barriers to seizure spread (hippocampal pyramidal cells and synaptically linked cell populations farther down the chain). We suggest that the only network changes needed for spontaneous granule cell epileptiform discharges to develop are extensive neuron loss in the hilus of the dentate gyrus and in the entorhinal cortex. We do not suggest that minor neuron loss is necessarily epileptogenic; only extensive hilar neuron loss is predicted to have a significant epileptogenic effect.²⁰

MOSSY FIBER SPROUTING

Few ideas are more conceptually appealing than the hypothesis of epileptogenic mossy fiber sprouting, as formulated originally by Nadler.^15,47 According to this scenario, normal granule cells have no recurrent excitatory connections, and therefore granule cells do not normally generate spontaneous population spikes or epileptiform discharges. Hilar mossy cell axotomy^48,49 or injury-induced mossy cell degeneration^15,50 denervates the granule cell dendrite segment normally innervated by mossy cells, and this loss of mossy cell input apparently triggers the re-innervation of granule cells by newly-formed granule cell axons (mossy fibers).⁵¹ The “mossy fiber sprouting” hypothesis posits that it is the formation of aberrant excitatory connections among normally unconnected granule cells that causes granule cells to become spontaneously “epileptic.”¹⁵ Many studies have replicated and extended the original findings of Nadler and colleagues,^15,47 but several critical issues were not considered in the many studies that were designed to support the hypothesis, and other observations suggest an alternate hypothesis.

First, the mossy fiber sprouting hypothesis ignores the network effects of the initial injury-induced neuron loss, and regards mossy cell loss as only a trigger for mossy fiber sprouting. However, the effects of neuron loss and reactive synaptic reorganization cannot be separated since these effects co-exist, with the former apparently causing the latter.⁵¹ Therefore, any parameters that have been correlated with mossy fiber sprouting in a multitude of studies could have been similarly correlated with hilar neuron loss although the role of the hilar neuron loss has rarely been discussed in studies that have sought to establish a causal link between mossy fiber sprouting and granule cell hyperexcitability. Second, we have found that the granule cell hyperexcitability observed in hippocampal slices from kainate-treated rats, and attributed to mossy fiber sprouting that takes weeks to develop,^15,30,31 is present in vivo immediately after kainate-induced injury, before mossy fiber sprouting develops.^30,45 Third, in rats in which immediate post-injury granule cell hyperexcitability was confirmed in our in vivo experiments, we have shown that the growth of mossy fiber sprouting was temporally associated with gradually increasing granule cell paired-pulse inhibition and an inability to evoke granule cell epileptiform discharges, rather than hyperexcitability.^30,45 Importantly, the judgment that granule cells were hyperexcitable shortly after injury, and later powerfully inhibited, was based not only on the assessment of responses to paired-pulse stimulation,^30,45 but also on the immediate appearance of multiple population spikes in response to single afferent stimuli, and the inability to evoke granule cell epileptiform discharges during the later synaptic reorganization phase, even by high frequency afferent excitation.³¹ Fourth, direct recording from the granule cell layers in awake pilocarpine- and kainate-treated animals revealed that the granule cells do not generate epileptiform discharges before the spontaneous behavioral seizures that develop in chemoconvulsant-treated rats.^31,52 This result is in contrast to the result of identical recordings made in perforant pathway-stimulated rats, in which all spontaneous behavioral seizures were preceded by granule cell discharges.^41,44 Fifth, and perhaps most definitively, in all studies in rats subjected to convulsive status epilepticus and then monitored continuously, spontaneous behavioral seizures precede mossy fiber sprouting.^38–41

Discussion of the “mossy fiber sprouting” hypothesis in the literature has focused almost exclusively on the aberrant autoinnervation of granule cells, and has largely ignored the possible impact of mossy cell death and mossy fiber sprouting on inhibitory basket cells.^30,45 In fact, mossy cell loss denervates all target cell dendrites in the dentate inner molecular layer, not just granule cells, and the re-innervation of vacated synaptic sites by newly formed granule cell axons would be predicted to target both inhibitory neurons and granule cells.^30,31,45 The consistent finding that synaptic reorganization does, in fact, re-innervate both granule cells and inhibitory neurons^30,31,45,54 (Figures 3 and 4) challenges the assumption that mossy fiber sprouting must be exclusively “excitatory” in nature.¹⁵ To the contrary, we interpret the available data as indicating that mossy fiber sprouting may play a mainly compensatory or restorative role.^30,45

Figure 3

Parvalbumin-positive inhibitory interneurons are targets of aberrant mossy fiber sprouting at the time of early recovery of granule cell paired-pulse inhibition. A: Timm staining in a control rat. Note that Timm-positive terminals surround and outline (more...)

Figure 4

GABA immunocytochemical electron microscopy of the dentate inner molecular layer 10 weeks after saline or kainate (KA)-induced status epilepticus (SE). A1,2: In control sections, proximal dendrites (D) of GABA-positive interneurons are contacted by small (more...)

GRANULE CELL DISPERSION

Granule cell dispersion in temporal lobe epilepsy, first described by Houser,⁵⁹ is another example of a frequently observed structural abnormality that may or may not play a role in altered dentate gyrus excitability.¹⁶ In the normal dentate gyrus, granule cells are tightly packed, forming a clearly delineated and relatively uniform cell layer. Although there is some structural variability among these neurons, particularly in the primate brain,⁶⁰ granule cells send their cone-shaped dendrites to the molecular layer, with virtually all dendrites reaching the hippocampal fissure. Mossy fiber axons emerge from the granule cell somata, and enter the hilus. This bipolarity of dentate granule cells clearly separates the input region of the cell from its output region, resulting in minimal granule cell-granule cell connectivity under normal conditions. In granule cell dispersion, the compact lamination of granule cell bodies is lost. As a result, the axons of granule cells dispersed into the molecular layer traverse the molecular layer for some distance and could contact the dendrites of more deeply located neurons, possibly increasing granule cell interconnectivity. Since granule cell dispersion is found in tissues from epileptic patients, the following scenario appears plausible: migration defects of granule cells during development, or dispersion of granule cells following hippocampal injury, might result in increased granule cell interconnectivity and hyperexcitability.

The results of studies conducted over the last ten years suggest a different scenario. In tissue samples from epileptic patients, the expression of the extracellular matrix protein Reelin was found to be significantly decreased.⁶¹ Moreover, it was noticed that the extent of granule cell dispersion correlated with the extent of decreased Reelin expression. Reelin is known for its role in layer formation in the cerebral cortex, cerebellum, and hippocampus,^62–68 and Reelin-deficient “reeler” mutants show a severe loss of granule cell lamination,^69–71 which is at least structurally reminiscent of granule cell dispersion in epilepsy. Thus, Reelin apparently stabilizes dentate gyrus architecture, and an injury-induced decrease in Reelin expression might cause granule cell dispersion. To test this hypothesis, Heinrich and colleagues⁷² infused a Reelin-neutralizing antibody into the dentate gyrus of mature mice. They unexpectedly observed granule cell dispersion at sites of Reelin antibody infusion, and this effect could not be induced when the Reelin antibody was replaced by a nonspecific IgG. The interpretation of these findings is that Reelin establishes or maintains the laminated organization of the dentate gyrus, and that decreased Reelin expression or antibody blockade of Reelin results in granule cell dispersion.

Unilateral granule cell dispersion is reliably produced in normal mice during the first month following unilateral intrahippocampal injections of the glutamate receptor agonist kainate,⁷³ and it has been shown that Reelin expression was dramatically decreased on the kainate-injected side, but not contralaterally.⁷² These results suggest that granule cell dispersion results from decreased Reelin expression, after cell loss or hypermethylation of the Reelin gene.⁷⁴ Although Reelin appears to play an important role in stabilizing cortical architecture in the mature brain,⁶⁸ it remains to be determined whether granule cell dispersion directly influences granule cell interconnectivity and excitability. In this regard, it is notable that reeler mice, which are deficient in Reelin and exhibit granule cell dispersion, do not show spontaneous seizures. In addition, recent recordings from dispersed granule cells in kainate-treated epileptic mice provided evidence of reduced, rather than increased, granule cell excitability.⁷⁵ Importantly, many patient hippocampi do not exhibit granule cell dispersion, and other animal models that exhibit confirmed granule cell-onset epilepsy do not show granule cell dispersion.^41,44 Although intrahippocampal kainate injection causes both granule cell dispersion and epilepsy in mice, perforant pathway stimulation-induced hippocampal injury in mice produces epilepsy without producing granule cell dispersion.⁷⁶ Thus, the pathophysiological implications of granule cell dispersion, if any, remain to be clarified.

THE LATENT PERIOD AND EPILEPTOGENESIS

The belief that all status epilepticus models exhibit a silent post-injury “latent period” during which seizures do not occur^33–36 is no longer tenable, for reasons cited above.^{31,37–41,57} Although there is no delay between injury and epilepsy after prolonged status epilepticus in animals, and a similar lack of any detectable latent period has been reported after prolonged status epilepticus in humans,⁷⁷ delays in the appearance of clinical seizures after injury usually exist.^34,57,78 Why do seizures sometimes develop in humans immediately after injury, but in other cases, only after years or decades? Unfortunately, estimates of the latency to the appearance of clinical epilepsy are inherently unreliable because they are almost always based on the time that elapses between a presumed injury and the observation of a significant behavioral event.⁷⁸ This is understandable, as the proper metric, i.e. the time of onset of the first focal epileptiform discharges, cannot be determined. For example, should a patient who has experienced a clinically unrecognized “aura” for thirty years, prior to a first clinical seizure, be regarded as having had epilepsy ever since the first aura, or only after the first clinical seizure has been observed and recognized? That is, does “epileptogenesis” really take thirty years to mature in this case, with the 30 year-long period of subclinical auras considered to be a “pre-epileptic” state during which a single epileptogenic process was slowly maturing? From a neurobiological perspective, the important event for the concept of epileptogenesis would seem to be the first onset of epileptiform discharges that produced the first aura (a focal seizure), regardless of whether the first discharges spread sufficiently to disrupt behavior or consciousness, and become clinically obvious. We contend that delays in the appearance of clinical seizures following brain injuries have been inferred incorrectly to indicate that “epileptogenesis” is a specific and progressive unitary process that is as long in duration as the time it takes for generalized or clinically obvious seizures to appear and be recognized. Clearly, the calculated latent period cannot reflect the length of “epileptogenesis” if the time to the appearance of the first epileptiform discharges cannot be accurately assessed. Regardless, recent results indicate that highly experimenter-controlled hippocampal injuries consistently produce immediate hippocampal epileptogenesis that is coincident with initial neuron loss.⁴¹

Based on our most recent studies,^41,44 we suggest that acquired temporal lobe epilepsy involves a relatively straightforward two-stage process. First, an injury causes changes that result in spontaneous principal cell epileptiform discharges, which are presumably subclinical in most cases. This stage (“epileptogenesis”) might be most effectively impeded in the immediate post-injury period by a neuroprotective treatment that minimizes the extent of initial neuron loss, some of which is delayed,⁴² and may be initially susceptible to treatment. Second, subclinical discharges increase gradually in duration, and invade and recruit other neuronal populations that act initially as barriers to seizure spread, ultimately causing clinical epilepsy, which includes subtle focal seizures.⁴¹ This distinct second phase (“epileptic maturation”) might be most effectively targeted by treatments that retard the kindling process, or that interfere with pro-epileptic secondary processes such as glial abnormalities, neurogenesis, synaptic reorganization, altered receptor expression, etc.

A “GRID CELL” HYPOTHESIS OF TEMPORAL LOBE EPILEPTOGENESIS

The data discussed above regarding the possible epileptogenicity of damage in the entorhinal cortex and dentate gyrus permit a conceptual synthesis that may have implications for thinking about temporal lobe epileptogenesis, and devising strategies to prevent it. If all animals subjected to a uniform insult exhibit pathological changes that closely resemble the human neurological condition, and if all of these animals exhibit spontaneous, hippocampal-onset seizures that develop without delay,⁴¹ then epilepsy in these animals is the likely result of the immediate effects of injury, rather than being the result of delayed secondary processes that develop after the animal is already epileptic.^41,57

If hippocampal injury causes immediate granule cell hyperexcitability, and no recurrent excitatory connections are necessary for epileptiform discharges to occur, why do granule cells spontaneously generate only brief intermittent seizure discharges and behavioral seizures,⁴¹ rather than continuous epileptiform discharges and status epilepticus? We hypothesize that some of the answers may lie in the nature of incomplete hilar and entorhinal cortex neuron loss, and the behavior of Layer II entorhinal neurons, which form the main excitatory input to the dentate granule cells.⁷⁹ Layer II entorhinal cortical neurons (EC2) are hyperexcitable following status epilepticus,⁴⁶ presumably as a consequence of cell loss in the adjacent Layers III (EC3) and V (EC5)^43,80,81 or other closely related nuclei. Recent studies in normal animals have shown that EC2 neurons constitute a system of “grid cells,” in which individual EC2 neurons form an environment-independent spatial coordinate system.^82,83 These “grid cells” normally discharge strictly independently in a spatial environment, and exhibit discrete inhibitory surrounds.⁸² Apparently, the independent firing patterns of EC2 grid cells constitute a universal map of the spatial environment, and these grid cells feed this information to their target cells in the dentate gyrus.⁸⁴ We predict that the pathology reliably produced in the EC3 and EC5 layers by prolonged convulsive- or non-convulsive status epilepticus^41,43,44 reduces the size of the grid cell inhibitory surround and decreases the location-based specificity of EC2 neuron discharges. EC2 neuron hyperexcitability may occur as a result of the loss of EC3 and EC5 neurons, which normally influence EC2 neurons.⁸⁵ Loss of EC2 neuronal inhibition would cause EC2 pyramidal cells to coalesce functionally, to disrupt the “grid” function that establishes normal spatial memory, and to generate abnormal synchronous discharges that propagate directly to the granule cell layers.⁸⁶ Therefore, we would predict that after extensive stimulation-induced injury of hilar neurons and EC3 and EC5 cells,^41,44 EC2 grid cells should lose their inhibitory surrounds and their spatial separation immediately, and begin to generate synchronous discharges that cause the spontaneous granule cell layer depolarizations, population spikes (apparent biomarkers of imminent epileptiform discharges), and epileptiform discharges that we have consistently recorded in awake rats prior to each granule cell-onset seizure (Figure 1).^41,44 Changes in the GABAergic projection from entorhinal cortex to hippocampus⁸⁷ might also affect hippocampal excitability.

CONCLUSION

In summary, we suggest that acquired temporal lobe epileptogenesis involves two causal pathologies: 1) extensive hilar neuron (mossy cell) loss, which reduces translamellar granule cell inhibition,²⁰ causing granule cell hyperexcitability to afferent input,^29,32 and; 2) entorhinal cortex damage, which causes a loss of functional separation in Layer II “grid cells,” resulting in abnormal and synchronous excitation of disinhibited granule cells, which generates spontaneous granule cell-onset seizures.^32,41,44 We hypothesize that widespread brain damage, such as that caused by prolonged convulsive status epilepticus, can result in immediate clinical epilepsy in both rats and humans^41,77 because all cortical and subcortical barriers to the spread of focal seizures are damaged or functionally altered by the initial insult. Conversely, after more limited injury in the entorhinal cortex and hilus, it may take time for EC2 neurons to begin to generate abnormal discharges. Once begun, it may require additional time for entorhinal discharges to overcome granule cell inhibition that remains as a result of incomplete hilar neuron loss. Thus, granule cell seizures would not start until the entorhinal cortex generates abnormal discharges capable of overcoming granule cell inhibition and evoking granule cell epileptiform activity. From this perspective, the inhibitory effects of mossy fiber sprouting may establish or extend the latency to the first granule cell-onset seizures by making granule cells more resistant to generating their first epileptiform discharges.^30,31,45

A post-injury delay of the first spontaneous granule cell discharges, plus additional time needed to recruit hippocampal pyramidal cells and to overcome downstream barriers to seizure spread, could explain a prolonged period of subclinical focal discharges that precedes the appearance of clinical, life-disrupting seizures. The hypothesis that the progression from subclinical- to clinical epilepsy involves a time-consuming kindling process⁸⁸ that should be targeted in the immediate post-injury period^41,44 is not original, but wholly consistent with the ideas of Graham Goddard, which were cogently summarized in Goddard’s obituary by Frank Morrell.⁸⁹ Thus, we suggest that two mechanisms should be primary targets for pharmacological treatment: 1) initial neuron loss, which has a delayed component⁴² that may constitute a therapeutic window that remains open for several days, and 2) a kindling process,⁸⁸ interruption of which could, at least theoretically, extend the latent period to clinical seizures indefinitely, even if epileptogenic neuron loss cannot be substantially reduced.

In addition to the need for experimental resolution of a variety of issues using animal models that reliably involve hippocampal epileptogenesis, it may also be worth considering that the terms “epileptogenesis,” “latent period,” and “kindling” are names of concepts created by the human mind, rather than being real and readily definable or identifiable neurobiological entities.⁹⁰ Giving the name “epileptogenesis” to a probably multifactorial process of unknown nature has implications for how we think about that process, and implies that it is something singular that can be prevented or aborted if only “it” can be identified. Similarly, the term “latent period” is a conception that confers significance on an ill-defined time period with an uncertain beginning and an unknown maturation date.^34,57 The use of the term “latent period” has significant implications for the importance we ascribe to a perceived interval during which nothing is observed. Clearly, an inability to hear a distant conversation is not evidence that nothing has been said. Given these considerations, the clarity of the discussion of the process that causes the brain to generate abnormal synchronized discharges for the first time (“epileptogenesis”), and of the secondary processes that enable focal epileptiform discharges to spread and become clinically detectable (“epileptic maturation”), might benefit from referring to the neurobiological processes themselves, rather than using created names applied to difficult-to-define subjective conceptions of those processes.⁹⁰

Finally, given that many endogenous homeostatic mechanisms normally limit excitation, epilepsy may be viewed as an unusually powerful network defect that, once established, cannot be completely suppressed by any combination of homeostatic mechanisms, and cannot be easily controlled pharmacologically. If any pharmacological intervention can successfully interfere with the processes of “epileptogenesis” and “epileptic maturation,” the combination of a neuroprotective compound and an anti-kindling compound in the immediate post-injury period, followed by long-term treatment with an anti-kindling compound alone, might be the most logical treatment approach.

ACKNOWLEDGEMENT

The authors gratefully acknowledge constructive criticism of the manuscript by Dr. Daniel H. Lowenstein, UCSF, Dr. Philip A. Schwartzkroin, UC Davis, and Dr. D. Steven Kerr, Otago University.

LITERATURE CITED

1.

Reid CA, Berkovic SF, Petrou S. Mechanisms of human inherited epilepsies. Prog Neurobiol. 2009;87:41–57. [PubMed: 18952142]

2.

Gaitanis JN, Walsh CA. Genetics of disorders of cortical development. Neuroimaging Clin N Am. 2004;14:219–229. [PubMed: 15182816]

3.

McNamara JO. Cellular and molecular basis of epilepsy. J Neurosci. 1994;14:3413–3425. [PMC free article: PMC6576925] [PubMed: 8207463]

4.

Chang BS, Lowenstein DH. Mechanisms of disease; epilepsy. N Engl J Med. 2003;349:1257–1266. [PubMed: 14507951]

5.

Fernández G, Effenberger O, Vinz B, Steinlein O, Elger CE, Döhring W, Heinze HJ. Hippocampal malformation as a cause of familial febrile convulsions and subsequent hippocampal sclerosis. Neurology. 1998;50:909–917. [PubMed: 9566371]

6.

Falconer MA, Taylor DC. Surgical treatment of drug-resistant epilepsy due to mesial temporal lobe sclerosis; etiology and significance. Arch Neurol. 1968;19:353–361. [PubMed: 5677186]

7.

Jasper HH, Pertuiset B, Flanigin H. EEG and cortical electrograms in patients with temporal lobe seizures. Arch Neurol Psychiatr. 1951;65:272–290. [PubMed: 14810279]

8.

Meldrum BS, Bruton CJ. Epilepsy. In: Adams JH, Duchen LW, editors. Greenfield’s Neuropathology. New York: Oxford University Press; 1992. pp. 1246–1283.

9.

Margerison JH, Corsellis JA. Epilepsy and the temporal lobes. A clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to the temporal lobes. Brain. 1966;89:499–530. [PubMed: 5922048]

10.

Lorente de Nó R. Studies on the structure of the cerebral cortex. II. Continuation of the study of the ammonic system. J Psychol Neurol. 1934;46:113–177.

11.

Amaral DG. A Golgi study of cell types in the hilar region of the hippocampus in the rat. J Comp Neurol. 1978;182:851–914. [PubMed: 730852]

12.

Berger TW, Semple-Rowland S, Bassett JL. Hippocampal polymorph neurons are the cells of origin for ipsilateral association and commissural afferents to the dentate gyrus. Brain Res. 1981;224:329–336. [PubMed: 6116526]

13.

Bakst I, Avendano C, Morrison JH, Amaral DG. An experimental analysis of the origins of somatostatin-like immunoreactivity in the dentate gyrus of the rat. J Neurosci. 1986;6:1452–1462. [PMC free article: PMC6568547] [PubMed: 2872280]

14.

Sloviter RS. Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy. Science. 1987;235:73–76. [PubMed: 2879352]

15.

Tauck DL, Nadler JV. Evidence of functional mossy fiber sprouting in hippocampal formation of kainic acid treated rats. J Neurosci. 1985;5:1016–1022. [PMC free article: PMC6565006] [PubMed: 3981241]

16.

Haas CA, Frotscher M. Reelin deficiency causes granule cell dispersion in epilepsy. Exp Brain Res. 2010;200:141–149. [PubMed: 19633980]

17.

Sloviter RS. “Epileptic” brain damage in rats induced by sustained electrical stimulation of the perforant path. I. Acute electrophysiological and light microscopic studies. Brain Res Bull. 1983;10:675–697. [PubMed: 6871737]

18.

Sloviter RS. Permanently altered hippocampal structure, excitability, and inhibition after experimental status epilepticus in the rat: the dormant basket cell hypothesis and its possible relevance to temporal lobe epilepsy. Hippocampus. 1991;1:41–66. [PubMed: 1688284]

19.

Lowenstein DH, Thomas MJ, Smith DH, McIntosh TK. Selective vulnerability of dentate hilar neurons following traumatic brain injury: a potential mechanistic link between head trauma and disorders of the hippocampus. J Neurosci. 1992;12:4846–4853. [PMC free article: PMC6575779] [PubMed: 1464770]

20.

Zappone CA, Sloviter RS. Translamellar disinhibition in the rat hippocampal dentate gyrus after seizure-induced degeneration of vulnerable hilar neurons. J Neurosci. 2004;24:853–864. [PMC free article: PMC6729823] [PubMed: 14749430]

21.

Buzsáki G, Eidelberg E. Commissural projection to the dentate gyrus of the rat: evidence for feed-forward inhibition. Brain Res. 1981;230:346–350. [PubMed: 7317783]

22.

Buzsáki G, Eidelberg E. Direct afferent excitation and long-term potentiation of hippocampal interneurons. J Neurophysiol. 1982;48:597–607. [PubMed: 6290613]

23.

Douglas RM, McNaughton BL, Goddard GV. Commissural inhibition and facilitation of granule cell discharge in fascia dentata. J Comp Neurol. 1983;219:285–294. [PubMed: 6311879]

24.

Bilkey DK, Goddard GV. Septohippocampal and commissural pathways antagonistically control inhibitory interneurons in the dentate gyrus. Brain Res. 1987;405:320–325. [PubMed: 3567610]

25.

Cossart R, Dinocourt C, Hirsch JC, Merchan-Perez A, De Felipe J, Ben-Ari Y, Esclapez M, Bernard C. Dendritic but not somatic GABAergic inhibition is decreased in experimental epilepsy. Nat Neurosci. 2001;4:52–62. [PubMed: 11135645]

26.

Sun C, Mtchedlishvili Z, Bertram EH, Erisir A, Kapur J. Selective loss of dentate hilar interneurons contributes to reduced synaptic inhibition of granule cells in an electrical stimulation-based animal model of temporal lobe epilepsy. J Comp Neurol. 2007;500:876–893. [PMC free article: PMC2844442] [PubMed: 17177260]

27.

Johansen FF, Zimmer J, Diemer NH. Early loss of somatostatin neurons in dentate hilus after cerebral ischemia in the rat precedes CA1 pyramidal cell loss. Acta Neuropathol. 1987;73:110–114. [PubMed: 2885998]

28.

Mody I, Otis TS, Bragin A, Hsu M, Buzsáki G. GABAergic inhibition of granule cells and hilar neuronal synchrony following ischemia-induced hilar neuronal loss. Neuroscience. 1995;69:139–150. [PubMed: 8637612]

29.

Jinde S, Zsiros V, Kohno K, Nakazawa K. Generation and characterization of inducible-dentate mossy cell ablation mice. Program number 645.11 2008, Neuroscience Meeting Planner. Chicago, IL: Society for Neuroscience; 2008. Online.

30.

Sloviter RS. Possible functional consequences of synaptic reorganization in the dentate gyrus of kainate-treated rats. Neurosci Lett. 1992;137:91–96. [PubMed: 1625822]

31.

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. 2005;488:441–462. [PubMed: 15973680]

32.

Sloviter RS. The functional organization of the hippocampal dentate gyrus and its relevance to the pathogenesis of temporal lobe epilepsy. Ann Neurol. 1994;35:640–654. [PubMed: 8210220]

33.

Wasterlain CG, Mazarati AM, Shirasaka Y, Thompson KW, Penix L, Liu H, Katsumori H. Seizure-induced hippocampal damage and chronic epilepsy: a Hebbian theory of epileptogenesis. Adv Neurol. 1999;79:829–843. [PubMed: 10514867]

34.

Bragin A, Wilson CL, Engel J Jr. Chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters: a hypothesis. Epilepsia. 2000;41(Suppl 6):S144–152. [PubMed: 10999536]

35.

Stables JP, Bertram E, Dudek FE, Holmes G, Mathern G, Pitkänen A, White HS. Therapy discovery for pharmacoresistant epilepsy and for disease-modifying therapeutics: summary of the NIH/NINDS/AES models II workshop. Epilepsia. 2003;44:1472–1478. [PubMed: 14636315]

36.

Kobayashi M, Buckmaster PS. Reduced inhibition of dentate granule cells in a model of temporal lobe epilepsy. J Neurosci. 2003;23:2440–2452. [PMC free article: PMC6741996] [PubMed: 12657704]

37.

Mazarati A, Bragin A, Baldwin R, Shin D, Wilson C, Sankar R, Naylor D, Engel J, Wasterlain CG. Epileptogenesis after self-sustaining status epilepticus. Epilepsia. 2002;43(Suppl 5):74–80. [PubMed: 12121299]

38.

Raol YH, Lund IV, Bandyopadhyay S, Zhang G, Roberts DS, Wolfe JH, Russek SJ, Brooks-Kayal AR. Enhancing GABA(A) receptor alpha 1 subunit levels in hippocampal dentate gyrus inhibits epilepsy development in an animal model of temporal lobe epilepsy. J Neurosci. 2006;26:11342–11346. [PMC free article: PMC6674546] [PubMed: 17079662]

39.

Goffin K, Nissinen J, Van Laere K, Pitkänen A. Cyclicity of spontaneous recurrent seizures in pilocarpine model of temporal lobe epilepsy in rat. Exp Neurol. 2007;205:501–505. [PubMed: 17442304]

40.

Jung S, Jones TD, Lugo JN, Sheerin JH, Miller JW, D’Ambrosio R, Anderson AE, Poolos NP. Progressive dendritic HCN channelopathy during epileptogenesis in the rat pilocarpine model of epilepsy. J Neurosci. 2007;27:13012–13021. [PMC free article: PMC3087381] [PubMed: 18032674]

41.

Bumanglag AV, Sloviter RS. Minimal latency to hippocampal epileptogenesis and clinical epilepsy after perforant pathway stimulation-induced status epilepticus in awake rats. J Comp Neurol. 2008;510:561–580. [PMC free article: PMC2562302] [PubMed: 18697194]

42.

Sloviter RS, Dean E, Sollas AL, Goodman JH. Apoptosis and necrosis induced in different hipocampal neuron populations by repetitive perforant path stimulation in the rat. J Comp Neurol. 1996;366:516–533. [PubMed: 8907362]

43.

Du F, Eid T, Lothman EW, Köhler C, Schwarcz R. Preferential neuronal loss in layer III of the medial entorhinal cortex in rat models of temporal lobe epilepsy. J Neurosci. 1995;15:6301–6313. [PMC free article: PMC6577998] [PubMed: 7472396]

44.

Norwood BA, Bumanglag AV, Osculati F, Sbarbati A, Marzola P, Nicolato E, Fabene PF, Sloviter RS. Classic hippocampal sclerosis and hippocampal-onset epilepsy produced by a single “cryptic” episode of focal hippocampal excitation in awake rats. J Comp Neurol. 2010;518:3381–3407. [PMC free article: PMC2894278] [PubMed: 20575073]

45.

Sloviter RS, Zappone CA, Harvey BD, Frotscher M. Kainic acid-induced recurrent mossy fiber innervation of dentate gyrus inhibitory interneurons: possible anatomical substrate of granule cell hyperinhibition in chronically epileptic rats. J Comp Neurol. 2006;494:944–960. [PMC free article: PMC2597112] [PubMed: 16385488]

46.

Kobayashi M, Wen X, Buckmaster PS. Reduced inhibition and increased output of layer II neurons in the medial entorhinal cortex in a model of temporal lobe epilepsy. J Neurosci. 2003;23:8471–8479. [PMC free article: PMC6740375] [PubMed: 13679415]

47.

Nadler JV. The recurrent mossy fiber pathway of the epileptic brain. Neurochem Res. 2003;28:1649–1658. [PubMed: 14584819]

48.

Laurberg S, Zimmer J. Lesion-induced sprouting of hippocampal mossy fiber collaterals to the fascia dentata in developing and adult rats. J Comp Neurol. 1981;200:433–459. [PubMed: 7276246]

49.

Frotscher M, Zimmer J. Lesion-induced mossy fibers to the molecular layer of the rat fascia dentata: identification of postsynaptic granule cells by the Golgi-EM technique. J Comp Neurol. 1983;215:299–311. [PubMed: 6189867]

50.

Nadler JV, Perry BW, Gentry C, Cotman CW. Loss and reacquisition of hippocampal synapses after selective destruction of CA3-CA4 afferents with kainic acid. Brain Res. 1980;191:387–403. [PubMed: 7378766]

51.

Jiao Y, Nadler JV. Stereological analysis of GluR2-immunoreactive hilar neurons in the pilocarpine model of temporal lobe epilepsy: correlation of cell loss with mossy fiber sprouting. Exp Neurol. 2007;205:569–582. [PMC free article: PMC1995080] [PubMed: 17475251]

52.

Queiroz CM, Gorter JA, Lopes da Silva FH, Wadman WJ. Dynamics of evoked local field potentials in the hippocampus of epileptic rats with spontaneous seizures. J Neurophysiol. 2009;101:1588–1597. [PubMed: 18842951]

53.

Sloviter RS, Zappone CA, Harvey BD, Bumanglag AV, Bender RA, Frotscher M. “Dormant basket cell” hypothesis revisited; relative vulnerabilities of dentate gyrus mossy cells and inhibitory interneurons after hippocampal status epilepticus in the rat. J Comp Neurol. 2003;459:44–76. [PubMed: 12629666]

54.

Kotti T, Riekkinen PJ Sr, Miettinen R. Characterization of target cells for aberrant mossy fiber collaterals in the dentate gyrus of epileptic rat. Exp Neurol. 1997;146:323–330. [PubMed: 9270041]

55.

Amaral DG, Witter MP. The three-dimensional organization of the hippocampal formation: a review of anatomical data. Neuroscience. 1989;31:571–591. [PubMed: 2687721]

56.

Sutula T, Zhang P, Lynch M, Sayin U, Golarai G, Rod R. Synaptic and axonal remodeling of mossy fibers in the hilus and supragranular region of the dentate gyrus in kainate-treated rats. J Comp Neurol. 1998;390:578–594. [PubMed: 9450537]

57.

Sloviter RS. Hippocampal epileptogenesis in animal models of mesial temporal lobe epilepsy with hippocampal sclerosis; the importance of the “latent period” and other concepts. Epilepsia. 2008;49(Suppl 9):85–92. [PubMed: 19087122]

58.

Sloviter RS, Dichter MA, Rachinsky TL, Dean E, Goodman JH, Sollas AL, Martin DL. Basal expression and induction of glutamate decarboxylase and GABA in excitatory granule cells of the rat and monkey hippocampal dentate gyrus. J Comp Neurol. 1996;373:593–618. [PubMed: 8889946]

59.

Houser CR. Granule cell dispersion in the dentate gyrus of humans with temporal lobe epilepsy. Brain Res. 1990;535:195–204. [PubMed: 1705855]

60.

Seress L, Frotscher M. Morphological variability is a characteristic feature of granule cells in the primate fascia dentata: A combined Golgi/electron microscope study. J Comp Neurol. 1990;293:253–267. [PubMed: 19189715]

61.

Haas CA, Dudeck O, Kirsch M, Huszka C, Kann G, Pollak S, Zentner J, Frotscher M. Role for reelin in the development of granule cell dispersion in temporal lobe epilepsy. J Neurosci. 2002;22:5797–5802. [PMC free article: PMC6757930] [PubMed: 12122039]

62.

Rakic P, Caviness VS Jr. Cortical development: view from neurological mutants two decades later. Neuron. 1995;14:1101–1104. [PubMed: 7605626]

63.

Frotscher M. Cajal-Retzius cells, Reelin, and the formation of layers. Curr Opin Neurobiol. 1998;8:570–575. [PubMed: 9811621]

64.

Rice DS, Curran T. Role of the reelin signaling pathway in central nervous system development. Annu Rev Neurosci. 2001:1005–1039. [PubMed: 11520926]

65.

Tissir F, Goffinet AM. Reelin and brain development. Nat Rev Neurosci. 2003;4:496–505. [PubMed: 12778121]

66.

Förster E, Zhao S, Frotscher M. Laminating the hippocampus. Nat Rev Neurosci. 2006;7:259–267. [PubMed: 16543914]

67.

Cooper JA. A mechanism for inside-out lamination in the neocortex. Trends Neurosci. 2008;31:113–119. [PubMed: 18255163]

68.

Frotscher M. Role for Reelin in stabilizing cortical architecture. Trends Neurosci. 2010. in press. [PubMed: 20598379]

69.

Stanfield BB, Cowan WM. The morphology of the hippocampus and dentate gyrus in normal and reeler mice. J Comp Neurol. 1979;185:393–422. [PubMed: 438366]

70.

Stanfield BB, Cowan WM. The development of the hippocampus and dentate gyrus in normal and reeler mice. J Comp Neurol. 1979;185:423–460. [PubMed: 86549]

71.

Drakew A, Deller T, Heimrich B, Gebhardt C, Del Turco D, Tielsch A, Förster E, Herz J, Frotscher M. Dentate granule cells in reeler mutants and VLDLR and ApoER2 knockout mice. Exp Neurol. 2002;176:12–24. [PubMed: 12093079]

72.

Heinrich C, Nitta N, Flubacher A, Müller M, Fahrner A, Kirsch M, Freiman T, Suzuki F, Depaulis A, Frotscher M, Haas CA. Reelin deficiency and displacement of mature neurons, but not neurogenesis, underlie the formation of granule cell dispersion in the epileptic hippocampus. J Neurosci. 2006;26:4701–4713. [PMC free article: PMC6674063] [PubMed: 16641251]

73.

Bouilleret V, Ridoux V, Depaulis A, Marescaux C, Nehlig A, Le Gal La, Salle G. Recurrent seizures and hippocampal sclerosis following intrahippocampal kainate injection in adult mice: electroencephalography, histopathology and synaptic reorganization similar to mesial temporal lobe epilepsy. Neuroscience. 1999;89:717–729. [PubMed: 10199607]

74.

Kobow K, Jeske I, Hildebrandt M, Hauke J, Hahnen E, Buslei R, Buchfelder M, Weigel D, Stefan H, Kasper B, Pauli E, Blümcke I. Increased Reelin promoter methylation is associated with granule cell dispersion in human temporal lobe epilepsy. J Neuropathol Exp Neurol. 2009;68:356–364. [PubMed: 19287316]

75.

Young CC, Stegen M, Bernard R, Müller M, Bischofberger J, Veh RW, Haas CA, Wolfart J. Upregulation of inward rectifier K+ (Kir2) channels in dentate gyrus granule cells in temporal lobe epilepsy. J Physiol. 2009;587:4213–4233. [PMC free article: PMC2754361] [PubMed: 19564397]

76.

Kienzler F, Norwood BA, Sloviter RS. Hippocampal injury, atrophy, synaptic reorganization, and epileptogenesis after perforant pathway stimulation-induced status epilepticus in the mouse. J Comp Neurol. 2009;515:181–196. [PMC free article: PMC2705826] [PubMed: 19412934]

77.

Mikaeloff Y, Jambaque I, Hertz-Pannier L, Zamfirescu A, Adamsbaum C, Plouin P, Dulac O, Chiron C. Devastating epileptic encephalopathy in school-aged children (DESC): a pseudo encephalitis. Epilepsy Res. 2006;69:67–79. [PubMed: 16469483]

78.

French JA, Williamson PD, Thadani VM, Darcey TM, Mattson RH, Spencer SS, Spencer DD. Characteristics of medial temporal lobe epilepsy: I. Results of history and physical examination. Ann Neurol. 1993;34:774–780. [PubMed: 8250525]

79.

Ruth RE, Collier TJ, Routtenberg A. Topography between the entorhinal cortex and the dentate septotemporal axis in rats: I. Medial and intermediate entorhinal projecting cells. J Comp Neurol. 1982;209:69–78. [PubMed: 7119174]

80.

Schwarcz R, Eid T, Du F. Neurons in layer III of the entorhinal cortex. A role in epileptogenesis and epilepsy. Ann NY Acad Sci. 2000;911:328–342. [PubMed: 10911883]

81.

Kumar SS, Buckmaster PS. Hyperexcitability, interneurons, and loss of GABAergic synapses in entorhinal cortex in a model of temporal lobe epilepsy. J Neurosci. 2006;26:4613–4623. [PMC free article: PMC6674073] [PubMed: 16641241]

82.

Hafting T, Fyhn M, Molden S, Moser MB, Moser EI. Microstructure of a spatial map in the entorhinal cortex. Nature. 2005;436:801–806. [PubMed: 15965463]

83.

Sargolini F, Fyhn M, Hafting T, McNaughton BL, Witter MP, Moser MB, Moser EI. Conjunctive representation of position, direction, and velocity in entorhinal cortex. Science. 2006;312:758–762. [PubMed: 16675704]

84.

van Strien NM, Cappaert NL, Witter MP. The anatomy of memory: an interactive overview of the parahippocampal-hippocampal network. Nat Rev Neurosci. 2009;10:272–282. [PubMed: 19300446]

85.

Quilichini P, Sirota A, Buzsáki G. Intrinsic circuit organization and theta-gamma oscillation dynamics in the entorhinal cortex of the rat. J Neurosci. 2010;30:11128–11142. [PMC free article: PMC2937273] [PubMed: 20720120]

86.

Scimemi A, Schorge S, Kullmann DM, Walker MC. Epileptogenesis is associated with enhanced glutamatergic transmission in the perforant path. J Neurophysiol. 2006;95:1213–1220. [PubMed: 16282203]

87.

Germroth P, Schwerdtfeger WK, Buhl EH. GABAergic neurons in the entorhinal cortex project to the hippocampus. Brain Res. 1989;494:187–192. [PubMed: 2765919]

88.

Goddard GV, McIntyre DC, Leech CK. A permanent change in brain function resulting from daily electrical stimulation. Exp Neurol. 1969;25:295–330. [PubMed: 4981856]

89.

Morrell F. In memoriam Graham Goddard: an appreciation. Epilepsia. 1987;28:717–719. [PubMed: 3319538]

90.

Sloviter RS. Apoptosis: a guide for the perplexed. Trends Pharmacol Sci. 2002;23:19–24. [PubMed: 11804647]

Footnote: This work was supported by grants NS18201 from the National Institute of Neurological Disorders and Stroke, National Institutes of Health, and Deutsche Forschungsgemeinschaft, Transregio Sonderforschungsbereich TR-3.