Departamento de Biología Celular y Fisiología, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, México, DF, México
Address correspondence and reprint requests to María Sitges, Departamento de Biología Celular y Fisiología, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Apartado Postal 70228, Ciudad Universitaria 04510, México D.F., México. E-mail: firstname.lastname@example.org
In the present study, the effects of the two classical anti-epileptic drugs, carbamazepine and valproic acid, and the non-classical anti-seizure drug vinpocetine were investigated on the expression of the pro-inflammatory cytokines IL-1β and TNF-α in the hippocampus of rats by PCR or western blot after the administration of one or seven doses. Next, the effects of the anti-seizure drugs were investigated on the rise in cytokine expression induced by lipopolysaccharides (LPS) inoculation in vivo. To validate our methods, the changes induced by the pro-convulsive agents 4-aminopyridine, pentylenetetrazole and pilocarpine were also tested. Finally, the effect of the anti-seizure drugs on seizures and on the concomitant rise in pro-inflammatory cytokine expression induced by 4-aminopyridine was explored. Results show that vinpocetine and carbamazepine reduced the expression of IL-1β and TNF-α from basal conditions, and the increase in both pro-inflammatory cytokines induced by LPS. In contrast, valproic acid failed to reduce both the expression of the cytokines from basal conditions and the rise in IL-1β and TNF-α expression induced by LPS. Tonic-clonic seizures induced either by 4-aminopyridine, pentylenetetrazole or pilocarpine increased the expression of IL-1β and TNF-α markedly. 4-aminopyridine-induced changes were reduced by all the tested anti-seizure drugs, although valproic acid was less effective. We conclude that the anti-seizure drugs, vinpocetine and carbamazepine, whose mechanisms of action involve a decrease in ion channels permeability, also reduce cerebral inflammation.
The mechanism of action of anti-seizure drugs like vinpocetine and carbamazepine involves a decrease in Na+ channels permeability. We here propose that this mechanism of action also involves a decrease in cerebral inflammation.
A role of brain pro-inflammatory cytokines in the generation and maintenance of seizures and in the establishment of chronic epileptic foci is indicated by several studies. For instance, increased levels of interleukin-1beta (IL-1β), tumor necrosis factor-alpha (TNF-α) and other pro-inflammatory cytokines in several animal models of epilepsy, as well as in the serum and cerebrospinal fluid from epileptic patients samples has been found (De Simoni et al. 2000; Plata-Salaman et al. 2000; Dube et al. 2005; Gorter et al. 2006; Ravizza and Vezzani 2006; Ravizza et al. 2008; Sinha et al. 2008; Vezzani et al. 2013). Interestingly, in mixed glial cell cultures, the blockage of Na+ channels by tetrodotoxin or by the anti-epileptic drug phenytoin, inhibited microglial activation and secretion of the pro-inflammatory cytokines IL-1β and TNF-α induced by lipopolysaccharides (LPS) (Black et al. 2009). The pharmacological down-modulation of voltage-sensitive Na+ channels, which are critical in the initiation and conduction of brain action potentials, is particularly effective for the control of epileptic seizures (Catterall 1999). Accordingly, anti-epileptic drugs that reduce voltage-sensitive Na+ channel permeability are among the most effective in seizure control (Taylor and Narasimhan 1997). Several in vitro studies have been devoted to exploring the effects of anti-epileptic drugs on brain cells responsible for inflammatory cascades (Pavone and Cardile 2003; Haghikia et al. 2008; Black et al. 2009; Stienen et al. 2011; Dambach et al. 2014). Nevertheless, the question of whether anti-seizure drugs are capable of affecting brain inflammation in vivo has not been directly addressed.
Vinpocetine (ethyl apovincamine-22-oate) is a neuroprotective drug that inhibits brain pre-synaptic Na+ channel-mediated responses more potently and effectively than several anti-epileptic drugs of the first and second generations including carbamazepine (Sitges et al. 2005, 2006, 2007a, 2011). In contrast to vinpocetine and carbamazepine, the anti-epileptic valproic acid, whose mechanism of action mainly involves an increase in GABAergic transmission (Löscher 2002), was unable to decrease presynaptic ionic channels permeability in a broad range of concentrations (Sitges et al. 2007b). In the guinea pig vinpocetine, like carbamazepine and other classic anti-epileptic drugs inhibited the epileptiform electroencephalographic (EEG) activity pharmacologically induced by different convulsive agents (Nekrassov and Sitges 2004, 2006, 2008; Sitges and Nekrassov 2004). In addition, vinpocetine has been shown to reduce nuclear factor-kappa B-mediated inflammation in vascular smooth muscle cells, endothelial cells, macrophages and epithelial cells (Jeon et al. 2010). Interestingly, to address the unmet medical need for the control of seizures in refractory epileptic patients, anti-inflammatory therapies are starting to be considered as prospective new potential treatments (Sinclair 2003; Verhelst et al. 2005; Reid et al. 2009; Maroso et al. 2010; Löscher et al. 2013).
In the present study, to explore whether anti-seizure drugs that differ in their mechanisms of action could affect brain inflammation differently, the effect of the in vivo administration of vinpocetine, carbamazepine and valproic acid on the expression of two amply recognized pro-inflammatory markers was investigated in the hippocampus. Because the hippocampus is a highly epileptogenic brain structure in which seizures have shown to increase pro-inflammatory cytokines (De Simoni et al. 2000; Plata-Salaman et al. 2000; Ravizza et al. 2008).
Materials and methods
Vinpocetine, carbamazepine and valproic acid were kindly donated by Psicofarma S.A. de C.V. (México). 4-aminopyridine, pilocarpine and lipopolysaccharide (LPS, Escherichia coli, serotype 0127:B8) were from Sigma-Aldrich (St. Louis, MO, USA). Pentylenetetrazole was from MP Biochemicals Inc (Aurora, OH, USA).
The vehicles used to dissolve the anti-seizure drugs were different: vinpocetine was dissolved in saline acidified with HCl and adjusted to pH 4 with NaOH; carbamazepine in dimethylsulfoxide and valproic acid in saline. 4-aminopyridine pentylenetetrazole, pilocarpine and LPS were dissolved in saline. All the drugs were administered i.p. at a small volume (1 mL/kg). In the case of carbamazepine, which was dissolved in an organic vehicle, the volume used was smaller (i.e. 0.5 mL/kg).
The dose of vinpocetine used of 5 mg/kg was chosen on the basis of our previous experience in the guinea pig showing that at this dose vinpocetine completely prevents the epileptiform EEG activity induced either by pentylenetetrazole or by 4-aminopyridine (Nekrassov and Sitges 2004, 2008; Sitges and Nekrassov 2004). In the case of carbamazepine we used a dose of 50 mg/kg to test its effect on the hippocampal inflammation markers, because in a previous study we found that even at a dose of 25 mg/kg carbamazepine completely prevented the epileptiform EEG activity induced by 4-aminopyridine and pentylenetetrazole in the rat (Sitges et al. 2012). In the case of valproic acid, we also used a dose of 50 mg/kg because in the guinea pig a lower dose (30 mg/kg) prevented the EEG epileptiform activity induced by pentylenetetrazole in 70% of the animals (Nekrassov and Sitges 2006).
The pro-convulsive agents, 4-aminopyridine and pentylenetetrazole were used at doses that in the rat induce seizures in 100% of the animals (Sitges et al. 2012), and the pilocarpine dose was chosen on the basis of previous studies. For instance see Ravizza et al. (2008). The dose of LPS used (100 μg/kg inoculated i.p.) was selected based on previous studies demonstrating that a single i.p. administration of LPS at that dose increases the expression of pro-inflammatory cytokines in hippocampus (Turrin et al. 2001; Frank et al. 2012).
In the present study, 126 male Wistar rats (291 ± 1.1 g initial weight) divided in 26 groups were included. Groups were defined by the substance(s) to be injected (Table S1). To test the effect of drugs on IL-1β and TNF-α mRNA expression in the hippocampus by reverse transcription-PCR 18 animal groups were formed. The other eight groups were used to test the effect of anti-seizure drugs on the expression of the IL-1β protein in the hippocampus by western blot. Animals were housed in a 12 h light–dark cycle in stable conditions of temperature and with access to food and water ad libitum. Animals were from the Instituto de Investigaciones Biomédicas Animal House at the Universidad Nacional Autónoma de México, and the Experiments were carried out in compliance with the Guidelines for Animal Experimentation and with the approval of the “Laboratory Animals Care and Use Committee”.
As there was no difference between the IL-1β and TNF-α mRNA expression in animals injected with the vehicles was found, all animals administered with the different vehicles were included in group 1 (veh). The effect of one or several doses of the anti-seizure drugs was tested in the following groups: group 2 that was injected once with vinpocetine at a dose of 5 mg/kg, group 3 injected once with carbamazepine at a dose of 50 mg/kg and group 4 injected once with valproic acid also at a dose of 50 mg/kg. The four groups used to test the effect of the repeated doses of vehicle or anti-seizure drug were: group 5, that was injected daily with the different vehicles for 1 week, and groups 6, 7 and 8, that were administered for 1 week with one daily injection of vinpocetine (5 mg/kg), carbamazepine (50 mg/kg) or valproic acid (50 mg/kg), respectively. The animals of these eight groups were sacrificed by decapitation 90 min after the single or the last injection. This time was chosen on the basis of previous studies showing that the anti-seizure drugs tested here reach an adequate brain concentration around this time (Vereczkey et al. 1979; Blotnik et al. 1996; Graumlich et al. 2000).
The effect of the anti-seizure drugs also was tested on the expression of the IL-1β protein by western blot. For this purpose eight additional groups were formed; namely, group 9 injected once with saline, group 10 injected once with vinpocetine (5 mg/kg), group 11 injected once with carbamazepine (50 mg/kg) and group 12 with valproic acid (50 mg/kg). Groups 13, 14, 15 and 16 got a daily injection for 1 week of saline, vinpocetine (5 mg/kg), carbamazepine (50 mg/kg) or valproic acid (50 mg/kg), respectively. The animals of these groups were sacrificed by decapitation also 90 min after the single or the last injection.
The effect of 100 μg/kg LPS on IL-1β and TNF-α mRNA expression in the hippocampus was determined by PCR in animals sacrificed 30 min, 1, 3 and 6 h following LPS. As shown in Figure S1, the maximum increase in IL-1β and TNF-α mRNA expression was found in the group exposed to LPS for 1 h. Thus, the effect of the anti-seizure drugs on the increase in cytokines mRNA expression induced by LPS was tested 1 h following LPS in three additional groups; namely group 18 that was pre-administered with 5 mg/kg vinpocetine 90 min before LPS, and groups 19 and 20 that were pre-administered with 50 mg/kg carbamazepine or 50 mg/kg valproic acid, respectively, 90 min before LPS.
The effect of seizures induced by different pro-convulsive agents on IL-1β and TNF-α mRNA expression in the hippocampus also was explored. For this purpose, three groups were formed: group 21 administered with 2.5 mg/kg 4-aminopyridine, group 22 with 50 mg/kg pentylenetetrazole and group 23 with 340 mg/kg pilocarpine. All animals exposed to the pro-convulsive agents were observed for 30 min after the first tonic-clonic seizure, and then they were sacrificed by decapitation. In the 30 min of observation none of the animals died in status epilepticus.
The last three groups were formed to test the effect of the anti-seizure drugs on seizures induced by 2.5 mg/kg 4-aminopyridine. We chose this pro-convulsive drug because its mechanism of action involves changes in voltage sensitive presynaptic Na+ channels permeability (Galvan and Sitges 2004). Thus, group 24 was pre-administered with vinpocetine (5 mg/kg) 90 min before 4-aminopyridine, and groups 25 and 26 with carbamazepine (50 mg/kg) or valproic acid (50 mg/kg) 90 min before 4-aminopyridine, respectively. All these animals were observed for 1 h following 4-aminopyridine injection and then sacrificed by decapitation.
Detection of IL-1β and TNF-α mRNA expression in the hippocampi by PCR
The brains of the animals submitted to the different experimental conditions were removed and the hippocampus of both hemispheres dissected out. For the total RNA extraction, the dissected hippocampi were placed in sterile tubes containing 1 mL of TRIzol Reagent (Invitrogen Life Technologies, Grand Island, NY, USA) and frozen at −80°C until used. Total RNA extraction was performed after hippocampus homogenization (15 strokes with a AA Teflon homogenizer, Thomas Scientific, Swedesboro, NJ, USA) according to the manufacturer's instructions. The RNA samples were suspended in 50 μL of nuclease-free water.
To determine the amount and purity of total RNA in each sample a nanodrop spectrophotometer (Thermo scientific, Wilmington, DE, USA) was used. The integrity of total RNA was assessed by agarose gel electrophoresis using ethidium bromide staining.
cDNA was obtained by reverse transcription of the total RNA using the kit SuperScript® III First-Strand Synthesis SuperMix (Invitrogen Life Technologies). For this purpose, a small aliquot containing 2 μg of RNA suspended in nuclease-free water was mixed with 0.5 μL Oligo (dT)20 (50 μM) and 0.5 μL annealing buffer and brought up to a final volume of 4 μL with nuclease-free water. This mixture was incubated at 65°C for 5 min and then chilled on ice. For reverse transcription, 1 μL of III/RNaseOUT™ enzyme mix and 5 μL of the 2X first-strand reaction mix (10 mM MgCl2 and 1 mM of each dNTP) were added before incubation at 50°C for 50 min. The reaction was stopped by heating the mixture at 85°C for 5 min. The cDNA resulting from this procedure was stored at −20°C until use.
IL-1β and TNF-α mRNA expression was evaluated by PCR using the kit GoTaq® DNA Polymerase. Briefly, 1.5 μL of cDNA (250 ng/μL) were amplified in a mixture containing 2 μL of 5X green buffer, 0.8 μL of MgCl2 (25 mM), 0.25 μL of PCR nucleotide mix (10 mM), 0.5 μL of the sense primer (10 pM), 0.5 μL of the antisense primer (10 pM), 0.05 μL of DNA Polymerase (5 u/μL) and 4.4 μL of sterile Milli-Q water (Merck Millipore, Billerica, MA, USA).
PCR reactions were done in an Eppendorf Mastercycler gradient (Eppendorf, Hauppauge, NY, USA). The temperature cycling conditions were: initial denaturation at 95°C for 5 min, followed by 34 cycles, including denaturation at 94°C for 30 s, primer annealing for 45 s at 58°C for IL-1β and β-actin or at 66°C for TNF-α, and primer extension at 72°C for 1 min. A final primer extension was performed at 72°C for 10 min after which the samples were immediately cooled at 4°C. The PCR primers used are shown in Table 1.
Table 1. Nucleotide sequence of the sense and antisense primers used for PCR
The amplified amount of pro-inflammatory cytokine mRNA was normalized by the amplified β-actin mRNA. A negative control in the absence of sample was run in parallel.
PCR products were separated by 1.5% agarose gel electrophoresis, stained with ethidium bromide and the signal intensity of the resulting bands measured by densitometry using a MiniBIS Pro Gel Documentation System (Bio-America, Miami, FL, USA) and the Image J software developed by Wayne Rasband, National Institutes of Health, Bethesda, MD, USA. Results are the relative expression of the pro-inflammatory cytokines mRNA (IL-1β mRNA/β-actin mRNA or TNF-α mRNA/β-actin mRNA ratio).
Detection of IL-1β protein expression in the hippocampus by western blot
The dissected hippocampi of groups 9 to 16 were placed in sterile tubes and frozen at −80°C until used. Total protein extraction was performed by hippocampus homogenization (10 strokes with a Thomas Scientific AA Teflon homogenizer) in 1 mL of lysis buffer (25 mM TrisHCl pH 7.4, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 1 mM phenylmethanesulfonyl-fluoride and 1 μL of the Halt™ Protease inhibitor cocktail, Thermo Scientific, Rockford, IL, USA). The mixture was incubated at 4°C for 1 h and then centrifuged at 10 000 g for 10 min at 4°C. The resulting supernatants were collected and stored at −80°C until protein determination by Lowry.
For electrophoresis 50 μg of total protein were loaded per line and separated on a 12% sodium dodecyl sulfate–polyacrylamide electrophoresis gel at 100 V for 90 min. Separated proteins were transferred to polyvinylidene fluoride (PVDF) microporous membranes at 200 mA for 60 min using an electro-blotting system (Bio-Rad, Hercules, CA, USA). The blots were stained (Ponseau S from Sigma-Aldrich) to confirm equal protein loading, and unstained with Tris-buffered saline Tween (TBST, 20 mM Tris, 150 mM NaCl, 0.1% Tween 20, pH 7.5). Non-specific binding to PVDF membranes was blocked by 5% Blotto non-fat dry milk (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) in TBST for 1 h at 24°C. Membranes were then incubated 20 h at 4°C with a rabbit anti-IL-1β polyclonal antibody (catalog number SC-7884, 31 kDa, Santa Cruz Biotechnology, Inc.) diluted 1 : 500 in 5% non-fat dry milk in TBST. After washing three times with TBST for 15 min, membranes were incubated with the secondary antibody coupled to horseradish peroxidase (goat anti-rabbit polyclonal antibody from Cell Signaling Technology, Inc., Danvers, MA, USA) diluted 1 : 1000 in 5% non-fat dry milk in TBST for 1 h.
β-Actin was used as loading control. To detect β-actin expression, the PVDF membranes were stripped (using a buffer containing 1.5% glycine, 0.1% sodium dodecyl sulfate and 1% Tween 20, pH 2.2) and re-probed with mouse anti-β-actin monoclonal antibody (Sigma-Aldrich) diluted 1 : 5000 in 5% non-fat dry milk in TBST for 1 h at 24°C. Then membranes were washed three times with TBST for 15 min before incubation with the secondary antibody coupled to horseradish peroxidase (goat anti-mouse polyclonal antibody from Santa Cruz Biotechnology, Inc.) diluted 1 : 10 000 in 5% non-fat dry milk in TBST for 1 h.
Proteins were visualized in Biomax light films (Sigma-Aldrich) using the Immobilon™ Western Chemiluminiscent horseradish peroxidase substrate (Millipore, Billerica, MA, USA). Band intensities were measured by densitometry also using the Image J software (National Institutes of Health, Bethesda, MD, USA). To adjust for possible loading errors, IL-1β band intensities were normalized against the β-actin band intensities, obtained in the corresponding lanes.
One-way anova followed by a post hoc Tukey test was used for the statistical evaluations. Statistical analyses were performed with SigmaPlot version 11.0 (Systat Software, Erkrath, Germany). The criterion for statistical significance was p < 0.05.
Acute and chronic effects of vinpocetine, carbamazepine and valproic acid on IL-1β and TNF-α mRNA expression in the hippocampus
The effect of the anti-seizure drugs injected once or daily for 1 week on IL-1β and TNF-α mRNA expression in the hippocampus is shown in Fig. 1. The upper part of this figure shows a similar decrease in IL-1β mRNA expression after one or after several doses. For instance, IL-1β expression in animals administered once or for 7 days with vehicles was lowered by the single dose of vinpocetine (G2) or carbamazepine (G3), as well as by the repeated doses of vinpocetine (G6) and carbamazepine (G7). In contrast, valproic acid failed to modify IL-1β mRNA expression from basal conditions after single (G4) or repeated doses (G8).
The lower part of Fig. 1 shows that also in the case of TNF-α messenger expression the anti-seizure drugs induced very similar decrease after 1 or 7 doses, except that repeated injections of carbamazepine were necessary to decrease TNF-α significantly.
A representative experiment showing bands of IL-1β, TNF-α and β-actin mRNA expression from animals injected with one (left gels) or seven doses of the anti-seizure drugs (right gels) is shown in Figure S2.
As shown in Figure S3 the vehicles used to dissolve the drugs tested did not modify the hippocampal mRNA expression of the pro-inflammatory cytokines. Therefore, all animals treated with the different drugs were compared against a general control that is referred to as ‘vehicle’ in the figures for simplicity. Although it is worthy to mention that the same statistic significance between groups injected with a drug and its respective vehicle was found.
Acute and chronic effects of vinpocetine, carbamazepine and valproic acid on the expression of IL-1β protein in the hippocampus
In order to investigate whether the potential anti-inflammatory effect of vinpocetine and carbamazepine at the doses tested also affected IL-1β protein levels in the hippocampus, the same treatment was used and IL-1β was detected by western blot. Fig. 2 shows that in animals injected once with vinpocetine (G10) or with carbamazepine (G11) the expression of IL-1β protein was below baseline values. In the groups injected with seven vinpocetine (G14) and carbamazepine (G15) doses the expression of IL-1β protein also was lower than in the group injected with seven doses of vehicle (G13).
Effect of vinpocetine, carbamazepine and valproic acid on LPS-induced increase in IL-1β and TNF-α mRNA expression
Figure 3 shows that the IL-1β and TNF-α mRNA expression detected in the hippocampus of the group of animals sacrificed 1 h following LPS (G17) was significantly reduced in the animals pre-administered with vinpocetine (G18) and carbamazepine (G19) before LPS, and unchanged in the group pre-administered with valproic acid before LPS (G20).
Tonic-clonic seizures and IL-1β and TNF-α mRNA expression changes induced by different pro-convulsive agents in the hippocampus
Animal groups injected with 4-aminopyridine (G21), pentylenetetrazole (G22) or pilocarpine (G23), all developed at least one generalized tonic–clonic seizure. Pentylenetetrazole induced the first tonic-clonic seizure with a latency of 1.5 ± 0.2 min, and 4-aminopyridine and pilocarpine with latencies of around 20 and 30 min, respectively. The generalized tonic-clonic seizures lasted around 1 min in all cases, but the number of generalized tonic-clonic seizures induced by each pro-convulsive agent within the 30 min following the first generalized seizure varied. In G21, 4 of the 6 animals administered with 4-aminopyridine presented a second tonic-clonic seizure, and in the G22 only one of the 7 animals injected with pentylenetetrazole presented a second tonic-clonic seizure. In the group administered with pilocarpine (G23), 3 additional tonic-clonic seizures were induced in all the animals within the 30 min following the first generalized tonic-clonic seizure, and 2 animals even presented a fifth tonic-clonic seizure within that time period. All pro-convulsive agents, 4-aminopyridine, pentylenetetrazole and pilocarpine, induced a significant increase in both IL-1β (Fig. 4a) and TNF-α mRNA expression (Fig. 4b) in the hippocampus of these animals.
Effect of vinpocetine, carbamazepine and valproic acid on tonic-clonic seizures and on the increase in IL-1β and TNF-α mRNA expression induced by 4-aminopyridine
To further test our hypothesis that the effect of the anti-seizure drugs on brain inflammation is linked to their action on ion channels, we also tested the effect of the anti-seizure drugs on the expression of the pro-inflammatory cytokines in the hippocampi of rats exposed to 4-aminopyridine, which induces seizures modifying the permeability of several presynaptic ionic channels (Galvan and Sitges 2004; Sitges et al. 2011). In the group administered with 4-aminopyridine, behavioral changes such as piloerection, grooming and wet dog shakes usually preceded the generalized tonic-clonic seizures with limb extensions. Table 2 shows that in the animals pre-injected with the anti-seizure drugs 90 min before 4-aminopyridine, the latency of these behavioral changes was increased. In the groups pre-administered with vinpocetine and carbamazepine the generalized tonic-clonic seizures were completely prevented and not all the animals presented wet dog shakes. In the group pre-administered with valproic acid, prevention of the behavioral changes was less conspicuous and 4-aminopyridine still induced a continuous tremor of the whole body musculature.
Table 2. Effect of vinpocetine, carbamazepine and valproic acid on tonic-clonic seizures induced by 4-aminopyridine
Latency in min to the indicated behavior
First tonic-clonic seizure
WDS, wet dog shakes.
Animals were observed for 60 min following 4-aminopyridine administration.
Vinpocetine, carbamazepine and valproic acid were injected at a dose of 5, 50 and 50 mg/kg, respectively.
The anti-seizure drugs were pre-administered 90 min before 4-aminopyridine.
Results are the mean ± SEM values of the indicated number (#) of animals per group.
**p < 0.01, ***p < 0.001 between 4-aminopyridine alone and in combination with the indicated drug.
continuous whole body tremor without hindlimb extensions
To test whether the effect of anti-convulsive drugs on 4-aminopyridine-induced seizures was reflected in the increase in IL-1β and TNF-α mRNA expression induced by 4-aminopyridine, the expression of these messengers in the hippocampus of the groups pre-administered once with saline, vinpocetine, carbamazepine or valproic acid before 4-aminopyridine was determined.
Figure 5a shows that the expression of IL-1β mRNA observed in G21, that was the group administered with the pro-convulsive agent 4-aminopyridine, was decreased in all the groups pre-administered with the anti-seizure drugs; namely, the vinpocetine (G24), the carbamazepine (G25) and the valproic acid (G26) groups. The decrease observed in the group pre-administered with valproic acid was, however, less pronounced. Moreover, in groups pre-administered with vinpocetine and carbamazepine before 4-aminopyridine the expression of IL-1β was drawn even below basal conditions. The expression of IL-1β in the control group injected only with vehicle (G1) was 0.50 ± 0.03, in the group injected with vinpocetine before 4-aminopyridine (G24) 0.36 ± 0.03; and in the group injected with carbamazepine before 4-aminopyridine (G25) 0.29 ± 0.02. The statistical difference between G1 and G24 was p = 0.04, and between G1 and G25 was p < 0.001.
Figure 5b shows that the rise in TNF-α mRNA expression induced by 4-aminopyridine was only significantly prevented in G24 and G25, namely the groups pre-administered with vinpocetine and carbamazepine.
The present study shows that an important property of some anti-seizure drugs is their capacity to exert cerebral anti-inflammatory effects. This conclusion is based on the decreased expression of the inflammation markers IL-1β and TNF-α under basal conditions exerted by vinpocetine and carbamazepine in the hippocampus. The sensitivity of the rise in the pro-inflammatory cytokines induced by LPS to vinpocetine and carbamazepine also supports this conclusion. In the present study, we focused on IL-1β and TNF-α because they are the main pro-inflammatory cytokines involved in triggering inflammatory responses, and because of the positive relationship between the increase in these particular cytokines and seizures (Dube et al. 2005; Balosso et al. 2013; Vezzani et al. 2013). However, other cytokines may also play a role in epilepsy and anti-epileptic treatment. For instance, IL-2, IL-4 and IL-6 have been shown to play a role in epilepsy too (Sinha et al. 2008); and anti-epileptic agents have been reported to influence IL-2 and IL-22 production (Himmerich et al. 2013).
It is amply recognized that the basis of the anti-seizure effects of vinpocetine and carbamazepine involves inhibition of Na+ channel permeability. Accordingly, vinpocetine and carbamazepine have been shown to inhibit Na+ channel-mediated Glu release in cerebral isolated nerve endings (Sitges et al. 2005, 2006, 2007a, 2011). In addition, they effectively prevented the EEG epileptiform activity induced by 4-aminopyridine (Sitges and Nekrassov 2004; Nekrassov and Sitges 2008; Sitges et al. 2012), which importantly involves several presynaptic ionic channels, including Na+ channels (Galvan and Sitges 2004; Sitges et al. 2011). Remarkably, our present study suggests that another mechanism of action of these two anti-seizure drugs involves brain anti-inflammatory effects. Thus, we would like to propose a contribution of the cerebral anti-inflammatory effect of vinpocetine and carbamazepine in their anti-seizure action. This idea is supported by observations that brain inflammation may contribute to the induction of epilepsy, as is also suggested by the epileptic properties of LPS and the induced cytokine production in the brain (Heida et al. 2004). Moreover, vinpocetine and carbamazepine might be reducing both neuronal excitability and glial pro-inflammatory cytokine release with the concomitant inhibition of brain inflammation in vivo via their action on Na+ channels permeability. Because in glial cell cultures the anti-epileptic phenytoin that also exerts its anti-seizure effect by inhibiting Na+ channels was shown to reduce IL-1β and TNF-α secretion (Black et al. 2009).
Using the body surface area normalization method for a 60 kg human (Reagan-Shaw et al. 2008), the 50 mg/kg of carbamazepine administered here to the rat correspond to a human equivalent dose of 486 mg. This dose is around the average carbamazepine dose given to epileptic patients in a day, and was able to reduce the expression of the pro-inflammatory markers in the rat hippocampus: a result suggesting that at therapy-relevant doses carbamazepine is capable of exerting anti-inflammatory effects in the brain. In the case of vinpocetine, the 5 mg/kg administered here to the rat would correspond to a human equivalent dose of 48.6 mg (using the body surface area normalization method for a 60 kg human), and there is evidence that at this dose vinpocetine controls seizures in epileptic patients (Dutov et al. 1986). The 50 mg/kg of valproic acid given to the rat here would correspond to a human equivalent dose of 486 mg for a 60 kg human; a dose that is around the lower dose of valproic acid given to epileptic patients. In a rat model of ischemia, valproic acid at a human equivalent dose of 2919 mg was found to reduce myeloperoxidase and ionized calcium binding adapter molecule 1 staining (Suda et al. 2013), the possibility that the failure of valproic acid to reduce the expression of the pro-inflammatory markers observed here in the hippocampus was related to the use of a low dose cannot be completely discarded. However, valproic acid administered to rats at a high human equivalent dose (2432 mg) for 3 days also failed to modify the expression of IL-1β and TNF-α in the hippocampus (Hsu et al. 2013).
Taking into account that vinpocetine is rapidly eliminated (125 min half-life) in the rat (Vereczkey et al. 1979), it is very likely that the similar decrease in IL-1β and TNF-α expression observed after the single and repeated vinpocetine doses was simply due to the daily vinpocetine clearance. In contrast, seven doses of carbamazepine were needed to decrease TNF-α expression, suggesting an additive effect of this drug. Consistent with this, carbamazepine is still found in rat brain tissue around 15 h after its i.p. administration (Graumlich et al. 2000). Blotnik et al. (1996) determined a 14 h half-life for valproic acid in the rat brain following an intravenous administration of 20 mg/kg, and found that this drug reaches a high concentration in the rat brain tissue already after 10 min. However, the i.p. administration of 50 mg/kg valproic acid was unable to modify IL-1β and TNF-α expression even after repeated doses.
The validity of the experimental design used here to unmask the cerebral anti-inflammatory effect of the anti-seizure drugs is strongly supported by our data showing that three different pro-convulsive agents were capable of exerting the opposite effect on the hippocampal expression of the pro-inflammatory markers. The observed increase produced by 4-aminopyridine, pentylenetetrazole and pilocarpine on pro-inflammatory cytokines also supports the idea that cerebral inflammation is a frequent phenomenon accompanying seizures. In addition, a positive relationship between the frequency and duration of tonic-clonic seizures and the severity of brain inflammation is suggested by our finding that the largest rise in pro-inflammatory cytokine expression in response to generalized tonic-clonic seizures was obtained in the group exposed to pilocarpine, the group that suffered the largest number of seizures.
4-aminopyridine was shown to increase Na+ and Ca2+ channel permeability (Galvan and Sitges 2004; Sitges et al. 2011). Vinpocetine and carbamazepine, which decrease presynaptic Na+ and Ca2+ channel permeability (Sitges et al. 2007a,b), were more effective than valproic acid in inhibiting the behavioral changes that precede the generalized tonic-clonic seizures induced by 4-aminopyridine. Supporting our interpretation that the decrease in ionic channel permeability reduces both neuronal excitability and glial pro-inflammatory cytokine release, vinpocetine and carbamazepine were also more effective than valproic acid in preventing the increase in TNF-α and IL-1β expression induced by 4-aminopyridine. However, as the decrease in brain excitability due to the rise in GABAergic transmission induced by valproic acid is also expected to indirectly reduce excitability (with the concomitant activation of cerebral ion channels and brain inflammation), in the animals pre-exposed to valproic acid, seizures induced by 4-aminopyridine were less severe and the increase in IL-1β expression induced by 4-aminopyridine was reduced.
Vinpocetine is known to decrease TNF-α and IL-1β in several peripheral tissues by inhibiting nuclear factor-kappa B activity (Jeon et al. 2010). Seizures have been shown to increase blood–brain barrier permeability (Fieschi et al. 1980; Marchi et al. 2007, 2009). Therefore, a certain contribution of systemic cytokines in the group administered with vinpocetine cannot be discarded. However, in the hippocampus of animals that were not exposed to 4-aminopyridine and in which blood-brain barrier permeability was undisrupted by seizures, vinpocetine also decreased IL-1β and TNF-α expression from basal conditions.
Finally, increased levels of pro-inflammatory cytokines in serum and cerebrospinal fluid samples from epileptic patients have been reported (Sinha et al. 2008). Although to characterize selective actions of specific anti-seizure drugs on brain inflammation markers, experimental animal models are likely to be more suitable. Because epileptic patients are usually medicated, and on the basis of present findings, seizures and some anti-epileptic drugs can both change the inflammatory markers expression.
In summary, the present results indicate that anti-seizure drugs with a mechanism of action that involves a decrease in Na+ channels permeability, like carbamazepine and the new anti-seizure drug vinpocetine, are highly effective in reducing the increased brain excitability accompanying seizures as well as in reducing cerebral inflammatory condition. Their action on both processes might make these anti-seizure drugs highly effective in suppressing or preventing epileptic seizures.
Acknowledgments and conflict of interest disclosure
The authors thank Araceli Guarneros, Luz María Chiu, and Mari-Carmen Basualdo for their excellent technical assistance, and Dr Robyn Hudson for kindly revising the text and her pertinent corrections. This study was partially supported by project IN200614 from Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT), Universidad Nacional Autónoma de México and by Psicofarma S.A. de C.V. Carlos Daniel Gómez was a PhD student in the Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México, and received a scholarship from Consejo Nacional de Ciencia y Tecnología (CONACYT), México.
All experiments were conducted in compliance with the ARRIVE guidelines. The authors declare that they have no conflicts of interest.