Epilepsy & Behavior 51 (2015) 112

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Long-term intracerebroventricular infusion of angiotensin II after

kainate-induced status epilepticus: Effects on epileptogenesis, brain
damage, and diurnal behavioral changes
Natasha M. Ivanova a, Dimitrina Atanasova a, Daniela M. Pechlivanova a, Rumyana Mitreva a, Nikolai Lazarov b,
Alexander G. Stoynev c, Jana D. Tchekalarova a,
a
b
c

Institute of Neurobiology, Bulgarian Academy of Sciences, Soa, Bulgaria

Department of Anatomy, Medical Faculty, MU-Soa, Bulgaria
Department of Pathophysiology, Medical Faculty, MU-Soa, Bulgaria

a r t i c l e

i n f o

Article history:
Received 14 May 2015
Revised 17 June 2015
Accepted 18 June 2015
Available online xxxx
Keywords:
Angiotensin II
Kainate
Diurnal rhythms
Behavior
Brain damage

a b s t r a c t
Our previous studies revealed that Angiotensin (Ang) II has anticonvulsant effects in acute seizure models.
However, data on its role in experimental models of epilepsy are missing. In the present study, we tested whether
posttreatment with Ang II after kainate (KA)-induced status epilepticus (SE) can affect epileptogenesis,
concomitant behavioral changes, and brain damage. The Wistar rats were intracerebroventricularly infused via
osmotic mini-pumps with Ang II (1.52 g/l/day for 28 days) after SE. Spontaneous motor seizures (SMS) were
video-recorded for up to three months. Locomotor activity, anxiety, and depression-like behavior were evaluated
during the last week of drug infusion, while spatial memory was assessed during the 3rd month after SE.
Angiotensin II decreased the latency for onset of the rst SMS and increased the frequency of SMS two months
after SE. The continuous peptide infusion exacerbated the KA-induced hyperactivity and caused depressionlike behavior. The reduced anxiety of KA-treated rats was alleviated by Ang II exposure. The KA-induced decit
in the hippocampal-dependent spatial memory was not inuenced by Ang II. However, Ang II partially prevented
the neuronal damage in the hippocampus, specically in the CA1 area. The role of AT1 and AT2 receptor activation
in the effects of the octapeptide is discussed.
2015 Elsevier Inc. All rights reserved.

1. Introduction
The reninangiotensin system (RAS) is known to participate in the
control of numerous physiological and behavioral functions including
regulation of blood pressure, release of pituitary gland hormones,
water and salt homeostasis, stress responses, cognitive processes, and
depression [1]. Most of these functions are mediated by angiotensin
(Ang) II type 1 (AT1) receptors [24]. Angiotensin II has been considered
as an active ligand at this receptor subtype. High AT1 receptor expression has been reported in brain areas involved in autonomic, hormonal,
cerebrovascular, and behavioral regulations, such as the pituitary gland,
area postrema, hypothalamus, circumventricular organs (CVOs), and
amygdala [5,6]. Besides the well-known classical physiological functions, literature data suggest that AT1 receptor subtype is involved in
the modulation of brain excitability, long-term potentiation, and control
of seizure susceptibility [710]. In this context, our studies [8] and that
Corresponding author at: Institute of Neurobiology, Acad. G. Bonchev Str., Bl. 23,
Bulgarian Academy of Sciences, Soa 1113, Bulgaria. Tel.: +359 2979 2172; fax: +359 2
719 109.
E-mail address: janetchekalarova@gmail.com (J.D. Tchekalarova).

http://dx.doi.org/10.1016/j.yebeh.2015.06.036
1525-5050/ 2015 Elsevier Inc. All rights reserved.

of Stragier et al. [11] have demonstrated that the biologically active

Ang peptides, Ang II, Ang III, and Ang IV, exert an anticonvulsant activity
in acute seizure models in rodents, as well as in pentylenetetrazol-,
bicuculline-, picrotoxine-, as well as pilocarpine-induced seizures.
Moreover, recently presented clinical data, which revealed an upregulation of 1 receptors and their mRNA expression in the cortex and hippocampus in both patients, diagnosed with temporal lobe epilepsy
(TLE) [12], support the presumption that the effects of Ang II on seizure
susceptibility might be mediated by the 1 receptor subtype. Repetitive seizures induced in an experimental model of TLE caused an upregulation of the components of RAS, ACE (angiotensin-converting
enzyme), and the AT1 receptor expression in the hippocampus of Wistar
audiogenic rats [13] and the pilocarpine model of TLE [14]. The longterm treatment with losartan after kainate (KA)-induced status epilepticus (SE) has been found to increase the latency for onset of the rst
spontaneous seizure and to prevent some of the deleterious consequences accompanying the chronic epileptic state in rats [15].
In view of the fact that, recently, we have found that acute injection of
Ang II possesses an anticonvulsant effect in a battery of seizure tests with
different mechanisms of action in mice, in the present study, we aimed
further to explore the efcacy of long-term intracerebroventricular

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

(i.c.v.) infusion of Ang II, started after KA-induced SE, on the development of epileptogenesis, deleterious behavioral consequences, and
brain damage in Wistar rats.
2. Material and methods
The procedures used in this study were in agreement with the
European Communities Council Directive 2010/63/EU. The experimental design was approved by the Institutional Ethics Committee at the
Institute of Neurobiology and the Ethics Committees for research at
the Soa Medical University under the contract No. 30/2011 for the
application grant DTK 02/56 20092012.
2.1. Subjects
The experiments were performed on male Wistar rats (sixty-day
old) (250300 g) obtained from an animal breeding facility of the
Institute of Neurobiology, Bulgarian Academy of Sciences. Following
arrival in the laboratory, the animals were individually housed under
standardized conditions (20 3 C, 4050% humidity; 12/12-h light/
dark cycle with lights on at 08:00 h) and habituated for 10 days. Food
and water were available ad libitum throughout the study except during
test procedures.
2.2. Experimental design
The rats were randomly distributed in four experimental groups
(n = 1014) as follows: Group I: control sham rats treated with vehicle
(C-sham); Group II: control rats treated with Ang II (C-Ang); Group III:
sham rats treated with KA (KAsham); and Group IV: rats treated with
KA + Ang II (KAAng). The observers, who counted and scored seizures,
as well as behavior in these animals, were not aware of the group to
which the animals belonged.
2.3. Induction of status epilepticus with kainic acid
Twenty-eight rats in total were subjected to KA injection. Kainic acid
was diluted in sterile saline (0.9% NaCl) at 2.5 mg/ml. The protocol of
KA-induced SE was executed according to [15]. In brief, SE was induced
by repetitive injections of KA (Abcam, UK) starting with a dose of 5
mg/kg, i.p. (1 ml/kg) at the rst hour of observation. Thereafter, KA
was delivered in half of the abovementioned dose every half an hour.
Matched controls were treated with an equivalent volume and number
of injections of saline. Seizure intensity was evaluated by a modied
Racine's scale [16] as previously [17,18]. Only rats which developed SE
(i.e., recurrent seizures with bilateral forelimb clonus for at least 3 h)
and survived thereafter were included in the subsequent analyses.
2.4. Implantation of osmotic mini-pumps
The surgery was performed under ketamine (40 mg/kg) and xylazine
(20 mg/kg, intraperitoneally i.p.) anesthesia ve days after SE. Following
local anesthesia with procaine 0.5% and xation on a stereotaxic device
(Narishige Sci. Inst. Labs, Japan), a midline incision over the skull was
made, and the skin and periosteum were removed with aseptic precautions. A 28-gauge, stainless steel cannula (Alzet Brain Infusion Kit 2,
Durect, Cupertino, CA) was implanted into the right lateral cerebral ventricle and xed on the skull with dental cement. The cannula was placed
1.0 mm posterior and 1.4 mm lateral to the bregma. The lower end of the
cannula was at a depth of 3.0 mm from the skull, and the upper end was
connected to an osmotic mini-pump (Alzet model 2004) for chronic i.c.v.
infusion at a pumping rate of 0.23 0.02 l/h for a 28-day period (model
2004). The pumps were lled with either Ang II (groups II and IV)
(Sigma-Aldrich, Bulgaria) (1.52 g/l/day) or 0.9% saline solution (groups
I and III) and placed subcutaneously on the back of the rats.

2.5. Video-monitoring of spontaneous recurrent seizures

Video-monitoring (24 h/day for 12 weeks starting 24 h after SE) was
executed by an infrared-sensitive color camera (S-2016, AVTECH,
Taiwan, no. AVC307R) connected to a computer. The recordings were
visually analyzed by two independent observers for the detection of
spontaneous motor seizures (SMS) of class IV or V (secondarily generalized seizures). Partial seizures of classes I and II were neglected because,
without simultaneous EEG recording, they could easily be missed. All
spontaneous seizures detected during the experimental manipulations
when the animals were outside of their boxes were also noted. Several
seizure parameters were evaluated: latent seizure-free period and
frequency of SMS.
2.6. Behavioral tests
The behavioral tests were performed during the last week of Ang II
infusion, i.e., between the 25th and the 33rd day after SE. The radial
arm maze test was executed between the 10th and the 12th week
after SE. The open eld (OF) test, elevated plus maze (EPM) test, and
forced swim test (FST) were performed at two time points 6 h after
lights on/off (at 15:00 p.m. and 03:00 a.m., respectively) under articial
diffused light during the light phase and in red dim light during the dark
phase. The behavioral experiments were conducted in a soundproof
room, where the animals were moved at least 30 min before each test.
The rats that exhibited SMS at least 1 h before starting the test were
excluded from the experimental procedure. The behavior in the OF and
EPM tests was recorded using an infrared sensitive CCD camera and a
video tracking system (SMART PanLab software, Harvard Apparatus,
USA).
2.6.1. Open eld test
The apparatus consisted of a gray polystyrene box (100 100
60 cm) divided into two zones: outer square (periphery) and inner square
(center). The rat was placed in the center of the box and was allowed to
explore it for 5 min. The calculated standard measures were as follows:
1) total distance traveled (cm); 2) time spent in the central zone (sec)
vs total time in %; and 3) an anxiety index calculated using the following
equation: Anxiety index = 1 [(Center time / Total time + (Center
distance / Total distance)) / 2]. Anxiety index values range from 0 to 1,
with a higher value indicating increased anxiety [19]. After each test,
the OF was thoroughly cleaned with 0.1% acetic acid solution to prevent
any odor traces.
2.6.2. Elevated plus maze test
The apparatus consisted of two open arms (50 10 cm), two enclosed
arms (50 10 50 cm), and a central platform (10 10 cm) elevated
50 cm above the oor level. At the beginning of the test, the rat was placed
on the central platform facing an open arm. The test lasted 5 min. The calculated standard measures were as follows: 1) total distance traveled
(cm); 2) time (sec) spent in the open arms vs total time in %; and 3) anxiety index, which unies all EPM parameters into one ratio. Anxiety
index = 1 [(Open arms time / Total time) + (Distance open arms /
Total distance) / 2]. After each test, the EPM was cleaned with 0.1% acetic
acid solution.
2.6.3. Sucrose preference test
The test was performed as described previously [18]. Taste preference was expressed as a percentage of the volume of sucrose solution
of the total volume of uid (sucrose plus regular water) consumed during 12 h (light phase 8:0020:00 h and dark phase 20:008:00 h).
2.6.4. Forced swim test
The despair-like behavior was evaluated by a classic forced swim
test [20]. The test was carried out in a clear and transparent cylinder
(50 cm tall25 cm diameter) lled to a level of 30 cm from the bottom

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

with 24 C tap water. The water in the apparatus was changed per rat
trial. Two swim sessions were conducted: 15 min on the 1st day and
5 min on the 2nd day. After each test, the rat was dried and kept
warm by a heating device for 10 min. Behavior during the 2nd day
(test) was recorded by two skilled experimenters, who were unaware
of the treatment conditions. The parameter measured was immobility
(in seconds) which occurred when the rat remained motionless or
made only movements necessary to keep its head above the water.
2.6.5. Radial arm maze (RAM) test
Visuospatial learning and memory was assessed using an 8-arm radial maze (RAM) (Harvard Biosci. Comp., USA). The stainless steel apparatus consisted of a central octagonal platform (30 cm in diameter) from
which eight identical arms (42 12 12) radiated from the platform.
The maze was elevated 50 cm above the oor-level. A variety of environmental cues (wall pictures, table, cupboards, door, and window)
were available for facilitation of spatial navigation. Seven days before
the start of training and during the test, rats were put on a diet for
15% reduction in b.w. All animals were habituated (shaping) for up to
3 days in the maze. A test for hippocampus-dependent spatial memory
was performed over 18 days with one trial per day. The test was
executed as previously described in detail [21]. Reentry into a baited
arm from which the food pellet had already been retrieved was scored
as working memory errors. The number of arms visited per session
was also recorded. After each trial, the maze was wiped down with 1%
acetic acid.
2.7. Histology
After obtaining a deep anesthesia with urethane (1500 mg/kg, i.p.,
Sigma-Aldrich) (n = 5 per group), rats were transcardially perfused initially with 0.05 M phosphate-buffered saline at pH 7.3 followed by 4%
paraformaldehyde in 0.1 M phosphate buffer (PB) at pH 7.3. The brains
were dissected out and postxed overnight at 4 C in the same xative
solution. After postxation, the brains were sliced in the coronal plane;
the tissue blocks were washed in PB, embedded in parafn, and cut into
5-m thick sections. The samples were then deparafnized with xylene
and ethanol and routinely stained with hematoxylin and eosin to
better identify pyknotic nuclei of damaged neurons. The sections were
investigated on a Nikon Eclipse 80i light microscope (Japan) and
photographed with a digital camera (Nikon DMX 1200). Sections were
analyzed for major cell loss with special attention to the dorsal and ventral hippocampi as previously described [22]. The staining intensity and
density of nerve cell bodies were estimated using Nikon's NIS Elements
Digital Imaging software. The relative neuronal densities of the selected
brain areas were quantied by determining the percentage of the
measurement grid occupied by stained cells. The resulting values

provide a relative index of the number of stained neurons in the selected

brain areas.
2.8. Statistical analysis
Parametric (for normally distributed data) or nonparametric tests
(for data not normally distributed) were used for the statistical analysis
(SigmaStat 11.0). Experimental data were evaluated by two-way
ANOVA (seizures and histological data) and three-way ANOVA (SPT,
OF, EPM, and RAM tests) with Condition (KA groups vs vehicle groups),
Drug (Ang II groups vs sham groups), Phase (light vs dark) as betweengroup factors, and number of sessions (RAM) as within-group factor.
Group differences after signicant ANOVAs were measured by post
hoc Bonferroni or Holm Sidak test. If data were not normally distributed,
ANOVA for nonparametric data (KruskalWallis on ranks) followed by
the MannWhitney U test was used. For the RAM test, with the exception of the rst day of session data, from blocks of three sessions were
combined for each animal to obtain more stable results as compared
to a single day of session. The data from the rst session day were presented separately, as this day reects the rst exposure of the rat to the
maze task. A p b 0.05 value was accepted as indicating statistically
signicant differences.
3. Results
3.1. Seizure activity
All of the KA-injected rats developed SE, and four (14%) of the rats
died after SE. The rats which survived were equally allocated to two
treatment groups (KAsham and KAAng II, n = 12/group). Longterm i.c.v. infusion of Ang II after SE signicantly shortened the latent
seizure-free period (median SD: 7 6.5 days; range: 625 days)
compared to that of the sham-treated rats (median SD: 15
8.3 days; range: 638 days) (MannWhitney Rank Sum Test: T =
120; p = 0.031).
A signicant increase in the seizure frequency as a function of time
was demonstrated both in the KAsham group (Op b 0.05 the 3rd vs
the 1st month) and in the KAAng II group (Op b 0.05 the 2nd and 3rd
vs the 1st month), respectively (Fig. 2). The KAAng II group demonstrated higher frequency of the SMS compared to the KAsham group
during the 2nd month after SE (p = 0.009). (See Fig. 1.)
3.2. Open eld test
Analysis of data by three-way ANOVA showed a main Condition
effect [F1,132 = 89.45, p b 0.001], a Drug effect [F1,132 = 19.66, p b
0.001], and a Phase effect [F1,132 = 8.60, p b 0.004] as well as

Fig. 1. Schematic illustration of the experimental protocol.

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 2. Seizure frequency in rats during 14, 58 and 912 weeks, respectively, after KA-induced SE. The data represent number of seizures in the KAsham group (n = 12) and KAAng II
group (n = 12), respectively. Ang II/vehicle infusion started 5 days after SE for a 28-day period. *p b 0.05 between groups (vs the KAsham group); Op b 0.05 within-group (one-way
ANOVA).

Condition Drug interaction [F1,132 = 15.33, p b 0.05] for the total distance traveled in the OF test. Post hoc test revealed that Ang II potentiated the increased locomotor activity observed in the KA-treated rats
without diurnal variations (p b 0.05) (Fig. 3). However, diurnal uctuations of total activity distance were detected in C-sham and C-Ang rats,
respectively (p b 0.05).
Three-way ANOVA revealed a main Condition effect [F1,136 = 24.62,
p b 0.001], a Drug effect [F1,136 = 8.76, p b 0.004], as well as Condition
Drug interaction [F1,136 = 5.33, p b 0.022] for the ratio of time spent in
the center vs total time. Angiotensin II exerted a phase-dependent
anxiogenic effect in the control rats (*p b 0.05), while in the KAtreated rats, the octapeptide was able to mitigate the low anxiety level
during both the light and the dark phase (p b 0.05) (Fig. 4A). Similarly,
a main Condition effect [F1,102 = 15.45, p b 0.001], a Drug effect [F1,102 =

24.53, p b 0.001], as well as Condition Drug [F1,102 = 12.89, p b 0.001]

were demonstrated for the anxiety index. Post hoc test showed that
both the C-Ang and the KAAng group were characterized by close to
the C-sham anxiety index, while the KAsham group showed a signicantly lower anxiety index than the C-sham group without diurnal
variations (p b 0.05) (Fig. 4B).

3.3. Elevated plus maze test

Three-way ANOVA revealed a main Condition effect [F1,105 = 50.66,
p b 0.001], a Drug effect [F1,105 = 11.17, p b 0.001], a Phase effect
[F1,105 = 9.66, p b 0.002], as well as Condition Drug interaction
[F1,105 = 4.43, p b 0.038]. Angiotensin II additionally enhanced the

Fig. 3. Effect of long-term i.c.v. Ang II infusion on locomotor activity in the open eld (OF) test indicated by total distance traveled in cm. Data are means SEM (n = 1014). *p b 0.05 vs
C-sham group, Op b 0.05 vs KAsham rats, #p b 0.05 within-group (15:00 h vs 03:00 h).

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 4. Effect of long-term i.c.v. Ang II infusion on anxiety level in the OF measured by time spent in center vs total time (A) and anxiety index (B) in %. Data are means SEM (n = 1014).
*p b 0.05 vs C-sham group, Op b 0.05 vs KAsham group, #p b 0.05 within-group (15:00 h vs 03:00 h).

KA-induced higher locomotion during the light and dark phases,

respectively (p b 0.05) (Fig. 5).
Analysis of data by three-way ANOVA indicated a main Condition effect [F1,87 = 63.98, p b 0.001], a Drug effect [F1,87 = 7.63, p b 0.007], a
Phase effect [F1,87 = 12.83, p b 0.001], as well as Condition Drug
Phase interaction [F1,72 = 14.55, p b 0.001] for the ratio of time spent in
the open arms vs total time. The KA-induced low anxiety level was alleviated to control level by Ang II during both the light and the dark phase
(p b 0.05) (Fig. 6A). Moreover, Ang II infusion exerted an anxiogenic effect
in the control group during the dark phase (p b 0.05).
A main Condition effect [F1,89 = 39.96, p b 0.001], a Drug effect
[F1,89 = 14.60, p b 0.001], and Condition Drug Phase interaction
[F1,89 = 4.799, p b 0.031] for the anxiety index in the EPM were evident.
In accordance with the ratio of time spent in the open arms vs total time,
the Ang II infusion exerted an anxiogenic effect with respect to the

anxiety index during the dark phase (p b 0.05) (Fig. 6B). Moreover,
the octapeptide was able to alleviate the KA-induced low anxiety
index to control level (p b 0.05).

3.4. Sucrose preference test

Analysis of data by three-way ANOVA indicated the following: a
main Condition effect [F1.109 = 13.82, p b 0.001], a Drug effect
[F1.109 = 16.46, p b 0.001], as well as Condition Phase interaction
[F1.109 = 5.12, p b 0.026]. Post hoc test showed that Ang II elicited
lower sucrose consumption in both the C-Ang and KAAng group
without diurnal variations (p b 0.05) (Fig. 7), while anhedonia-like behavior was detected only during the light phase in the KAsham
group (p b 0.05).

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 5. Effect of long-term i.c.v. Ang II infusion on elevated plus maze (EPM) test indicated by total distance traveled in cm. Data are means SEM (n = 1014). *p b 0.05 vs C-sham group,
O
p b 0.05 vs KAsham group, #p b 0.05 within-group (15:00 h vs 03:00 h).

3.5. Forced swim test

Three-way ANOVA demonstrated no signicant effect of any factor
except for Condition Drug Phase interactions [F1,83 = 7.91, p b
0.006]. Unlike the KAsham group, which exhibited depressive-like behavior only during the light phase, Ang IIsham and KAAng II groups
were characterized by a signicantly increased immobility time during
the dark phase (p b 0.05) (Fig. 8).
3.6. Radial arm maze test
3.6.1. Working memory error (WME)
Three-way ANOVA showed a signicant main effect of Condition
[F1,242 = 17.715, p b 0.001] and Session [F1,242 = 4.219, p b 0.001] but
no interaction among the three factors. Both control groups (C-sham
and C-Ang II) successfully learned the task and decreased the WME
along the 18-day period of training [one-way ANOVA: H = 26.809,
p = b 0.001 and F6,54 = 6.409, p b 0.001, respectively]. Contrary to the
controls, both epileptic groups, KAsham and KAAng, were unable to
learn the RAM task since no signicant difference among trial days
was observed within groups (p N 0.05) (Fig. 9A).
3.6.2. Number of arms visited per session
Three-way ANOVA showed a signicant main Condition effect
[F1,230 = 51.571, p b 0.001], a Drug effect [F1,230 = 5.646, p b 0.018],
and Condition Drug Session interaction [F1,230 = 3.835, p b 0.001].
Unlike the epileptic groups, the control groups (C-sham and C-Ang II)
decreased the number of arms visited per session along the 18-day period of training [one-way ANOVA: H = 27.824, p = b0.001 and F6,54 =
8.712, p b 0.001] (Fig. 9B).
3.7. Histopathological changes
Analysis of data by two-way ANOVA indicated the following: (1) for
CA1 area (septotemporal), a main Condition effect [F1,9 = 7.264, p b
0.031]; (temporal): a main Condition effect [F1,9 = 278.735, p b 0.001]
and a Drug effect [F1,9 = 120.05, p b 0.001]; (2) for CA3a area
(septotemporal): a main Condition effect [F1,9 = 7.264, p b 0.031];
(temporal): a main Condition effect [F1,9 = 21.397, p b 0.002]

and Drug effect [F1,9 = 5.837, p b 0.046]; and (3) for CA3c area
(septotemporal): a main Condition effect [F1,9 = 20.764, p b 0.002];
(temporal): a main Condition effect [F1,9 = 8.003, p b 0.022]. A severe
neuronal loss in the CA1 (septotemporal and temporal), CA3a
(septotemporal and temporal), and CA3c (septotemporal and temporal) areas of the hippocampus was detected in the KAveh group
(p b 0.05 vs C-sham) (Figs. 10, 11A). The long-term Ang II infusion
during epileptogenesis caused neuroprotection selectively in the CA1
area of the hippocampus in the KA-treated rats (p b 0.05), while the
neuronal loss in the temporal CA3 area was not as severe as in the
KAsham group (Fig. 11A).
Analysis of data by two-way ANOVA in the hilus of dentate gyrus
indicated a main Condition effect for the septotemporal [F1.8 =
20.795, p b 0.002] and the temporal hilus [F1.8 = 5.995, p b 0.040],
respectively. The septotemporal hilus of the dentate gyrus showed a
more severe neuronal loss as a consequence of KA-induced neurotoxicity (p b 0.05) (Fig. 11B).
4. Discussion
The present study demonstrated, for the rst time, the effect of longterm i.c.v. infusion of Ang II (1.52 g/l/day for a 28-day period), started
after the KA-induced SE, on epileptogenesis, behavioral impairment,
and brain damage in Wistar rats. Surprisingly, our ndings revealed
that, although Ang II exacerbated epileptogenesis and concomitant
behavioral changes, it exerted a neuroprotective effect in the CA1 area
of the hippocampus.
Contrary to our previous work, where a single i.c.v. injection
with Ang II showed anticonvulsant effect in a number of acute seizure
tests and the pentylenetetrazol-kindling model in mice [22,23], the
long-term exposure to this octapeptide not only did not prevent
epileptogenesis but also aggravated its development in the KA model
of TLE in rats. The discrepancy in the effects of Ang II can be attributed
to the different route of administration (single injection vs long-term infusion) and seizure tests in nave animals vs chronic model of epilepsy
and species (mice vs rats). Therefore, these results suggest that the
abovementioned differences in the study protocol are important and
crucial for the expression of either anticonvulsant or proepileptogenic
effects of this neuropeptide and its specic receptor activation. The
present results are in accordance with clinical ndings considering the

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 6. Effect of long-term i.c.v. Ang II infusion on anxiety level in the EPM measured by time (sec) spent in open arms vs total time (A) and anxiety index (B) in %. Data are means SEM
(n = 1014). *p b 0.05 vs C-sham group, Op b 0.05 vs KAsham group, #p b 0.05 within-group (15:00 h vs 03:00 h).

close link between increased levels of angiotensin peptides and TLE

[24]. We did not measure the level of Ang II in our experimental protocol but there are controversial data concerning the levels of endogenous
peptide in the three phases of the pilocarpine model of TLE. Thus, while
Hong et al. [25] reported that Ang II levels are increased at the end of the
acute phase and during the latent phase, Gouveia et al. [14] found higher
levels only during the chronic phase. Therefore, further studies are
needed to establish when exactly the levels of Ang II in the hippocampus are changed in models of TLE, including the KA model of TLE.
A number of literature data demonstrated a dual excitatory/inhibitory
effect of Ang II on neuronal excitability in different brain regions [2629]
suggesting a ne-tuning modulatory mechanism, which might underlie
the effects of this neuropeptide through direct activation of AT1/AT2 receptors, or presynaptic or postsynaptic interaction with either inhibitory

or excitatory neurotransmitter systems such as GABA or glutamate, as

well as dopamine and norepinephrine.
Recently, we reported that long-term treatment with the selective
AT1 receptor antagonist losartan, which started during the KA-induced
SE, increased the seizure-free latent period and alleviated seizure activity during the chronic phase of epilepsy [15]. The antioxidant and
antiinammatory effects exerted by the treatment with losartan during
the acute epileptic phase, which include suppression of oxidative stress
and increase in HSP 70, specically in the hippocampus of rats [30], are
in line with the hypothesis that a blockade of AT1 receptors might represent a putative neuroprotective mechanism against the KA-induced
neurotoxicity. In this respect, the neuroprotective efcacy of the longterm losartan treatment after SE was demonstrated specically in the
CA1 area of the hippocampus in epileptic rats [15]. On the contrary,

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 7. Effect of long-term i.c.v. Ang II infusion on sugar consumption in KA-treated rats. Taste preference is expressed as in % over a 12-h period (light and dark phases). Data are means
SEM (n = 1014). *p b 0.05 vs C-sham group, Op b 0.05 vs KAsham group, #p b 0.05 within-group (15:00 h vs 03:00 h).

AT1 receptor-mediated responses of Ang II involve an enhanced expression of inammatory and oxidant substances such as leukotrienes, free
radicals, C-reactive protein, and prostaglandins [31]. The involvement
of AT1 receptors in TLE is also supported by the ndings that these receptors are upregulated in the hippocampus both in experimental
models of epilepsy [13,14] and in patients diagnosed with TLE [12].
These data suggest that a blockade of AT1 receptors might represent
an important tool for the prevention of epileptogenesis.
A putative role of RAS, including Ang II in the regulation of anxiety,
depression, and memory consolidation has been suggested earlier [9].
Recently, we reported that the long-term blockade of AT1 receptor attenuated hyperactivity in epileptic rats during the light phase [15]. In

the present study, long-term i.c.v. exposure to Ang II exacerbated the

KA-induced increased locomotion in rats in the OF and EPM tests,
without diurnal variations, suggesting an involvement of AT1 receptors
in the effects of this octapeptide. However, the precise mechanism
of Ang II receptor-mediated responses is uncertain and needs additional
studies.
We found that Ang II alleviated the low anxiety index to control level
in the KA-treated rats without diurnal variations in the OF and EPM
tests. Interestingly, in our previous work, losartan aggravated this behavioral index during the light phase, while showing a response similar
to that of Ang II during the dark phase. These results suggest
phase-dependent AT1/AT2 receptor-mediated responses of exogenous

Fig. 8. Effect of long-term i.c.v. Ang II infusion on immobility time (sec) in the forced swim test. Data are means SEM (n = 1014). *p b 0.05 vs C-sham group, Op b 0.05 vs KAsham
group, #p b 0.05 within-group (15:00 h vs 03:00 h).

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 9. Effect of long-term i.c.v. Ang II infusion on working memory error (A) and in the number of arms visited per session (B) tested in radial arm maze over an 18-day trial period. Data are
means SEM (n = 1014). *p b 0.05 vs C-sham group, +p b 0.05 vs 1st session in the same group.

Ang II in the KA-treated rats. In this respect, the hypothesis that

AT1 receptor activation is responsible for Ang II effects on anxiety
during epileptogenesis corresponds with the ndings of Marinzalda
et al. [32] that emotional responses might be under AT1 receptormediated control. The octapeptide has similar afnity for AT1 and
AT2 receptors, and literature data support the idea that AT1/AT2 receptor activation determines the opposite effects of Ang II [33]. Our
present behavioral data agree with previous reports indicating
circadian-dependent responses of Ang II [34]. The involvement of
the angiotensin system in the modulation of circadian uctuations
in the periphery has been suggested earlier to be mediated through
AT1 receptor activation [35].
In agreement with our previous studies, the KA-treated rats showed
a depression-like behavior in SPT and FST, which was demonstrated

during the light phase. The octapeptide not only did not prevent this
emotional disturbance in the KA-treated rats during the light phase
but also caused depression during the dark phase, too. The alleviating
effect of Ang II on increased immobility in the KA-treated rats during
the light phase might represent a false-positive response because of
an increased motor activity in the KAAng II group. Previously, we demonstrated that long-term AT1 receptor blockade is able to prevent the
KA-induced depression during the light phase [15] suggesting AT1
receptor-mediated depression-like behavior in Ang II-infused rats.
The present results are in line with experimental and clinical data
suggesting that angiotensin-converting enzyme inhibitors have
potential antidepressant activity [3638].
Angiotensin II infusion exerted pronounced neuroprotection
selectively in the CA1 area and partly reduced the neuronal damage in

10

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

Fig. 10. Representative conventionally H&E-stained sections of the hippocampal formation of a C-sham rat (A), a KAsham rat (B), and KAAng II rat (C). The representative images on the
right panels are higher magnications of the boxed areas in the left images from the CA1 and CA3c areas of the hippocampus, respectively. Please note that the KAsham rats (B) showed
severe neuronal loss in CA1 and CA3 pyramidal cell layers, and the hilus of the dentate gyrus (DG) when compared to the C-sham rats (A). Scale bars = 200 m (AC, left) and 75 m in
high-resolution insets (AC, right).

the temporal CA3 area of the hippocampus. This result was unexpected
in view of the fact that the octapeptide aggravated behavior related to
locomotion and depression. Although in this study we were unable to
measure nonconvulsive seizures (stages III), our results revealed that
the frequency of motor seizures was increased after discontinuation of
the long-term octapeptide infusion. We started the long-term Ang II
infusion several days following SE suggesting that the neuroprotective
effect of the peptide in the CA1 area was associated with an active
cell death requiring time for the lethal initiation phase to occur [39]. Interestingly enough, recently, we reported that long-term treatment
with losartan also reduced the KA-induced damage specically in the
CA1 area of the hippocampus [15]. There are at least two explanations

for these ndings. One is that the favorable effects of this octapeptide
might be mediated by the AT2 receptor activation. In support of this
suggestion is the report of Wilms et al. [40] that AT1 and AT2 receptors
demonstrate opposite effects on neuronal injury caused by stroke or
brain trauma with protection resulting from AT2 receptor activation.
Furthermore, recently, Bar-Klein et al. [41] hypothesized that the
blocking of TGF-b signaling, AT1 signaling, or a combination of both is involved in the effect of losartan on suppression of seizures after exposure
to albumin and after vascular injury.
Angiotensin AT1 and AT2 receptors are missing or expressed at a low
density in mammalian brain structures [9,33]. However, they are
reported to be upregulated under pathological conditions in the

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

11

Fig. 11. Effect of long-term i.c.v. Ang II infusion on KA-induced brain damage. Neuronal damage in the hippocampus (septotemporal and temporal) CA1, CA3a, and CA3c pyramidal cell
layers (A) and in the hilus of dentate gyrus (septotemporal and temporal) (B). *p b 0.05 vs C-group; Op b 0.05 vs KA-treated rats.

hippocampus of a rat model of epilepsy [13] as well as in patients with TLE

[12]. The other explanation might be that the exogenous Ang II exerts an
indirect modulatory effect on inhibitory neurotransmission [2629].
The KAsham rats exhibited impaired spatial learning over the
whole testing period of 18 days which is in agreement with our previous
data with epileptic rats [30]. Our data do not suggest a clear-cut involvement of Ang II in the control of spatial memory whether in sham controls or in KA-treated rats. Previous ndings demonstrated that
endogenous Ang II does not participate in spatial memory formation
[42]. Duchemin et al. [43] reported that chronic infusion with a higher
dose of Ang II that induced a malignant hypertension in mice for a period of 14 or 21 days caused learning and memory decits in the Morris
water maze test, conducted during the same period of peptide infusion,
i.e., after the third week of exposure. The higher frequency of arms visited per session in the Ang II-treated rats, which is also evident in the

KA-treated rats with impaired learning capacity, can result from impulsive behavior and amnestic effect of KA neurotoxicity.
Although in this study blood pressure was not measured, following
previous data [44,45], our experimental protocol of drug delivery suggests that Ang II did not cause obvious changes in blood pressure of
KA-treated rats. Our preliminary experiments did not demonstrate a
signicant effect on BP after the KA-induced SE (unpublished data).
However, epileptogenesis might exacerbate the effect of Ang II infusion
on blood pressure, which needs further clarication.
5. Conclusion
Whereas the long-term i.c.v. infusion of Ang II after the KA-induced
SE aggravated epileptogenesis and worsened behavioral changes including hyperlocomotion and depression, it caused a neuroprotection

12

N.M. Ivanova et al. / Epilepsy & Behavior 51 (2015) 112

in the CA1 area of the hippocampus. Our present results and previous
ndings on the effects of long-term AT1 receptor suggest a ne-tuning
modulatory mechanism, which might underlie the effects of this neuropeptide through direct activation of AT1/AT2 receptors.
Acknowledgments
This work was supported by the Medical Science Council, Medical
University, Soa, Bulgaria (research grant 28/2012) and the National
Science Fund (research grant # DTK 02/56 20092012). The authors
thank Sabina Mitova, Medical Department, Medical University, Soa
for her expert technical assistance with the study.
Disclosure/conicts of interest
None of the authors has any conict of interest concerning this
manuscript.
References
[1] Wright JW, Harding JW. Brain reninangiotensina new look at an old system. Prog
Neurobiol 2011;95:4967.
[2] Allen AM, Zhuo J, Mendelsohn FA. Localization and function of angiotensin AT1
receptors. Am J Hypertens 2000;13:31S8S.
[3] de Gasparo M, Catt KJ, Inagami T, Wright JW, Unger T. International union
of pharmacology. XXIII. The angiotensin II receptors. Pharmacol Rev 2000;52:41572.
[4] Thomas WG, Mendelsohn FA. Angiotensin receptors: form and function and distribution. Int J Biochem Cell Biol 2003;35:7749.
[5] Rowe BP, Saylor DL, Speth RC. Analysis of angiotensin II receptor subtypes in
individual rat brain nuclei. Neuroendocrinology 1992;55:56373.
[6] Tsutsumi K, Saavedra JM. Angiotensin-II receptor subtypes in median eminence and
basal forebrain areas involved in regulation of pituitary function. Endocrinology
1991;129:30018.
[7] ukawski K, Janowska A, Jakubus T, Tochman-Gawda A, Czuczwar SJ. Angiotensin
AT1 receptor antagonists enhance the anticonvulsant action of valproate in the
mouse model of maximal electroshock. Eur J Pharmacol 2010;640:1727.
[8] Tchekalarova J, Georgiev V. Angiotensin peptides modulatory system: how is it
implicated in the control of seizure susceptibility? Life Sci 2005;76:95570.
[9] Wright JW, Yamamoto BJ, Harding JW. Angiotensin receptor subtype mediated
physiologies and behaviors: new discoveries and clinical targets. Prog Neurobiol
2008;84:15781.
[10] von Bohlen und Halbach O, Albrecht D. Angiotensin II inhibits longterm potentiation
within the lateral nucleus of the amygdala through AT1 receptors. Peptides 1998;19:
10316.
[11] Stragier B, Clinckers R, Meurs A, De Bundel D, Sarre S, Ebinger G, et al. Involvement of
the somatostatin-2 receptor in the anticonvulsant effect of angiotensin IV against
pilocarpine-induced limbic seizures in rats. J Neurochem 2006;98:110013.
[12] Argaaraz GA, Konno AC, Perosa SR, Santiago JF, Boim MA, Vidotti DB, et al. The renin
angiotensin system is upregulated in the cortex and hippocampus of patients with temporal lobe epilepsy related to mesial temporal sclerosis. Epilepsia 2008;49:134857.
[13] Pereira M, Becari C, Oliviera J, Salgadoc O, Garcia-Cairasco N, Costa-Net C. Inhibition
of the reninangiotensin system prevents seizures in a rat model of epilepsy. Clin Sci
2010;119:47782.
[14] Gouveia TL, Frangiotti MI, de Brito JM, de Castro Neto EF, Sakata MM, Febba AC, et al.
The levels of reninangiotensin related components are modied in the hippocampus
of rats submitted to pilocarpine model of epilepsy. Neurochem Int 2012;61:5462.
[15] Tchekalarova J, Ivanova N, Pechlivanova D, Atanasova D, Lazarov N, Kortenska L,
et al. Antiepileptogenic and neuroprotective effect of losartan in kainate model of
temporal lobe epilepsy. Pharmacol Biochem Behav 2014;127:2736.
[16] Racine RJ. Modication of seizure activity by electrical stimulation: II. Motor seizure.
Electroencephalogr Clin Neurophysiol 1972;32:28194.
[17] Tchekalarova J, Pechlivanova D, Itzev D, Lazarov N, Markova P, Stoynev A. Diurnal
rhythms of spontaneous recurrent seizures and behavioral alterations of Wistar
and spontaneously hypertensive rats in kainate model of epilepsy. Epilepsy Behav
2010;17:2332.
[18] Tchekalarova J, Pechlivanova D, Atanasova Ts, Markova P, Lozanov V, Stoynev Al.
Diurnal variations of depressive-like behavior of Wistar and spontaneously
hypertensive rats in kainate model of temporal lobe epilepsy. Epilepsy Behav
2011;20:27785.
[19] Mazor A, Matar MA, Kaplan Z, Kozlovsky N, Zohar J, Cohen H. Gender-related
qualitative differences in baseline and post-stress anxiety responses are not
reected in the incidence of criterion-based PTSD-like behaviour patterns. World J
Biol Psychiatry 2009;10:85669.
[20] Porsolt RD, Bertin A, Blavet N, Deniel M, Jalfre M. Immobility induced by forced
swimming in rats: effects of agents which modify central catecholamine and serotonin activity. Eur J Pharmacol 1979;57:20110.

[21] Tchekalarova J, Petkova Z, Pechlivanova D, Moyanova Sl, Kortenska L, Mitreva R,

et al. Prophylactic treatment with melatonin after status epilepticus: effects on
epileptogenesis, neuronal damage and behavioral changes in kainate model of
temporal lobe epilepsy. Epilepsy Behav 2013;27:17487.
[22] Tchekalarova J, Ivanova N, Pechlivanova D, Atanasova D, Lazarov N, Kortenska L,
et al. Antiepileptogenic and neuroprotective effect of losartan in kainate model of
temporal lobe epilepsy. Pharmacol Biochem Behav 2014;127:2736.
[23] Tchekalarova J, Georgiev V. Adenosineangiotensin II interactions in pentylenetetrazol seizure threshold in mice. J Physiol Paris 1999;93:1917.
[24] Almeida SS, Naffah-Mazzacoratti MG, Guimares PB, Wasinski F, Pereira FEG,
Canzian M, et al. Carbamazepine inhibits angiotensin I-converting enzyme, linking
it to the pathogenesis of temporal lobe epilepsy. Transl Psychiatry 2012;2:e93.
[25] Hong S, HaiQin W, Xin Y, GuiLian Z, Ru Z, ShuQin Z, et al. Angiotensin II and its
receptor in activated microglia enhanced neuronal loss and cognitive impairment
following pilocarpine-induced status epilepticus. Mol Cell Neurosci 2015;65:5867.
[26] Albrecht D, Broser M, Krger H. Excitatory action of angiotensins II and IV on
hippocampal neuronal activity in urethane anesthetized rats. Regul Pept 1997;70:
1059.
[27] Armstrong DL, Garcia EA, Ma T, Quinones B, Wayner MJ. Angiotensin II blockade of
long-term potentiation at the perforant path-granule cell synapse in vitro. Peptides
1996;17:68993.
[28] Haas HL, Felix D, Celio MR, Inagami T. Angiotensin II in the hippocampus. A
histochemical and electrophysiological study. Experientia 1980;36:13945.
[29] Palovcik RA, Phillips MI. Saralasin increases activity of hippocampal neurons
inhibited by angiotensin II. Brain Res 1984;323:3458.
[30] Tchekalarova J, Ivanova N, Pechlivanova D, Illieva K, Atanasova M. Strain-dependent
effects of chronically infused losartan against kainic acid induced seizures, oxidative
stress and heat shock protein 72 expression. Cell Mol Neurobiol 2013;34:13342.
[31] Das UN. Is angiotensin II an endogenous pro-inammatory molecule? Med Sci
Monit 2005;11:15562.
[32] Marinzalda M de L, Prez PA, Gargiulo PA, Casarsa BS, Bregonzio C, et al.
Fear-potentiated behaviour is modulated by central amygdala angiotensin II AT1 receptors stimulation. Biomed Res Int 2014. http://dx.doi.org/10.1155/2014/183248.
[33] von Bohlen und Halbach O, Albrecht D. The CNS reninangiotensin system. Cell
2006;326:599616.
[34] Braga AN, da Silva Lemos M, da Silva JR, Fontes WR, dos Santos RA. Effects of angiotensins on day-night uctuations and stress-induced changes in blood pressure. Am
J Physiol Regul Integr Comp Physiol 2002;282:R166371.
[35] Nonaka H, Emoto N, Ikeda K, Fukuya H, Rohman MS, Yagita K, et al. Angiotensin II
induces circadian gene expression of clock genes in cultured vascular smooth
muscle cells. Circulation 2001;104:17468.
[36] Germain L, Chouinard G. Treatment of recurrent unipolar major depression with
captopril. Biol Psychiatry 1988;23:63741.
[37] Giardina WJ, Ebert DM. Positive effects of captopril in the behavioral despair swim
test. Biol Psychiatry 1989;25:697702.
[38] Zubenko GS, Nixon RA. Mood-elevating effect of captopril in depressed patients. Am
J Psychiatry 1984;141:1101.
[39] Sloviter RS. Apoptosis: a guide for the perplexed. Trends Pharmacol Sci 2002;23:1924.
[40] Wilms H, Rosenstiel P, Unger T, Deuschl G, Lucius R. Neuroprotection with angiotensin receptor antagonists: a review of the evidence and potential mechanisms. Am J
Cardiovasc Drugs 2005;5:24553.
[41] Bar-Klein G, Cacheaux LP, Kamintsky L, Prager O, Weissberg I, Schoknecht K, et al.
Losartan prevents acquired epilepsy via TGF-b signaling suppression. Ann Neurol
2014;75:86475.
[42] Chalas A, Conway EL. No evidence for involvement of angiotensin II in spatial
learning in water maze in rats. Behav Brain Res 1996;81:199205.
[43] Duchemin S, Belanger E, Wu R, Ferland G, Girouard H. Chronic perfusion of angiotensin
II causes cognitive dysfunctions and anxiety in mice. Physiol Behav 2013;109:638.
[44] Irvine RJ, White JM. The effects of central and peripheral angiotensin on hypertension and nociception in rats. Pharmacol Biochem Behav 1997;57:3741.
[45] Zhang Q, Yao F, O'Rourke S, Qian S, Sun C. Angiotensin II enhances GABAB receptormediated responses and expression in nucleus tractus solitarii of rats. Am J Physiol
Heart Circ Physiol 2009;297:H183744.

Glossary
Ang: angiotensin
KA: kainate
SE: status epilepticus
SMS: spontaneous motor seizures
RAS: reninangiotensin system
CVOs: circumventricular organs
TLE: temporal lobe epilepsy
ACE: angiotensin-converting enzyme
OF: open eld
EPM: elevated plus maze
FST: forced swim test
RAM: radial arm maze
BP: blood pressure