JN AJP: Endocrinology and Metabolism
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Neurophysiol 80: 230-240, 1998;
0022-3077/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fountain, N. B.
Right arrow Articles by Lothman, E. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fountain, N. B.
Right arrow Articles by Lothman, E. W.

The Journal of Neurophysiology Vol. 80 No. 1 July 1998, pp. 230-240
Copyright ©1998 by the American Physiological Society

Responses of Deep Entorhinal Cortex are Epileptiform in an Electrogenic Rat Model of Chronic Temporal Lobe Epilepsy

Nathan B. Fountain1, Jonathan Bear2, Edward H. Bertram III1, and Eric W. Lothman2, ✠

1 Department of Neurology and 2 Neuroscience Program, University of Virginia Health Sciences Center, Charlottesville, Virginia 22908

    ABSTRACT
Abstract
Introduction
Methods
Results
Discussion
References

Fountain, Nathan B., Jonathan Bear, Edward H. Bertram III, and Eric W. Lothman. Responses of deep entorhinal cortex are epileptiform in an electrogenic rat model of chronic temporal lobe epilepsy. J. Neurophysiol. 80: 230-240, 1998. We investigated whether entorhinal cortex (EC) layer IV neurons are hyperexcitable in the post-selfsustaining limbic status epilepticus (post-SSLSE) animal model of temporal lobe epilepsy. We studied naive rats (n = 44), epileptic rats that had experienced SSLSE resulting in spontaneous seizures (n = 45), and electrode controls (n = 7). There were no differences between electrode control and naive groups, which were pooled into a single control group. Intracellular and extracellular recordings were made from deep layers of EC, targeting layer IV, which was activated by stimulation of the superficial layers of EC or the angular bundle. There were no differences between epileptic and control neurons in basic cellular characteristics, and all neurons were quiescent under resting conditions. In control tissue, 77% of evoked intracellular responses consisted of a short-duration [8.6 ± 1.3 (SE) ms] excitatory postsynaptic potential and a single action potential followed by gamma -aminobutyric acid-A (GABAA) and GABAB inhibitory post synaptic potentials (IPSPs). Ten percent of controls did not contain IPSPs. In chronically epileptic tissue, evoked intracellular responses demonstrated prolonged depolarizing potentials (256 ± 39 ms), multiple action potentials (13 ± 4), and no IPSPs. Ten percent of epileptic responses were followed by rhythmic "clonic" depolarizations. Epileptic responses exhibited an all-or-none response to progressive increases in stimulus intensity and required less stimulation to elicit action potentials. In both epileptic and control animals, intracellular responses correlated precisely in morphology and duration with extracellular field potentials. Severing the hippocampus from the EC did not alter the responses. Duration of intracellular epileptic responses was reduced 22% by the N-methyl-D-aspartate (NMDA) antagonist D(-)-2-amino-5-phosphonovaleric acid (APV), but they did not return to normal and IPSPs were not restored. Epileptic and control responses were abolished by the non-NMDA antagonist 6,7-dinitroquinoxaline-2-3-dione (DNQX). A monosynaptic IPSP protocol was used to test connectivity of inhibitory interneurons to primary cells by direct activation of interneurons with a stimulating electrode placed near the recording electrode in the presence of APV and DNQX. Using this protocol, IPSPs similar to control (P > 0.05) were seen in epileptic cells. The findings demonstrate that deep layer EC cells are hyperexcitable or "epileptiform" in this model. Hyperexcitability is not due to interactions with the hippocampus. It is due partially to augmented NMDA-mediated excitation. The lack of IPSPs in epileptic neurons may suggest inhibition is impaired, but we found evidence that inhibitory interneurons are connected to their target cells and are capable of inducing IPSPs.

    INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References

There is direct evidence that entorhinal cortex (EC) is involved in human temporal lobe epilepsy (TLE) and animal models of epilepsy. Seizures may arise independently from EC or may arise simultaneously in EC and other limbic areas in both patients (Spencer and Spencer 1994) and in the post-selfsustaining limbic status epilepticus (post-SSLSE) model (Bertram 1997), which shares many features with human TLE (Bertram et al. 1990; Lothman et al. 1990). Cell loss in EC, especially layer III, is present in human TLE (Du et al. 1993; Kim et al. 1990) and in several animal models, including the post-SSLSE model (Du and Schwarcz 1992; Du et al. 1995). EC is involved in acutely induced epileptiform discharges (Bear and Lothman 1993; Jones and Lambert 1990; Pare et al. 1992; Rafiq et al. 1993 1995; Stringer and Lothman 1992; Walther et al. 1986; Wilson et al. 1988). Therefore EC may be a site of seizure initiation or propagation.

There is also indirect evidence implicating EC in epilepsy. Control of TLE by anterior temporal lobectomy typically is attributed to removal of the hippocampus, but EC is usually unavoidably removed during this surgery (Sperling et al. 1996). Furthermore, outcome from temporal lobectomy to control TLE correlates with the amount of parahippocampal gyrus (including entorhinal cortex) resected; the more parahippocampal gyrus resected, the better the outcome (Siegel et al. 1990). Therefore, it is possible that removal of EC contributes to control of seizures.

The organization of EC may provide a basis for its role in TLE. EC consists of several cell types formed into six layers (Amaral and Insausti 1990; Swanson et al. 1987). Layer II has a distinctive appearance because it contains clusters or rosettes of "star pyramids," or stellate cells, whereas layer IV contains large pyramidal-like cells that form a continuous layer. However, other layers of the cortex are less well demarcated. Thus layers I-III often are termed superficial layers, whereas layers IV-VI are termed deep layers. Physiology and connectivity follow a similar generalization, such that deep layers are the primary site of input into EC, whereas superficial layers provide the primary output of EC. The EC is reciprocally connected to the hippocampus (Hampson and Deadwyler 1992; Jones 1993; Swanson et al. 1987; Witter 1993; Witter et al. 1989), which provides the potential for a reverberatory circuit that may initiate or perpetuate seizure activity (Lothman et al. 1991). Strong connections with limbic cortex may provide a pathway for seizure spread (Brothers and Finch 1985; Finch et al. 1986, 1988; Lingenhohl and Finch 1991; White et al. 1990).

In a previous study of hippocampal-parahippocampal slices from epileptic post-SSLSE rats, we found hyperexcitability of EC layer II neurons, the primary output layer of the EC (Bear et al. 1996). In this model, these neurons have prolonged depolarizing envelopes and multiple action potentials (APs) in response to a single shock. They also lack inhibitory postsynaptic potentials (IPSPs), possibly suggesting an impairment of inhibition.

EC layer IV has several characteristics that could allow it to participate in seizures. It is important as the primary site of input to the EC, where it may participate in a reverberatory circuit with the hippocampus or with layer II. Acutely induced epileptiform activity in vitro may be recorded in layer IV (Bear and Lothman 1993; Jones and Heinemann 1988; Rafiq et al. 1995). Finally, layer IV neurons have unique intrinsic properties that may predispose them to hyperexcitability to an even greater degree than layer II cells because they may have a prominent N-methyl-D-aspartate (NMDA)-mediated excitatory component and a less prominent gamma -aminobutyric acid (GABA)-mediated inhibitory component, possibly lacking GABAB IPSPs (Jones 1987, 1989, 1994; Jones and Heinemann 1988).

We hypothesized that EC layer IV neurons would be hyperexcitable in this model of chronic TLE. We sought to determine whether this hyperexcitability was due to excess excitation mediated by NMDA mechanisms, whether there was evidence of impaired inhibition, and whether this activity was dependent on connections with the hippocampus. Some of these results were previously presented in abstract form (Fountain et al. 1996).

    METHODS
Abstract
Introduction
Methods
Results
Discussion
References

Animal groups

Three groups of adult male Sprague-Dawley rats were studied: naive animals, which did not have electrodes implanted; electrode control animals, which were implanted with electrodes but not stimulated; and epileptic animals, which were implanted with electrodes, received continuous hippocampal stimulation (CHS), and developed SSLSE and subsequent spontaneous seizures.

For the electrode control and post-SSLSE groups, the experimenter was blinded with respect to the nature of the experimental group from which the slices were derived until after analyses were complete. To avoid the possible confounding issue of electrode placement effects, only slices from the side of the brain contralateral to electrode implantation were studied.

Induction of epilepsy by SSLSE

Methods for the induction of SSLSE by CHS were performed as previously described (Lothman et al. 1989). Briefly, for each animal, a pair of bipolar stimulating electrodes were implanted under anesthesia in the left posterior ventral hippocampus (AP -3.6, ML -4.0, DV -5.0 from dura; incisor bar at +5.0). After 1 wk of recovery, stimulus trains (50 Hz of 1-ms, 400-µA biphasic square wave pulses for 10 s) were delivered every 13 s for 90 min to induce SSLSE. Only animals displaying SSLSE for >= 2 h, as identified by electroencephalographic (EEG) criteria previously described (Lothman et al. 1990), were studied further, because these animals were more likely to develop spontaneous recurrent seizures than animals with shorter SSLSE (Bertram and Cornett 1993). A period of >= 1 mo occurred between SSLSE and electrophysiological study. All post-SSLSE rats underwent prolonged EEG monitoring in a monitoring unit designed for 24-h continuous monitoring (Bertram and Cornett 1993, 1994). EEG recording started 4 wk after SSLSE and continued until at least one spontaneous seizure was recorded. All post-SSLSE rats had at least one seizure and most had many seizures. A parallel group of six electrode control rats was monitored for 3 mo, and none demonstrated seizures during EEG monitoring or during handling (Bear et al. 1996).

Hippocampal-parahippocampal slice preparation

Rats were anesthetized with 5% halothane in air and decapitated. The brains were removed from the skull, and the cerebellum and rostral pole of the brain were dissected away. The brain was mounted to a vibratome chuck using cyanoacrylate glue. An agar block provided support during slicing. The brains were mounted at a 30° angle to the base of the brain to maintain connectivity of the EC with the hippocampus (Bear et al. 1996; Jones and Heinemann 1988; Rafiq et al. 1993) and consisted of temporal cortex, hippocampus, and EC (Fig. 1). Slices (350- to 400-µm thick) were cut on a vibratome. All dissection was done in 4°C artificial cerebrospinal fluid (ACSF) saturated with carbogen gas (95% O2-5% CO2). Thereafter slices were maintained submerged in carbogen-bubbled ACSF at 22°C until use, at which time they were transferred to an interface recording chamber (Schwartzkroin 1975) where they were maintained at 37°C in ACSF perfused at 2.0-3.0 ml/min and humidified carbogen gas (5 l/min). Ionic composition of the ACSF was as follows (in mM): 153 Na+, 1.5 Mg2+, 1.5 Ca2+, 1.5 SO2-4, 1.1 PO3-4, 132 Cl-, 3 K+, and 10 glucose, pH = 7.4 by bicarbonate buffering.


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 1. Line drawing of hippocampal-parahippocampal slice at the level of the posterior commissure. Locations for placement of recording electrodes in deep layers [entorhinal cortex (EC) IV] and stimulating electrodes in superficial layers (EC II) and angular bundle (AB), are shown. Hippocampal areas CA1, CA3, and dentate gyrus are labeled for orientation.

To avoid including unhealthy slices, only cells with a resting membrane potential (RMP) of at least -50 mV and exhibiting evoked APs were included in the study. Evoked response measurements were made at the RMP. To avoid effects induced by a varying RMP, only cells with a stable RMP, varying by <3 mV, were included.

Stimulation and recording

Recording electrodes were targeted for the region of layer IV (Fig. 1). However individual layers cannot be identified in the unstained section in the recording chamber and cells were not identified histologically, so that we could only resolve electrode placement to be in the "deep" layers, containing layers IV, V, and VI. Stimulating electrodes were placed in the angular bundle of the perforant pathway to stimulate afferent input to layer IV orthodromically and in superficial layers of EC to stimulate axons of layer IV cells antidromically. Antidromic stimulation was desired to stimulate recurrent collaterals onto inhibitory interneurons. Stimulating electrodes were oriented parallel to layer II in the region adjacent to the recording electrode.

Intracellular recording electrodes were filled with 2 M potassium methylsulfate (80-120 MOmega ). Single-barrel glass extracellular recording electrodes were filled with 2 M NaCl (approx 5 MOmega ). Bipolar stimulating electrodes were used for stimulation. Stimuli were controlled by a digital timer (Winston T-10) and delivered from a stimulus isolation unit. Stimuli were 100-µs square waves ranging in intensity from 1 to 100 V. Recording electrodes were led through high-impedance, capacitance-compensated amplifiers (Axoprobe, Axon Instruments) and then in parallel to a computerized data acquisition system (pClamp, Axon Instruments) and a digital oscilloscope.

"Excitability" was assessed by the AP stimulation threshold (the minimum stimulus intensity necessary to evoke an AP), number of APs elicited, and amplitude and duration of EPSPs in control animals and of depolarizing envelopes in epileptic animals. In controls, width at half height was measured in matched EPSPs at ~4 mV amplitudes, subthreshold for AP generation, to control for possible changes in AP threshold. In epileptic responses, total duration of the deoplarizing response was measured at the lowest stimulus intensity that evoked a response because the responses were all-or-none (see RESULTS). Inhibition was assessed by presence of IPSPs and their characteristics, including time to peak, amplitude, duration, half-time of decay, kinetic parameters, and reversal potentials. Reversal potentials were determined by measuring the amplitude of the maximum IPSPs during sequential current steps and plotting these values as a function of membrane potential. All measurements from evoked responses were in response to a single shock. Measurements of excitatory postsynaptic potentials (EPSPs) and IPSPs were made at the RMP. Basic cellular characteristics (RMP, input resistance, etc.) were measured only once per cell. Parameters measured in evoked responses (e.g., response duration) were measured once per stimulus site because the response to stimulation at one site could be different from the response to stimulation at a different site, mediated by different synapses.

Maximum amplitudes and durations of intracellular and extracellular responses were obtained from input/output function (I/O) curves. I/O curves were constructed by plotting stimulus voltage versus response amplitude over a range of stimulus intensities (Fig. 9), from very low to high intensity (Fig. 5). Stimulus intensity was increased progressively until the maximum amplitude of the response was produced; increasing the stimulus intensity further did not increase the amplitude of the response. The amplitude of all responses, including IPSP amplitudes in the presence of excitatory blockade (monosynaptic IPSP protocol), always were measured at this plateau.


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 9. Example of input/output (I/O) curve for GABAA and GABAB IPSPs measured in a response from an epileptic animal using the monosynaptic IPSP protocol. Absolute amplitude is plotted against stimulus intensity. Measurements for comparisons were taken at the plateau of the curve.


View larger version (18K):
[in this window]
[in a new window]
  		FIG. 5. Simultaneous intracellular (top trace in each pair) and extracellular (bottom trace in each pair) recording of the response to very small incremental increases in stimulus intensity from an epileptic animal. Note lack of an increasing EPSP and instead an "all-or-none" response to only a 0.1-V increase in stimulus intensity from 0.2 to 0.3 V.

A monosynaptic IPSP protocol assessed whether inhibitory interneurons synapse onto primary cells (Bear et al. 1996; Bekenstein and Lothman 1993; Williams et al. 1993). Recording electrodes were placed as detailed above. Stimulating electrodes were placed in the same layer and within 100 µm of the recording electrode to directly stimulate local inhibitory interneurons, which typically synapse at the cell body of primary cells in EC (Wouterlood et al. 1995). To mask any contribution from normal or pathological excitation, glutamate-mediated ionotropic neurotransmission was blocked by D(-)-2-amino-5-phosphonovaleric acid (APV) and 6,7-dinitroquinoxaline-2-3-dione (DNQX). IPSPs evoked in this manner were measured as noted above.

Response pharmacology

Drug effects on responses were determined by addition of drugs to the bathing ACSF. APV (50 µM; Cambridge Research Pharmaceuticals) was used to block NMDA-mediated glutamate neurotransmission. DNQX (5 or 10 µM; Cambridge Research Pharmaceuticals) in ACSF, prepared from 5% dimethylsulfoxide (DMSO) in water (0.01% DMSO vol/vol), was used to block non-NMDA-mediated glutamate neurotransmission. This concentration of DMSO had no effect on responses, similar to previous work (Bear et al. 1996). Picrotoxin (100 µM; Sigma) was used to block GABAA-mediated neurotransmission and 2-hydroxysaclofen (100 µM; RBI) was used to block GABAB-mediated neurotransmission. Responses were measured at 3-min intervals after the drug was added to the bath. Data for analysis was collected at the point when a plateau occurred in the response and the drug had no additional effect. If no effect was seen after 60 min, data for analysis was collected at that time. Drugs were washed out of the bathing solution to demonstrate restoration of predrug response characteristics.

Statistical analysis

Data for continuous variables are expressed as means ± SE. Analysis of variance was used to determine the significance of differences in the means of multiple continuous variables. Fisher's exact test was used to determine the significance of differences between groups for proportional variables. Data were assessed for differences between the naive and electrode control groups and between the control and epileptic groups using unpaired t-tests. Paired t-tests were used to determine the significance of differences between response characteristics of a cell before and after drug wash-in. P < 0.05 was considered statistically significant.

    RESULTS
Abstract
Introduction
Methods
Results
Discussion
References

Characteristics of the naive and electrode control groups

There were no differences between the electrode control and naive groups. All parameters measured were not statistically different (P > 0.05) between naives (n = 44) and electrode controls (n = 7) with regard to intrinsic membrane properties, characteristics of intracellular responses to stimulation of the angular bundle or superficial layers of EC (morphology, presence of IPSPs and quantitation of response characteristics), and characteristics of extracellular responses. In addition, electrode controls did not differ electrophysiologically from naives in other areas previously studied (Bear et al. 1996; Bekenstein and Lothman 1993; Rempe et al. 1995). Therefore, data from the naive and electrode control groups were pooled into a single control group for comparison with the epileptic group in all analyses.

Basic cellular properties of the control and epileptic groups

The basic cellular properties listed in Table 1 were not statistically different between the control and epileptic groups. In response to current steps, most cells demonstrated spike frequency adaptation (SFA) without inward rectifying currents or rebound APs (Fig. 2B). However, some cells contained inward rectifying currents (Fig. 2, A and D) or rebound APs (Fig. 2A) or lacked SFA (Fig. 2D). There was no relationship between which cells demonstrated inward rectification, rebound APs, or SFA. AP firing patterns were qualitatively identical in epileptics and controls. To roughly quantitate this, we counted the number of APs during the first 100 ms and the second 100 ms in epileptics and controls in response to a 0.4-nA depolarizing current. The number of APs in the first 100-ms epoch was not statistically different between epileptics (2.4 ± 0.4) and controls (3.0 ± 0.4), and the number of APs in the second 100-ms epoch was identical in epileptics (1.3 ± 0.4) and controls (1.3 ± 0.4). However, more APs were present in the first 100-ms epoch than the second 100-ms epoch in both epileptics and controls (P < 0.01), consistent with SFA.

 
View this table:
[in this window] [in a new window]
  		TABLE 1. Basic cellular properties of EC deep layer cells


View larger version (43K):
[in this window]
[in a new window]
  		FIG. 2. Examples of intracellular deep layer entorhinal cortex responses to 0.1-nA current steps. They inconsistently contained spike frequency adaptation (A and B), an inward rectifying current (A and D), and rebound depolarization with an action potential (A). A and B are from control and C and D from epileptic animals.

Evoked intracellular responses of the control and epileptic groups

There were profound differences between control and epileptic groups with regard to evoked intracellular responses (Table 2). All orthodromic responses in controls contained a short-duration EPSP, which progressively increased in amplitude with increasing stimulus intensity until a single AP was induced. Most cells contained an early IPSP and a late IPSP (Fig. 3A). Early IPSPs were identified as GABAA mediated based on their time to peak near 20 ms and reversal potential near -70 mV. Late IPSPs were identified as GABAB mediated based on their time to peak near 110 ms, reversal potential near -90 mV, and abolition by saclofen (n = 3). Importantly, some cells contained only early IPSPs without late IPSPs (Fig. 3B) or did not contain any IPSPs (Fig. 3C). In controls, both angular bundle and superficial layer stimulation most often produced orthodromic responses, which were identical in all respects, regardless of stimulation site. Antidromic responses consisting of an AP and IPSPs, without EPSPs, were recorded in 30% of responses to superficial layer stimulation and 32% of responses to angular bundle stimulation (P = 1.00 by Fisher's exact test). The only difference between angular bundle and superficial layer stimulation was the AP stimulation threshold, which was much lower for superficial stimulation than for angular bundle stimulation (P = 0.0007), probably because the superficial layer stimulation site was much closer than the angular bundle stimulation site to the recording electrode.

 
View this table:
[in this window] [in a new window]
  		TABLE 2. Characteristics of deep layer EC intracellular evoked responses


View larger version (16K):
[in this window]
[in a new window]
  		FIG. 3. Typical intracellular deep layer entorhinal cortex responses. A-C: responses from control animals always contained a short-duration excitatory postsynaptic potential (EPSP) and a single action potential and most responses contained an early gamma -aminobutyric acid-A (GABAA)-mediated inhibitory postsynaptic potential (IPSP; IP1) and a late GABAB-mediated IPSP (IP2) as in A, but some responses contained only a GABAA-mediated IPSP as in B, and some contained no IPSPs as in C. D-F: responses from epileptic animals were always very prolonged with multiple action potentials, and no IPSPs were observed. In this and all subsequent figures, responses were evoked by a single shock and stimulation artifact has been removed.

Evoked responses of the epileptic group were morphologically and statistically very different from controls in all response parameters measured (Table 2). All responses were very prolonged compared with controls. None were <100 ms in duration, and there was no overlap with the range of response duration for controls (Fig. 3, D-F). All responses had multiple APs, and no early or late IPSPs were seen. A few cells had afterhyperpolarization, but this was not consistent (Fig. 3D). Some epileptic responses were very complex and multiphasic, demonstrating an initial sustained depolarization followed by rhythmic depolarizations, which may be termed "clonic" (Fig. 4). Responses were morphologically identical regardless of stimulation site within one animal but varied in duration and complexity between different animals. Responses evoked by angular bundle stimulation were identical to superficial stimulation in all parameters measured, but the AP stimulation threshold was less for superficial stimulation (P = 0.03), similar to controls.


View larger version (20K):
[in this window]
[in a new window]
  		FIG. 4. Typical extracellular (ExC) deep layer entorhinal cortex response with simultaneous intracellular (IC) recording from an epileptic animal in response to a single shock delivered to the angular bundle. IC depolarizations correlate with ExC, primarily negative, deflections. Note initial prolonged "tonic" depolarization followed by "clonic" depolarizations. Total response duration of this cell to a single stimulus was 800 ms.

In the epileptic group, most APs occurred during the first several milliseconds after stimulation (Fig. 3, D and E). AP firing patterns demonstrated a progressive decrement in spike frequency during evoked responses, which was more robust but generally similar to SFA during depolarizing current steps. To roughly quantitate the progressive decrement, we compared the number of APs in response to a 0.4-nA depolarizing current step to the number evoked by angular bundle stimulation. Analysis was limited to a sample of 12 epileptic cells with total response durations of >200 ms. During the first 100 ms, more APs were present in the evoked response (4.3 ± 0.29) than were induced by the current step (2.4 ± 0.36), with P < 0.05 by paired t-tests. No APs were present by the second 100-ms epoch in 9 (75%) of the current steps and 10 (83%) of the evoked responses. Two of three cells without SFA during current steps had sustained repetitive firing during the evoked response. Figure 2D is a cell without SFA in response to current steps, and Fig. 3F is the same cell's response to stimulation.

Responses were elicited with less stimulation and evolved in a different manner in epileptic slices despite electrode placement identical to controls. The AP stimulation threshold to angular bundle stimulation was lower than in controls (Table 2) so that responses were elicited with less stimulation. The AP stimulation threshold to superficial layer stimulation was also lower than in controls, but this did not reach statistical significance, probably because the absolute voltages delivered were so small that it was difficult to deliver them in sufficiently small steps to distinguish a difference between epileptic and control. Epileptic responses did not demonstrate a progressive increase in amplitude with progressive increases in stimulus intensity. Instead, all cells demonstrated an all-or-none response to stimulation. The first response recorded with a progressive increase in stimulus intensity was of maximum magnitude despite very small increments in the intensity. For example, Fig. 5 illustrates the response to small increases in stimulus intensities obtained from voltage steps of 0.1 V from 0 to 0.4 V. No response was obtained at 0.2 V, and a maximum response was obtained at 0.3 V; no increase in response magnitude was obtained from increases greater than this. Paradoxically, increases in stimulus intensity inconsistently slightly decreased the duration of the response.

Evoked extracellular responses of the control and epileptic groups

Control responses had simple biphasic or triphasic morphology and usually consisted of a brief short latency negative spike (termed E1), followed by a longer latency negative spike (termed E2) and a positive or negative slow potential (termed E3; Fig. 6A). E1 had features suggestive of an antidromic population spike, including no decrement in response to repetitive stimulation at 100 Hz and preservation during blockade of excitatory neurotransmission by APV and DNQX. E2 and E3 had features that suggested they were orthodromically mediated, including a decrement in amplitude over ~10 stimuli during 100-Hz stimulation and abolition by blocking of excitatory neurotransmission by APV and DNQX. Simultaneous intracellular and extracellular recording demonstrated a close parallel between intracellular and extracellular potentials. E1 occurred at the same time as an antidromic AP and E2 at the same time as an orthodromic AP. E3 had a less clear relationship to intracellular potentials. However, not all extracellular responses contained all three components. Some responses only contained a prominent E1 and a slow negative potential, possibly suggesting more influence of antidromic stimulation. Regardless of morphology, all total response durations were relatively short (50 ± 7 ms, n = 18). Response morphology was not influenced by stimulation site; stimulation at the angular bundle produced "antidromic" responses as often as stimulation at superficial EC layers.


View larger version (9K):
[in this window]
[in a new window]
  		FIG. 6. Typical extracellular (ExC) deep layer entorhinal cortex responses with simultaneous intracellular (IC) recording from controls. A: most common morphology of the ExC response, containing a short-latency and short-duration negative potential (E1), a longer latency but short-duration negative potential (E2), and a longer duration positive potential (E3) in response to stimulation of the agnular bundle. Note E2 corresponds with an orthodromic action potential in IC tracing. B: ExC response containing a negative E3 overriding E2 in response to stimulation of the superficial layers. Note E1 corresponds to an antidromic action potential and E2 to an orthodromic action potential in IC tracing.

Extracellular responses in the epileptic group were very prolonged and multiphasic (Fig. 4) compared with controls with total durations similar to intracellular responses. The responses were often complex, with irregular positive and negative phases. As anticipated from intracellular recordings, no identifiable E1, E2, or E3 was present. The peak amplitude of the extracellular response from epileptic animals (0.5 ± 0.2 mV) was less than even the lowest amplitude component of the response from controls (0.85 ± 0.1 mV for the E2 component, P < 0.02). Extracellular responses closely paralleled intracellular responses (Figs. 4 and 5), and paroxysmal depolarizations correlated with extracellular spikes whether or not an AP was recorded.

Effect of isolating EC from hippocampus

To remove potential hippocampal reverberatory influences on the increased duration of epileptic responses, EC was severed from the rest of the slice by a fine knife cut along a line through the hippocampal sulcus, perforant pathway, and subiculum during recording of responses to superficial layer stimulation. There were no changes in the recorded extracellular responses in the control (n = 3) or epileptic (n = 3) groups. Responses to angular bundle stimulation could not be consistently assessed because of the proximity of the stimulation site to the knife cut. Intracellular responses were not assessed because the intracellular electrode was consistently displaced from the cell during the lesioning procedure.

Effect of ionotropic glutamate antagonists on intracellular responses

The NMDA antagonist APV was washed into the bathing ACSF during intracellular recording in six cells from epileptic animals (Fig. 7, left). The depolarizing envelope decreased in duration by an average of 22%, from 104 ± 17.8 to 81.8 ± 12.3 ms (P = 0.03 by paired t-tests). However, responses did not return to normal and no IPSPs appeared. A similar effect was noted on total response durations in extracellular recordings. DNQX alone or in combination with APV abolished all orthodromic responses to angular bundle or superficial layer stimulation. In a few cells, antidromic action potentials persisted. APV also was applied to six cells from control animals (Fig. 7, right) ,and no statistically significant difference was noted in EPSP width (8.4 ± 3.0 ms before application and 9.4 ± 2.7 ms after application).


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 7. Effect of the N-methyl-D-aspartate (NMDA) antagonist D(-)-2-amino-5-phosphonovaleric acid (APV) on simultaneous intracellular (IC) and extracellular (ExC) deep layer entorhinal cortex responses from epileptic (left) and control (right) animals. Responses before drug administration (top) and after 30- to 60-min application of 50 µM APV (bottom). Note that total response duration is decreased in epileptics but response does not return to normal. Control responses were not altered.

Monosynaptic IPSP protocol in epileptic and control groups

The lack of IPSPs in epileptic responses suggests the possibility that inhibitory neurons are dysfunctional. To determine whether local inhibitory interneurons were connected to primary cells, we used the monosynaptic IPSP protocol to directly stimulate local inhibitory interneurons by "near-site" stimulation in the presence of blockade of excitatory neurotransmission with APV and DNQX. Figure 8 illustrates intracellular responses using the monosynaptic IPSP protocol in control and epileptic animals. Before drug administration, near-site stimulation evoked responses identical to far-site stimulation; IPSPs were seen in control responses but not in epileptic responses. In the presence of APV and DNQX, IPSPs persisted in the controls and appeared in the epileptic group (Fig. 8). The kinetic parameters and amplitudes of IPSPs elicited from the epileptic group in this manner (Fig. 9) were not statistically different from controls (Table 3). Early IPSPs were GABAA mediated and late IPSPs were GABAB mediated based on their kinetic parameters, reversal potentials, and pharmacological responsiveness.


View larger version (11K):
[in this window]
[in a new window]
  		FIG. 8. Intracellular deep layer entorhinal cortex responses using the "monosynaptic protocol" in control (left) and epileptic animals (right). In response to far-site stimulation before drug administration, typical short-duration responses are seen in control and prolonged responses in epileptic animals. Responses to near-site stimulation before drug administration are identical to far-site stimulation. In response to the monosynaptic IPSP protocol consisting of near-site, direct inhibitory interneuron stimulation in the presence of the NMDA antagonist APV and the non-NMDA antagonist 6,7-dinitroquinoxaline-2-3-dione (DNQX), IPSPs are seen in both control and epileptic animals. An early GABAA-mediated IPSP was consistently present and a late GABAB-mediated IPSP was inconsistently present in both control and epileptic animals. In the example shown, the early IPSP may appear to decay faster in the epileptic trace, but there was no statistically significant difference between epileptic and controls in the half-time of decay.

 
View this table:
[in this window] [in a new window]
  		TABLE 3. IPSP characteristics using the monosynaptic IPSP protocol

    DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References

Deep layer EC cells are hyperexcitable or epileptiform

The most important finding of this study is marked hyperexcitability of deep layer EC cells, which is not due to interactions with the hippocampus or to alterations in the basic membrane characteristics examined and is only partially attributable to NMDA-mediated mechanisms. In addition, we found evidence that IPSPs may be present but hidden beneath prolonged excitatory responses.

Responses from epileptic animals were profoundly abnormal, containing very prolonged depolarizations with multiple APs and no IPSPs; this is similar to the classic paroxysmal depolarization shift (PDS) which underlies the epileptic spike recorded by surface electroencephalograms (EEG) (Ayala et al. 1973). In contrast, control animals demonstrated relatively brief EPSPs, a single AP, and inconstant IPSPs, similar to previous reports in normal animals (Jones and Heinemann 1988). Despite being more robust, less stimulation was required to elicit epileptic responses, so the effect is clearly not the result of higher stimulation intensities in epileptic animals. They also demonstrated an all-or-none response to progressive increases in stimulus intensity. These findings may be characterized as hyperexcitable because they predispose the cell to discharge more robustly.

There was a consistent close correlation between intracellular depolarizations and extracellularly recorded paroxysmal potentials in simultaneous intracellular and extracellular recordings. However, field responses were of lower amplitude in epileptic animals, possibly for several reasons. First, there are fewer cells in the EC of the epileptic rat to generate a response. Second, there may be temporal dispersion from multiple cells firing slightly dysynchronously. Third, the responses were very complex and multiphasic, so it was difficult to find the maximum amplitude at which a sustained response occurred.

Many intracellular responses, which were evoked by only a single stimulation, were very complex and contained an initial sustained depolarization followed by rhythmic bursts of depolarization. This progression is remarkably similar to the scalp EEG progression of a tonic-clonic seizure, during which there is a brief period of sustained fast activity during the tonic phase and then rhythmic bursts of long duration spikes during the clonic phase. Therefore, the intracellularly recorded hyperexcitability may be reflected on the scalp EEG as interictal or ictal discharges, which may be termed epileptiform.

The current findings are similar to our previous findings in epileptic superficial EC, which also demonstrated prolonged depolarizations without IPSPs (Bear et al. 1996), but differ in some respects. First, some deep EC cells from controls lacked IPSPs, something that has been reported by others (Jones and Heinemann 1988). Second, some deep EC cells had rebound APs, which raises altered AP generation as a possible mechanism of hyperexcitability. However, epileptic cells did not have a greater tendency to rebound APs or an altered response to depolarizing current, which suggests that this does not contribute to hyperexcitability. Third, we did not record any substantial IPSPs in epileptic tissue. The findings are also similar to other workers (Bear and Lothman 1993; Jones 1989; Jones and Heinemann 1988; Jones and Lambert 1990), who found prolonged depolarization of superficial and deep EC cells in low magnesium.

EC appears to be more hyperexcitable, as defined in previous sections, than other areas studied. The responses are much longer in duration and are more complex than those of hippocampal region CA1 or dentate gyrus in this model (Bekenstein and Lothman 1993; Mangan et al. 1995; Rempe et al. 1995) or hippocampus in the postkainic acid model (Ashwood et al. 1986; Franck and Schwartzkroin 1985). What role this has in the pathophysiology of seizures is uncertain. However, it is notable that this is the first region studied where responses are sufficiently prolonged to explain interictal epileptic spike and sharp waves, which are 20-200 ms in duration when recorded by scalp EEG in humans (Chatrian et al. 1974).

Overall, passive membrane characteristics of both control and epileptic cells were similar to layer IV cells described in detail by Jones and Heinemann (1988), who identified three types of cells based on their AP firing patterns. Most of our cells correlate with "nonbursting" neurons, which were described with a slightly more hyperpolarized RMP of -71 mV, a similar IR of 44 MOmega , demonstrated inward rectification in 11% of cells with hyperpolarizing current injection, usually lacked a postspike train AHP, and had one or two spikes in response to subicular stimulation. The main difference is that we found more cells with SFA. Similar to their report, we found only a few cells without any SFA and with rapid spiking, correlating with "fast-spiking" neurons, which were too few to characterize in detail. We did not identify any "bursting" neurons, but it is possible that we did not depolarize cells enough to detect this characteristic.

Mechanisms of hyperexcitability

Hyperexcitability of EC in this model is not due to changes in basic cellular properties, including RMP, input resistance, SFA, inward rectifying currents, and incidence of rebound APs because they were not altered in epileptic cells. The same number of APs and AP firing patterns were present during current steps in epileptics and controls, which suggests that AP propagation is not altered in epileptics. More APs were evoked by stimulation than by a current step within epileptic cells, supporting the contention that synaptically mediated mechanisms, and not intrinsic AP propagation, are responsible for hyperexcitability. However, sharp electrode recordings have several limitations in measuring basic cellular properties because currents with reversal potentials near the RMP or AP threshold, including some calcium, potassium, and sodium currents, may be masked so that other alterations in basic cellular properties could be present but not detected by the methods used here. Finally, hyperexcitability also is not due to interactions with the hippocampus because responses were not changed by severing the hippocampus from EC.

NMDA-mediated glutamatergic neurotransmission is a common and well-studied form of pathological or "excess" excitation in human epilepsy and animal models including kindling, electrical status epilepticus, and kainate- and pilocarpine-induced seizures (Bertram and Lothman 1990; Croucher et al. 1982; Fariello et al. 1989; Meldrum 1994; Millan et al. 1988; Mody and Heinemann 1987; Okazaki et al. 1989; Piredda and Gale 1986). We found a reduction in duration of epileptic responses by NMDA blockade with APV, but responses did not return to normal and IPSPs did not appear. This makes it unlikely that NMDA-mediated mechanisms are solely responsible for hyperexcitability. We did not find a significant effect of APV on control responses, but our analysis was powered only to detect very profound effects. In addition, NMDA receptors are activated preferentially by sustained depolarization, as occurred with epileptic responses, so that they may not be detectable by these methods during normal synaptic transmission. Therefore, we cannot determine whether preexisting NMDA receptors are activated by other excitatory mechanisms in the epileptic state or whether they are expressed pathologically in the epileptic state. We can only determine that they mediate some of the prolonged depolarization. Other mechanism of excitation that may contribute to hyperexcitability include metabotropic glutamate receptors (Akiyama et al. 1992), non-NMDA-mediated calcium currents (Iino et al. 1990), or excitatory GABA neurotransmission (Olsen and Avoli 1997).

It is possible that altered local network connections could contribute to hyperexcitability by selective loss of inhibitory interneurons or by sprouting of local recurrent collaterals which synapse onto primary cells, analogous to synaptic reorganization of dentate mossy fibers (Cronin et al. 1992), although there has been no evidence reported of this occurring in EC. We did not collect anatomic data or seizure frequency so we cannot correlate the degree of cellular hyperexcitability with degree of cell loss.

The lack of IPSPs in epileptic responses may suggest that inhibition is impaired and allows normal or pathological excitation to continue unchecked; however, we found evidence that impaired inhibition is not the sole source of hyperexcitability. First, we found some normal deep layer EC cells do not have IPSPs, similar to other authors (Jones 1987, 1989; Jones and Heinemann 1988). A lack of IPSPs cannot be the sole pathological mechanism of hyperexcitability for these cells because they do not contain IPSPs normally. Second, the monosynaptic IPSP protocol revealed IPSPs in epileptic neurons in response to near-site (direct interneuron) stimulation when excitatory neurotransmission was blocked; this demonstrates that some inhibitory interneurons are present in deep EC, and they form GABAergic synapses onto primary cells. Similar findings have been reported in this model in EC superficial layers (Bear et al. 1996) and hippocampus (Bekenstein and Lothman 1993). This also demonstrates that near-site stimulation effectively stimulates local inhibitory interneurons, yet IPSPs were not seen in response to near-site stimulation when excitation was not blocked. This suggests that GABAergic neurons properly synapse onto primary cells, but their effects are masked or overwhelmed by pathological excess excitation. Antagonism of NMDA-mediated excitation may be expected to reveal IPSPs if they are masked by this activity, but depolarization remained prolonged after addition of APV (81 ± 12 ms) and extended beyond the typical duration of GABAA-mediated IPSPs, so that they still may be masked.

Inhibition may be impaired even if IPSPs are "hidden" in prolonged depolarization and inhibitory neurons are present (Nakajima et al. 1991; Sloviter 1987, 1991; Williams et al. 1993). Several potential mechanisms of impaired inhibition are not tested by the monosynaptic IPSP protocol. First, there may be a selective loss of GABAergic neurons resulting in fewer inhibitory synapses. Second, subtle GABA receptor changes, such as altered kinetics or affinity may occur. Although GABAB IPSP characteristics were not statistically different between control and epileptic groups, two of five epileptic cells lacked GABAB IPSPs. Therefore, the data do not definitively exclude a loss of GABAB IPSPs in the epileptic group. Third, novel local network changes could occur in the epileptic state, such that inhibitory interneurons themselves are "epileptiform" and synapse onto adjacent inhibitory interneurons, resulting in inhibition of inhibition (Colder et al. 1996). We cannot determine whether we recorded from any interneurons because we did not identify neurons histologically and there are no consistent electrophysiologic characteristics to distinguish primary cells from interneurons in EC (Alonso and Klink 1993; Jones 1994). However, every cell we examined from epileptic animals demonstrated an epileptic response so that if we did record from interneurons, then the response was abnormal.

Conclusions

These results demonstrate that deep layers of EC are hyperexcitable and epileptiform in the post-SSLSE rat model of chronic temporal lobe epilepsy with very prolonged depolarizations in response to a single shock, resembling the classic PDS or possibly even a tonic-clonic seizure. NMDA-mediated excitation accounts for only a portion of hyperexcitability. The lack of IPSPs may suggest impaired inhibition, but the monosynaptic IPSP protocol demonstrated that inhibitory interneurons in the deep layers of EC are connected to their target cells by inhibitory synapses, suggesting that the primary location of inhibitory dysfunction is not in this connection. Additional studies are needed to determine whether other forms of excess excitation are present and the degree and location of impaired inhibition in the epileptic EC and whether the findings are applicable to human epilepsy. However, the findings further emphasize the importance of extrahippocampal sites in temporal lobe epilepsy.

    ACKNOWLEDGEMENTS

  We thank J. Williamson for expert technical assistance.

  This work was supported by National Institute of Neurological Disorders and Stroke Grant NS-25605 and the Epilepsy Foundation of America.

    FOOTNOTES

✠   Deceased 15 April 1995

  Address for reprint requests: Nathan B. Fountain, Dept. of Neurology, PO Box 394, University of Virginia Health Sciences Center, Charlottesville, VA 22908.

  Received 29 September 1997; accepted in final form 25 February 1998.

    REFERENCES
Abstract
Introduction
Methods
Results
Discussion
References
0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society



This article has been cited by other articles:


Home page
 J. Neurophysiol.Home page
C. M. Queiroz, J. A. Gorter, F. H. Lopes da Silva, and W. J. Wadman
Dynamics of Evoked Local Field Potentials in the Hippocampus of Epileptic Rats With Spontaneous Seizures
J Neurophysiol, March 1, 2009; 101(3): 1588 - 1597.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
P. de Guzman, Y. Inaba, E. Baldelli, M. de Curtis, G. Biagini, and M. Avoli
Network hyperexcitability within the deep layers of the pilocarpine-treated rat entorhinal cortex
J. Physiol., April 1, 2008; 586(7): 1867 - 1883.
[Abstract] [Full Text] [PDF]

Home page
 J. Pharmacol. Exp. Ther.Home page
A. W. Hendricson, R. E. Maldve, A. G. Salinas, J. W. Theile, T. A. Zhang, L. M. Diaz, and R. A. Morrisett
Aberrant Synaptic Activation of N-Methyl-D-aspartate Receptors Underlies Ethanol Withdrawal Hyperexcitability
J. Pharmacol. Exp. Ther., April 1, 2007; 321(1): 60 - 72.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurophysiol.Home page
E. Labyt, L. Uva, M. de Curtis, and F. Wendling
Realistic Modeling of Entorhinal Cortex Field Potentials and Interpretation of Epileptic Activity in the Guinea Pig Isolated Brain Preparation
J Neurophysiol, July 1, 2006; 96(1): 363 - 377.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
E. A. Tolner, F. Kloosterman, E. A. van Vliet, M. P. Witter, F. H. L. da Silva, and J. A. Gorter
Presubiculum Stimulation In Vivo Evokes Distinct Oscillations in Superficial and Deep Entorhinal Cortex Layers in Chronic Epileptic Rats
J. Neurosci., September 21, 2005; 25(38): 8755 - 8765.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurophysiol.Home page
D. X. Zhang and E. H. Bertram
Different Reactions of Control and Epileptic Rats to Administration of APV or Muscimol on Thalamic or CA3-Induced CA1 Responses
J Neurophysiol, November 1, 2003; 90(5): 2875 - 2883.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
M. Kobayashi, X. Wen, and P. S. Buckmaster
Reduced Inhibition and Increased Output of Layer II Neurons in the Medial Entorhinal Cortex in a Model of Temporal Lobe Epilepsy
J. Neurosci., September 17, 2003; 23(24): 8471 - 8479.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
M. Rizzi, S. Caccia, G. Guiso, C. Richichi, J. A. Gorter, E. Aronica, M. Aliprandi, R. Bagnati, R. Fanelli, M. D'Incalci, et al.
Limbic Seizures Induce P-Glycoprotein in Rodent Brain: Functional Implications for Pharmacoresistance
J. Neurosci., July 15, 2002; 22(14): 5833 - 5839.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurophysiol.Home page
M. D'Antuono, R. Benini, G. Biagini, G. D'Arcangelo, M. Barbarosie, V. Tancredi, and M. Avoli
Limbic Network Interactions Leading to Hyperexcitability in a Model of Temporal Lobe Epilepsy
J Neurophysiol, January 1, 2002; 87(1): 634 - 639.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurophysiol.Home page
M. Barbarosie, J. Louvel, I. Kurcewicz, and M. Avoli
CA3-Released Entorhinal Seizures Disclose Dentate Gyrus Epileptogenicity and Unmask a Temporoammonic Pathway
J Neurophysiol, March 1, 2000; 83(3): 1115 - 1124.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fountain, N. B.
Right arrow Articles by Lothman, E. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fountain, N. B.
Right arrow Articles by Lothman, E. W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online