Physiology & Behavior 83 (2004) 165 171

Physical training reverts hippocampal electrophysiological changes in rats

submitted to the pilocarpine model of epilepsy
Ricardo Mario Aridaa,b,*, Emilio Rafael Garrido Sanabriaa, Andre Cesar da Silvaa,
Leonardo Coutinho Fariaa, Fulvio Alexandre Scorzaa,b, Esper Abrao Cavalheiroa
a
Laboratorio de Neurologia Experimental, Universidade Federal de Sao Paulo-Escola Paulista de Medicina (UNIFESP/EPM), Rua Botucatu 862,
Vila Clementino, CEP 04023-900 Sao Paulo, Brazil
b
Laboratorio de Neurociencias, Nucleo de Pesquisas Tecnologicas-Universidade de Mogi das Cruzes (NPT/UMC), Sao Paulo, Brazil

Received 4 May 2004; received in revised form 9 August 2004; accepted 12 August 2004

Abstract

Physical exercise and fitness programs in patients with epilepsy are still a matter of controversy. Effects of physical exercise in animals
with epilepsy have been demonstrated. To further investigate the possible mechanisms by which physical activity interferes with
epileptogenesis, the present work was aimed to study the effect of aerobic exercise on bin vitroQ hippocampal electrophysiological parameters
observed in rats submitted to the pilocarpine model of epilepsy. Electrophysiological changes were monitored by extracellular field potentials
recorded from CA1 area. Control rats and rats with epilepsy were submitted to an aerobic exercise program. The number of population spikes
(PS) and slope of field excitatory postsynaptic potentials (fEPSP) were analyzed. Trained rats with epilepsy exhibited a reduction in PS when
compared with nontrained rats with epilepsy in different concentrations of extracellular potassium or bicuculline. Physical training also
enhanced the late phase of LTP in rats with epilepsy. Our results indicate that physical training reduces CA1 hyperresponsiveness and can
modify synaptic plasticity in rats submitted to the pilocarpine model of limbic epilepsy.
D 2004 Published by Elsevier Inc.

Keywords: Physical training; Exercise; Epilepsy; Pilocarpine; Long-term potentiation; Population spikes

1. Introduction objectives to assess physical fitness by using standardized

tests of physical endurance [13] or physical training
The relationship of physical exercise and epilepsy, and programs [4]. But there is no clear evidence for short- and
the implications of epilepsy for physical fitness programs long-term effects of physical exercise on epileptogenesis.
have been controversial. Several patients have been Previous studies from our laboratory [5,6] have analyzed the
unnecessarily cautioned against participation in physical effect of physical exercise on two experimental models of
activities. temporal lobe epilepsy. In a first study, we observed that
Most of the relevant literature has compared physical and chronic physical activity in rats was able to retard the
social activities in patients with epilepsy based on ques- development of the amygdala kindling, an experimental
tionnaires and/or clinical studies. They also directed their model of progressive epileptogenesis [5]. In a subsequent
study, by using the pilocarpine model of epilepsy in rats [7],
it was shown that physical training (aerobic physical
program) reduces the frequency of spontaneous recurrent
* Corresponding author. Laboratorio de Neurologia Experimental, seizures during the chronic period of this epilepsy model
Universidade Federal de Sao Paulo-Escola Paulista de Medicina, Rua
Botucatu 862, Vila Clementino, CEP 04023-9000, Sao Paulo, SP. Brazil.
[6]. Aiming to understand the basic mechanisms by which
Tel.: +55 11 55764508; fax: +55 11 55739304. physical activity would modify the natural course of
E-mail address: [email protected] (R.M. Arida). epilepsy in animals, the objective of the present work was
0031-9384/$ - see front matter D 2004 Published by Elsevier Inc.
doi:10.1016/j.physbeh.2004.08.008
166 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171

to study the effects of an aerobic physical program on the groups were submitted to hippocampal electrophysiological
hippocampal electrophysiological characteristics of rats with analysis. Accordingly, rats were deeply anaesthetized with
epilepsy using the pilocarpine model. halothane before decapitation. The brain was quickly
removed and trimmed before being placed in the vibratome
(Vibraslice, Campden Instruments). Transverse hippocam-
2. Materials and methods pal slices (400 um) were cut in modified sucrose-based
artificial cerebrospinal fluid containing (in mM): sucrose,
Adult Wistar rats weighing 200280 g were housed 129.0; KCl, 3.0; NaH2PO4, 1.25; NaHCO3, 22; CaCl2, 1;
under environmentally controlled conditions (07001900 h MgSO4, 5 and d-glucose, 10. Slices were placed in a
light/dark cycle; 2224 8C) and permitted free access to holding chamber under incubation with standard ACSF for
food and water throughout the experiment. The procedures at least 1 h before recording. Then, slices were transferred
involving the animals and their care at the Experimental to the interface recording chamber with a temperature
Neurology Laboratory of the Federal University of Sao control system at 33F0.3 8C and flow rate fixed around
Paulo respected the Institutions guidelines, which comply 1.31.5 ml min 1. The ACSF medium was continuously
with national and international rules and policies. prewarmed at 30 8C and pregassed with a mixture of 95%
For the induction of epilepsy, rats were prepared O2 and 5% CO2. The standard ACSF (pH 7.4) was
according to established protocols [8]. Accordingly, sus- composed (in mM) of: NaCl, 129.0; KCl, 3.0; NaH2PO4,
tained seizures were induced by a single intraperitoneal 1.25; NaHCO3, 22; CaCl2, 1.8; MgSO4, 1.8 and d-
administration of pilocarpine hydrochloride (350 mg/kg; glucose, 10. For LTP experiments, field excitatory post-
Sigma, St. Louis, MO). Scopolamine methylnitrate synaptic potentials (fEPSPs) were obtained from apical
(Sigma) was injected (1 mg/kg, s.c.) 30 min before dendritic layer of CA1 area (stratum radiatum). Ortho-
pilocarpine in order to reduce peripheral cholinergic effects dromic population spikes (PS) were recorded with micro-
[8]. Following status epilepticus, the surviving animals were electrode positioned in stratum pyramidale. Those
continuously monitored by a video system during 24 h for extracellular recordings were obtained using glass micro-
detection of recurrent spontaneous seizures. Chronically pipettes filled with 1 M NaCl and having resistances of 4
epileptic rats suffering at least three seizures were submitted 6 MV. The recording electrode is placed approximately
to physical training sessions (trained rats with epilepsy; 200 Am below the top surface of the slice and at a distance
n=10). In brief, after familiarizing with the training of approximately 600 Am from the stimulation electrode in
apparatus for three consecutive days by placing them on a the stratum radiatum of CA1. Population spikes (PS) were
treadmill (Columbus Instruments) for 10 min/day at a speed studied in slices taken from 10 animals in each exper-
of 12 m/min at 0% degree incline, rats were submitted to an imental group. These responses were evoked by electrical
aerobic exercise program of 45 sessions on a treadmill, 6 stimulation of CA1 afferent fibers (Schaffer collaterals and
days per week. Each training session started with a 5-min commissural pathway) using a bipolar Teflon-insulated
warm-up at 1215 m/min. Running time and speed iridiumplatinum microelectrode gently placed a stratum
gradually increased from 30 min at 18 m/min during the radiatum towards CA3. Stimulus was delivered using
first 3 days to 60 min at 2225 m/min during the subsequent programmable-stimulator Master-8 (AMPI, Jerusalem,
days. The intensity of exercise (60% VO2max) was Israel) and stimulator isolator NL800 (Digitimer, England).
determined for each animal by the maximum O2 uptake The slice was stimulated at submaximal intensity until a
test. Maximum O2 uptake (VO2max) was measured with an stable response (13 mV fEPSP amplitude) was obtained.
open-circuit system (Columbus Instruments Oxymax Sys- The stimulus intensity is then adjusted so that the evoked
tem). The testing protocol for determination of VO2max was fEPSP apparently does not trigger postsynaptic cells to fire
as follows. After a warm-up at 12 m/min on a 0% grade for an action potential; that is, no contamination with PS is
5 min, treadmill speed was increased by 3 m/min every 3 observed. Test stimuli (0.05 Hz) were delivered using an
min until animals refused to run. The VO2 measurements isolation unit and ranged 100150 As duration pulses and
were realized before and at the end of the physical training 330 V. An inputoutput function was obtained using test
program. Another group of rats with epilepsy not submitted pulses at different incrementing intensity levels. For fEPSP
to training sessions (nontrained rats with epilepsy; n=10) and LTP experiments, test stimuli intensity was adjusted at
were used as control of the effect of physical activity on 3545% of the intensity that produced maximum slope
hippocampal electrophysiological parameters. Two other changes. In epileptic slices, stimulation of high intensities
groups of naive animals, one submitted to training sessions can trigger polyspiking activity (multiple population
as described above (control trained; n=10) and another spikes) that can mask the fEPSP. To avoid this problem,
group free of any treatment (control; n=10) were used as stimulus was set at the intensity in which no apparent
control of the influence of the epileptic condition on the contamination from PS appeared in the fEPSP. For PS,
hippocampal in vitro study. stimulus intensity was adjusted to evoke a PS about 50%
As already mentioned, following epilepsy induction and of maximal amplitude. During epileptiform responses,
the 45 days of physical training, animals from the four multiple PS commonly emerged from slow positively.
 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171 167

The number of PS was counted by inspecting the VO2max before and after the aerobic physical program
individual peak negativities in every response (usually among groups.
six sweeps at different stimulus intensities). Different
concentrations of potassium (3.5 and 8 mM) and 10 AM 3.3. Population spike responses in CA1 area
bicuculline were bath-applied to change excitability. Only
one slice was studied at the time in the interface chamber. The number of population spikes (PS) was compared
Two slices per animal were used. Control traces were at a stimulus intensity that evoked maximal-amplitude
recorded at test stimuli for at least 30 min in order to initial PS. Extracellular responses in CA1 stratum
establish baseline responses. To induce LTP, tetanic stimuli radiatum in control tissue were characterized by a single
were applied through the same bipolar electrode and population spike at different stimulus intensities as
consisted in two trains of high-frequency stimulation previously described [911]. There were no significant
(HFS) at 100 Hz (1-s duration) separated by 25-s interval. differences in number of PS between slices from control
After HFS, fEPSP responses at single test stimuli (0.05 and trained control rats when compared at several
Hz) were followed over time further than 1 h. The concentrations of extracellular potassium (3.5 mM K+
electrophysiological signals were amplified using AXO- and 8 mM K+) or 10 AM bicuculline (3.5 mM K+).
CLAMP2-B system (Axon Instruments), digitized (DIG- However, there was an increase in the number of PS in
IDATA 1200) and recorded in a personal computer using nontrained rats with epilepsy when compared to control
PCLAMP6 software. The population spike amplitude of rats at 3.5 mM K+ (nontrained rats with epilepsy=
the evoked responses was measured between the negative 3.0F0.3; controls=1.0F0.0, meanFS.E.M.) and 8 mM
peak and a line drawn on the top of the two positive peaks. K+ (nontrained rats with epilepsy=3.83F0.3; controls=
For fEPSP analysis, the rising slope of negative population 1.6F0.2, meanFS.E.M., pb0.05, ANOVA) concentrations
EPSP was measured by placing two cursors (2 ms (Fig. 1A and B). The analysis of number of population
separation) at about 25 ms following stimulation (at the spikes confirmed previous data showing hyperresponsive-
middle of fEPSP rising phase) using Clampfit 6 software. ness of epileptic tissue when compared with control
All fEPSPs data during the LTP experiments were divided slices. This abnormality was observed in slices of all
by mean fEPSP slope calculated during the pre-LTP animals with epilepsy but not submitted to the physical
(baseline) period. Successive normalized slope values were training program. Spontaneous epileptiform discharges as
then averaged and plotted versus time in a graph. brief paroxysms of multiple population spikes were
Statistical comparisons were made using paired or observed sporadically in nontrained rats with epilepsy
unpaired t-test or ANOVA with repeated measures as and trained rats with epilepsy, but this variable was not
appropriated. consistently studied in the present work. Trained rats with
epilepsy exhibited a significant reduction in hyperrespon-
siveness (3.5 mM K+=1.2F0.2; 8 mM K+=2.2F0.2;
3. Results Bic=2.6F0.2, meanFS.E.M.) when compared with non-
trained rats with epilepsy (3.5 mM K+=3.0F0.3; 8 mM
3.1. Behavioral data K + =3.8F0.3; Bic=4.71F0.4, meanFS.E.M., pb0.05,
ANOVA; Fig. 1A and B).
As already observed in previous experiments [6],
physical training was able to reduce the number of 3.4. Long-term potentiation of fEPSPs
spontaneous seizures in rats with epilepsy during the 45
days of observation (1.0F0.7 seizures per day for the Normal fEPSPs were recorded in control and trained
trained group versus 1.7F1.2 seizures per day for the control animals, consisting in negative outgoing potentials
nontrained group, meanFS.E.M., pb0.05). after Schaffer collateral stimulation. Baseline responses
for about 30 min were stable in all groups. The
3.2. VO2max data development of LTP was followed in all experimental
groups. There was no difference in the magnitude of LTP
After 45 days of exercise, there was a significant in control groups (Students t-test, p=0.12). Nontrained
increase in VO2max from 39.1 ml/kg/min [3344, 95% rats with epilepsy exhibited a significant reduction in
confidence interval (CI)] to 48.2 ml/kg/min (4453, 95% posttetanic potentiation as measured during 2 min
CI) ( pb0.01ANOVA analysis) in control trained rats. following LTP induction (ANOVA, pN0.05; Fig. 2).
Trained rats with epilepsy presented similar VO2max Posttetanic potentiation was significantly reduced in
values, and significant increase in VO2max from 38.2 trained rats with epilepsy when compared to control
ml/kg/min [3241, 95% confidence interval (CI)] to 47.6 and trained control rats (ANOVA, pb0.05) but not
ml/kg/min (4151, 95% CI) ( pb0.01ANOVA analysis). significantly when compared to nontrained rats with
VO2max generally occurred while rats ran at speeds of 32 epilepsy (ANOVA, pN0.05). The deficit in LTP was
38 m/min. No significant differences were found in sustained during the course of the experiment in rats with
168 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171

Fig. 1. Population spikes recorded in CA1 area under different experimental manipulations. Panel A shows electrophysiological recordings in control (a) and
control trained (b), nontrained rats with epilepsy (c) and trained rats with epilepsy (d). Notice a single population spike under 3.5 mM K+ and normal
responsiveness to 8 mM K+ and bicuculline application in (a) and (b). Nontrained rats with epilepsy (c) exhibiting hyperesponsiveness characterized by multiple
(*) population spikes in normal 3.5 mM K+ and higher responsiveness to 8 mM K+ and bicuculline. Trained rats with epilepsy (d) show hyperexcitability, and
apparently normal responsiveness to 8 mM K+ and bicuculline. Panel B shows statistical analysis confirming a reduction in the number of population spikes in
trained rats with epilepsy when compared to nontrained rats with epilepsy in extracellular 3.5 mM K+, 8 mM K+ and 3.5 mM K+ bicuculline concentrations
(*pb0.05). An increase in number of PS in nontrained rats with epilepsy was observed when compared to control rats at 3.5 mM K+ and 8 mM K+ concentrations
(**pb0.05). Analysis used repeated-measures ANOVA followed by multiple comparisons was used. Data are expressed as meanFS.E.M.

epilepsy (trained and nontrained). Analysis of time course 4. Discussion

of LTP from pooled data of trained rats with epilepsy
shows that, although a decrement of LTP was still present 4.1. Physical exercise reduces hyperexcitability in rats with
compared to control rats and trained control rats epilepsy
( pb0.05), it was significantly different in the late phase
of LTP (90 min) when compared with nontrained rats The main observation in our experiments is related
with epilepsy (Students t-test, pb0.05). Trained control to the fact that rats with epilepsy exposed to regular
rats exhibited higher levels of potentiation characterized physical training exhibited marked changes in epilepsy-
by a significant recovery of fEPSP slopes during this induced hippocampal electrophysiological abnormalities
phase (Fig. 2). when compared to their respective controls (nontrained
 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171 169

Fig. 2. Analysis of LTP in hippocampal slices from different groups at different time intervals. (A) Representative traces from LTP experiments during baselines
( 2 min prior to high-frequency stimulation) and posttetanic stimulation (2, 60 and 90 min following stimulation). (B) The graph shows no differences between
slices from control and control trained rats before (2 min) and after high-frequency stimulation (2, 30, 60 and 90 min). Nontrained rats with epilepsy exhibit a
significant reduction in fEPSP slope when compared with the control group (*pb0.05). Trained rats with epilepsy present higher slope measurements after
high-frequency stimulation when compared to nontrained rats with epilepsy (**pb0.05). Those values were not significantly different compared to control
slices. Analysis used repeated-measures ANOVA followed by multiple comparisons. Data are expressed as meanFS.E.M.

rats with epilepsy). These changes were mainly charac- temporal lobe epilepsy [1214] and the present findings of
terized by a reduction in hippocampal population spike a reduction in CA1 hyperresponsiveness might be an bin
number in response to variable potassium extracellular vitroQ indicator of the chronic effect of physical exercise in the
concentration and to the addition of bicuculline in the epileptogenic process as previously observed bin vivoQ [5,6].
bathing milieu. These findings might be underlying the
processes by which physical training was able to reduce 4.2. Physical exercise partially rescue LTP impairment in
the number of spontaneous seizures in rats with rats with epilepsy
epilepsy [6]. However, a question that remains unsolved
and that deserves further investigation is related to the Our observations of deficits in the early phases of
mechanisms by which physical exercise could influence hippocampal LTP and of persistent deficit in posttetanic
or modify the electrophysiological properties of neuronal potentiation in trained rats with epilepsy are similar to those
networks. described by other authors in animals submitted to different
Hippocampal circuitry changes seem to mediate the origin models of chronic epilepsy [15,16]. However, late LTP
of chronic epileptogenesis in experimental models of hippocampal phases were significantly improved in our
170 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171

experimental animals when compared with the long-lasting Acknowledgements

deficits in LTP observed in nontrained rats with epilepsy.
Recent studies indicate changes in neuronal plasticity after Research supported by CNPq, FAPESP, FINESP and
physical exercise [1719] and voluntary running has PRONEX (Brazil). A.C. Silva and L.C. Faria are fellow-
resulted in long-lasting survival of newborn cells and ships from CNPq.
increased dentate gyrus LTP in mice [20,21].
In these lines, several studies have shown that the type of
exercise may produce different effects on brain function References
[2224]. To better elucidate the mechanisms by which
physical exercise interfered on the reduction on seizure [1] Roth DL, Goode KT, Williams VL, Faught E. Physical exercise,
frequency, we utilized the same protocol of forced exercise stressful life experience, and depression in adults with epilepsy.
as used in our previous study [6]. In addition, the treadmill Epilepsia 1994;35:1248 55.
protocol induced significant changes in VO2max in rats with [2] Steinhoff BJ, Neususs K, Thegeder H, Reimers CD. Leisure time
activity and physical fitness in patients with epilepsy. Epilepsia 1996;
epilepsy, as commonly observed in control trained rats. 37:1221 7.
Complex mechanisms involving the activation of NMDA [3] Jalava M, Sillanpaa M. Physical activity, health-related fitness, and
receptors are relevant for LTP induction in CA1 area [25,26] health experience in adults with childhood-onset epilepsy: a controlled
and as known, these receptors are changed during hippo- study. Epilepsia 1997;38:424 9.
campal epileptogenesis [2729]. In addition, rats submitted [4] Nakken KO, Bjorholt PG, Johannessen SI, Loyning T, Lind E. Effect
of physical training on aerobic capacity, seizure occurrence, and serum
to the pilocarpine model of epilepsy exhibit a complex level of antiepileptic drugs in adults with epilepsy. Epilepsia 1990;31:
pattern of neuronal death and synaptic reorganization in 88 94.
hippocampal area [7] that further disrupts the functional [5] Arida RM, Vieira AJ, Cavalheiro EA. Effect of physical exercise on
hippocampal plasticity. It is quite probable that such kindling development. Epilepsy Res 1998;30:127 32.
[6] Arida RM, Scorza FA, Santos NF, Peres CA, Cavalheiro EA. Effect of
irreversible morpho-functional changes allow aerobic exer-
physical exercise on seizure occurrence in a model of temporal lobe
cise to revert only partially the deficits observed in the LTP epilepsy in rats. Epilepsy Res 1999;37:45 52.
of rats with epilepsy. [7] Cavalheiro EA, Leite JP, Bortolotto ZA, Turski WA, Ikonomidou C,
The molecular mechanisms by which exercise can Turski L. Long-term effects of pilocarpine in rats: structural damage of
influence the structure/function of the central nervous the brain triggers kindling and spontaneous recurrent seizures.
system, particularly those involved in epilepsy have not Epilepsia 1991;32:778 82.
[8] Turski WA, Cavalheiro EA, Schwarz M, Czuczwar SJ, Kleinrok Z,
been defined. Neurotrophins in hippocampal neurons are Turski L. Limbic seizures produced by pilocarpine in rats: behav-
regulated by several neurotransmitters [30], increased by ioural, electroencephalographic and neuropathological study. Behav
neuronal depolarization and the opening of voltage-gated Brain Res 1983;9:315 35.
calcium channels [31]. The synthesis of neurotrophins, in [9] Doller HJ, Weight FF. Perforant path activation of hippocampal CA1
particular brain-derived neurotrophic factor (BDNF) and stratum pyramidale neurons: electrophysiological evidence for a direct
pathway. Brain Res 1982;237:1 13.
nerve growth factor (NGF), is regulated by neuronal activity [10] Doller HJ, Weight FF. Perforant pathway-evoked potentiation of CA1
[32]. Neurotrophin mRNA levels are up-regulated bin vitroQ neurons in the hippocampal slice preparation. Brain Res 1985;333:
and bin vivoQ by stimulation parameters which are known to 305 10.
induce hippocampal LTP [32,33]. Moreover, physical [11] Anderson WW, Swartzwelder HS, Wilson WA. The NMDA receptor
activity has also been considered to influence neurotrophic antagonist 2-amino-5-phosphonovalerate blocks stimulus train-
induced epileptogenesis but not epileptiform bursting in the rat
factors. Neeper et al. [18] demonstrated increased BDNF hippocampal slice. J Neurophysiol 1987;57:1 21.
and NGF mRNA expression in hippocampal neurons after [12] Shumate MD, Lin DD, Gibbs JW, Holloway KL, Coulter DA.
exercise, suggesting that physiological levels of neural GABA(A) receptor function in epileptic human dentate granule cells:
activity can alter neurotrophin expression, which is in comparison to epileptic and control rat. Epilepsy Res 1998;32:114 28.
agreement to seizure studies [3436]. These results have [13] Sanabria ERG, Su H, Yaari Y. Initiation of network bursts by Ca2+-
dependent intrinsic bursting in the rat pilocarpine model of temporal
potentially important practical as well as theoretical lobe epilepsy. J Physiol 2001;532:205 16.
implications, suggesting that the alterations observed after [14] Smith BN, Dudek FE. Network interactions mediated by new
physical exercise treatment could interfere in synaptic excitatory connections between CA1 pyramidal cells in rats with
physiology and LTP phenomena from trained rats with kainate-induced epilepsy. J Neurophysiol 2002;87:1655 8.
[15] Bernard C, Wheal HV. Plasticity of AMPA and NMDA receptor-
epilepsy.
mediated epileptiform activity in a chronic model of temporal lobe
In summary, the present findings indicate that physical epilepsy. Epilepsy Res 1995;21:95 107.
training reduces CA1 hyperresponsiveness normally [16] Bernard C, Marsden DP, Wheal HV. Changes in neuronal excitability
observed in rats with epilepsy. Moreover, long-term aerobic and synaptic function in a chronic model of temporal lobe epilepsy.
physical exercise partially recovers from prominent epi- Neuroscience 2001;103:17 26.
lepsy-induced impairment of hippocampal LTP. The mech- [17] Neeper AS, Gomez-Pinilla F, Choi J, Cotman CW. Exercise and brain
neurotrophins. Nature 1995;373:109.
anisms by which physical training is able to induce such [18] Neeper AS, Gomez-Pinilla F, Choi J, Cotman CW. Physical activity
changes are not completely understood and deserve further increases mRNA for brain-derived neurotrophic factor and nerve
investigations. growth factor in rat brain. Brain Res 1996;726:49 56.
 R.M. Arida et al. / Physiology & Behavior 83 (2004) 165171 171

[19] Kempermann G, Kuhn HG, Gage FH. More hippocampal neurons in [29] Mathern GW, Pretorius JK, Mendoza D, Lozada A, Kornblum HI.
adult mice living in an enriched environment. Nature 1996;386: Hippocampal AMPA and NMDA mRNA levels correlate with
493 5. aberrant fascia dentata mossy fiber sprouting in the pilocarpine model
[20] Van Praag H, Christie BR, Sejnowski TJ, Gage FH. Running enhances of spontaneous limbic epilepsy. J Neurosci Res 1998;54:734 53.
neurogenesis, learning, and long-term potentiation in mice. Proc Natl [30] Gwag BJ, Springer JE. Activation of NMDA receptors increases
Acad Sci 1999;96:13427 31. brain-derived neurotrophic factor (BNDF) mRNA expression in the
[21] Van Praag H, Kempermann G, Gage FH. Running increases cell hippocampal formation. Neuroreport 1993;5:125 8.
proliferation and neurogenesis in the adult mouse dentate gyrus. Nat [31] Zafra F, Lindholm D, Castren E, Hartikka J, Thoenen H. Regulation of
Neurosci 1999;2:266 70. brain-derived neurotrophic factor and nerve growth factor m RNA in
[22] Brown BS, Payne T, Kim C, Moore G, Krebs P, Martin W. Chronic primary cultures of hippocampal neurons and astrocytes. J Neurosci
response of rat brain norepinephrine and serotonin levels to endurance 1992;12:4793 9.
training. J Appl Physiol 1979;46:19 23. [32] Castren E, Pitkanen M, Sirvio J, Parsadanian A, Lindholm D,
[23] Ostman I, Nyback H. Adaptive changes in central and peripheral Thoenen H, et al. The induction of LTP increases BDNF and NGF
noradrenergic neurons in rats following chronic exercise. Neuro- mRNA but decreases NT-3 mRNA in the dentate gyrus. Neuroreport
science 1976;1:41 7. 1993;4(7):895 8.
[24] Brown BS, Van Huss W. Exercise and rat brain catecholamines. J Appl [33] Dragunow M, Beilharz E, Mason B, Lawlor P, Abraham W,
Physiol 1973;34:665 9. Gluckman P. Brain-derived neurotrophic factor expression after
[25] Reymann KG, Matthies HK, Schulzeck K, Matthies H. N-methyl-d- long-term potentiation. Neurosci Lett 1993;169:232 6.
aspartate receptor activation is required for the induction of both early [34] Gall CM, Isackson PJ. Limbic seizures increase neuronal production of
and late phases of long-term potentiation in rat hippocampal slices. messenger RNA for nerve growth factor. Science 1989;245:758 61.
Neurosci Lett 1989;96:96 101. [35] Gall C. Seizure-induced changes in neurotrophin expression: impli-
[26] Grosshans DR, Clayton DA, Coultrap SJ, Browning MD. LTP leads to cations for epilepsy. Exp Neurol 1993;124:150 66.
rapid surface expression of NMDA but not AMPA receptors in adult [36] Isackson PJ, Huntsman MM, Murray KD, Gall CM. BDNF mRNA
rat CA1. Nat Neurosci 2002;5:27 33. expression is increased in adult rat forebrain after limbic seizures:
[27] Martin D, McNamara JO, Nadler JV. Kindling enhances sensitivity of temporal patterns of induction distinct from NGF. Neuron 1991;6:
CA3 hippocampal pyramidal cells to NMDA. J Neurosci 1992;12: 937 48.
1928 35.
[28] Bernard C, Wheal HV. A role for synaptic and network plasticity in
controlling epileptiform activity in CA1 in the kainic acid-lesioned rat
hippocampus in vitro. J Physiol 1996;495(Pt 1):127 42.