Epilepsy Research
Epilepsy Research
Epilepsy Research
Epilepsy Research
journal homepage: www.elsevier.com/locate/epilepsyres
a r t i c l e i n f o a b s t r a c t
Article history: About 30% of the patients with epilepsy do not respond to clinically established anticonvulsants, despite
Received 6 June 2016 having effective concentrations of the antiepileptic drug in plasma. Therefore, new preclinical models
Received in revised form 7 October 2016 of epilepsy are needed to identify more efficacious treatments. We describe here a new drug-resistant
Accepted 24 October 2016
seizure model in mice to be used at the early stages of pre-clinical trials. This model consists in inducing
Available online 27 October 2016
daily generalized seizures for 23 consecutive days by administration of 3-mercaptopropionic acid (MP).
As a result, 100% of animals become resistant to phenytoin and 80% to phenobarbital. Such resistance
Keywords:
is strongly associated with the overexpression of P-glycoprotein (Pgp), observed in cerebral cortex, hip-
Refractory epilepsy
Animal models
pocampus and striatum while resistance to Pgp nonsubstrate drugs such as carbamazepine, diazepam
P-glycoprotein and levetiracetam is not observed.
Drug resistance This model could be useful for screening novel anticonvulsant drugs with a potential effect on phar-
3-Mercaptopropionic acid macoresistant seizures treatment.
Mice © 2016 Elsevier B.V. All rights reserved.
1. Introduction tissues (including the brain) are regulated by drug efflux trans-
porters of the ATP-binding cassette (ABC) superfamily, such as
Epilepsy is a chronic neurological disorder characterized by P-glycoprotein (Pgp), the multidrug resistance associated proteins
recurrent and spontaneous seizures (Meldrum, 1984). About 30% (MDRPs), and Breast Cancer Resistance Protein (BCRP) (Kuteykin-
of the patients with epilepsy do not respond to clinically estab- Teplyakov et al., 2009). Back in 1995, Tishler observed increased
lished anticonvulsants despite effective antiepileptic drug (AED) expression of Pgp in drug-resistant patients, suggesting that the
plasma concentrations, defining a multidrug resistance (MDR) phe- loss of response may be caused by limited brain bioavailability of
notype (Kwan and Brodie, 2000). Therefore, there is a genuine need the AEDs (Tishler et al., 1995). Thereafter, several reports indicated
to incorporate new models of refractory epilepsy (RE) at the pre- high levels of Pgp expression in epileptogenic brain tissue from
clinical stage of drug development, in order to identify new AEDs patients with RE (Sisodiya et al., 2002; Lazarowski et al., 1997;
that overcome the problem of drug resistance (White, 2003). Aronica et al., 2004; Löscher and Potschka, 2005a, b).
The inability of the AEDs to reach their molecular targets and A diversity of experimental models of pharmacoresistant
the high frequency and severity of seizures have been proposed epilepsy in rats has demonstrated the inducible Pgp overexpression
as explanations to drug resistant epilepsy (Lazarowski et al., 2007; at the blood brain barrier (BBB) which is associated to the MDR phe-
Potschka, 2010; Rogawski, 2013). Drug concentrations in several notype (Seegers et al., 2002; Rizzi et al., 2002; Jing et al., 2010; Volk
and Löscher, 2005; Bankstahl and Löscher, 2008). In these models,
Pgp levels in vessel-related cells and neurons correlated with the
loss of protective effects of phenytoin (PHT), a proven Pgp substrate
∗ Corresponding author at: Laboratorio de Investigación y Desarrollo de Bioac- (Zhang et al., 2012). These models require several weeks of treat-
tivos (LIDeB), Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas, ment and only about 40% of the surviving animals become resistant
Universidad Nacional de La Plata, La Plata, Argentina. to drugs (Löscher et al., 1993; Löscher, 2011; Potschka, 2012). These
E-mail address: andreaenrique@biol.unlp.edu.ar (A. Enrique).
http://dx.doi.org/10.1016/j.eplepsyres.2016.10.012
0920-1211/© 2016 Elsevier B.V. All rights reserved.
A. Enrique et al. / Epilepsy Research 129 (2017) 8–16 9
M e a n s e iz u r e s c o r e
Brain Pgp expression pattern has previously been studied in a
rat model of seizure induced by 3-mercaptopropionic acid (MP) 5
(Lazarowski et al., 2004). MP is a classic seizure-inductor as a con-
sequence of the inhibition of gamma aminobutyric acid (GABA) 4
synthesis. (De Sarro et al., 2003; Giraldez et al., 1999; Giraldez and
Girardi, 2000; Girardi et al., 1989, 2004; Sprince et al., 1969). The 3
repetitive seizure activity induced by daily administration of MP in
rats during 7 days induces an increase in Pgp expression in capillary 2
endothelial cells of the blood brain barrier, neurons, and glial cells
from cortex, striatum, and hippocampus (Lazarowski et al., 2004). 1
This situation is associated with astrocyte hypertrophy, character-
ized by an increase in number and thickness of branching and in 0
soma size (Girardi et al., 2004) and lower brain penetration of PHT 1 3 5 7 9 11 13 15 17 19 21 23
(Hocht et al., 2007).
Although the repetitive administration of MP in rat repre- T r e a tm e n t d a y s
sents a good experimental model to reproduce pharmacoresistant
Fig. 1. MP treatment. Swiss albino mice were administered once a day with a variable
seizures, an important disadvantage is that experiments in rats
dose of MP from 30 to 36 mg/kg. The dose was increasing according to animal behav-
require high amount of the experimental drugs.Concerning this ior in order to maintain the generalized seizure response. Mean seizure severity
issue, mouse represents an attractive cost-efficient model. score ± SEM vs MP treatment days (n = 47).
The aim of this study was the development of a new model of
pharmacoresistant seizures linked to Pgp overexpression induced
by repetitive administration of MP in mice. This model could be 2.2. 3-Mercaptopropionic acid (MP) treatment
used to screen for drugs capable of controlling RE in short time-
periods. Initially, mice received a daily administration of saline (0.1 ml
The model was validated by evaluating the protective effects of intraperitoneal [i.p.]) during 7 days. This procedure allows the ani-
standard AEDs: PHT and Phenobarbital (PB), (proven Pgp substrate mals’ habituation to handling, avoiding down regulation of GABA
AEDs (Zhang et al., 2012)) and Carbamazepine (CBZ), Levetirac- receptors induced by acute handling, which could affect the suscep-
etam (LEV) and Diazepam (DZP) (Pgp nonsubstrates). Based on the tibility to convulsions (Skilbeck et al., 2010). Thereafter they were
hypothesis that there is an overexpression of Pgp in animals treated daily administered with i.p. MP for 23 consecutive days. MP dose
with MP, we expect to observe drug resistance to Pgp-substrate was increased through the treatment, starting from 30 mg/kg on
AEDs, and a protective response to those AEDs that are not Pgp day 1 to 36 mg/kg on day 23 in order to obtain generalized seizures
substrate. The effect of co-administration of nimodipine (NIMO) (a throughout the 23 days (Fig. 1). The MP starting dose (30 mg/kg) is
Pgp blocker) with PHT and PB was also studied. the necessary dose to induce generalized seizures without causing
the death of the animals and it was chosen from preliminary dose-
response experiments. 23 MP treatment days was the optimal time
to avoid PHT anticonvulsant effect. (See Section 2.2.1).
MP was daily prepared in saline (5 ml/kg) and neutralized with
trizma base immediately before its administration. After each MP
injection, the mice were placed individually in transparent Plexi-
2. Materials and methods glass cages and they were observed for 30 min. Seizure activity was
classified according to a behavioral scale adapted to the MP induced
2.1. Animals and drugs seizures (Racine, 1972):
0- no response.
We employed Swiss albino male mice (25–35 g) provided by the 1- ear and facial twitching, nodding, piloerection, chewing.
Faculty of Veterinary, National University of La Plata. The animals 2- myoclonic jerks.
were housed in colonies of ten, under a regime of 12 h light/dark 3- forelimb clonus.
with water and food ad libitum. Every effort was made to mini- 4- kangaroo position with bilateral clonic jerks of the forelimbs.
mize animal stress. The animal care for this experimental protocol 5- generalized clonic seizures with falling.
was conducted in accordance with the NIH guidelines for the Care 6- sudden running fits and generalized clonic seizures.
and Use of Laboratory Animals and it was approved by the Ethical 7- generalized tonic-clonic seizures.
Committe of Exact Sciences Faculty of University of La Plata.
MP was acquired from Merck (Hohenbrunn, Germany). LEV was 2.2.1. Determination of the initial dose of MP and treatment time
provided by Glaxosmithline. CBZ and NIMO were a generous gift To determine initial MP dose, 3 different mice groups (n = 4)
from Bagó Laboratories. DZP was acquired from Roche, PB from were administered with 27, 30 and 33 mg/kg during 4 days. The per-
Serva (Heidelberg, Germany) and PHT from Sigma-Aldrich. centages of animals with generalized seizures and mortality were
PHT was dissolved in saline and the pH = 11.2 was adjusted with recorded each day.
NaOH. Carboxymethylcellulose 1% (CMC) was used as vehicle for MP convulsant dose was administered i.p. once a day for 0, 7,
CBZ, NIMO and PB. Finally, DZP and LEV were dissolved in saline. 10, 13, 19 or 23 consecutive days to determine the necessary treat-
All AEDs were evaluated at their time of maximum response. All ment time to develop PHT resistance. MP19 and MP23 groups were
drugs except MP were administered at 10 ml/kg. injected from Monday to Friday; the other groups were adminis-
The study was not performed blinded to treatment groups. Ani- tered from Monday to Sunday. 24 h after the last administration
mals were randomized to treatment. of MP, PHT 18 mg/kg, i.p., and 2 h later, MP convulsant dose were
10 A. Enrique et al. / Epilepsy Research 129 (2017) 8–16
administered. The animals were observed for 30 min. Seizure score MP23–vehicle group (n = 6): animals were handled as described
were recorded. Animals that had generalized seizure (stage 4 or as others groups but received vehicle on day 24. It was a seizure
higher) were considered resistant. control group.
S23-PB group (n = 10): mice were managed as described above
2.2.2. Evaluation of Pgp overexpression induced by repetitive MP for the MP23-PB group except that they received saline instead of
administration MP for 23 consecutive days. On day 24 mice were administered with
MP23 group (n = 22): Mice were manipulated as described in PB (15 mg/kg, i.p.) and, an hour later, MP (30 mg/kg, i.p.)
Section 2.2. during 23 days. 24 h later the last MP administration, S23-vehicle group (n = 6): mice were managed as described
they were sacrificed by cervical dislocation. Four of them were above for the MP23-PB group except that they received vehicle for
used for immunohistochemistry studies (see below). The remaining 23 consecutive days as pretreatment.On day 24 mice were admin-
eighteen mice were used for western blot studies (see below). istered with vehicle and, an hour later, MP (30 mg/kg, i.p.).
Control group (n = 22): Mice were managed as described for These procedures are summarized in Fig. 3.
the MP23 group, but they received saline (0.15 ml) instead of MP
during 23 days. Four mice were used for immunohistochemistry 2.2.5. Evaluation of the anticonvulsant effects of Pgp
studies and eighteen mice were used for western blot studies. non-substrate AEDs after repetitive administration of MP
CBZ, DZP and LEV have been reported as Pgp nonsubstrates
when they were evaluated in in vivo animal models (Owen et al.,
2.2.3. Evaluation of PHT resistance induced by repetitive MP 2001; Mealey et al., 2010; Zhang et al., 2012). These experiments
administration were designed to evaluate if repetitive MP administration induces
These experiments were focused to determine if the repetitive resistance to these AEDs. On day 24, MP was administered at time
administration of MP induces resistance to PHT, which is a rec- of maximum response to each AED.
ognized Pgp susbtrate (Stepien et al., 2012). On day 24, MP was MP23-CBZ group (n = 9): after habituation, mice were daily
administered at time of maximum response to PHT. injected with MP during 23 days as described previously. 24 h after
MP23-PHT group (n = 6): after habituation, mice were daily the last MP injection, mice received CBZ (25 mg/kg, i.p.) and 15 min
injected with MP during 23 days as described previously. 24 h after later, MP (36 mg/kg, i.p.).
the last MP injection, mice received PHT (18 mg/kg, i.p.) and 2 h MP23-DZP group (n = 6): MP was daily injected during 23 days,
later MP (36 mg/kg, i.p.). At this dose, preliminary experiments 24 h after the last MP administration mice received DZP (0.5 mg/kg,
from our laboratory showed that PHT was able to prevent stage i.p.) and, an hour later, MP (36 mg/kg, i.p.).
4–7 of seizures induced by a single administration of MP in control MP23-LEV group (n = 5): MP was daily injected during 23 days,
mice. After MP administration, mice were observed during 30 min. 24 h after the last MP administration mice received LEV (33 mg/kg,
Seizure severity and latencies were registered. i.p.) and, an hour later, MP (36 mg/kg, i.p.).
MP23-PHT + NIMO group (n = 6): animals were managed as MP23-saline group (n = 9): MP was daily injected during
described for the MP23-PHT group, but they received NIMO 23 days, 24 h after the last MP administration mice received saline
(3.5 mg/kg, i.p., a Pgp blocker) 30 min before PHT administration. and, an hour later, MP (36 mg/kg, i.p.).
MP23-NIMO group (n = 6): mice were manipulated as described S23-CBZ, S23-DZP and S23-LEV (control groups; n = 6): For
above for the MP23-PHT + NIMO group, except that they received each respective group animals received saline during 23 days. On
vehicle instead of PHT. day 24 they received CBZ (25 mg/kg) or DZP (0.5 mg/kg) or LEV
MP23-vehicle group (n = 6): animals were handled as described (33 mg/kg) and after the time indicated above for each drug, MP
above for the MP23-PHT, except that they received vehicle instead (30 mg/kg, i.p.).
of PHT on day 24. This group represented the seizure control group. S23-saline group (n = 6): Saline was daily injected during
S23 − PHT group (n = 6): mice were managed as described above 23 days. On 24th day they received saline and, an hour later, MP
for the MP23-PHT group except that they received saline instead of (30 mg/kg, i.p.).
MP for 23 consecutive days. On 24th day mice were administered These procedures are summarized in Fig. 4.
with PHT (18 mg/kg, i.p.) and, 2 h later, MP (30 mg/kg, i.p.).
S23-NIMO group (n = 6): mice were managed as described 2.3. Evaluation of Pgp expression by immunohistochemistry
above for MP23-NIMO group except that they received saline
instead of MP for 23 consecutive days. This technique was applied as previously described (Gori and
These procedures are summarized in Fig. 2 Girardi, 2013) with certain modifications.
Brains were removed and immediately embedded in 1, 2-
2.2.4. Evaluation of PB resistance induced by repetitive MP dichloroethane (Merck Millippore) and snap frozen in dry ice.
administration Frozen brains were stored at −80 ◦ C for at least overnight. Coro-
PB has previously been reported as Pgp substrate (Stepien et al., nal brain sections (20 m) were cut by means of a Leica CM 1850
2012). This experiment aims to assess whether the resistance to cryostat, mounted on gelatin-coated slides and fixed in cold 100%
PHT is also extended to PB. On day 24, MP was administered at acetone for 10 min.
time of maximum response to PB. Brain sections of both treated and control groups were simulta-
MP23-PB group (n = 10): after habituation, mice were daily neously processed for Pgp-170 immunostaining. In order to inhibit
injected with MP during 23 days as described previously. 24 h after endogenous peroxidase activity, tissue sections were incubated
the last MP injection, mice received PB (15 mg/kg, i.p.) and 1 h in 0.5% v/v H2 O2 in phosphate buffer saline (PBS) for 30 min at
later MP (36 mg/kg, i.p.). Preliminary experiments from our lab- room temperature. Brain sections were blocked during 1 h with
oratory showed that PB 15 mg/kg was able to prevent seizures 3% v/v normal sheep serum and 0.3% triton in PBS, then the sec-
corresponding to stage 2 or higher. After MP administration, mice tions were incubated for 24 h at 4 ◦ C with 1:100 anti-Pgp-170
were observed during 30 min. Seizure severity and latencies were primary monoclonal antibody C219 (Calbiochem) or C494 (Signet
registered. Laboratories, Dedham, MA). After washing with PBS-X (PBS 0.02
MP23-PB + NIMO group (n = 10): animals were managed as 5% Triton X-100) tissue sections were incubated for 1 h at room
described for the MP23-PB group, but they received NIMO temperature with 1:100 anti mouse biotinilated antibody (Sigma-
(3.5 mg/kg, i.p.) 30 min before the PB administration. Aldrich) and then for 1 h with 1:200 streptavidin-peroxidase com-
A. Enrique et al. / Epilepsy Research 129 (2017) 8–16 11
Treatment
Pre-treatment Group
0 min 30 min 2h
(during 23 days)
Fig. 2. Treatment scheme: PHT activity in mice treated with MP or saline during 23 days.
Treatment
Pre-treatment Group
0 min 30 min 1h
(during 23 days)
Fig. 3. Treatment scheme: PB activity in mice treated with MP or saline during 23 days.
Treatment
Pre-treatment Group
0 min 15 min 1h
(during 23 days)
Fig. 4. Treatment scheme: CBZ, DZP and LEV activity in mice treated with MP or saline during 23 days.
plex (Sigma-Aldrich). Reaction was carried out with 0.035% w/v The supernatants were spun at 100,000 x g for 30 min. The result-
3, 3-diaminobenzidine plus 2.5% w/v nickel ammonium sulphate ing pellets containing the crude membranes were resuspended
and 0.1% v/v H2 O2 dissolved in acetate buffer. Sections were cover- in 20 mM Tris–HCl, 0.25 M sucrose, and 0.5 mM EDTA (Auzmendi
slipped using DPX (Fluka) for light microscopic observation. et al., 2009). Protein concentration was determined using Bradford
Negative controls were processed simultaneously by omitting reagent (Sigma) with bovine serum albumin as standard (Bradford,
the primary antibodies. 1976).
All antibodies, as well as streptavidin complex, were dissolved For gel electrophoresis, aliquots of tissue sample were diluted
in PBS containing 1% v/v normal sheep serum and 0.3% v/v triton 4 fold with dodecyl sulfate (SDS) sample buffer and denatures at
X-100, pH 7.4. 100 ◦ C. 30 g of protein was loaded on each lane. After separation
Images of inmunoperoxidase sections were obtained using an by SDS-PAGE 8%, the resolved proteins were electro-transferred
Axiophot Zeiss light microscope, equipped with a video cam- onto polyvinylidene difluoride (PVDF) membranes (Amersham
era (Olympus Q5). Images obtained from the light microscope Hybond-P GE Healthcare). Membranes were blocked with 4% non-
were analyzed with ImageJ software (National Institute of Health, fat milk and 2% Glycine in buffer phosphate saline (PBS-Tween
USA). The resolution of each pixel was 256 gray levels (8 bits). 0.05%) and then incubated with anti-Pgp (1:300 C219 antibody
Immunoblot values represent the means of 4 mice per each exper- Calbiochem) overnight at 4 ◦ C. After three washes whit PBS-T
imental condition. The percentages of marked surface in twenty the membranes were incubated with horseradish peroxidase-
to twenty five fields from each region were measured. The whole conjugated secondary antibody in buffer (1:500 Amersham GE
hippocampus, cerebral cortex and striatum were measured. Healthcare) for one hour and them again washed three times with
PBS-T. To visualize immunoreactivity, membranes were incubated
with the ECL chemiluminescense detection system (Amersham ECL
2.4. Evaluation of Pgp expression by western blot
Western blotting detection reagents and analysis system from GE
Healthcare, Buckinghamshire,UK) and exposed to X-ray blue films
Hippocampus, striatum and cerebral cortices (pooled from two
(Agfa, Argentina). Developed films were scanned and optical den-
mice) were dissected in cold environment. Tissue was homoge-
sity quantified by Image Studio Light software of Li-Cor.
nized in a solution containing 0.32 M sucrose neutralized with Tris
As a protein load control, a polyclonal anti--actin antibody
base solution (0.2 M) pH 7.2 and 0.05 ml/g proteinase inhibitor
(Sigma) was used at a dilution 1:1000 in the same membranes after
cocktail at 10% (w/v), in a Teflon glass Potter–Elvehjem homoge-
stripping.
nizer. The homogenate was centrifuged at 900 x g for 10 min at 4 ◦ C.
12 A. Enrique et al. / Epilepsy Research 129 (2017) 8–16
Table 1
Determination of MP treatment time. Percentage of mice that presented generalized
***
seizure after different times of MP treatment (n = 6 in all groups).
8 ** *
3.1. MP repetitive administration: mice behavior 3.1.3. Repetitive MP-induced seizure trigger PHT resistance
When vehicle was administered to MP23 (MP23-vehicle group)
Each MP administration induced piloerection, nodding and 100% of mice presented generalized seizures with score 5, indicat-
myoclonus with a latency of 3.11 ± 0.55 min. Eventually, animals ing that MP23 mice were responsive to MP on day 24. In S23-PHT
developed forelimb clonus and kangaroo position with bilateral group none of the mice had generalized seizures. They presented
clonic jerks of the forelimbs. Subsequently, animals presented partial seizures with a score between 0 and 3 (1.6 ± 0.62; p < 0.001
seizures characterized by sudden circle running, ending with a vs. MP23-PHT), so that PHT had anticonvulsant effect in S23 mice.
tonic-clonic seizure with a latency of 4.52 ± 0.94 min. This type of All animals (100%) from MP23-PHT group showed generalized
seizure activity was induced during the 23 days of MP treatment. seizures with score 6 after the last MP administration; therefore
On day 23 all animals had a seizure score 5–7 with a survival rate PHT did not show anticonvulsant effect. Concerning the animals
of 80%. receiving NIMO before PHT (MP23-PHT + NIMO), they did not show
neither generalized nor partial crisis, with score 0 or 1 (0.4 ± 0.36;
3.1.1. Determination of the initial dose of MP and treatment time p < 0.001 vs. MP23-PHT), indicating that acquired resistance to PHT
The high dose (33 mg/kg) induced generalized seizure in 100% can be reversed by prior administration of a Pgp blocker. 100% of
of mice with a mortality of 50% at first day. This group was not eval- mice from MP23-NIMO and S23-NIMO groups showed generalized
uated the 3 following days. 27 mg/kg induced generalized seizure seizures with score between 5 and 7 (6.4 ± 0.36 and 5 ± 0 respec-
only in 50% of mice. 30 mg/kg was the optimal dose due to it induced tively); NIMO itself had no anticonvulsant effect in neither MP23
generalized seizure in 100% of mice and 0% mortality at 4th day. nor S23 mice. (Fig. 8A and B)
PHT resistance increased across treatment days (Table 1). 100%
of mice displayed PHT resistance after 19 and 23 days of MP 3.1.4. Repetitive MP-induced seizure elicit PB resistance
treatment. However, after 19 days of treatment seizure score was S23-PB group showed anticonvulsant effects in 100% of mice
4.33 ± 0.02, while after 23 days it was 5.66 ± 0.02 (p < 0.05) (Fig. 5). (p < 0.001 vs S23-vehicle group) with a score of 0 or 1 (mean of
Therefore 23 days of MP treatment was chosen. 0.18 ± 0.12; p < 0.001 vs S23-saline group). Score 1 included nod-
ding, eye blinking and immobility; from these results, the criterion
3.1.2. Repetitive administration of MP in mice induces Pgp on resistance is the presence of seizures with score ≥ 2.
overexpression in brain 80% of mice from MP23-PB group showed score ≥ 2 (p < 0.01 vs
Pgp immunoreactivity in the control group was weakly positive S23-PB group) with a mean of 3.4 ± 0.68 (p < 0.001 vs S23-PB group),
in blood vessels from cerebral cortex, hippocampus and striatum being resistant to the PB anticonvulsant effects.
(Fig. 6A and B) and western blot studies showed Pgp expression in Animals that received NIMO and PB (MP23-PB + NIMO) had
these areas (Fig. 7A). Brain sections from MP23 animals showed an seizure score 2.7 ± 0.78, being 60% ≥ 2 (p < 0.05 vs MP23-vehicle
intense Pgp immunoreactivity in blood vessels from cerebral cor- group). This group did not differ from the MP23-PB group. (Fig. 9A
tex, hippocampus, and striatum. Parietal, piriform and perirhinal and B)
A. Enrique et al. / Epilepsy Research 129 (2017) 8–16 13
Fig. 6. Pgp expression in cerebral cortex, hippocampus and striatum by immunohistochemistry. A- Immunostaining pictures from different brain areas in MP23 and control
mice. (Primary magnification 100×). Parietal cerebral cortex, CA1 hippocampal subarea and striatum are shown. B- Primary magnification 200×. Arrows indicate Pgp stained
vessels. Insets show magnification at 400×. C- The percentage of stained surface ± SEM in MP23 and control mice was graphed. *** p < 0.001.
3.1.5. Repetitive MP-induced seizure does not decrease the to develop RE. Pgp expression can be induced in RE experimental
anticonvulsant effect of non-Pgp substrate AEDs models, in order to investigate the potential advantage of novel
MP23 and S23 mice administered with saline on day 24 showed drugs compared with clinically established AEDs (Potschka, 2012).
generalized seizure (score 5 or 6; 5 ± 0 and 5.11 ± 0.11 respectively) In this investigation we have developed a mice model with Pgp
in all mice after MP administration. CBZ administration on S23 overexpression by means of daily administration of the convulsant
(S23-CBZ group) protected the mice from generalized seizure at agent MP, achieving a high percentage of resistant mice.
evaluated doses showing a significantly lower mean seizure score This seizure model requires the progressive increase of MP dose
respect to S23-saline (1.4 ± 0.3; p < 0.001). In the same way, MP23- (30–36 mg/kg) through the experimental procedure to induce daily
CBZ decreased the score respect to MP23-saline group (2.11 ± 0.35; generalized seizures during 23 consecutive days. This experimen-
p < 0.001) (Fig. 10). These results indicate that CBZ can induce anti- tal procedure is in agreement with previous reports from Világi
convulsant effects despite the observed Pgp overexpression. et al. (2009). They employed an increasing dose of 4 amino- pyri-
LEV and DZP did not show anticonvulsant effect on S23 groups at dine to produce seizure stage 5 according to Racine scale. The need
evaluated doses but they reduced the seizure score in both MP23- to increase the MP dose could be related to a rised seizure thresh-
LEV (2.4 ± 0.2; p < 0.05 vs. MP23-saline) and MP23-DZP (1 ± 0.44; p old (tolerance to convulsant stimuli); in contrast to the progressive
< 0.001 vs. MP23-saline) groups indicating a greater anticonvulsant decrease in seizure threshold observed in kindling models (Mason
activity on MP23 compared to S23 brains (Fig. 10). While the MP23- and Cooper, 1972; Leclercq et al., 2014; Racine, 1972).
DZP group showed a significant difference compared to S23-DZP It is worth mentioning that the increase of MP dose along
group, MP23-LEV group did not differ from the S23-LEV group. treatment might be due to extrusion of this convulsant drug by
overexpressed Pgp, however there is no data about this at present.
We have observed an Pgp overexpression after repetitive
4. Discussion
seizures, by immuno histochemical and western blot studies, asso-
ciated with a low response to PHT and PB anticonvulsant effect,
The overexpression of Pgp has previously been shown to be a
which was reversed by NIMO (a Pgp blocker), according with pre-
consequence of seizures (Bauer et al., 2008; Miller, 2008, 2015),
vious studies on rat brain (Hocht et al., 2007, 2009).
thus uncontrolled seizures in patients represent a high risk factor
14 A. Enrique et al. / Epilepsy Research 129 (2017) 8–16
***
B ***
6 *
n=10 MP23
n=6
S23
n=10
4 n=10
Score
Fig. 7. Pgp expression in cerebral cortex, hippocampus and striatum by western blot (A) 2
Western blot for Pgp (170 KDa) and ␣- actine (43 KDa). Each lane represents 30 g
of pooled membrane proteins. (B) The bar graphs illustrate% of relative expression n=6
respect to control ± SEM. *p < 0.05; ***p < 0.001.
0
lo
PB
lo
PB
O
cu
cu
IM
3-
3-
hí
hí
N
P2
S2
ve
ve
+
M
PB
3-
3-
A P2
S2
**
3-
M
P2
150 **
M
MP23
% generalized seizure
**
** ** S23
n=6 n=6 n=6 n=6 Fig. 9. Repetitive MP induced seizure elicit PB resistance. A) Percentage of mice with
100
seizure score ≥2 vs. treatment. B) Seizure score mean ± SEM vs. treatment. *p < 0.05;
**p < 0.01; ***p < 0.001.
50
NIMO administration before PB reduced partially resistance to
n=6 n=6 PB, but NIMO was needed to see PB protective effect in MP23 mice.
0
The low percentage of PB resistance reversion by NIMO may be
O
T
T
e
O
PH
IM
IM
IM
hi
N
N
3-
3-
ve
3-
3-
S2
3-
S2
P2
T
M
P2
PH
AEDs, like CBZ, DZP and LEV. The fact that DZP and LEV elicit pro-
M
n=6 S23 cantly suppressed the development and acquisition of PTZ kindling,
6 n=6 n=6 and they found an increase in the hippocampal levels of SV2A pro-
tein (a primary action site of LEV) in fully kindled animals (Ohno
4 et al., 2009).
n=6 It is important to note that current models of drug resistant
2 epilepsy require extremely long times for its setting up, due to
n=6
the resistance induction period and the subsequent treatment used
0 for the selection of nonresponders. Models generating epileptic
animals having spontaneous seizures after induction of a status
T
T
O
O
O
e
cl
PH
PH
IM
IM
IM
hi
3-
3-
ve
3-
3-
S2
P2
3-
S2
P2
T
P2
PH
M
3-
seizures and then, another 6 weeks for the selection of resistant ani-
Fig. 8. Repetitive MP induced seizure trigger PHT resistance. A) Percentage of mice mals (Bankstahl and Löscher, 2012). Kindling model requires about
with generalized seizures vs. treatment. B) Seizure score mean ± SEM vs. treatment. 8 weeks after implantation of the electrodes until the kindled state
** p < 0.01; *** p < 0.001. is reached, and between 6 and 8 additional weeks for selection of
resistant animals (Löscher, 2006). MP23 model can be an excellent
A. Enrique et al. / Epilepsy Research 129 (2017) 8–16 15
***
*
***
6 *** **
n=6
n=9 n=6 n=6 S23
n=9 n=5
2 n=6
n=6
0
Saline CBZ LEV DZP
Treatment
Fig. 10. Evaluation of Pgp non-substrate AEDs on MP23 and S23 mice. CBZ, LEV and DZP were evaluated. Mean seizure score ± SEM vs. treatment was represented. *p < 0.05;
**p < 0.01; ***p < 0.001.
alternative for evaluating new drugs, considering that it has the 5. Conclusion
advantage of generating animals with pharmacoresistant seizures
in only 23 days. 100% of mice are PHT- resistant and 80% are PB- In this investigation we have developed a new mice model of
resistant. The animals still remain resistant for at least a week after pharmacoresistant seizures, which is capable of identifying drug
the end of treatment with 3MP (results not shown). treatments for the control of Pgp-mediated drug resistant seizures.
MP23 have a mortality rate of 20%, other chemoconvulsant mod- This new model involves 23 consecutive daily MP administrations;
els have shown different percentages. Curia et al. have reviewed it has 100% PHT resistance and 80% PB resistance at the end of
mortality rate in different laboratories, investigating pilocarpine treatment while the Pgp nonsubstrate CBZ, LEV, DZP have anticon-
and litium pilocarpine models. Pilocarpine dose of 320–360 mg/kg vulsant effect.
had a mean of 37.3% (17–55%) of death and 22% (5–31%) was reg- We have evidenced the Pgp overexpression by means of western
istered with the dose of 380 mg/kg while in litium-pilocarpine blotting assays in cerebral cortex, striatum and hippocampus. An
treatment 56% (24–100%) was reported (Curia et al., 2008). In kainic increase of Pgp expression in blood capillaries was observed in all
acid mice model Ahmad et al. reported 50% of mortality and only regions studied by immunochemistry.
44% of all mice get an epileptic stage; reducing the animal per- This model promises to be useful for screening novel anticon-
formance to 22%. These authors mention 20 to 50% of mortality vulsant drugs with a potential effect on pharmacoresistant seizures
depending on the mice strain (Ahmad et al., 2013). treatment.
Electrical kindling models have lower mortality than chemocon-
vulsant models (Kandratavicius et al., 2014), but not all the animals
get a fully kindled stage. All MP23 survival animals are useful for Acknowledgments
anticonvulsant assays.
Taking into account other epilepsy models with animals treated The authors are grateful to Dr. Marina Vaccotto for her helpful
with a single AED, it is worth mentioning that the models of advice regarding western blot technique. This work was sup-
spontaneous seizures achieve about 35% of PB-resistant animals, ported by the National University of La Plata (UNLP) (Incentivos
considering only the animals which develop spontaneous seizures X-597), the National Council for Scientific and Technical Research
(Bankstahl and Löscher, 2012; Brandt and Löscher, 2014), while (CONICET) (PIP 11220090100603 and Res 3646/14) and National
amygdala kindling model achieves between 12 and 23% of PHT- Agency for Scientific and Technological Promotion (ANPCyT) (PICTs
resistant animals (Löscher et al., 1993; Löscher, 2011; Potschka, 2010-1774, 2013-3175) Convenio de cooperación internacional
2012). CONICET-CONCYT, D1800. Convenio de cooperación internacional
The high percentage of resistance achieved in our model avoids MINCYT-CONACYT-México-MX/11/02.
selecting non-responders by administering AEDs before the new
candidate evaluation, assuming that all the animals are resistant.
References
This is an advantage because AEDs used in the selection stage could
alter the response to the test drug (e.g. due to altered pharmacoki- Ahmad, N., Yow, H., Makmor-Bakry, M., et al., 2013. Seizure attainment and
netics induced by the used drug in the selection stage). mortality of mice in kainite-induced status epilepticus. Int. J. Pharm. Pharm.
Additionally, the new model can be used to study the mecha- Sci. 5 (3), 145–147.
Aronica, E., Gortes, J.A., Ramkema, M., Redeker, S., Özbas-Gerçeker, F., Van Vliet,
nisms of Pgp overexpression induced by recurrent seizures, as well E.A., Scheffer, G.L., Scheper, R.J., Der Valk, P.V., Baayen, J.C., Troost, D., 2004.
as for the evaluation of new drugs for RE treatment. For exam- Expression and cellular distribution of multidrug resistance-related protein in
ple, it can be employed to evaluate the effectiveness of drugs in the hippocampus of patients with mesial temporal lobe epilepsy. Epilepsia 45
(5), 441–451.
a high Pgp expression model (evaluation at 24th day), or to test the
Auzmendi, J., Gonzalez, N., Girardi, E., 2009. The NMDAR Subunit NR2 B Expression
drug potential to prevent the development of resistance due to Pgp is modified in Hippocampus after Repetitive Seizures. Neurochem. Res. 34,
overexpression (daily administration). 819–826.
Bankstahl, J.P., Löscher, W., 2008. Resistance to antiepileptic drugs and expression
We propose the use of this model in the early stages of screening
of P-glycoprotein in two rat models of status epilepticus. Epilepsy Res. 82,
new Pgp nonsubstrate AEDs, selecting the promising compounds 70–85.
to be evaluated later in RE model. Further studies will be made to Bankstahl, M., Löscher, W., 2012. Inter-individual variation in the anticonvulsant
elucidate other pharmacoresistance mechanisms. effect of phenobarbital in the pilocarpine rat model of temporal lobe epilepsy.
Exp. Neurol. 234 (1), 70–84.
16 A. Enrique et al. / Epilepsy Research 129 (2017) 8–16
Bauer, B., Hartz, A., Pekcec, A., Toellner, K., Miller, D., Potschka, H., 2008. Lazarowski, A., Czornyj, L., Lubienieki, F., Girardi, E., Vazquez, S., D’Giano, C., 2007.
Seizure-induced up-regulation of P-glycoprotein at the blood-brain barrier ABC transporters during epilepsy and mechanisms underlying multidrug
through glutamate and cyclooxygenase-2 signaling. Mol. Pharmacol. 73, resistance in refractory epilepsy. Epilepsia 48 (5), 140–149.
1444–1453. Leclercq, K., Matagne, A., Kaminski, R.M., 2014. Low potency and limited efficacy of
Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of antiepileptic drugs in the mouse 6 Hz corneal kindling model. Epilepsy Res.
microgram quantities of protein utilizing the principle of protein-dye binding. 108, 675–683.
Anal. Biochem. 12, 248–254. Luna Tortós, C., Fedrowitz, M., Löscher, W., 2008. Several major antiepileptic drugs
Brandt, C., Löscher, W., 2014. Antiepileptic efficacy of lamotrigine in Phenobarbital are substrates for human P-glycoprotein. Neuropharmacology 55, 1364–1375.
resistant and responsive epileptic rats: a pilot. Epilepsy Res. 108, 1145–1157. Mason, C.R., Cooper, R.M., 1972. A permanent change in convulsive threshold in
Curia, G., Longo, D., Biagini, G., Jones, R.S., Avoli, M., 2008. The pilocarpine model of normal and brain-damaged rats with repeated small doses of
temporal lobe epilepsy. J. Neurosci. Methods 172 (2), 143–157. pentylenetetrazol. Epilepsia 13, 663–674.
De Sarro, G., Ferreri, G., Gareri, P., Russo, E., De Sarro, A., Gitto, R., Chimirri, A., 2003. Mealey, K.L., Waiting, D., Raunig, D.L., Schmidt, K.R., Nelson, F.R., 2010. Oral
Comparative anticonvulsant activity of some 2,3-benzodiazepine derivates in bioavailability of P-glycoprotein substrate drugs do not differ between ABCB1-
rodents. Pharmacol. Biochem. Behav. 74, 595–602. 1D and ABCB1 wild type dogs. J. Vet. Pharmacol. Ther. 33, 453–460.
Giraldez, L., Girardi, E., 2000. Effects of an adenosine analogue administration on Meldrum, B., 1984. Amino acid neurotransmitters and new approaches to
the striatal NMDA receptors in an experimental model of epilepsy. anticonvulsant drug action. Epilepsia 25 (2), 140–149.
Neurochem. Int. 36, 243–247. Miller, D.S., 2008. Regulation of P-glycoprotein and other ABC drug transporters at
Giraldez, L., Zanetti, F., Girardi, E., 1999. Striatum adenosine A2 receptors are the blood-brain barrier. Trends Pharmacol. Sci. 31 (6), 246–254.
modified during seizure: effect of cyclopentyladenosine administration. Miller, D.S., 2015. Regulation of ABC transporters at the blood-brain barrier. Clin.
Neurochem. Res. 24, 1219–1225. Pharmacol. Ther. 97 (4), 395–403.
Girardi, E., Perez Raffo, G., Rodriguez de Lores Arnaiz, G., 1989. Increase of Ohno, Y., Ishihara, S., Terada, R., Kikuta, M., Sofue, N., Kawai, Y., Serikawa, T., Sasa,
5-nucleotidase activity in some brain subcellular fractions after the M., 2009. Preferential increase in the hippocampal synaptic vesicle protein 2A
administration of the convulsant 3-mercaptopropionic acid. Neurochem. Int. (SV2A) by pentylenetetrazole kindling. Biochem. Biophys. Res. Commun. 390,
14, 331–335. 415–420.
Girardi, E., Ramos, A.J., Vanore, G., Brusco, A., 2004. Astrocytic response in Ohno, Y., Ishihara, S., Terada, R., Serikawa, T., Sasa, M., 2010. Antiepileptogenic and
hippocampus and cerebral cortex in an experimental epilepsy model in rat. anticonvulsive actions of levetiracetam in a pentylenetetrazole kindling
Neurochem. Res. 29, 371–377. model. Epilepsy Res. 89, 360–364.
Gori, M.B., Girardi, E., 2013. 3-mercaptopropionic acid-induced repetitive seizures Owen, A., Pirmohamed, M., Tettey, J.N., Morgan, P., Chadwick, D., Park, B.K., 2001.
increase GluN2A expression in rat hippocampus: a potential neuroprotective Carbamazepine is not a substrate for P-glycoprotein. Blackwell Sci. Ltd. Br. J.
role of cyclopentyladenosine. Cell. Mol. Neurobiol. 33, 803–813. Clin. Pharmacol. 51, 345–349.
Hocht, C., Lazarowski, A., Gonzalez, N., Auzmendi, J., Opezzo, J., Bramuglia, G., Taira, Potschka, H., 2010. Transporter hypothesis of drug-resistant epilepsy: challenges
C., Girardi, E., 2007. Nimodipine restores the altered hippocampal phenytoin for pharmacogenetic approaches. Pharmacogenomics 11 (10), 1427–1438.
pharmacokinetics in a refractory epileptic model. Neurosci. Lett. 413, 168–172. Potschka, H., 2012. Animal models of drug-resistant epilepsy. Epileptic Disord. 14
Hocht, C., Lazarowski, A., Gonzalez, N., Mayer, M., Opezzo, J., Taira, C., Girardi, E., (3), 226–234.
2009. Differential hippocampal pharmacokinetics of phenobarbital and Racine, R.J., 1972. Modification of seizure activity by electrical stimulation. II.
carbamazepine in repetitive seizures induced by 3-mercaptopropionic acid. Motor seizure. Electroencephalogr. Clin. Neurophysiol. 32 (3), 281–294.
Neurosci. Lett. 453, 54–57. Rizzi, M., Caccia, S., Guiso, G., Richichi, C., Gorter, J.A., Aronica, E., Aliprandi, M.,
Jing, X., Liu, X., Wen, T., Xie, S., Yao, D., Liu, X., Wang, G., Xie, L., 2010. Combined Bagnati, R., Fanelli, R., D’Incalci, M., Samanin, R., Vezzani, A., 2002. Limbic
effects of epileptic seizure and phenobarbital induced overexpression of seizures induce P-glycoprotein in rodent brain: functional implications for
P-glycoprotein in brain of chemically kindled rats. Br. J. Pharmacol. 159, pharmacoresistance. J. Neurosci. 22, 5833–5839.
1511–1522. Rogawski, M.A., 2013. The intrinsic severity hypothesis of pharmacoresistance to
Kandratavicius, L., Alves Balista, P., Lopes-Aguiar, C., Naime Ruggiero, R., Umeoka, antiepileptic drugs. Epilepsia 54 (2), 33–40.
E.H., Garcia-Cairasco, N., Soares Bueno-Junior, L., Pereira Leite, J., 2014. Animal Seegers, U., Potschka, H., Loscher, W., 2002. Expression of the multidrug
models of epilepsy: use and limitations. Neuropsychiatr. Dis. Treat. 10, transporter P-glycoprotein in brain capillary endothelial cells and brain
1693–1705. parenchyma of amygdala-kindled rats. Epilepsia 43, 675–684.
Kuteykin-Teplyakov, K., Brandt, C., Hoffmann, K., Loscher, W., 2009. Complex Sisodiya, S.M., Lint, W.R., Harding, B.N., Squier, M.V., Thom, M., 2002. Drug
time-dependent alterations in the brain expression of different drug efflux resistance in epilepsy: human epilepsy. Novartis Found. Symp. 243, 167–174
transporter genes alter status epilepticus. Epilepsia 50 (4), 887–897. (discussion 174-9, 180-5).
Kwan, P., Brodie, M.J., 2000. Early identification of refractory epilepsy. N. Engl. J. Skilbeck, K., Jhonston, G., Hinton, T., 2010. Stress and GABAa receptors. J.
Med. 342, 314–319. Neurochem. 112, 1115–1130.
Löscher, W., Potschka, H., 2005a. Drug resistance in brain diseases and the role of Sprince, H., Parker, C.M., Josephs Jr., J.A., Magazino, J., 1969. Convulsant activity of
drug efflux transporters. Nat. Rev. Neurosci. 6, 591–602. homocysteine and other short-chain mercaptoacids: protection therefrom.
Löscher, W., Potschka, H., 2005b. Role of drug efflux transporters in the brain for Ann. N. Y. Acad. Sci. 166, 323–325.
drug disposition and treatment of brain diseases. Prog. Neurobiol. 76, 22–76. Stepien, K., Tomaszewski, M., Tomaszewska, J., Czuczwar, S., 2012. The multidrug
Löscher, W., Rundfeldt, C., Hönack, D., 1993. Pharmacological characterization of transporter P-glycoprotein in pharmacoresistance to antiepileptic drugs.
phenytoin-resistant amygdala-kindled rats, a new model of drug-resistant Pharmacol. Rep. 64, 1011–1019.
partial epilepsy. Epilepsy Res. 15 (3), 207–219. Tishler, D., Weinberg, K., Hinton, D., Barbaro, N., Geralyn, A., Corey, R., 1995. MDR1
Löscher, W., 2006. Animal models of drug-refractory epilepsy. In: Pitkänen, Asla, Gene expression in brain of patient with medically intractable epilepsy.
Schwartzkroin, Philip A., Moshé, Solomon L. (Eds.), Models of Seizures and Epilepsia 36, 1–6.
Epilepsy. , 1st edn. Elsevier, Burlington, USA. Világi, I., Dobó, E., Borbély, S., Czégé, D., Molnár, E., Mihály, A., 2009. Repeated
Löscher, W., 2011. Critical Review of current animal models of seizure and epilepsy 4-aminopyridine induced seizures diminish the efficacy of glutamatergic
used in discovery and development of new antiepileptic drugs. Seizure 20, transmission in the neocortex. Exp. Neurol. 219, 136–145.
359–368. Volk, H.A., Löscher, W., 2005. Multidrug resistance in epilepsy: rats with
Lazarowski, A., Riveros, D., Sevlever, G., Masssaro, M., Rabinowicz, A., 1997. High drug-resistant seizures exhibit enhanced brain expression of P-glycoprotein
expression of multidrug resistance gene (MDR-1) and persistent low levels of compared with rats with drug-responsive seizures. Brain 128, 1358–1368.
phenytoin (PHT) on a patient with refractory epilepsy due to tuberous sclerosis White, H.S., 2003. Preclinical development of antiepileptic drugs: past, present,
(TS). J. Neurol. Sci. 150 (S28) (1-17-22). and future directions. Epilepsia 44 (7), 2–8.
Lazarowski, A., Ramos, A., García-Rivello, H., Brusco, A., Girardi, E., 2004. Neuronal Zhang, C., Kwan, P., Zuo, Z., Baum, L., 2012. The transport of antiepileptic drugs by
and glial expression of the multidrug resistance gene product in an P-glycoprotein. Adv. Drug Deliv. Rev. 64, 930–942.
experimental epilepsy model. Cell. Mol. Neurobiol. 24 (1), 77–85.