- Email: [email protected]
Study of spontaneous recurrent seizures and morphological alterations after status epilepticus induced by intrahippocampal injection of pilocarpine
Epilepsy & Behavior 20 (2011) 257–266
Contents lists available at ScienceDirect
Epilepsy & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ye b e h
Study of spontaneous recurrent seizures and morphological alterations after status epilepticus induced by intrahippocampal injection of pilocarpine M.A. Furtado 1, O.W. Castro, F. Del Vecchio, J.A. Cortes de Oliveira, N. Garcia-Cairasco ⁎ Physiology Department, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, São Paulo, Brazil
a r t i c l e
i n f o
Article history: Received 28 June 2010 Revised 22 November 2010 Accepted 25 November 2010 Available online 14 January 2011 Keywords: Epilepsy Hippocampus Neurodegeneration Temporal lobe epilepsy Animal model
a b s t r a c t Epileptic seizures are clinical manifestations of neuronal discharges characterized by hyperexcitability and/or hypersynchrony in the cortex and other subcortical regions. The pilocarpine (PILO) model of epilepsy mimics temporal lobe epilepsy (TLE) in humans. In the present study, we used a more selective approach: microinjection of PILO into the hilus of the dentate gyrus (H-PILO). Our main goal was to evaluate the behavioral and morphological alterations present in this model of TLE. Seventy-six percent of all animals receiving H-PILO injections had continuous seizures called status epilepticus (SE). A typical pattern of evolution of limbic seizures during the SE with a latency of 29.3 ± 16.3 minutes was observed using an analysis of behavioral sequences. During the subsequent 30 days, 71% of all animals exhibited spontaneous recurrent seizures (SRSs) during a daily 8-hour videotaping session. These SRSs had a very conspicuous and characteristic pattern detected by behavioral sequences or neuroethological analysis. Only the animals that had SE showed positive Neo-Timm staining in the inner molecular layer of the dentate gyrus (sprouting) and reduced cell density in Ammon's horn pyramidal cell subfield CA1. However, no correlation between the intensity of sprouting and the mean number and total number of SRSs was found. Additionally, using FluoroJade staining, we observed neurodegeration in the hilus and pyramidal cell subfields CA3 and CA1 24 hours after SE. These data indicate that H-PILO is a reliable, selective, efficient, low-mortality model that mimics the acute and chronic behavioral and morphological aspects of TLE. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Seizures induced by the systemic injection of pilocarpine (PILO) are a known model of temporal lobe epilepsy (TLE). In the current study, we used a more specific approach: a model of seizure induction established by microinjection of PILO into the hilus of the dentate gyrus (H-PILO). Our main goal was to associate behavioral and morphological alterations of status epilepticus (SE) and spontaneous recurrent seizures (SRSs) induced by H-PILO. To analyze behavioral seizures, we used neuroethological tools because it is not possible to detect specific behavioral interactions using only the Racine seizure severity index [2]. We have used neuroethology to study different models of epilepsy, such as acute audiogenic seizures [3–5], audiogenic kindling [6], and SE induced with systemic pilocarpine (S-PILO) [7]. More recently, we also used ⁎ Corresponding author. University of São Paulo—USP, Ribeirão Preto School of Medicine, Av. Bandeirantes, 390014049-900 Ribeirão Preto, SP, Brazil. Fax: + 55 16 3633 0017. E-mail addresses: [email protected], [email protected] (M.A. Furtado), [email protected] (N. Garcia-Cairasco). 1 Current address: Clinical Research Management, Inc., Department of Closed Head Injury, Brain Dysfunction and Blast Injury Division, Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910, USA. Fax: + 1 301 319 9810. 1525-5050/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2010.11.024
neuroethology to characterize complex partial seizures in patients with TLE [8]. In particular, the model of SE induced by S-PILO has its behavioral seizure progression well characterized through neuroethology and interstrain differences can also be discriminated [7]. However, the high mortality rate in the model of S-PILO is a limiting factor. We used Neo-Timm, Nissl histochemistry and Fluoro-Jade (FJ) staining [9] to assess neuroplastic changes and cellular degeneration after H-PILO. Mossy fiber sprouting (MFS), detected as exacerbated positive Neo-Timm staining [10], is observed in the inner molecular layer of the dentate gyrus of epileptic animals [11] and humans with TLE [12,13]. This MFS seems to occur after mossy cell death and deafferentation of the basket cells located in the hilus [14]. FJ + cells are neurodegenerating neurons and, in this case, were detected in subfields of the hippocampal formation. Considering the wide range of morphological modifications that occur after SE induced with S-PILO, it is controversial to argue specifically that the synaptic rearrangements in the hippocampus are responsible for the SRSs. However, some authors maintain that the MFS of the dentate granule cells contributes to epileptogenesis in animals that exhibit SRS. The functional consequence of MFS remains unknown. This synaptic reorganization in the hippocampal formation might be implicated in the consequent epileptogenesis in animal models of epilepsy and patients with epilepsy [15]. In contrast, MFS
258
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
could also represent a compensatory or inhibitory response [16,17]. A great deal of controversy surrounds the interpretation of the functional role of this sprouting [18–23]. Thus, in the context of the aforementioned studies, determining the morphological alterations at the synaptic and molecular levels that occur in epileptic animals is still challenging. In the current study, we investigated the acute neuroanatomical changes induced by H-PILO using FJ and whether SE chronically induced MFS and cell loss, using Neo-Timm and Nissl staining, respectively. The presence of SRS was studied, and their behavioral patterns were measured through neuroethological tools. Finally, in light of all the controversy over the role of MFS in the occurrence of SRSs, we determined if the frequency of SRS was related to MFS. We tested the following hypotheses: (1) severe neurodegeneration can be observed 24 hours after SE induced with H-PILO, using FJ staining; (2) SE and SRSs induced with H-PILO can be characterized by a typical progression of limbic behavioral patterns; (3) MFS occurs in parallel with the occurrence of SRSs but is not related to the frequency or total number of SRSs; (4) epileptic animals exhibit a reduction in the cellular density of the CA1 region of Ammon´s horn. 2. Methods 2.1. Animals and surgical procedures All experiments were performed in accordance with the Brazilian Society for Neuroscience and Behavior guidelines based on those from the Society for Neuroscience and consistent with the guidelines set by the Colegio Brasileiro de Experimentação Animal (COBEA) and Institutional Ethics Committee (CETEA) Protocol 195/2005. Every effort was made to avoid unnecessary pain or stress to the animals. Adult male Wistar rats (n = 50) were anesthetized with tribromoethanol (Avertin, Sigma–Aldrich; St. Louis, MO, USA; 2,5%, 10 mL/kg, ip), placed in a stereotaxic apparatus (David Kopf Instruments, Tujunga, CA, USA), and implanted with a guide cannula in the left hippocampus (AP 6.30 mm, L 4.5 mm, V 4.5 mm, bregma reference) [24] by stereotaxic surgery. Animals were allowed to recover for 7 days before the administration of H-PILO through a cannula. 2.2. Microinjection procedures We used a 5-μL syringe (Hamilton, Sigma, Reno, NV, USA) connected to a microinjection pump (Harvard Apparatus, 2000, Holliston, MA, USA) adjusted to a flow of 0.5 μL/minute. After the end of microinjection, the syringe remained connected to the cannula for approximately 1 minute to avoid reflux of the drug. All animals were gently immobilized during the microinjection procedures. The experimental group received H-PILO (Sigma–Aldrich) through the cannula (n = 34; 2.4 mg/mL, 1 μL) [25]. Ninety minutes after SE onset, the animals were injected with diazepam (5 mg/kg, Roche, Sao Paulo, Brazil). A control group (n = 16) was injected with saline (1 μL). 2.3. Behavioral assessment of status epilepticus and spontaneous recurrent seizures Immediately after either saline or H-PILO microinjection, we recorded the behavioral activity of all animals on VHS tapes with a camcorder (Gradiente GCP-180CR, Sao Paulo, Brazil) up to 2 hours after SE onset. Behavioral sequences were obtained from the video analysis (see Neuroethological Evaluation of Status Epilepticus and Spontaneous Recurrent Seizures). We used a set of acrylic cages that allowed the simultaneous observation of six animals. To detect SRSs, we used the same cages mentioned above. The behavioral activity was also recorded on VHS tapes 8 hours per day for up to 30 days after SE.
We chose the daylight period because SRSs occur more frequently during the day [26]. 2.4. Neuroethological evaluation of status epilepticus and spontaneous recurrent seizures In this study, we used neuroethology-based methods on the premise that the analysis of behavioral sequences can provide us with more information than isolated behavioral measurements. Neuroethology allowed us to study the movement grammar and its temporal and hierarchical structure [27–29]. For a detailed analysis of behavioral sequences, we determined observation periods of 60 seconds according to specific behaviors based on the scale established by Racine [2] and modified by Pinel and Rovner [30] (see Table 1). These specific behaviors were called behavioral triggers. Behavioral triggers for SE are shown in Fig. 1A and were defined as follows: stage A, head myoclonus; stage B, forelimb myoclonus; stage C, rearing followed by falling; and stage D, jump. Behavioral triggers for SRSs are shown in Fig. 1B and were defined as follows: stage A, rearing; stage B, one rearing followed by falling; and stage C, three or more rearings followed by falling. This approach was used to provide markers for recording the onset of behaviors already validated in the evolution of severity indexes but that are expressed at variable times in different animals. These triggers have been used successfully in the characterization of SE induced with S-PILO in rats [7]. Using a dictionary of behavioral items [6,7,31], we recorded every behavior observed during each 60-second period using a second-bysecond analysis. We recorded the behaviors with the aid of a video capture card (ATI All in Wonder PRO, Elizabeth, NJ, USA), and the video was reviewed several times. For statistical analysis, we used a program called ETHOMATIC [31]. The ETHOMATIC program calculates and displays the mean frequency and mean duration of each behavioral item in the given observation window. The software also performs statistical analysis, verifying significant associations between pairs (dyads) of behavioral items and calculating a χ2 value. Using the data obtained with ETHOMATIC, we constructed flowcharts to graphically represent all statistically significant data (Fig. 2). Microsoft Power Point was used for the graphical representation of flowcharts. Fig. 2 shows the behavioral samples used to construct the flowcharts. The entire behavioral sequences are shown in Figs. 3 and 4. 2.5. Histological procedures All animals received a lethal dose of thionembutal (100 mg/kg). For FJ staining, we euthanized animals 24 hours after SE (n = 10, experimental group; n = 6 control group) as described by Schmued et al. [9]. Animals were transcardially perfused with 100 mL of phosphate buffer followed by paraformaldehyde (500 mL, 4%) dissolved in phosphate buffer. After transcardiac perfusion, all brains were removed, frozen, and sectioned in slices of 30 μm for FJ histochemistry and 20 μm for Nissl
Table 1 Limbic seizure severity indexes.a Class
Behavior
0 1 2 3 4 5 6 7 8
Immobility Facial automatisms Head myoclonus Forelimb myoclonus Rearing Falling More than three falls Wild running Tonic–clonic seizures
a Racine scale [2] or limbic index (LI) as modified by Pinel and Rovner [30].
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
259
Fig. 1. Behavioral sampling methods used in the construction of flowcharts. The sequential letters represent oscillating behavioral triggers corresponding to specific behaviors of the seizure scale [30]. (A) Status epilepticus: first head myoclonus = stage A; first forelimb myoclonus = stage B; first rear followed by falls = stage C; first jump = stage D. (B) Spontaneous recurrent seizures: first rear = stage A; first rear followed by fall = stage B; first rear followed by several falls = stage C.
staining. Brain sections were obtained using a Cryostat (Microm HM505-E, Microm International, Walldorf, Germany). For better quality and conservation of the tissue, the sections were mounted directly inside the cryostat onto gelatin-coated slides. Subsequently, the slides were stored in a freezer at –20 °C. For Nissl and Neo-Timm staining, we euthanized animals at 30 days (n = 12, experimental group; n = 5, control group) and 15 days (n = 12, experimental group; n = 5, control group) after SE. When the 15- and 30-day groups were combined, the experimental group comprised 24 animals and the control group 10 animals. Animals were transcardially perfused with 100 mL of Millonig's phosphate buffer (100 mL) followed by Millonig's phosphate buffer with sodium sulfide (100 mL, 0.1%), glutaraldehyde (100 mL, at 3%), and again Millonig's phosphate buffer with sodium sulfide (200 mL, 0.1%). 2.5.1. Fluoro-Jade staining Fluoro-Jade histochemistry was performed as follows: tissue was immersed in 100% ethanol for 3 minutes, followed by 1 minute in 70% ethanol and 1 minute in distilled water. After this step, slides were transferred to a solution of 0.06% potassium permanganate for 15 minutes and were gently shaken on a rotating platform. Slides were rinsed three times for 1 minute in distilled water and were then transferred to the FJ staining solution and gently shaken for 30 minutes. The 0.0001% working solution of FJ was prepared by adding 1 mL of stock FJ solution (0.01%) to 99 mL of 0.1% acetic acid in
distilled water. After staining, sections were rinsed three times (1 minute) in distilled water and slides were coverslipped. 2.5.2. Neo-Timm staining To study axonal sprouting in the hippocampus, we used Neo-Timm histochemistry [10,11]. This technique is based on the histochemical detection of zinc vesicles [32]. The aforementioned perfusion with Millonig's phosphate buffer with sodium sulfide is necessary to increase the affinity of silver nitrate for the zinc during the staining. 2.5.3. Nissl staining and assessment of cellular density Every adjacent section was stained with cresyl violet and was used to determine the general size, shape, and distribution of neuronal cell bodies, as well as localization of the cannula and confirmation of the lesion. All images were obtained using the same system employed for histochemistry (see above). Figs. 6C and D illustrate the sampling method used for optical density measurements of Neo-Timm staining, and Fig. 6H shows the sampling method used for Nissl-stained cell counts. Morphological criteria described by Williams and Rakic [33] were used to estimate the Nissl-stained cells. Pyramidal cells counted in the CA1, CA2, and CA3 subfields have a size and shape that are easily identifiable. Furthermore, the nuclei of neurons have very evident nucleoli. In contrast, glial cells usually have small and dark nucleoli. Neurons contain more abundant and granular Nissl substance than other types of cells. Cell bodies and nuclei of neurons have a sectional
260
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
Fig. 2. (A) Example of a short behavioral sequence (time–item) and its graphic representation as a flowchart. (B) Flowchart structure and calibration. The height of the rectangles represents the frequency of a behavioral item, and the base corresponds to the duration of the behavior in an observation window. The arrows represent statistical values (χ2 N 3:84, P b 0:05) equivalent to the degree of association between pairs of behaviors (dyads) in a sequence. The colors identify the main behavioral clusters. The use of colors and circles (B; see also Fig. 3) and the calibration of rectangles in the flowcharts are for illustration purposes and do not have any impact on the statistical analysis (see Methods for details). A complete description of behaviors is given in the legend to Fig. 3. Reproduced, with permission, from Garcia-Cairasco et al. [7].
area two to three times greater than that of glial cells, even in the early developmental stages. Although neurons belonging to a local circuitry were slightly larger than glial cells, neurons had more prominent nucleoli. In addition, nuclei of neurons have a rounded or oval shape, whereas the nuclei of glial cells have an irregular shape [33].
light intensity was maintained in the optical microscope, and we used only slices stained in the same batch for the Neo-Timm staining reaction. This particular care was taken to avoid random differences in the measured density of gray levels. The data were exported to Microsoft Excel for Windows software to perform statistical analysis.
2.6. Statistics 3. Results Full statistical analyses were performed using SIGMA-STAT Version 3.5 (Systat Software Inc, San Jose, CA, USA). Student's t test was used to compare different animal groups, according to number of cells (Nissl staining) and gray level (Neo-Timm staining). Pearson's correlation was calculated to verify the association between gray level and frequency of seizures. 2.7. Image analysis All images were digitized using an Optronics CCD camera (Santa Barbara, CA, USA) connected to a BX-60 Olympus microscope (Olympus; Melville, NY, USA). For quantitative analyses, we converted all images to a gray scale of 8 bits. The freeware Image Tool program (University of Texas, Version 1.27) was used to measure gray levels in the inner molecular layer of the dentate gyrus. The values obtained in the ventral thalamus were used as background. Using a gray scale of 256 values, we measured the integrated density of gray levels per area. In the next step, we subtracted the values obtained in the thalamus from the values obtained in the hippocampus. The same
3.1. Behavioral assessment of status epilepticus and spontaneous recurrent seizures Only animals for which cannula position was confirmed by histology were used. Among all the animals, 76% exhibited SE within 29.3 ± 16.3 minutes (mean ± SEM). Additionally, 71% of animals displaying SE had SRSs and survived until the day of sacrifice. In all groups, the acute behavioral effects caused by the administration of H-PILO consisted of initial piloerection followed by periods of whole-body tremor and/or myoclonic twitching restricted to the head and face, accompanied by salivation. SE was characterized by repeated cycles of head and forelimb myoclonus, rearing, and falling. As expected, no such events were observed in saline-injected animals. During SRSs, animals exhibited orofacial automatisms, head and forelimb myoclonus, and eventually rearing and falling. The evolution of the severity of SRSs resembles the severity observed during amygdala kindling [2], and the majority of seizures occurred on awakening. Animals with SE had elevated levels
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
261
of exploration and reactivity compared with control rats (qualitative observation; data not shown) during the first week after SE induction. 3.2. Neuroethological evaluation of status epilepticus and spontaneous recurrent seizures
Fig. 3. Flowcharts representing sequential stages of SE induced with H-PILO. (A–D) Stages A–D. AF, atonic fall; ELE, elevation; ERE, erect posture; FALL, falling (class 5 limbic seizures); IM, immobility (pause); JP, Jumping; MT, masticatory movements; MYO, myoclonic jerks (limbic); MYO1, myoclonic jerks, forelimb; MYOh, myoclonic jerks, head; MYOt, myoclonic jerks, trunk; REAR, rearing (class 4 limbic seizure); SCA, scanning; SNF, sniffing; STRA, straub tail; TCV, tonic seizures (opisthotonus); WA, walking; WDS, wet dog shakes.
Fig. 4. Flowcharts representing sequential stages of SRS induced with H-PILO. (A–C) Stages A–C. Abbreviations as for Fig. 3.
Throughout the neuroethological analysis of SE, we observed exploratory behaviors of low frequency and low duration during stage A (Fig. 3, with one head myoclonus as the behavioral trigger). Statistically significant interactions (arrows, χ2 log) between sniffing and scanning were also observed. The associations between the behaviors of head myoclonus, sniffing, and masticatory movements were also significant; however, the interactions between masticatory movements and head myoclonus were more prominent. Wet dog shakes occurred only during stage A (Fig. 3), but there were no significant interactions with other behaviors. During stage B (Fig. 3; with one forelimb myoclonus as the behavioral trigger), exploratory behaviors of low frequency and low duration also occurred. Reciprocal and significant interactions between limbic behaviors, such as between head myoclonus and forelimb myoclonus and between head myoclonus and masticatory movements, were observed. In stage C (Fig. 3; with one rearing followed by falling as the behavioral trigger), exploratory behaviors were rare. Limbic behaviors, such as head myoclonus and forelimb myoclonus, rearing, and falling, were common. Forelimb myoclonus–head myoclonus, myoclonic jerks (trunk)–head myoclonus, and rearing–falling interactions were statistically significant. During stage D (Fig. 3; with one jump as the behavioral trigger), exploratory behaviors were scarce. Reciprocal forelimb–head myoclonus interactions were statistically significant. The associations between walking, jumping, and falling were significant. Throughout the neuroethological analysis of SRSs, we observed (Fig. 4, with rearing as the behavioral trigger) a cluster of exploratory behaviors in stage A characterized by statistically significant interactions (arrows, χ2 log) between sniffing and scanning. Other behaviors were present in a pattern of reverberation. For example, we noticed strong interactions between masticatory movements, forelimb myoclonus, and head myoclonus closing a loop or behavioral cluster with masticatory movements; we also observed mutual interactions between rearing and head myoclonus and straub tail and forelimb myoclonus. Behaviors such as masticatory movements, head myoclonus and forelimb myoclonus were characterized by an elevated frequency. Immobility was observed to have an elevated duration and was weakly associated with masticatory movements. In stage B (Fig. 4, with one rearing followed by falling as the behavioral trigger), we noticed a cluster of exploratory behaviors characterized by interactions between sniffing and scanning. We also discriminated interactions between falling and straub tail. Interestingly, the associations between head myoclonus and forelimb myoclonus were mutual and more intense than during stage A (Fig. 4). The pattern of reverberation was not as clear as during stage A (Fig. 4), but forelimb myoclonus was characterized by elevated frequency and duration. Head myoclonus was characterized by an elevated frequency. We observed a slight association between immobility and masticatory movements. During stage C (Fig. 4, with three or more rearings followed by falling as the behavioral trigger), the cluster of exploratory behaviors was reduced, and we discriminated strong interactions between rearing and falling. The pattern of reverberation was very clear and was characterized by mutual interactions between masticatory movements and forelimb myoclonus and head myoclonus and masticatory movements, closing a loop. The interactions between head myoclonus and masticatory movements were less intense during stage C than during stage A and absent during stage B (Fig. 4). The frequency of head myoclonus was lower than than during stages A and B (Fig. 4).
262
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
Interestingly, in all stages, we observed the behaviors alert state and lying, and especially in stage A (Fig. 4), there was a statistically significant interaction between these behaviors.
saline and animals that did not exhibit SE did not have positive FJ staining (upper right inset Fig. 5A′).
3.3. Fluoro-Jade staining
3.4. Gray-level analysis of the axonal sprouting in the dorsal hippocampus
Animals (n = 16) were euthanized 24 hours after SE. A reduced background allowed us to easily identify stained neurons. We observed strong FJ staining in the hilus, CA3, and CA1 (Fig. 5) and also the amygdala (images not shown). Control animals treated with
Animals injected with saline (Figs. 6A and B) and animals that did not exhibit SE did not have sprouting. All animals with SE had sprouting in the inner molecular layer of the dentate gyrus (Figs. 6C–F). Apparently there was no association between the degree of sprouting
Fig. 5. Fluoro-Jade staining. FJ + neurodegeneration after 24 hours of SE induced with H-PILO. Confocal microscopy images: (A) hilus of dentate gyrus, (A′) zoom in hilus; (B) CA1, (B') zoom in CA1; (C) CA3, (C') zoom in CA3. Calibration bar: (A–C) 75 μm, (A´–C´) 25 μm. Note the absence of FJ staining in a control animal in the upper right inset of A´.
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
263
Fig. 6. Neo-Timm staining. (A, B) Control animals. (C, D) Epileptic animals. (A, C) Brain sections of animals sacrificed 15 days after SE. (B, D) Brain sections of animals sacrificed 30 days after SE. Calibration bar: 200 μm. Observe details in the insets (E, F). Calibration bar: 40 μm. (G) There is no difference in the gray levels in the inner molecular layer of the dentate gyrus of animals injected with H-PILO between the left (dark bars) and right (light bars) hippocampus. Animals that displayed SE (PILO 15 and 30 days) had increased gray level from either left or right hippocampus when compared with the respective control groups (*P b 0.01; Student's t test). (H) Sampling pattern used for cell counting in the dorsal hippocampus [24]. (I–L) Nissl staining. Sections in the coronal plane located approximately –3.3 mm from bregma. (I, K) Control animal: CA1 and CA3c, respectively. (J, L) Animal exhibiting SE: CA1 and CA3c, respectively. Arrow points to example of the reduction of cellular density in an epileptic animal (CA1). Calibration bar: 200 μm.
and the frequency of SRSs. Additionally, qualitative evaluation revealed that some animals had stronger staining in the site ipsilateral to PILO injection and in the ventral hippocampus.
Quantitative analysis of the sprouting (Fig. 6G) from both the left and right hippocampus revealed a statistically significant difference between the animals exhibiting SE and the controls (Student's t test,
Table 2 Mean numbers of pyramidal cells per area. Group
Mean (SEM) number of pyramidal cells/μm2 CA1 subfield Control
Left Right
0.0025 (0.0002) 0.0028 (0.0002)
CA2 subfield 15 days
30 days
a
a
0.0019 (0.0004) 0.0020b (0.0003)
0.0015 (0.0003) 0.0016b (0.0003)
CA3 subfield
Control
15 days
30 days
Control
15 days
30 days
0.0020 (0.0002) 0.0025 (0.0003)
0.0019 (0.0003) 0.0019 (0.0003)
0.0012 (0.0005) 0.0018 (0.0004)
0.0013 (0.0002) 0.0015 (0.0002)
0.0011 (0.0001) 0.0012 (0.0001)
0.0012 (0.0001) 0.0012 (0.0001)
a
P b 0.05, comparing the cellular densities in the left CA1 hippocampal subfield of the animals sacrificed 15 and 30 days after SE and control animals. P b 0.05, comparing the cellular densities in the right CA1 hippocampal subfield of the animals sacrificed 15 and 30 days after SE and control animals. Statistical analyses were performed using Student's t test: b
Left CA1: control vs 15 days, P = 0.041; control vs 30 days, P = 0.021. Right CA1: control vs 15 days, P = 0.048; control vs 30 days, P = 0.037. Left CA2: control vs 15 days, P = 0.892; control vs 30 days, P = 0.09. Right CA2: control vs 15 days, P = 0.159; control vs 30 days, P = 0.093. Left CA3: control vs 15 days, P = 0.097; control vs 30 days, P = 0.401. Right CA3: control vs 15 days, P = 0.071; control vs 30 days, P = 0.088.
264
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
P b 0.01). Also, there were no differences in the level of sprouting between the 15- and 20-day groups and the left (dark bars) and right (light bars) hippocampus (Student's t test, P = 0.795). We did not find any correlation between gray level and frequency and total number of seizures of each animal (Pearson's correlation, P N 0.05). 3.5. Cellular density analysis in the dorsal hippocampus Analysis of sections (Fig. 6H) stained for Nissl histochemistry indicated that, when compared with the control group (Fig. 6I; Table 2; P b 0.05, Student's t test), the experimental group (Fig. 6J) had a reduced density of cells in the CA1 field. Especially in the CA1 subfield of epileptic animals, we observed cells with pyknotic nuclei and an altered form. Some areas contained vacuoles, and disruption of the tissue and could not be analyzed. The CA3 subfield did not statistically significantly differ between control (Fig. 6K) and experimental (Fig. 6L) animals. We did not find any difference in cellular density between the 15and 30-day groups in the left and right hippocampus (Table 2). 4. Discussion A single dose of H-PILO (2.4 mg/μL) induced SE in the majority of animals with the typical evolution pattern described in our previous study [1], and the seizures resembled those induced with S-PILO [34,35] and PILO injected into the hippocampus with a microdialysis probe [36] or intracerebroventricularly [37]. Our results support the H-PILO model as a quite localized and efficient way to mimic experimentally TLE with low mortality. The daylight period was preferred for observations because evidence in the literature indicates that SRSs occur with a higher frequency during the day [26]. We verified that the severity of seizures increased with time. In fact, it has been suggested by several authors that a kindling mechanism operates during the development of SRSs [38–40]. In a previous study, we showed the behavioral evolution of SE and SRSs induced with H-PILO using known seizure severity indexes [1]. Throughout the current analysis of the flowcharts, we observed specific behavioral interactions not noticed previously when we used only the Racine [2] seizure severity scale. We also observed a strong variability in the onset of SE and SRSs. On this basis, we selected the onset of the observation windows according to the severity of seizures exhibited by each animal (i.e., behavioral triggers; see Fig. 1). We analyzed the behavioral associations occurring during SE and SRSs induced with H-PILO using neuroethological tools. Neuroethology has been validated in other studies that have evaluated models of epilepsy such as the development of audiogenic kindling in Wistar Audiogenic Rats (WARs, a genetically selected inbred strain [5,6]), seizures induced with S-PILO in WARs [7], and seizures induced by injection of crude venom from the spider Parawixia bistriata [41]. Leite and co-workers [42], reviewing the kainic acid and PILO models of TLE, proposed the use of this neuroethological technique to evaluate new antiepileptic agents. More recently, we used the same methodology to characterize seizures of patients with TLE [8]. Although its application is labor intensive and time consuming, the results obtained are worthwhile and allow the assessment of subtle alterations caused by antiepileptic drugs. In fact, there have been several studies of spontaneous seizures in humans using statistical tools [43–45]. Analysis of the flowcharts constructed using sequences from SE and SRSs induced with H-PILO suggested that a reverberant circuitry could contribute to the genesis of the observed limbic seizures. The strong statistical associations (χ2 log) observed at each stage of SE and SRSs corroborate our hypothesis (e.g., Fig. 3 stage B, interactions between MYOh, and MYO1; and Fig. 4 stage A, interactions between MT, MYOh, and MYO1). During stage A (Fig. 3), the sequence of head
myoclonus and wet dog shakes suggests that displaced behaviors (see also Fig. 4) are expressed probably because of the activation of endogenous anticonvulsant mechanisms [46]. When these mechanisms fail, the SE is fully installed (Figs. 4B–D). Interestingly, during stages C and D (Fig. 3), the occurrence of audiogenic-like behaviors, such as tonic–clonic convulsions, jumps, and atonic falling, suggests that brainstem structures are being activated during SE. During stages A and B (Fig. 4), the presence of strong exploratory clusters, now during the period of SRSs, could be indicative of the activation of endogenous antiseizure mechanisms. In other words, the exploratory behaviors might compete with ictal behaviors because we did not observe severe seizures in stages A and B (Fig. 4). In fact, in previous studies, we observed facial automatisms, compulsive grooming, and exploration during acute and chronic acoustic stimulation of animals that are not susceptible to audiogenic seizures [3,6]. In general, we observed that limbic behaviors increased in severity according to the Racine scale [2] as modified by Pinel and Rovner [30]. However, subtle alterations, such as the cluster of exploratory behaviors mentioned above, were undetectable using only linear scales or behavioral scores. Because the extension of the neural circuits involved in seizures and their behavioral expression may be related to synchronous activation of different brain networks, this epileptogenic network–behavior relationship is almost impossible to demonstrate using isolated behavioral observations. Therefore, neuroethology is useful for the detection of the development of specific behavioral associations and reverberative clusters. For example, the interactions between forelimb myoclonus (MYO1) and head myoclonus (MYOh) were more intense during stage C than during stage A (Fig. 4). The behaviors lying and alert state showed that the majority of seizures occurred after a sleep period. Such data have already been reported by Leite et al. [47] in a model of seizures induced with a high dose of S-PILO. We used FJ staining to determine the acute neurodegeneration pattern after H-PILO microinjection. We observed neuronal damage in CA1, CA3, hilus (Fig. 5), and amygdala (not shown). In fact, this is obviously only a partial account of the wider lesion described in rats by Turski et al. [48] after injection of the muscarinic agonist bethanecol into CA3. Briefly, these authors mention lesions of the piriform cortex, entorhinal cortex, olfactory tubercle, anterior olfactory nucleus, subiculum, amygdaloid complex, temporoparietal cortex, and hypothalamic nuclei. There are additional data indicating that S-PILO (in adult and young rats) can lead to neurodegeneration in several other brain regions 3 months after SE [34,49,50]. After analyzing the acute and chronic structural modifications in the dorsal hippocampus (FJ and Nissl data), we concluded that the pattern of cell lesions seen after H-PILO in our study confirm hilar, CA3, and CA1 neurodegeneration. Because in the current study we did not examine regions other than the hippocampal formation, we cannot say there were no distant lesions such as those induced after S-PILO in mice and detected, in addition to the hippocampus, in the amygdala, thalamus, olfactory cortex, neocortex, and substantia nigra by Turski et al. [51]. However, we have also shown, in a comparative study of S-PILO versus H-PILO, that lesions in the H-PILO model are narrower than those from the S-PILO model [52]. Interestingly, Romcy-Pereira and Garcia-Cairasco [53] showed that animals displaying audiogenic kindling have no FJ + cells. This result suggests that short seizures do not induce neurodegeneration, but prolonged SE is capable of activating neuroexcitotoxic mechanisms. Also, the duration of SE is proportional to the degree of neurodegeneration [54], indicating that a rapid blockade of SE is essential in avoiding neuronal loss and epileptogenesis. Finally, Colombo et al. [55] showed that FJ also stains reactive astroglia in the cerebral cortex of nonhuman primates. However, this limitation did not constitute a problem because astroglia can be easily distinguished from pyramidal cells and other types of neurons. We did
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
not find evidence of neuronal loss in animals injected with saline, indicating that the cannula causes only mechanical damage restricted to the site of implantation. The initial cholinergic overstimulation leads to a subsequent excessive activation of NMDA, which, in turn, increases the calcium influx that can lead to a cascade of molecular events causing neuronal death [56,57], mainly by necrosis [58]. This neuronal death is more evident after prolonged limbic seizures than after low-severity or atypical seizures [59,60]. We did not find significant cell reduction in CA3 with Nissl staining at the chronic time window, although we saw clear FJ labeling in this structure 24 hours after SE. The absence of a significant difference in cell density between the left and right hippocampus suggests that the seizures propagated to the contralateral site of injection. The lack of differences between the 15- and 30-day groups indicates that only the initial injury was necessary for the morphological alterations observed [38]. Mossy fiber sprouting was observed in all animals that exhibited SE. Furthermore, 72% of these animals had SRSs. The MFS observed in human TLE has been described in several studies [12,13] and occurs in parallel with SRSs. Also, MFS is a common finding in animals that have SRSs [61]. However, neither Sutula et al. [62] nor Mathern [13] observed associations between the duration of epilepsy and the density of MFS. When comparing the groups of animals sacrificed 15 and 30 days after H-PILO, we did not observe any significant difference in MFS. The absence of a correlation between MFS intensity and total number or frequency of SRSs suggested that MFS is not related to the pattern of SRSs. In fact, the absence of a correlation between the frequency of SRSs and MFS has been reported [11]. Furthermore, SRSs can occur prior to the development of MFS [63]. Although blockade of MFS by cycloheximide injection does not prevent the occurrence of SRSs in the model of systemic seizures induced with PILO [18] or in the model of seizures induced by injection of kainic acid into the hippocampus [19], Williams et al. [23] did not confirm those results. In the current study, we combined the neuroethological analysis of behavioral sequences with Neo-Timm, Nissl, and FJ histochemistry, which allowed the evaluation of behavioral alterations, density of cell loss, neurodegeneration, and density of MFS, respectively, in the dorsal hippocampus. Future experiments are underway to study the associations between behavioral and electroencephalographic alterations in limbic structures during SE and SRSs induced with H-PILO. In conclusion, neuroethological tools allowed us to identify every step of the motor acts to elucidate complex and simple sequences of movements. Although behavioral indexes are very useful, they should not be used alone but in combination with more precise behavioral evaluations, such as the sequential analysis of the semiological patterns of seizures induced with H-PILO. Combination of precise neuroethology tools and other methods, such as histochemistry (NeoTimm, Nissl, and FJ) and EEG, is necessary to improve the prediction of the neural substrates of epileptic seizures shown in the current study of H-PILO. Therefore, we consider the induction of SE with H-PILO a reliable TLE model with low mortality and a more restricted pattern of neurodegeneration than that with S-PILO. Acknowledgments This study was supported by FAPESP, FAPESP-Cinapce, CAPES (PROAP-PROEX), CNPq, PRONEX and FAEPA. M.A. Furtado was a PROAP-CAPES-Brazil Ph.D fellow. O.W. Castro is a CNPq-Brazil PhD fellow. N. Garcia-Cairasco and J.A.C. Oliveira hold CNPq-Brazil Research and Technician fellowships, respectively. References [1] Furtado MA, Braga GK, Oliveira JA, Del VF, Garcia-Cairasco N. Behavioral, morphologic, and electroencephalographic evaluation of seizures induced by intrahippocampal microinjection of pilocarpine. Epilepsia 2002;43(Suppl 5):37–9.
265
[2] Racine RJ. Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroencephalogr Clin Neurophysiol 1972;32:281–94. [3] Garcia-Cairasco N, Sabbatini RM. Role of the substantia nigra in audiogenic seizures: a neuroethological analysis in the rat. Braz J Med Biol Res 1983;16:171–83. [4] Doretto MC, Garcia-Cairasco N. Differential audiogenic seizure sensitization by selective unilateral substantia nigra lesions in resistant Wistar rats. Physiol Behav 1995;58:273–82. [5] Doretto MC, Fonseca CG, Lobo RB, Terra VC, Oliveira JA, Garcia-Cairasco N. Quantitative study of the response to genetic selection of the Wistar audiogenic rat strain (WAR). Behav Genet 2003;33:33–42. [6] Garcia-Cairasco N, Wakamatsu H, Oliveira JA, Gomes EL, Del Bel EA, Mello LE. Neuroethological and morphological (Neo-Timm staining) correlates of limbic recruitment during the development of audiogenic kindling in seizure susceptible Wistar rats. Epilepsy Res 1996;26:177–92. [7] Garcia-Cairasco N, Rossetti F, Oliveira JA, Furtado MA. Neuroethological study of status epilepticus induced by systemic pilocarpine in Wistar audiogenic rats (WAR strain). Epilepsy Behav 2004;5:455–63. [8] Dal-Col ML, Terra-Bustamante VC, Velasco TR, Oliveira JA, Sakamoto AC, GarciaCairasco N. Neuroethology application for the study of human temporal lobe epilepsy: from basic to applied sciences. Epilepsy Behav 2006;8:149–60. [9] Schmued LC, Albertson C, Slikker Jr W. Fluoro-Jade: a novel fluorochrome for the sensitive and reliable histochemical localization of neuronal degeneration. Brain Res 1997;751:37–46. [10] Timm F. Histochemistry of zinc. Dtsch Z Gesamte Gerichtl Med 1958;47:428–31. [11] Mello LE, Cavalheiro EA, Tan AM, et al. Circuit mechanisms of seizures in the pilocarpine model of chronic epilepsy: cell loss and mossy fiber sprouting. Epilepsia 1993;34:985–95. [12] Babb TL, Pretorius JK, Mello LE, Mathern GW, Levesque MF. Synaptic reorganizations in epileptic human and rat kainate hippocampus may contribute to feedback and feedforward excitation. Epilepsy Res Suppl 1992;9:193–202. [13] Mathern GW, Leite JP, Babb TL, et al. Aberrant hippocampal mossy fiber sprouting correlates with greater NMDAR2 receptor staining. NeuroReport 1996;7:1029–35. [14] McNamara JO. Cellular and molecular basis of epilepsy. J Neurosci 1994;14: 3413–25. [15] Tauck DL, Nadler JV. Evidence of functional mossy fiber sprouting in hippocampal formation of kainic acid-treated rats. J Neurosci 1985;5:1016–22. [16] Elmer E, Kokaia Z, Kokaia M, Lindvall O, McIntyre DC. Mossy fibre sprouting: evidence against a facilitatory role in epileptogenesis. NeuroReport 1997;8:1193–6. [17] Xu B, McIntyre DC, Fahnestock M, Racine RJ. Strain differences affect the induction of status epilepticus and seizure-induced morphological changes. Eur J Neurosci 2004;20:403–18. [18] Longo BM, Mello LE. Blockade of pilocarpine- or kainate-induced mossy fiber sprouting by cycloheximide does not prevent subsequent epileptogenesis in rats. Neurosci Lett 1997;226:163–6. [19] Longo BM, Mello LE. Supragranular mossy fiber sprouting is not necessary for spontaneous seizures in the intrahippocampal kainate model of epilepsy in the rat. Epilepsy Res 1998;32:172–82. [20] Longo BM, Mello LE. Effect of long-term spontaneous recurrent seizures or reinduction of status epilepticus on the development of supragranular mossy fiber sprouting. Epilepsy Res 1999;36:233–41. [21] Longo BM, Sanabria ER, Gabriel S, Mello LE. Electrophysiologic abnormalities of the hippocampus in the pilocarpine/cycloheximide model of chronic spontaneous seizures. Epilepsia 2002;43(Suppl 5):203–8. [22] Longo B, Covolan L, Chadi G, Mello LE. Sprouting of mossy fibers and the vacating of postsynaptic targets in the inner molecular layer of the dentate gyrus. Exp Neurol 2003;181:57–67. [23] Williams PA, Wuarin JP, Dou P, Ferraro DJ, Dudek FE. Reassessment of the effects of cycloheximide on mossy fiber sprouting and epileptogenesis in the pilocarpine model of temporal lobe epilepsy. J Neurophysiol 2002;88:2075–87. [24] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. San Diego: Academic Press; 1996. [25] Croiset G, De WD. Proconvulsive effect of vasopressin; mediation by a putative V2 receptor subtype in the central nervous system. Brain Res 1997;759:18–23. [26] Arida RM, Scorza FA, Peres CA, Cavalheiro EA. The course of untreated seizures in the pilocarpine model of epilepsy. Epilepsy Res 1999;34:99–107. [27] Fentress JC, Stilwell FP. Letter: grammar of a movement sequence in inbred mice. Nature 1973;244:52–3. [28] Norton S. The study of sequences of motor behavior. In: Iversen LC, Iversen SD, Snyder SH, editors. Handbook of psychopharmacology: principles of behavioral pharmacology, vol. 7. New York: Plenum; 1977. p. 82–105. [29] Garcia-Cairasco N. A critical review on the participation of inferior colliculus in acoustic–motor and acoustic–limbic networks involved in the expression of acute and kindled audiogenic seizures. Hear Res 2002;168:208–22. [30] Pinel JP, Rovner LI. Experimental epileptogenesis: kindling-induced epilepsy in rats. Exp Neurol 1978;58:190–202. [31] Garcia-Cairasco N, Doretto MC, Prado RP, Jorge BP, Terra VC, Oliveira JA. New insights into behavioral evaluation of audiogenic seizures: a comparison of two ethological methods. Behav Brain Res 1992;48:49–56. [32] Danscher G. Histochemical demonstration of heavy metals: a revised version of the sulphide silver method suitable for both light and electronmicroscopy. Histochemistry 1981;71:1–16. [33] Williams RW, Rakic P. Elimination of neurons from the rhesus monkey's lateral geniculate nucleus during development. J Comp Neurol 1988;272:424–36. [34] Turski WA, Cavalheiro EA, Schwarz M, Czuczwar SJ, Kleinrok Z, Turski L. Limbic seizures produced by pilocarpine in rats: behavioural, electroencephalographic and neuropathological study. Behav Brain Res 1983;9:315–35.
266
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
[35] Lemos T, Cavalheiro EA. Suppression of pilocarpine-induced status epilepticus and the late development of epilepsy in rats. Exp Brain Res 1995;102:423–8. [36] Kohane DS, Holmes GL, Chau Y, Zurakowski D, Langer R, Cha BH. Effectiveness of muscimol-containing microparticles against pilocarpine-induced focal seizures. Epilepsia 2002;43:1462–8. [37] Gulec G, Isbil-Buyukcoskun N. Arginine vasopressin does not contribute to seizures induced by intracerebroventricularly-injected pilocarpine. Neuropeptides 2005;39:575–9. [38] Cavalheiro EA, Leite JP, Bortolotto ZA, Turski WA, Ikonomidou C, Turski L. Longterm effects of pilocarpine in rats: structural damage of the brain triggers kindling and spontaneous recurrent seizures. Epilepsia 1991;32:778–82. [39] Bertram EH, Cornett J. The ontogeny of seizures in a rat model of limbic epilepsy: evidence for a kindling process in the development of chronic spontaneous seizures. Brain Res 1993;625:295–300. [40] Bertram EH, Cornett JF. The evolution of a rat model of chronic spontaneous limbic seizures. Brain Res 1994;661:157–62. [41] Rodrigues MC, Guizzo R, dos Santos WF, Cairasco NG. A comparative neuroethological study of limbic seizures induced by Parawixia bistriata venom and kainic acid injections in rats. Brain Res Bull 2001;55:79–86. [42] Leite JP, Garcia-Cairasco N, Cavalheiro EA. New insights from the use of pilocarpine and kainate models. Epilepsy Res 2002;50:93–103. [43] Wieser HG. Electroclinical features of the psychomotor seizure. Stuttgart: Fischer London: Butterworths; 1983. [44] Kotagal P, Luders HO, Williams G, Nichols TR, McPherson J. Psychomotor seizures of temporal lobe onset: analysis of symptom clusters and sequences. Epilepsy Res 1995;20:49–67. [45] Manford M, Fish DR, Shorvon SD. An analysis of clinical seizure patterns and their localizing value in frontal and temporal lobe epilepsies. Brain 1996;119(Pt 1):17–40. [46] Rodrigues MC, Rossetti F, Foresti ML, et al. Correlation between shaking behaviors and seizure severity in five animal models of convulsive seizures. Epilepsy Behav 2005;6:328–36. [47] Leite JP, Bortolotto ZA, Cavalheiro EA. Spontaneous recurrent seizures in rats: an experimental model of partial epilepsy. Neurosci Biobehav Rev 1990;14:511–7. [48] Turski WA, Cavalheiro EA, Turski L, Kleinrok Z. Intrahippocampal bethanechol in rats: behavioural, electroencephalographic and neuropathological correlates. Behav Brain Res 1983;7:361–70. [49] Poirier JL, Capek R, De KY. Differential progression of Dark Neuron and Fluoro-Jade labelling in the rat hippocampus following pilocarpine-induced status epilepticus. Neuroscience 2000;97:59–68. [50] Kubova H, Druga R, Haugvicova R, Suchomelova L, Pitkanen A. Dynamic changes of status epilepticus-induced neuronal degeneration in the mediodorsal nucleus of the
[51]
[52]
[53]
[54]
[55] [56] [57] [58] [59]
[60]
[61]
[62]
[63]
thalamus during postnatal development of the rat. Epilepsia 2002;43(Suppl 5): 54–60. Turski WA, Cavalheiro EA, Bortolotto ZA, Mello LM, Schwarz M, Turski L. Seizures produced by pilocarpine in mice: a behavioral, electroencephalographic and morphological analysis. Brain Res 1984;321:237–53. Castro OW, Furtado MA, Fernandes A, Tilelli CQ, Garcia Cairasco N. Comparative neuroanatomical and temporal characterization of Fluorojade-positive neurodegeneration after status epilepticus induced by systemic and intrahippocampal pilocarpine in Wistar rats. Proceedings, American Epilepsy Society Meeting, Boston, MA, 50. Epilepsia; 2009. p. 333. Romcy-Pereira RN, Garcia-Cairasco N. Hippocampal cell proliferation and epileptogenesis after audiogenic kindling are not accompanied by mossy fiber sprouting or Fluoro-Jade staining. Neuroscience 2003;119:533–46. Gorter JA, Goncalves Pereira PM, van Vliet EA, Aronica E, da Silva FH Lopes, Lucassen PJ. Neuronal cell death in a rat model for mesial temporal lobe epilepsy is induced by the initial status epilepticus and not by later repeated spontaneous seizures. Epilepsia 2003;44:647–58. Colombo JA, Yanez A, Puissant V, Lipina S. Long, interlaminar astroglial cell processes in the cortex of adult monkeys. J Neurosci Res 1995;40:551–6. Wasterlain CG, Shirasaka Y. Seizures, brain damage and brain development. Brain Dev 1994;16:279–95. Shih TM, McDonough Jr JH. Neurochemical mechanisms in soman-induced seizures. J Appl Toxicol 1997;17:255–64. Niquet J, Liu H, Wasterlain CG. Programmed neuronal necrosis and status epilepticus. Epilepsia 2005;46(Suppl 5):43–8. Mohapel P, Ekdahl CT, Lindvall O. Status epilepticus severity influences the longterm outcome of neurogenesis in the adult dentate gyrus. Neurobiol Dis 2004;15: 196–205. Tilelli CQ, Del VF, Fernandes A, Garcia-Cairasco N. Different types of status epilepticus lead to different levels of brain damage in rats. Epilepsy Behav 2005;7: 401–10. Dudek FE, Obenaus A, Schweitzer JS, Wuarin JP. Functional significance of hippocampal plasticity in epileptic brain: electrophysiological changes of the dentate granule cells associated with mossy fiber sprouting. Hippocampus 1994;4:259–65. Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol 1989;26: 321–30. Nissinen J, Lukasiuk K, Pitkanen A. Is mossy fiber sprouting present at the time of the first spontaneous seizures in rat experimental temporal lobe epilepsy? Hippocampus 2001;11:299–310.
Contents lists available at ScienceDirect
Epilepsy & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ye b e h
Study of spontaneous recurrent seizures and morphological alterations after status epilepticus induced by intrahippocampal injection of pilocarpine M.A. Furtado 1, O.W. Castro, F. Del Vecchio, J.A. Cortes de Oliveira, N. Garcia-Cairasco ⁎ Physiology Department, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, São Paulo, Brazil
a r t i c l e
i n f o
Article history: Received 28 June 2010 Revised 22 November 2010 Accepted 25 November 2010 Available online 14 January 2011 Keywords: Epilepsy Hippocampus Neurodegeneration Temporal lobe epilepsy Animal model
a b s t r a c t Epileptic seizures are clinical manifestations of neuronal discharges characterized by hyperexcitability and/or hypersynchrony in the cortex and other subcortical regions. The pilocarpine (PILO) model of epilepsy mimics temporal lobe epilepsy (TLE) in humans. In the present study, we used a more selective approach: microinjection of PILO into the hilus of the dentate gyrus (H-PILO). Our main goal was to evaluate the behavioral and morphological alterations present in this model of TLE. Seventy-six percent of all animals receiving H-PILO injections had continuous seizures called status epilepticus (SE). A typical pattern of evolution of limbic seizures during the SE with a latency of 29.3 ± 16.3 minutes was observed using an analysis of behavioral sequences. During the subsequent 30 days, 71% of all animals exhibited spontaneous recurrent seizures (SRSs) during a daily 8-hour videotaping session. These SRSs had a very conspicuous and characteristic pattern detected by behavioral sequences or neuroethological analysis. Only the animals that had SE showed positive Neo-Timm staining in the inner molecular layer of the dentate gyrus (sprouting) and reduced cell density in Ammon's horn pyramidal cell subfield CA1. However, no correlation between the intensity of sprouting and the mean number and total number of SRSs was found. Additionally, using FluoroJade staining, we observed neurodegeration in the hilus and pyramidal cell subfields CA3 and CA1 24 hours after SE. These data indicate that H-PILO is a reliable, selective, efficient, low-mortality model that mimics the acute and chronic behavioral and morphological aspects of TLE. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Seizures induced by the systemic injection of pilocarpine (PILO) are a known model of temporal lobe epilepsy (TLE). In the current study, we used a more specific approach: a model of seizure induction established by microinjection of PILO into the hilus of the dentate gyrus (H-PILO). Our main goal was to associate behavioral and morphological alterations of status epilepticus (SE) and spontaneous recurrent seizures (SRSs) induced by H-PILO. To analyze behavioral seizures, we used neuroethological tools because it is not possible to detect specific behavioral interactions using only the Racine seizure severity index [2]. We have used neuroethology to study different models of epilepsy, such as acute audiogenic seizures [3–5], audiogenic kindling [6], and SE induced with systemic pilocarpine (S-PILO) [7]. More recently, we also used ⁎ Corresponding author. University of São Paulo—USP, Ribeirão Preto School of Medicine, Av. Bandeirantes, 390014049-900 Ribeirão Preto, SP, Brazil. Fax: + 55 16 3633 0017. E-mail addresses: [email protected], [email protected] (M.A. Furtado), [email protected] (N. Garcia-Cairasco). 1 Current address: Clinical Research Management, Inc., Department of Closed Head Injury, Brain Dysfunction and Blast Injury Division, Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910, USA. Fax: + 1 301 319 9810. 1525-5050/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2010.11.024
neuroethology to characterize complex partial seizures in patients with TLE [8]. In particular, the model of SE induced by S-PILO has its behavioral seizure progression well characterized through neuroethology and interstrain differences can also be discriminated [7]. However, the high mortality rate in the model of S-PILO is a limiting factor. We used Neo-Timm, Nissl histochemistry and Fluoro-Jade (FJ) staining [9] to assess neuroplastic changes and cellular degeneration after H-PILO. Mossy fiber sprouting (MFS), detected as exacerbated positive Neo-Timm staining [10], is observed in the inner molecular layer of the dentate gyrus of epileptic animals [11] and humans with TLE [12,13]. This MFS seems to occur after mossy cell death and deafferentation of the basket cells located in the hilus [14]. FJ + cells are neurodegenerating neurons and, in this case, were detected in subfields of the hippocampal formation. Considering the wide range of morphological modifications that occur after SE induced with S-PILO, it is controversial to argue specifically that the synaptic rearrangements in the hippocampus are responsible for the SRSs. However, some authors maintain that the MFS of the dentate granule cells contributes to epileptogenesis in animals that exhibit SRS. The functional consequence of MFS remains unknown. This synaptic reorganization in the hippocampal formation might be implicated in the consequent epileptogenesis in animal models of epilepsy and patients with epilepsy [15]. In contrast, MFS
258
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
could also represent a compensatory or inhibitory response [16,17]. A great deal of controversy surrounds the interpretation of the functional role of this sprouting [18–23]. Thus, in the context of the aforementioned studies, determining the morphological alterations at the synaptic and molecular levels that occur in epileptic animals is still challenging. In the current study, we investigated the acute neuroanatomical changes induced by H-PILO using FJ and whether SE chronically induced MFS and cell loss, using Neo-Timm and Nissl staining, respectively. The presence of SRS was studied, and their behavioral patterns were measured through neuroethological tools. Finally, in light of all the controversy over the role of MFS in the occurrence of SRSs, we determined if the frequency of SRS was related to MFS. We tested the following hypotheses: (1) severe neurodegeneration can be observed 24 hours after SE induced with H-PILO, using FJ staining; (2) SE and SRSs induced with H-PILO can be characterized by a typical progression of limbic behavioral patterns; (3) MFS occurs in parallel with the occurrence of SRSs but is not related to the frequency or total number of SRSs; (4) epileptic animals exhibit a reduction in the cellular density of the CA1 region of Ammon´s horn. 2. Methods 2.1. Animals and surgical procedures All experiments were performed in accordance with the Brazilian Society for Neuroscience and Behavior guidelines based on those from the Society for Neuroscience and consistent with the guidelines set by the Colegio Brasileiro de Experimentação Animal (COBEA) and Institutional Ethics Committee (CETEA) Protocol 195/2005. Every effort was made to avoid unnecessary pain or stress to the animals. Adult male Wistar rats (n = 50) were anesthetized with tribromoethanol (Avertin, Sigma–Aldrich; St. Louis, MO, USA; 2,5%, 10 mL/kg, ip), placed in a stereotaxic apparatus (David Kopf Instruments, Tujunga, CA, USA), and implanted with a guide cannula in the left hippocampus (AP 6.30 mm, L 4.5 mm, V 4.5 mm, bregma reference) [24] by stereotaxic surgery. Animals were allowed to recover for 7 days before the administration of H-PILO through a cannula. 2.2. Microinjection procedures We used a 5-μL syringe (Hamilton, Sigma, Reno, NV, USA) connected to a microinjection pump (Harvard Apparatus, 2000, Holliston, MA, USA) adjusted to a flow of 0.5 μL/minute. After the end of microinjection, the syringe remained connected to the cannula for approximately 1 minute to avoid reflux of the drug. All animals were gently immobilized during the microinjection procedures. The experimental group received H-PILO (Sigma–Aldrich) through the cannula (n = 34; 2.4 mg/mL, 1 μL) [25]. Ninety minutes after SE onset, the animals were injected with diazepam (5 mg/kg, Roche, Sao Paulo, Brazil). A control group (n = 16) was injected with saline (1 μL). 2.3. Behavioral assessment of status epilepticus and spontaneous recurrent seizures Immediately after either saline or H-PILO microinjection, we recorded the behavioral activity of all animals on VHS tapes with a camcorder (Gradiente GCP-180CR, Sao Paulo, Brazil) up to 2 hours after SE onset. Behavioral sequences were obtained from the video analysis (see Neuroethological Evaluation of Status Epilepticus and Spontaneous Recurrent Seizures). We used a set of acrylic cages that allowed the simultaneous observation of six animals. To detect SRSs, we used the same cages mentioned above. The behavioral activity was also recorded on VHS tapes 8 hours per day for up to 30 days after SE.
We chose the daylight period because SRSs occur more frequently during the day [26]. 2.4. Neuroethological evaluation of status epilepticus and spontaneous recurrent seizures In this study, we used neuroethology-based methods on the premise that the analysis of behavioral sequences can provide us with more information than isolated behavioral measurements. Neuroethology allowed us to study the movement grammar and its temporal and hierarchical structure [27–29]. For a detailed analysis of behavioral sequences, we determined observation periods of 60 seconds according to specific behaviors based on the scale established by Racine [2] and modified by Pinel and Rovner [30] (see Table 1). These specific behaviors were called behavioral triggers. Behavioral triggers for SE are shown in Fig. 1A and were defined as follows: stage A, head myoclonus; stage B, forelimb myoclonus; stage C, rearing followed by falling; and stage D, jump. Behavioral triggers for SRSs are shown in Fig. 1B and were defined as follows: stage A, rearing; stage B, one rearing followed by falling; and stage C, three or more rearings followed by falling. This approach was used to provide markers for recording the onset of behaviors already validated in the evolution of severity indexes but that are expressed at variable times in different animals. These triggers have been used successfully in the characterization of SE induced with S-PILO in rats [7]. Using a dictionary of behavioral items [6,7,31], we recorded every behavior observed during each 60-second period using a second-bysecond analysis. We recorded the behaviors with the aid of a video capture card (ATI All in Wonder PRO, Elizabeth, NJ, USA), and the video was reviewed several times. For statistical analysis, we used a program called ETHOMATIC [31]. The ETHOMATIC program calculates and displays the mean frequency and mean duration of each behavioral item in the given observation window. The software also performs statistical analysis, verifying significant associations between pairs (dyads) of behavioral items and calculating a χ2 value. Using the data obtained with ETHOMATIC, we constructed flowcharts to graphically represent all statistically significant data (Fig. 2). Microsoft Power Point was used for the graphical representation of flowcharts. Fig. 2 shows the behavioral samples used to construct the flowcharts. The entire behavioral sequences are shown in Figs. 3 and 4. 2.5. Histological procedures All animals received a lethal dose of thionembutal (100 mg/kg). For FJ staining, we euthanized animals 24 hours after SE (n = 10, experimental group; n = 6 control group) as described by Schmued et al. [9]. Animals were transcardially perfused with 100 mL of phosphate buffer followed by paraformaldehyde (500 mL, 4%) dissolved in phosphate buffer. After transcardiac perfusion, all brains were removed, frozen, and sectioned in slices of 30 μm for FJ histochemistry and 20 μm for Nissl
Table 1 Limbic seizure severity indexes.a Class
Behavior
0 1 2 3 4 5 6 7 8
Immobility Facial automatisms Head myoclonus Forelimb myoclonus Rearing Falling More than three falls Wild running Tonic–clonic seizures
a Racine scale [2] or limbic index (LI) as modified by Pinel and Rovner [30].
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
259
Fig. 1. Behavioral sampling methods used in the construction of flowcharts. The sequential letters represent oscillating behavioral triggers corresponding to specific behaviors of the seizure scale [30]. (A) Status epilepticus: first head myoclonus = stage A; first forelimb myoclonus = stage B; first rear followed by falls = stage C; first jump = stage D. (B) Spontaneous recurrent seizures: first rear = stage A; first rear followed by fall = stage B; first rear followed by several falls = stage C.
staining. Brain sections were obtained using a Cryostat (Microm HM505-E, Microm International, Walldorf, Germany). For better quality and conservation of the tissue, the sections were mounted directly inside the cryostat onto gelatin-coated slides. Subsequently, the slides were stored in a freezer at –20 °C. For Nissl and Neo-Timm staining, we euthanized animals at 30 days (n = 12, experimental group; n = 5, control group) and 15 days (n = 12, experimental group; n = 5, control group) after SE. When the 15- and 30-day groups were combined, the experimental group comprised 24 animals and the control group 10 animals. Animals were transcardially perfused with 100 mL of Millonig's phosphate buffer (100 mL) followed by Millonig's phosphate buffer with sodium sulfide (100 mL, 0.1%), glutaraldehyde (100 mL, at 3%), and again Millonig's phosphate buffer with sodium sulfide (200 mL, 0.1%). 2.5.1. Fluoro-Jade staining Fluoro-Jade histochemistry was performed as follows: tissue was immersed in 100% ethanol for 3 minutes, followed by 1 minute in 70% ethanol and 1 minute in distilled water. After this step, slides were transferred to a solution of 0.06% potassium permanganate for 15 minutes and were gently shaken on a rotating platform. Slides were rinsed three times for 1 minute in distilled water and were then transferred to the FJ staining solution and gently shaken for 30 minutes. The 0.0001% working solution of FJ was prepared by adding 1 mL of stock FJ solution (0.01%) to 99 mL of 0.1% acetic acid in
distilled water. After staining, sections were rinsed three times (1 minute) in distilled water and slides were coverslipped. 2.5.2. Neo-Timm staining To study axonal sprouting in the hippocampus, we used Neo-Timm histochemistry [10,11]. This technique is based on the histochemical detection of zinc vesicles [32]. The aforementioned perfusion with Millonig's phosphate buffer with sodium sulfide is necessary to increase the affinity of silver nitrate for the zinc during the staining. 2.5.3. Nissl staining and assessment of cellular density Every adjacent section was stained with cresyl violet and was used to determine the general size, shape, and distribution of neuronal cell bodies, as well as localization of the cannula and confirmation of the lesion. All images were obtained using the same system employed for histochemistry (see above). Figs. 6C and D illustrate the sampling method used for optical density measurements of Neo-Timm staining, and Fig. 6H shows the sampling method used for Nissl-stained cell counts. Morphological criteria described by Williams and Rakic [33] were used to estimate the Nissl-stained cells. Pyramidal cells counted in the CA1, CA2, and CA3 subfields have a size and shape that are easily identifiable. Furthermore, the nuclei of neurons have very evident nucleoli. In contrast, glial cells usually have small and dark nucleoli. Neurons contain more abundant and granular Nissl substance than other types of cells. Cell bodies and nuclei of neurons have a sectional
260
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
Fig. 2. (A) Example of a short behavioral sequence (time–item) and its graphic representation as a flowchart. (B) Flowchart structure and calibration. The height of the rectangles represents the frequency of a behavioral item, and the base corresponds to the duration of the behavior in an observation window. The arrows represent statistical values (χ2 N 3:84, P b 0:05) equivalent to the degree of association between pairs of behaviors (dyads) in a sequence. The colors identify the main behavioral clusters. The use of colors and circles (B; see also Fig. 3) and the calibration of rectangles in the flowcharts are for illustration purposes and do not have any impact on the statistical analysis (see Methods for details). A complete description of behaviors is given in the legend to Fig. 3. Reproduced, with permission, from Garcia-Cairasco et al. [7].
area two to three times greater than that of glial cells, even in the early developmental stages. Although neurons belonging to a local circuitry were slightly larger than glial cells, neurons had more prominent nucleoli. In addition, nuclei of neurons have a rounded or oval shape, whereas the nuclei of glial cells have an irregular shape [33].
light intensity was maintained in the optical microscope, and we used only slices stained in the same batch for the Neo-Timm staining reaction. This particular care was taken to avoid random differences in the measured density of gray levels. The data were exported to Microsoft Excel for Windows software to perform statistical analysis.
2.6. Statistics 3. Results Full statistical analyses were performed using SIGMA-STAT Version 3.5 (Systat Software Inc, San Jose, CA, USA). Student's t test was used to compare different animal groups, according to number of cells (Nissl staining) and gray level (Neo-Timm staining). Pearson's correlation was calculated to verify the association between gray level and frequency of seizures. 2.7. Image analysis All images were digitized using an Optronics CCD camera (Santa Barbara, CA, USA) connected to a BX-60 Olympus microscope (Olympus; Melville, NY, USA). For quantitative analyses, we converted all images to a gray scale of 8 bits. The freeware Image Tool program (University of Texas, Version 1.27) was used to measure gray levels in the inner molecular layer of the dentate gyrus. The values obtained in the ventral thalamus were used as background. Using a gray scale of 256 values, we measured the integrated density of gray levels per area. In the next step, we subtracted the values obtained in the thalamus from the values obtained in the hippocampus. The same
3.1. Behavioral assessment of status epilepticus and spontaneous recurrent seizures Only animals for which cannula position was confirmed by histology were used. Among all the animals, 76% exhibited SE within 29.3 ± 16.3 minutes (mean ± SEM). Additionally, 71% of animals displaying SE had SRSs and survived until the day of sacrifice. In all groups, the acute behavioral effects caused by the administration of H-PILO consisted of initial piloerection followed by periods of whole-body tremor and/or myoclonic twitching restricted to the head and face, accompanied by salivation. SE was characterized by repeated cycles of head and forelimb myoclonus, rearing, and falling. As expected, no such events were observed in saline-injected animals. During SRSs, animals exhibited orofacial automatisms, head and forelimb myoclonus, and eventually rearing and falling. The evolution of the severity of SRSs resembles the severity observed during amygdala kindling [2], and the majority of seizures occurred on awakening. Animals with SE had elevated levels
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
261
of exploration and reactivity compared with control rats (qualitative observation; data not shown) during the first week after SE induction. 3.2. Neuroethological evaluation of status epilepticus and spontaneous recurrent seizures
Fig. 3. Flowcharts representing sequential stages of SE induced with H-PILO. (A–D) Stages A–D. AF, atonic fall; ELE, elevation; ERE, erect posture; FALL, falling (class 5 limbic seizures); IM, immobility (pause); JP, Jumping; MT, masticatory movements; MYO, myoclonic jerks (limbic); MYO1, myoclonic jerks, forelimb; MYOh, myoclonic jerks, head; MYOt, myoclonic jerks, trunk; REAR, rearing (class 4 limbic seizure); SCA, scanning; SNF, sniffing; STRA, straub tail; TCV, tonic seizures (opisthotonus); WA, walking; WDS, wet dog shakes.
Fig. 4. Flowcharts representing sequential stages of SRS induced with H-PILO. (A–C) Stages A–C. Abbreviations as for Fig. 3.
Throughout the neuroethological analysis of SE, we observed exploratory behaviors of low frequency and low duration during stage A (Fig. 3, with one head myoclonus as the behavioral trigger). Statistically significant interactions (arrows, χ2 log) between sniffing and scanning were also observed. The associations between the behaviors of head myoclonus, sniffing, and masticatory movements were also significant; however, the interactions between masticatory movements and head myoclonus were more prominent. Wet dog shakes occurred only during stage A (Fig. 3), but there were no significant interactions with other behaviors. During stage B (Fig. 3; with one forelimb myoclonus as the behavioral trigger), exploratory behaviors of low frequency and low duration also occurred. Reciprocal and significant interactions between limbic behaviors, such as between head myoclonus and forelimb myoclonus and between head myoclonus and masticatory movements, were observed. In stage C (Fig. 3; with one rearing followed by falling as the behavioral trigger), exploratory behaviors were rare. Limbic behaviors, such as head myoclonus and forelimb myoclonus, rearing, and falling, were common. Forelimb myoclonus–head myoclonus, myoclonic jerks (trunk)–head myoclonus, and rearing–falling interactions were statistically significant. During stage D (Fig. 3; with one jump as the behavioral trigger), exploratory behaviors were scarce. Reciprocal forelimb–head myoclonus interactions were statistically significant. The associations between walking, jumping, and falling were significant. Throughout the neuroethological analysis of SRSs, we observed (Fig. 4, with rearing as the behavioral trigger) a cluster of exploratory behaviors in stage A characterized by statistically significant interactions (arrows, χ2 log) between sniffing and scanning. Other behaviors were present in a pattern of reverberation. For example, we noticed strong interactions between masticatory movements, forelimb myoclonus, and head myoclonus closing a loop or behavioral cluster with masticatory movements; we also observed mutual interactions between rearing and head myoclonus and straub tail and forelimb myoclonus. Behaviors such as masticatory movements, head myoclonus and forelimb myoclonus were characterized by an elevated frequency. Immobility was observed to have an elevated duration and was weakly associated with masticatory movements. In stage B (Fig. 4, with one rearing followed by falling as the behavioral trigger), we noticed a cluster of exploratory behaviors characterized by interactions between sniffing and scanning. We also discriminated interactions between falling and straub tail. Interestingly, the associations between head myoclonus and forelimb myoclonus were mutual and more intense than during stage A (Fig. 4). The pattern of reverberation was not as clear as during stage A (Fig. 4), but forelimb myoclonus was characterized by elevated frequency and duration. Head myoclonus was characterized by an elevated frequency. We observed a slight association between immobility and masticatory movements. During stage C (Fig. 4, with three or more rearings followed by falling as the behavioral trigger), the cluster of exploratory behaviors was reduced, and we discriminated strong interactions between rearing and falling. The pattern of reverberation was very clear and was characterized by mutual interactions between masticatory movements and forelimb myoclonus and head myoclonus and masticatory movements, closing a loop. The interactions between head myoclonus and masticatory movements were less intense during stage C than during stage A and absent during stage B (Fig. 4). The frequency of head myoclonus was lower than than during stages A and B (Fig. 4).
262
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
Interestingly, in all stages, we observed the behaviors alert state and lying, and especially in stage A (Fig. 4), there was a statistically significant interaction between these behaviors.
saline and animals that did not exhibit SE did not have positive FJ staining (upper right inset Fig. 5A′).
3.3. Fluoro-Jade staining
3.4. Gray-level analysis of the axonal sprouting in the dorsal hippocampus
Animals (n = 16) were euthanized 24 hours after SE. A reduced background allowed us to easily identify stained neurons. We observed strong FJ staining in the hilus, CA3, and CA1 (Fig. 5) and also the amygdala (images not shown). Control animals treated with
Animals injected with saline (Figs. 6A and B) and animals that did not exhibit SE did not have sprouting. All animals with SE had sprouting in the inner molecular layer of the dentate gyrus (Figs. 6C–F). Apparently there was no association between the degree of sprouting
Fig. 5. Fluoro-Jade staining. FJ + neurodegeneration after 24 hours of SE induced with H-PILO. Confocal microscopy images: (A) hilus of dentate gyrus, (A′) zoom in hilus; (B) CA1, (B') zoom in CA1; (C) CA3, (C') zoom in CA3. Calibration bar: (A–C) 75 μm, (A´–C´) 25 μm. Note the absence of FJ staining in a control animal in the upper right inset of A´.
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
263
Fig. 6. Neo-Timm staining. (A, B) Control animals. (C, D) Epileptic animals. (A, C) Brain sections of animals sacrificed 15 days after SE. (B, D) Brain sections of animals sacrificed 30 days after SE. Calibration bar: 200 μm. Observe details in the insets (E, F). Calibration bar: 40 μm. (G) There is no difference in the gray levels in the inner molecular layer of the dentate gyrus of animals injected with H-PILO between the left (dark bars) and right (light bars) hippocampus. Animals that displayed SE (PILO 15 and 30 days) had increased gray level from either left or right hippocampus when compared with the respective control groups (*P b 0.01; Student's t test). (H) Sampling pattern used for cell counting in the dorsal hippocampus [24]. (I–L) Nissl staining. Sections in the coronal plane located approximately –3.3 mm from bregma. (I, K) Control animal: CA1 and CA3c, respectively. (J, L) Animal exhibiting SE: CA1 and CA3c, respectively. Arrow points to example of the reduction of cellular density in an epileptic animal (CA1). Calibration bar: 200 μm.
and the frequency of SRSs. Additionally, qualitative evaluation revealed that some animals had stronger staining in the site ipsilateral to PILO injection and in the ventral hippocampus.
Quantitative analysis of the sprouting (Fig. 6G) from both the left and right hippocampus revealed a statistically significant difference between the animals exhibiting SE and the controls (Student's t test,
Table 2 Mean numbers of pyramidal cells per area. Group
Mean (SEM) number of pyramidal cells/μm2 CA1 subfield Control
Left Right
0.0025 (0.0002) 0.0028 (0.0002)
CA2 subfield 15 days
30 days
a
a
0.0019 (0.0004) 0.0020b (0.0003)
0.0015 (0.0003) 0.0016b (0.0003)
CA3 subfield
Control
15 days
30 days
Control
15 days
30 days
0.0020 (0.0002) 0.0025 (0.0003)
0.0019 (0.0003) 0.0019 (0.0003)
0.0012 (0.0005) 0.0018 (0.0004)
0.0013 (0.0002) 0.0015 (0.0002)
0.0011 (0.0001) 0.0012 (0.0001)
0.0012 (0.0001) 0.0012 (0.0001)
a
P b 0.05, comparing the cellular densities in the left CA1 hippocampal subfield of the animals sacrificed 15 and 30 days after SE and control animals. P b 0.05, comparing the cellular densities in the right CA1 hippocampal subfield of the animals sacrificed 15 and 30 days after SE and control animals. Statistical analyses were performed using Student's t test: b
Left CA1: control vs 15 days, P = 0.041; control vs 30 days, P = 0.021. Right CA1: control vs 15 days, P = 0.048; control vs 30 days, P = 0.037. Left CA2: control vs 15 days, P = 0.892; control vs 30 days, P = 0.09. Right CA2: control vs 15 days, P = 0.159; control vs 30 days, P = 0.093. Left CA3: control vs 15 days, P = 0.097; control vs 30 days, P = 0.401. Right CA3: control vs 15 days, P = 0.071; control vs 30 days, P = 0.088.
264
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
P b 0.01). Also, there were no differences in the level of sprouting between the 15- and 20-day groups and the left (dark bars) and right (light bars) hippocampus (Student's t test, P = 0.795). We did not find any correlation between gray level and frequency and total number of seizures of each animal (Pearson's correlation, P N 0.05). 3.5. Cellular density analysis in the dorsal hippocampus Analysis of sections (Fig. 6H) stained for Nissl histochemistry indicated that, when compared with the control group (Fig. 6I; Table 2; P b 0.05, Student's t test), the experimental group (Fig. 6J) had a reduced density of cells in the CA1 field. Especially in the CA1 subfield of epileptic animals, we observed cells with pyknotic nuclei and an altered form. Some areas contained vacuoles, and disruption of the tissue and could not be analyzed. The CA3 subfield did not statistically significantly differ between control (Fig. 6K) and experimental (Fig. 6L) animals. We did not find any difference in cellular density between the 15and 30-day groups in the left and right hippocampus (Table 2). 4. Discussion A single dose of H-PILO (2.4 mg/μL) induced SE in the majority of animals with the typical evolution pattern described in our previous study [1], and the seizures resembled those induced with S-PILO [34,35] and PILO injected into the hippocampus with a microdialysis probe [36] or intracerebroventricularly [37]. Our results support the H-PILO model as a quite localized and efficient way to mimic experimentally TLE with low mortality. The daylight period was preferred for observations because evidence in the literature indicates that SRSs occur with a higher frequency during the day [26]. We verified that the severity of seizures increased with time. In fact, it has been suggested by several authors that a kindling mechanism operates during the development of SRSs [38–40]. In a previous study, we showed the behavioral evolution of SE and SRSs induced with H-PILO using known seizure severity indexes [1]. Throughout the current analysis of the flowcharts, we observed specific behavioral interactions not noticed previously when we used only the Racine [2] seizure severity scale. We also observed a strong variability in the onset of SE and SRSs. On this basis, we selected the onset of the observation windows according to the severity of seizures exhibited by each animal (i.e., behavioral triggers; see Fig. 1). We analyzed the behavioral associations occurring during SE and SRSs induced with H-PILO using neuroethological tools. Neuroethology has been validated in other studies that have evaluated models of epilepsy such as the development of audiogenic kindling in Wistar Audiogenic Rats (WARs, a genetically selected inbred strain [5,6]), seizures induced with S-PILO in WARs [7], and seizures induced by injection of crude venom from the spider Parawixia bistriata [41]. Leite and co-workers [42], reviewing the kainic acid and PILO models of TLE, proposed the use of this neuroethological technique to evaluate new antiepileptic agents. More recently, we used the same methodology to characterize seizures of patients with TLE [8]. Although its application is labor intensive and time consuming, the results obtained are worthwhile and allow the assessment of subtle alterations caused by antiepileptic drugs. In fact, there have been several studies of spontaneous seizures in humans using statistical tools [43–45]. Analysis of the flowcharts constructed using sequences from SE and SRSs induced with H-PILO suggested that a reverberant circuitry could contribute to the genesis of the observed limbic seizures. The strong statistical associations (χ2 log) observed at each stage of SE and SRSs corroborate our hypothesis (e.g., Fig. 3 stage B, interactions between MYOh, and MYO1; and Fig. 4 stage A, interactions between MT, MYOh, and MYO1). During stage A (Fig. 3), the sequence of head
myoclonus and wet dog shakes suggests that displaced behaviors (see also Fig. 4) are expressed probably because of the activation of endogenous anticonvulsant mechanisms [46]. When these mechanisms fail, the SE is fully installed (Figs. 4B–D). Interestingly, during stages C and D (Fig. 3), the occurrence of audiogenic-like behaviors, such as tonic–clonic convulsions, jumps, and atonic falling, suggests that brainstem structures are being activated during SE. During stages A and B (Fig. 4), the presence of strong exploratory clusters, now during the period of SRSs, could be indicative of the activation of endogenous antiseizure mechanisms. In other words, the exploratory behaviors might compete with ictal behaviors because we did not observe severe seizures in stages A and B (Fig. 4). In fact, in previous studies, we observed facial automatisms, compulsive grooming, and exploration during acute and chronic acoustic stimulation of animals that are not susceptible to audiogenic seizures [3,6]. In general, we observed that limbic behaviors increased in severity according to the Racine scale [2] as modified by Pinel and Rovner [30]. However, subtle alterations, such as the cluster of exploratory behaviors mentioned above, were undetectable using only linear scales or behavioral scores. Because the extension of the neural circuits involved in seizures and their behavioral expression may be related to synchronous activation of different brain networks, this epileptogenic network–behavior relationship is almost impossible to demonstrate using isolated behavioral observations. Therefore, neuroethology is useful for the detection of the development of specific behavioral associations and reverberative clusters. For example, the interactions between forelimb myoclonus (MYO1) and head myoclonus (MYOh) were more intense during stage C than during stage A (Fig. 4). The behaviors lying and alert state showed that the majority of seizures occurred after a sleep period. Such data have already been reported by Leite et al. [47] in a model of seizures induced with a high dose of S-PILO. We used FJ staining to determine the acute neurodegeneration pattern after H-PILO microinjection. We observed neuronal damage in CA1, CA3, hilus (Fig. 5), and amygdala (not shown). In fact, this is obviously only a partial account of the wider lesion described in rats by Turski et al. [48] after injection of the muscarinic agonist bethanecol into CA3. Briefly, these authors mention lesions of the piriform cortex, entorhinal cortex, olfactory tubercle, anterior olfactory nucleus, subiculum, amygdaloid complex, temporoparietal cortex, and hypothalamic nuclei. There are additional data indicating that S-PILO (in adult and young rats) can lead to neurodegeneration in several other brain regions 3 months after SE [34,49,50]. After analyzing the acute and chronic structural modifications in the dorsal hippocampus (FJ and Nissl data), we concluded that the pattern of cell lesions seen after H-PILO in our study confirm hilar, CA3, and CA1 neurodegeneration. Because in the current study we did not examine regions other than the hippocampal formation, we cannot say there were no distant lesions such as those induced after S-PILO in mice and detected, in addition to the hippocampus, in the amygdala, thalamus, olfactory cortex, neocortex, and substantia nigra by Turski et al. [51]. However, we have also shown, in a comparative study of S-PILO versus H-PILO, that lesions in the H-PILO model are narrower than those from the S-PILO model [52]. Interestingly, Romcy-Pereira and Garcia-Cairasco [53] showed that animals displaying audiogenic kindling have no FJ + cells. This result suggests that short seizures do not induce neurodegeneration, but prolonged SE is capable of activating neuroexcitotoxic mechanisms. Also, the duration of SE is proportional to the degree of neurodegeneration [54], indicating that a rapid blockade of SE is essential in avoiding neuronal loss and epileptogenesis. Finally, Colombo et al. [55] showed that FJ also stains reactive astroglia in the cerebral cortex of nonhuman primates. However, this limitation did not constitute a problem because astroglia can be easily distinguished from pyramidal cells and other types of neurons. We did
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
not find evidence of neuronal loss in animals injected with saline, indicating that the cannula causes only mechanical damage restricted to the site of implantation. The initial cholinergic overstimulation leads to a subsequent excessive activation of NMDA, which, in turn, increases the calcium influx that can lead to a cascade of molecular events causing neuronal death [56,57], mainly by necrosis [58]. This neuronal death is more evident after prolonged limbic seizures than after low-severity or atypical seizures [59,60]. We did not find significant cell reduction in CA3 with Nissl staining at the chronic time window, although we saw clear FJ labeling in this structure 24 hours after SE. The absence of a significant difference in cell density between the left and right hippocampus suggests that the seizures propagated to the contralateral site of injection. The lack of differences between the 15- and 30-day groups indicates that only the initial injury was necessary for the morphological alterations observed [38]. Mossy fiber sprouting was observed in all animals that exhibited SE. Furthermore, 72% of these animals had SRSs. The MFS observed in human TLE has been described in several studies [12,13] and occurs in parallel with SRSs. Also, MFS is a common finding in animals that have SRSs [61]. However, neither Sutula et al. [62] nor Mathern [13] observed associations between the duration of epilepsy and the density of MFS. When comparing the groups of animals sacrificed 15 and 30 days after H-PILO, we did not observe any significant difference in MFS. The absence of a correlation between MFS intensity and total number or frequency of SRSs suggested that MFS is not related to the pattern of SRSs. In fact, the absence of a correlation between the frequency of SRSs and MFS has been reported [11]. Furthermore, SRSs can occur prior to the development of MFS [63]. Although blockade of MFS by cycloheximide injection does not prevent the occurrence of SRSs in the model of systemic seizures induced with PILO [18] or in the model of seizures induced by injection of kainic acid into the hippocampus [19], Williams et al. [23] did not confirm those results. In the current study, we combined the neuroethological analysis of behavioral sequences with Neo-Timm, Nissl, and FJ histochemistry, which allowed the evaluation of behavioral alterations, density of cell loss, neurodegeneration, and density of MFS, respectively, in the dorsal hippocampus. Future experiments are underway to study the associations between behavioral and electroencephalographic alterations in limbic structures during SE and SRSs induced with H-PILO. In conclusion, neuroethological tools allowed us to identify every step of the motor acts to elucidate complex and simple sequences of movements. Although behavioral indexes are very useful, they should not be used alone but in combination with more precise behavioral evaluations, such as the sequential analysis of the semiological patterns of seizures induced with H-PILO. Combination of precise neuroethology tools and other methods, such as histochemistry (NeoTimm, Nissl, and FJ) and EEG, is necessary to improve the prediction of the neural substrates of epileptic seizures shown in the current study of H-PILO. Therefore, we consider the induction of SE with H-PILO a reliable TLE model with low mortality and a more restricted pattern of neurodegeneration than that with S-PILO. Acknowledgments This study was supported by FAPESP, FAPESP-Cinapce, CAPES (PROAP-PROEX), CNPq, PRONEX and FAEPA. M.A. Furtado was a PROAP-CAPES-Brazil Ph.D fellow. O.W. Castro is a CNPq-Brazil PhD fellow. N. Garcia-Cairasco and J.A.C. Oliveira hold CNPq-Brazil Research and Technician fellowships, respectively. References [1] Furtado MA, Braga GK, Oliveira JA, Del VF, Garcia-Cairasco N. Behavioral, morphologic, and electroencephalographic evaluation of seizures induced by intrahippocampal microinjection of pilocarpine. Epilepsia 2002;43(Suppl 5):37–9.
265
[2] Racine RJ. Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroencephalogr Clin Neurophysiol 1972;32:281–94. [3] Garcia-Cairasco N, Sabbatini RM. Role of the substantia nigra in audiogenic seizures: a neuroethological analysis in the rat. Braz J Med Biol Res 1983;16:171–83. [4] Doretto MC, Garcia-Cairasco N. Differential audiogenic seizure sensitization by selective unilateral substantia nigra lesions in resistant Wistar rats. Physiol Behav 1995;58:273–82. [5] Doretto MC, Fonseca CG, Lobo RB, Terra VC, Oliveira JA, Garcia-Cairasco N. Quantitative study of the response to genetic selection of the Wistar audiogenic rat strain (WAR). Behav Genet 2003;33:33–42. [6] Garcia-Cairasco N, Wakamatsu H, Oliveira JA, Gomes EL, Del Bel EA, Mello LE. Neuroethological and morphological (Neo-Timm staining) correlates of limbic recruitment during the development of audiogenic kindling in seizure susceptible Wistar rats. Epilepsy Res 1996;26:177–92. [7] Garcia-Cairasco N, Rossetti F, Oliveira JA, Furtado MA. Neuroethological study of status epilepticus induced by systemic pilocarpine in Wistar audiogenic rats (WAR strain). Epilepsy Behav 2004;5:455–63. [8] Dal-Col ML, Terra-Bustamante VC, Velasco TR, Oliveira JA, Sakamoto AC, GarciaCairasco N. Neuroethology application for the study of human temporal lobe epilepsy: from basic to applied sciences. Epilepsy Behav 2006;8:149–60. [9] Schmued LC, Albertson C, Slikker Jr W. Fluoro-Jade: a novel fluorochrome for the sensitive and reliable histochemical localization of neuronal degeneration. Brain Res 1997;751:37–46. [10] Timm F. Histochemistry of zinc. Dtsch Z Gesamte Gerichtl Med 1958;47:428–31. [11] Mello LE, Cavalheiro EA, Tan AM, et al. Circuit mechanisms of seizures in the pilocarpine model of chronic epilepsy: cell loss and mossy fiber sprouting. Epilepsia 1993;34:985–95. [12] Babb TL, Pretorius JK, Mello LE, Mathern GW, Levesque MF. Synaptic reorganizations in epileptic human and rat kainate hippocampus may contribute to feedback and feedforward excitation. Epilepsy Res Suppl 1992;9:193–202. [13] Mathern GW, Leite JP, Babb TL, et al. Aberrant hippocampal mossy fiber sprouting correlates with greater NMDAR2 receptor staining. NeuroReport 1996;7:1029–35. [14] McNamara JO. Cellular and molecular basis of epilepsy. J Neurosci 1994;14: 3413–25. [15] Tauck DL, Nadler JV. Evidence of functional mossy fiber sprouting in hippocampal formation of kainic acid-treated rats. J Neurosci 1985;5:1016–22. [16] Elmer E, Kokaia Z, Kokaia M, Lindvall O, McIntyre DC. Mossy fibre sprouting: evidence against a facilitatory role in epileptogenesis. NeuroReport 1997;8:1193–6. [17] Xu B, McIntyre DC, Fahnestock M, Racine RJ. Strain differences affect the induction of status epilepticus and seizure-induced morphological changes. Eur J Neurosci 2004;20:403–18. [18] Longo BM, Mello LE. Blockade of pilocarpine- or kainate-induced mossy fiber sprouting by cycloheximide does not prevent subsequent epileptogenesis in rats. Neurosci Lett 1997;226:163–6. [19] Longo BM, Mello LE. Supragranular mossy fiber sprouting is not necessary for spontaneous seizures in the intrahippocampal kainate model of epilepsy in the rat. Epilepsy Res 1998;32:172–82. [20] Longo BM, Mello LE. Effect of long-term spontaneous recurrent seizures or reinduction of status epilepticus on the development of supragranular mossy fiber sprouting. Epilepsy Res 1999;36:233–41. [21] Longo BM, Sanabria ER, Gabriel S, Mello LE. Electrophysiologic abnormalities of the hippocampus in the pilocarpine/cycloheximide model of chronic spontaneous seizures. Epilepsia 2002;43(Suppl 5):203–8. [22] Longo B, Covolan L, Chadi G, Mello LE. Sprouting of mossy fibers and the vacating of postsynaptic targets in the inner molecular layer of the dentate gyrus. Exp Neurol 2003;181:57–67. [23] Williams PA, Wuarin JP, Dou P, Ferraro DJ, Dudek FE. Reassessment of the effects of cycloheximide on mossy fiber sprouting and epileptogenesis in the pilocarpine model of temporal lobe epilepsy. J Neurophysiol 2002;88:2075–87. [24] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. San Diego: Academic Press; 1996. [25] Croiset G, De WD. Proconvulsive effect of vasopressin; mediation by a putative V2 receptor subtype in the central nervous system. Brain Res 1997;759:18–23. [26] Arida RM, Scorza FA, Peres CA, Cavalheiro EA. The course of untreated seizures in the pilocarpine model of epilepsy. Epilepsy Res 1999;34:99–107. [27] Fentress JC, Stilwell FP. Letter: grammar of a movement sequence in inbred mice. Nature 1973;244:52–3. [28] Norton S. The study of sequences of motor behavior. In: Iversen LC, Iversen SD, Snyder SH, editors. Handbook of psychopharmacology: principles of behavioral pharmacology, vol. 7. New York: Plenum; 1977. p. 82–105. [29] Garcia-Cairasco N. A critical review on the participation of inferior colliculus in acoustic–motor and acoustic–limbic networks involved in the expression of acute and kindled audiogenic seizures. Hear Res 2002;168:208–22. [30] Pinel JP, Rovner LI. Experimental epileptogenesis: kindling-induced epilepsy in rats. Exp Neurol 1978;58:190–202. [31] Garcia-Cairasco N, Doretto MC, Prado RP, Jorge BP, Terra VC, Oliveira JA. New insights into behavioral evaluation of audiogenic seizures: a comparison of two ethological methods. Behav Brain Res 1992;48:49–56. [32] Danscher G. Histochemical demonstration of heavy metals: a revised version of the sulphide silver method suitable for both light and electronmicroscopy. Histochemistry 1981;71:1–16. [33] Williams RW, Rakic P. Elimination of neurons from the rhesus monkey's lateral geniculate nucleus during development. J Comp Neurol 1988;272:424–36. [34] Turski WA, Cavalheiro EA, Schwarz M, Czuczwar SJ, Kleinrok Z, Turski L. Limbic seizures produced by pilocarpine in rats: behavioural, electroencephalographic and neuropathological study. Behav Brain Res 1983;9:315–35.
266
M.A. Furtado et al. / Epilepsy & Behavior 20 (2011) 257–266
[35] Lemos T, Cavalheiro EA. Suppression of pilocarpine-induced status epilepticus and the late development of epilepsy in rats. Exp Brain Res 1995;102:423–8. [36] Kohane DS, Holmes GL, Chau Y, Zurakowski D, Langer R, Cha BH. Effectiveness of muscimol-containing microparticles against pilocarpine-induced focal seizures. Epilepsia 2002;43:1462–8. [37] Gulec G, Isbil-Buyukcoskun N. Arginine vasopressin does not contribute to seizures induced by intracerebroventricularly-injected pilocarpine. Neuropeptides 2005;39:575–9. [38] Cavalheiro EA, Leite JP, Bortolotto ZA, Turski WA, Ikonomidou C, Turski L. Longterm effects of pilocarpine in rats: structural damage of the brain triggers kindling and spontaneous recurrent seizures. Epilepsia 1991;32:778–82. [39] Bertram EH, Cornett J. The ontogeny of seizures in a rat model of limbic epilepsy: evidence for a kindling process in the development of chronic spontaneous seizures. Brain Res 1993;625:295–300. [40] Bertram EH, Cornett JF. The evolution of a rat model of chronic spontaneous limbic seizures. Brain Res 1994;661:157–62. [41] Rodrigues MC, Guizzo R, dos Santos WF, Cairasco NG. A comparative neuroethological study of limbic seizures induced by Parawixia bistriata venom and kainic acid injections in rats. Brain Res Bull 2001;55:79–86. [42] Leite JP, Garcia-Cairasco N, Cavalheiro EA. New insights from the use of pilocarpine and kainate models. Epilepsy Res 2002;50:93–103. [43] Wieser HG. Electroclinical features of the psychomotor seizure. Stuttgart: Fischer London: Butterworths; 1983. [44] Kotagal P, Luders HO, Williams G, Nichols TR, McPherson J. Psychomotor seizures of temporal lobe onset: analysis of symptom clusters and sequences. Epilepsy Res 1995;20:49–67. [45] Manford M, Fish DR, Shorvon SD. An analysis of clinical seizure patterns and their localizing value in frontal and temporal lobe epilepsies. Brain 1996;119(Pt 1):17–40. [46] Rodrigues MC, Rossetti F, Foresti ML, et al. Correlation between shaking behaviors and seizure severity in five animal models of convulsive seizures. Epilepsy Behav 2005;6:328–36. [47] Leite JP, Bortolotto ZA, Cavalheiro EA. Spontaneous recurrent seizures in rats: an experimental model of partial epilepsy. Neurosci Biobehav Rev 1990;14:511–7. [48] Turski WA, Cavalheiro EA, Turski L, Kleinrok Z. Intrahippocampal bethanechol in rats: behavioural, electroencephalographic and neuropathological correlates. Behav Brain Res 1983;7:361–70. [49] Poirier JL, Capek R, De KY. Differential progression of Dark Neuron and Fluoro-Jade labelling in the rat hippocampus following pilocarpine-induced status epilepticus. Neuroscience 2000;97:59–68. [50] Kubova H, Druga R, Haugvicova R, Suchomelova L, Pitkanen A. Dynamic changes of status epilepticus-induced neuronal degeneration in the mediodorsal nucleus of the
[51]
[52]
[53]
[54]
[55] [56] [57] [58] [59]
[60]
[61]
[62]
[63]
thalamus during postnatal development of the rat. Epilepsia 2002;43(Suppl 5): 54–60. Turski WA, Cavalheiro EA, Bortolotto ZA, Mello LM, Schwarz M, Turski L. Seizures produced by pilocarpine in mice: a behavioral, electroencephalographic and morphological analysis. Brain Res 1984;321:237–53. Castro OW, Furtado MA, Fernandes A, Tilelli CQ, Garcia Cairasco N. Comparative neuroanatomical and temporal characterization of Fluorojade-positive neurodegeneration after status epilepticus induced by systemic and intrahippocampal pilocarpine in Wistar rats. Proceedings, American Epilepsy Society Meeting, Boston, MA, 50. Epilepsia; 2009. p. 333. Romcy-Pereira RN, Garcia-Cairasco N. Hippocampal cell proliferation and epileptogenesis after audiogenic kindling are not accompanied by mossy fiber sprouting or Fluoro-Jade staining. Neuroscience 2003;119:533–46. Gorter JA, Goncalves Pereira PM, van Vliet EA, Aronica E, da Silva FH Lopes, Lucassen PJ. Neuronal cell death in a rat model for mesial temporal lobe epilepsy is induced by the initial status epilepticus and not by later repeated spontaneous seizures. Epilepsia 2003;44:647–58. Colombo JA, Yanez A, Puissant V, Lipina S. Long, interlaminar astroglial cell processes in the cortex of adult monkeys. J Neurosci Res 1995;40:551–6. Wasterlain CG, Shirasaka Y. Seizures, brain damage and brain development. Brain Dev 1994;16:279–95. Shih TM, McDonough Jr JH. Neurochemical mechanisms in soman-induced seizures. J Appl Toxicol 1997;17:255–64. Niquet J, Liu H, Wasterlain CG. Programmed neuronal necrosis and status epilepticus. Epilepsia 2005;46(Suppl 5):43–8. Mohapel P, Ekdahl CT, Lindvall O. Status epilepticus severity influences the longterm outcome of neurogenesis in the adult dentate gyrus. Neurobiol Dis 2004;15: 196–205. Tilelli CQ, Del VF, Fernandes A, Garcia-Cairasco N. Different types of status epilepticus lead to different levels of brain damage in rats. Epilepsy Behav 2005;7: 401–10. Dudek FE, Obenaus A, Schweitzer JS, Wuarin JP. Functional significance of hippocampal plasticity in epileptic brain: electrophysiological changes of the dentate granule cells associated with mossy fiber sprouting. Hippocampus 1994;4:259–65. Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol 1989;26: 321–30. Nissinen J, Lukasiuk K, Pitkanen A. Is mossy fiber sprouting present at the time of the first spontaneous seizures in rat experimental temporal lobe epilepsy? Hippocampus 2001;11:299–310.