The antiepileptogenic effect of electrical stimulation at different low frequencies is accompanied with change in adenosine receptors gene expression in rats
Summary
Purpose: Previous studies have shown that the anticonvulsant effects of low‐frequency stimulation (LFS) can be affected by activation of adenosine receptors. In the present study, the effect of LFS at different frequencies on kindling rate and adenosine receptors gene expression was investigated.
Methods: Animals were kindled by perforant path stimulation in a rapid kindling manner. LFS (0.5, 1, and 5 Hz) was applied after termination of each kindling stimulation. Seizure severity was measured according to behavioral and electrophysiologic parameters. At the end of the experiments, adenosine A1 and A2A receptor gene expression were measured.
Results: The inhibitory effect of LFS on kindling acquisition was observed at all frequencies. In addition, the inhibitory action of LFS on enhancement of field excitatory postsynaptic potential slope and population spike amplitude during kindling acquisition was not affected by the LFS frequency. However, the effects of LFS on paired‐pulse recordings were greater at frequency of 5 Hz. Application of LFS during kindling acquisition also prevented the kindling induced decrease in the A1 receptor gene expression and attenuated the level of A2A receptor gene expression in the dentate gyrus. These effects were also greater at the frequency of 5 Hz.
Discussion: According to these data, it may be suggested that the antiepileptogenic effects of LFS, developed through inhibition of synaptic transmission in the dentate gyrus, is mediated somehow through preventing the decrease of A1 receptor and through attenuating the A2A receptor gene expression. These effects might be dependent on the frequency of LFS.
Many patients with temporal lobe epilepsy remain resistant to current therapies, and there is a need to find new, effective, and safe alternative treatments. The recent progress achieved by using electrical stimulation to treat movement disorders has led to renewed research effort to develop direct stimulation of the brain as therapeutic strategies for epilepsy (Goodman et al., 2005).
Deep brain stimulation may be effective in reducing the severity of the seizure induced by different laboratory models, including kindling (Gaito et al., 1980; Velisek et al., 2002). Kindling refers to the gradual electrographic and behavioral development in response to repeated application of brief intermitted and low‐intensity trains of electrical stimulation (Goddard et al., 1969). Although some investigators have focused on the efficiency of high‐frequency stimulation (Boon et al., 2007a,b; Van Roost et al., 2007), new experimental (Gaito & Gaito, 1981; Goodman et al., 2005; Mohammad‐Zadeh et al., 2007) and clinical (Yamamoto et al., 2002, 2006; Kinoshita et al., 2004) evidence shows that low‐frequency stimulation (LFS) has also inhibitory effect on seizure.
Kindling potentiates excitatory and inhibitory synaptic transmission in the dentate gyrus (Adamec et al., 1981; de Jonge & Racine, 1987; Maru & Goddard, 1987; Gilbert, 1991). It has been shown that the slope of population excitatory postsynaptic potential (pEPSP) and amplitude of population spike (PS) increase following kindling (Robinson et al., 1991; Ruthrich et al., 2001). In addition, kindling of the dentate gyrus induces an increase in early (10–50 ms intervals) and late (150–1,000 ms intervals) paired‐pulse depression, which may reflect the potentiation of synaptic transmission of inhibitory neurons (Gilbert, 1998). Kindling also produces a reduction of paired‐pulse facilitation (70–100 ms intervals) (Maru & Goddard, 1987; Gilbert, 1998). Our recent experiments have demonstrated that application of LFS on the perforant path prevents kindling acquisition and inhibits kindling induced changes in the synaptic transmission (Mohammad‐Zadeh et al., 2007).
The quenching effects of LFS on kindled seizures are mediated by inhibitory neurotransmitters/ neuromodulators such as adenosine (Mohammad‐Zadeh et al., 2009) and galanin (Sadegh et al., 2007). Adenosine is a neuromodulator that mainly inhibits synaptic transmission and decreases neuronal excitability in the hippocampus through activation of the predominant adenosine A1 receptors (Ribeiro, 1995; de Mendonca & Ribeiro, 1997). The adenosine A2A receptors are less abundant than A1 receptors in the hippocampus (Sebastiao & Ribeiro, 1992; Cunha et al., 1994). Despite their low abundance, A2A receptors play a role in different noxious brain situations, in particular during convulsive behavior (Adami et al., 1995; Huber et al., 2002; Zeraati et al., 2006; Hosseinmardi et al., 2007). It has been reported that electrically kindled seizures increase the level of adenosine A1 receptors shortly after convulsive behavior (Angelatou et al., 1990, 1993). In few studies, a decrease in the adenosine A2A receptors level has also been reported during seizure activity (Aden et al., 2004).
Our recent study showed that activation of adenosine A1 receptors has a role in mediating the anticonvulsant effects of LFS on perforant path kindling acquisition so that microinfusion of an adenosine A1 receptor antagonist suppressed this inhibitory effect of LFS. On the other hand, A2A adenosine receptors blockade could not prevent the inhibitory action of LFS on kindling acquisition (Mohammad‐Zadeh et al., 2009). Therefore, considering (1) the important role of adenosine receptors in depotentiating effect of LFS on long‐term potentiation (LTP) (Fujii et al., 1997, 2000), (2) preventing effect of LFS on kindling‐induced potentiation (Mohammad‐Zadeh et al., 2007), and (3) the involvement of adenosine receptors in mediating the inhibitory effect of LFS on kindling acquisition (Mohammad‐Zadeh et al., 2009), it may be postulated that LFS application results in a change in adenosine receptors gene expression.
The exact antiepileptic mechanisms of LFS and the optimal stimulus parameters for preventing or disrupting seizure activity remain to be elucidated. It has been shown that LFS effectiveness depends on its frequency (Gaito, 1980b; Kemp & Bashir, 2001). Therefore, in the present study we investigated the effect of electrical stimulation at different low frequencies on their antiepileptogenic effects and on adenosine A1 and A2A receptors gene expression in perforant path kindled seizures. LFS was applied in three frequencies: 0.5 Hz and 5 Hz as the lower and higher range of LFS frequencies and 1 Hz as the most common LFS frequency.
Methods
Animals
Forty‐eight adult male Wistar rats (280–300 g at the time of surgery) obtained from the Pasteur Institute of Iran (Tehran, IR Iran) were maintained in a colony room kept at a constant temperature with an artificial 12‐h light/dark schedule. The lights were turned on at 7:00 am. Animals were housed in individual cages with woodchip bedding and permitted free access to standard food and water. Efforts were made to minimize animal suffering and to minimize the number of animals used. All studies were performed in accordance with the ethical guidelines set by the “Ethical Committee of School of Medical Sciences, Tarbiat Modares University,” which completely coincides with the “NIH Guide for the Care and Use of Laboratory Animals.” All experiments were done at the same time (8:00 am to 2:00 pm).
Surgical procedure
Surgical procedure was done as described previously (Mohammad‐Zadeh et al., 2007, 2009). Under sodium pentobarbital anesthesia (50 mg/kg, i.p.) animals underwent stereotaxic implantation with a bipolar stimulating electrode in the perforant path (coordinates: A, −6.9 mm; L, 4.1 mm; and, V, 2.0–2.5 mm below dura) and a monopolar recording electrode in the dentate gyrus (coordinates: A, −2.8 mm; L, 1.8 mm; and, V, 2.5–3.0 mm below dura) of the right hemisphere (Paxinos & Watson, 2004). The incisor bar was set 3.3 mm below the interaural line. The depth of the recording and stimulating electrodes was adjusted to maximize the population spike amplitude in the dentate gyrus in response to the perforant path stimulation and to confirm correct electrodes location. The stimulating electrode consisted of two twisted electrodes with a tip distance of 0.5 mm. Electrodes were stainless steel, Teflon coated, 127 μm in diameter, and insulated except at their tips (A‐M Systems, Inc., Carlsborg, WA, U.S.A.). Stainless steel screws were also positioned in the skull above the frontal and occipital cortices, and they served as reference and ground electrodes. All electrodes were connected to pins of a lightweight multichannel miniature socket as a head‐stage and fixed to the skull with dental acrylic. Electrophysiologic experiments were done after at least 10 days for recovery.
Stimulation and recording
All the recordings were performed after the rat had been transferred from the home cage to a recording box (30 × 30 × 30 cm). The head‐stage of the rat was connected to a flexible, shielded cable. Evoked responses were recorded in freely moving rat while the animal was awake with its eyes open as we explained previously (Mohammad‐Zadeh et al., 2007, 2009).
Input–output curves
To obtain the input–output curves, single 0.1‐ms monophasic square wave pulses were delivered through a Nihon Kohden (Tokyo, Japan) stimulator and Nihon Kohden SS‐202J constant‐current stimulus isolation unit every 10 s and applied at varying intensities (100–800 μA) to the perforant path while the evoked field potentials were monitored in the dentate gyrus. At each time point, 12 evoked responses were averaged. Both pEPSP slope and PS amplitude were monitored. The PS amplitude was measured by averaging the height from the peak of the pEPSPs to the base of PS. By means of input–output curve, the maximum PS amplitude was determined for each individual animal and all potentials employed as baseline criteria were evoked at a stimulus intensity that produced 50% of this maximum response (i.e., test pulse). The measured test pulse for different animals was between 100 and 500 μA. Responses were evoked, amplified, and digitized (at 10 kHz) using a PC‐based data acquisition system (D3107; World of Science Instruments Co., Tehran, Iran) and custom‐designed software, averaged, and were continuously monitored and stored on disk.
Rapid kindling procedure
Following at least 10 days post surgical recovery, the afterdischarge (AD) threshold was determined by 1‐ms monophasic square wave of 50 Hz with 5‐s train duration as described previously (Mohammad‐Zadeh et al., 2007; Sadegh et al., 2007). The stimulations were initially delivered at 30 μA and then at 5‐min intervals of increasing stimulus intensity in increments of 10 μA. The minimum intensity sufficient to induce ADs for at least 10 s was designated as the AD threshold and used for stimulation. In this study, the AD threshold intensity of different animals ranged from 80 to 150 μA. Rats were stimulated at AD threshold 12 times a day with 5‐min intervals. Epileptiform ADs were continuously recorded from the dentate gyrus after each kindling stimulation. The afterdischarge duration (ADD) and the behavioral progression of kindling (stages 1–5; according to Racine scores) were monitored (Racine, 1972). During kindling stimulations, the maximum seizure stage of animals on each day was considered as the kindling seizure stage.
Field potential recording
Stimulation intensity (200–500 μA) for the field potential recording was set at test pulse, and 120 sweeps were averaged every 10 s. Field potential recording was done before kindling stimulations on the days 1, 4, and 7 during kindling acquisition and both pEPSP slope and PS amplitude were monitored.
Paired‐pulse tests
Following the recovery period, paired‐pulse tests were run on days 1, 4, and 7 during kindling acquisition. Stimulation intensity (200–500 μA) for the paired‐pulse tests was set at the test pulse, and six sweeps were averaged at each of nine interpulse intervals: 10, 20, 30, 50, 70, 100, 300, 500, and 1,000 ms. These intervals tested randomly, and pulse pairs were separated by 10 s (0.1 Hz). The paired‐pulse index was determined for each animal by calculating the percent ratio of the second PS (test) to the first (conditioning). Each day, the responses to paired‐pulse stimulations were recorded before kindling stimulations.
Gene expression study
RNA preparation and reverse transcription
For gene expression study, animals were sacrificed and the dentate gyrus was extracted and immediately preserved in liquid nitrogen. Total RNA was isolated based on isothiocyanate–phenol–chloroform protocol (Ausubel et al., 2002), using RNX+ reagent (CinnaGen, Tehran, Iran) according to the manufacturer’s instruction. The final total RNA pellet was suspended in 30 μl of DEPC (diethyl‐pyrocarbonate)–treated water (Fermentas, Vilnius, Lithuania). One microliter of total RNA was used for spectrophotometric determination of the RNA concentration. Two microliters of total RNA were used for determination of RNA integrity and quality by electrophoresis on agarose gel. The remaining RNA was stored at −70°C after adding 1 μl of RNase inhibitor (Fermentas). For each sample, cDNA synthesis was performed using 2 μg of total RNA, Oligo‐dT primer (Fermentas) and M‐MuLV reverse transcriptase (Fermentas) based on the manufacturer’s instruction.
Primer design and semiquantitative RT‐PCR
Primer sequences for adenosine A1 receptor (A1AR), adenosine A2A receptor (A2AAR) and β‐actin, designed on the basis of the published sequences in GenBank, were as follows: A1AR forward, 5′‐GCTGGAACAACCTGAGTGT‐3′; A1AR reverse, 5′‐GTGGGACAGGGAGAACGT‐3′; A2AAR forward, 5′‐CCTGTCTGTTCGAGGACGT‐3′; A2AAR reverse, 5′‐ACATCCCGAGGAGAGCCTT‐3′; β‐actin forward, 5′‐CCCAGAGCAAGAGAGGCATC‐3′; and β‐actin reverse, 5′‐CTCAGGAGGAGCAATGATCT‐3′. Polymerase chain reaction (PCR) cycles were optimized to achieve a linear relation between cDNA concentrations entered into the reaction and the amounts of PCR products. Based on the results of optimizing studying, segments of A1AR, A2AAR, and β‐actin (internal control) cDNA were amplified for 25, 27 and 22 cycles, respectively. Three microliters of the synthesized cDNA was used as template. PCR was performed using specific primers and TaqDNA polymerase (CinnaGen) based on the manufacturer’s instruction. The PCR was started at 95°C for 10 min followed by mentioned cycles of amplification, each containing a denaturizing step (95°C for 1 min), annealing step (at 59°C for 1 min), and extension step (72°C for 1 min). A final extension step at 72°C for 5 min was also applied. To check for contaminating genomic DNA, PCR was performed using the product of a reverse transcription (RT) reaction prepared in the absence of RT enzyme. Ten microliters of amplified product was run on 1% agarose (Roche, Penzberg, Germany). Agarose gels were stained by ethidium bromide and visualized under an ultraviolet (UV) light. A 100‐bp DNA ladder was used as a molecular size marker. Semiquantitative analyses of PCR products were done by band densitometry using a computerized image analyzing system (Lab Works software, UVP, Cambridge, United Kingdom).
Experimental design
Experiment 1: Effect of different LFS frequencies on perforant path kindling acquisition and synaptic excitability
In this experiment animals were divided into sham‐operated, LFS, kindled, and kindled + LFS (KLFS) groups. In KLFS groups, the animals were stimulated according to rapid kindling protocol and LFS (0.1 ms pulses at AD threshold intensity, i.e., 80–150 μA) was applied daily during the intervals between kindling stimulations for 200 s at 0.5, 1, and 5 Hz (named as KLFS 0.5 Hz, KLFS 1 Hz, and KLFS 5 Hz, respectively). In the kindled group, animals were followed using the same protocol but they did not receive LFS. Animals of the LFS groups received only LFS at 0.5, 1, and 5 Hz (without kindling stimulations) and sham‐operated animals did not receive any kind of stimulations after surgery. For baseline recording, field potential parameters were recorded on the days 1, 4, and 7 just before kindling stimulations for 20 min. As mentioned previously, paired‐pulse indices were also determined on the same days after field potential recordings. In the kindled group, two animals exhibited at least one stage 5 seizure on the sixth day and four animals on the seventh day of the experiment (the mean number of stimulations was 72.83 ± 2.72). Therefore, the parameters of the other groups were also recorded for 7 days. Fig. 1 shows the time line of stimulation and recording paradigms of different groups. Animals were killed 24 h after the last stimulation, and their brains were used in experiment 2.
Time line diagram showing the experimental protocol used in kindled and kindled + low‐frequency stimulation (KLFS) groups.
Experiment 2: Effect of kindling and LFS on A1AR and A2AAR gene expression
At the end of experiment 1, each animal was killed. Its brain was removed and the dentate gyrus of the stimulated side was extracted for A1AR and A2AAR gene expression measurements.
Statistics
Data were averaged and expressed as mean ± standard error of the mean (SEM) and accompanied by the number of observations. For each time point during the experiment, average and SEM were calculated from the data on 12 (for basal synaptic response experiments) or 6 (for input–output curve recordings and paired‐pulse experiments) successive evoked responses. A mean value of responses at 10 time points on day 1 was defined as the baseline (100%). Subsequent data were expressed as the percent change from the baseline. One‐way and two‐way analyses of variance (ANOVAs) were used to determine changes in measured parameters in kindled and KLFS groups during different days, and statistically significant differences were evaluated further by a Tukey’s post hoc test. The effect of LFS on behavioral seizure scores was analyzed by using the nonparametric Kruskal‐Wallis and Mann‐Whitney U test. The values of A1AR and A2AAR band density achieved from gel analysis and band densitometry were calculated and expressed as A1AR/β‐actin and A2AAR/β‐actin for each sample, respectively, and the averages for different groups were compared by using a one‐way ANOVA, followed by the Tukey’s test. The probability level interpreted as statistically significant was p < 0.05.
Results
Experiment 1: Effect of different LFS frequencies on perforant path kindling acquisition and synaptic excitability
At the first day of experiments there was no significant difference between the AD thresholds of different groups. In addition the average ADD after the first kindling stimulation did not show any significant difference between kindled and KLFS groups. It means that there was no significant difference in seizure susceptibility between these two groups. In addition, there was no significant difference between test‐pulse intensity of sham‐operated, kindled, KLFS, and LFS groups on the first day of stimulation. It means that there was no difference in synaptic sensitivity of different groups at the beginning of the experiments. Measuring the pEPSP slope and PS amplitude in sham‐operated or LFS groups showed no significant difference during different days (data not shown). During experiments, animals were stimulated by the same test pulse that was measured on the first day.
Effect of different LFS frequencies on perforant path kindling rate
LFS application prevented the increase in daily ADD (the summation of ADDs recorded after 12 daily stimulations) during kindling development. A two‐way ANOVA followed by Tukey’s post hoc test showed a significant difference in daily ADD between the kindled and KLFS groups on the sixth and seventh days of the stimulation (Fig. 2A). However, there was no significant difference between different KLFS groups. Similar to our previous reports (Mohammad‐Zadeh et al., 2007, 2009), LFS application also prevented the progress in behavioral seizure stages, so that after 7 days the animals of KLFS groups still showed no more than stage 3 seizure. Statistical analysis showed a significant decrease in seizure stage on the sixth and seventh days of stimulation (Fig. 2B). Again, there was no significant difference between three KLFS groups in achieving different seizure stages.
The effect of different low‐frequency stimulation (LFS) frequencies on daily afterdischarge duration (ADD) (A) and seizure stages (B) during kindling development. There is a significant decrease in both parameters on the sixth and seventh days of kindling procedure. Values are mean ± standard error of the mean (SEM) (n = 6). *p < 0.05, **p < 0.01, and ***p < 0.001 when compared to kindled group.
Effect of different LFS frequencies on field potential recording
During kindling acquisition kindling‐induced potentiation was detectable as an increase in pEPSP slope and PS amplitude in the kindled group. As Fig. 3 shows, there was 82.3 ± 1.3% increase in pEPSP and 85.01 ± 1.0 increase in PS amplitude of kindled group during the first 7 days of the kindling procedure. LFS application clearly suppressed the potentiation effects of kindling on basal synaptic transmission, as reported previously (Mohammad‐Zadeh et al., 2007, 2009). This suppressing effect of LFS was not frequency dependent (Fig. 3B and 3C). No significant difference was observed in field potential parameters during 7 days of kindling procedure when LFS applied at the frequencies of 0.5, 1 and 5 Hz (there was 20.9 ± 0.5, 17.9 ± 0.7, and 8.5 ± 0.9% increase in pEPSP and 8.5 ± 0.7, 5.6 ± 1.6 and 8.1 ± 0.7% increase in PS amplitude in KLFS 0.5Hz, KLFS 1 Hz, and KLFS 5 Hz, respectively, Fig. 3C).
(A) Electrographic example of wave‐forms taken on days 1 and 7 in kindled and kindled + low‐frequency stimulation (KLFS) groups. Kindling produced an increase in population spike (PS) amplitude and population excitatory postsynaptic potential (pEPSP) slope. Application of LFS prevented these changes in field potential recordings. (B) Time course diagrams showing the changes in PS amplitude and pEPSP slope in kindled and KLFS groups (n = 6) on the first, fourth, and seventh days of kindling acquisition. Each graph shows the baseline perforant path evoked responses recorded for 20 min before kindling stimulation. Application of LFS at the frequency of 0.5, 1, and 5 Hz inhibited kindling‐induced potentiation (revealed as increase in PS amplitude and pEPSP slope in kindled group). However, there was no significant difference between the effectiveness of different LFS frequencies. The number above each panel shows the time at which the electrographic samples of part A were recorded. (C) The percentage changes of PS amplitude and pEPSP slope on day 7 (relative to day 1) showing the preventing effect of different frequencies of LFS on increase of PS amplitude and pEPSP slope at the seventh day of kindling procedure. Values are mean ± standard error of the mean (SEM), ***p < 0.001 when compared to kindled group.
Effect of different frequencies of LFS on paired‐pulse measurements
Typical examples of paired‐pulse responses recorded from the dentate gyrus of kindled and KLFS 5‐Hz groups at interpulse intervals of 30 ms (in which early paired‐pulse depression occurs), 100 ms (in which paired‐pulse facilitation occurs), and 300 ms (in which late paired‐pulse depression occurs) have been shown in Fig. 4A. As the figure shows, LFS prevented the kindling‐induced changes in paired‐pulse responses.
(A) Examples of paired‐pulse responses recorded from the dentate gyrus of kindled and kindled + low‐frequency stimulation (KLFS) groups at three interpulse intervals (30, 100, and 300 ms). Kindling resulted in an increase in early and late paired‐pulse depression on day 7 of kindling procedure (kindled group). Application of LFS (in KLFS group) prevented the changes in paired‐pulse responses. (B) Plot of paired‐pulse index (second PS/first PS×100) values obtained for different interpulse intervals in kindled and KLFS groups (n = 6). LFS application prevented the increase in early and late paired‐pulse depression and also prevented the decrease in paired‐pulse facilitation. (C) The percentage changes of paired‐pulse index on day 7 (relative to day 1) at three selected interpulse intervals in kindled and KLFS groups. LFS prevented the kindling‐induced increase in early paired‐pulse depression (at 30 ms inter‐pulse interval) only at the frequency of 5 Hz. The inhibitory effect of LFS on kindling‐induced decrease of paired pulse facilitation (at 100 ms interpulse interval) was applied at the frequency of 1 and 5 Hz. All LFS frequencies had inhibitory effect on kindling‐induced increase in late paired‐pulse depression. There was a significant difference between the inhibitory effects of KLFS 0.5‐Hz and KLFS 5‐Hz groups. Values are mean ± standard error of the mean (SEM), *p < 0.05, **p < 0.01 and ***p < 0.001 when compared to kindled group and §p < 0.05 when compared the KLFS 0.5‐Hz and KLFS 5‐Hz groups.
Paired‐pulse index (based upon PS amplitude)/interpulse interval curves were measured in all groups. As it has been reported previously (de Jonge & Racine, 1987; Maru & Goddard, 1987; Gilbert, 1991; Mohammad‐Zadeh et al., 2007, 2009), kindling increased the early and late paired‐pulse depression (Fig. 4B). LFS application prevented potentiation in both early and late paired‐pulse depression during kindling development (Fig. 4B). In Fig. 4C, the percentage changes of paired‐pulse response on day 7 (relative to day 1), were compared in three selected interpulse intervals in kindled and KLFS groups. As the figure shows, there was a significant difference between the kindled and KLFS 5‐Hz groups at the early paired‐pulse depression (interpulse interval of 30 ms) and between the kindled group and all three KLFS groups at the late paired‐pulse depression (interpulse interval of 300 ms). In addition, kindling caused a significant decrease in paired‐pulse facilitation in kindled group during kindling acquisition. Application of LFS prevented the decrease in paired‐pulse facilitation during kindling acquisition only in KLFS 1‐Hz and KLFS 5‐Hz groups, but not in KLFS 0.5 Hz (Fig. 4B, C). As observed in Fig. 4C, there was a significant difference between KLFS 0.5‐Hz and KLFS 5‐Hz groups at the early paired‐pulse depression and between KLFS 0.5 and both KLFS 1 Hz and KLS 5 Hz at paired‐pulse facilitation.
Experiment 2: Effect of kindling and different LFS frequencies on A1AR and A2AAR gene expression
Right dentate gyrus was extracted from the animals used in experiment 1. As explained before, all animals were killed 24 h after the last stimulation. Semiquantitative PCR technique was used to estimate A1AR and A2AAR mRNA levels in tissue samples, normalized to an internal standard (β‐actin). Ethidium bromide staining of PCR products showed a single band of the predicted size of 618, 640, and 830 bp for A1AR, A2AAR, and β‐actin mRNA, respectively. Expression of A1AR was decreased in kindled rats compared to control group (p < 0.05). Application of LFS in KLFS groups significantly prevented this decrease, so that no difference was observed in A1AR mRNA level between KLFS and control groups. However, there was no significant difference between the three KLFS groups (Fig. 5). In addition, although there was no significant change in A2AAR mRNA level between kindled and control groups, A2AAR mRNA level was decreased significantly as a result of LFS application in all KLFS groups compared to the control and kindled groups. In addition, statistical analysis showed significant difference in A2AAR mRNA level between KLFS 0.5‐Hz and KLFS 5‐Hz groups (Fig. 5). Application of LFS alone (in LFS group) had no significant effect on adenosine receptor gene expression.
(A) Effect of different low‐frequency stimulation (LFS) frequencies on adenosine A1 receptor (A1) and adenosine A2A receptor (A2A) mRNA level in kindled and kindled + LFS (KLFS) groups. (B) The A1 (left) and A2A (right) gene expression has been shown as the percentage (%) of β‐actin band density. Kindling acquisition reduced A1 gene expression significantly but did not exert a significant effect on A2A mRNA level in the kindled group. Application of LFS prevented the reduction of A1 gene expression and obviously reduced A2A mRNA level in the KLFS groups. There was a significant difference between KLFS 0.5 HZ and 5 Hz in reduction of A2A gene expression. Values are mean ± standard error of the mean (SEM) (n = 6). *p < 0.05 and ***p < 0.001 when compared to control group; †p < 0.05 and †††p < 0.001 when compared to respective kindled group; §p < 0.05 when compared KLFS 0.5 Hz and KLFS 5 Hz.
Discussion
Current findings demonstrated that anticonvulsant effects of LFS during rapid perforant path kindling could be exerted at three different frequencies (1, 3, and 5 Hz). These effects were accompanied with an increase in adenosine A1 and a decrease in adenosine A2A receptor gene expression.
Previous studies showed that application of LFS at different frequencies [0.1–1 Hz (D’Arcangelo et al., 2005), 1–3 Hz (Gaito, 1980a; Gaito et al., 1980; Kulla et al., 1999), and 4–6 Hz (Gwinn & Spencer, 2004)] shortly after kindling stimulation is antiepileptogenic. Changes in frequency can change the mechanism of LFS action (Kemp et al., 2000). However, to our knowledge, no comparison was made between the anticonvulsant effectiveness of different LFS frequencies in previous studies, except for one study by Gaito (1980b), who showed that the interference of sine wave LFS with the production of kindled convulsions is decreased by increasing the LFS frequency. However, because there are a lot of differences between Gaito’s experiments and ours we cannot compare these two experiments completely.
The finding of kindling suppression effect by 5‐Hz stimulation in the present study is an interesting if considered that low‐frequency stimulation of 3 Hz was successfully used for kindling induction (e.g., Corcoran & Cain, 1980; Minabe et al., 1986; Emori et al., 1990). Of course, similar anticonvulsant effects were also observed with high‐frequency stimulations (Boon et al., 2007a,b; Van Roost et al., 2007).
Application of different LFS frequencies delayed kindling procedure and prevented kindling‐induced potentiation in synaptic transmission. These effects were not frequency dependent. Considering the more inhibitory effect of LFS on preventing the acquisition of stage 4 and 5 seizures (Velisek et al., 2002; Mohammad‐Zadeh et al., 2007; Sadegh et al., 2007), it may be postulated that some frequency dependency would be observed if the animals were stimulated more than 7 days (to achieve stage 4 and 5 seizure). In addition, LFS application prevented the kindling‐induced increase in early (in KLFS 5‐Hz group) and late (in KLFS 0.5‐, 1‐, and 5‐Hz groups) paired‐pulse depression. LFS also prevented the decrement in paired‐pulse facilitation. Here, there was a significant difference between the effectiveness of LFS at different frequencies, so that LFS at the frequencies of 1 and 5 Hz were more effective than at the frequency of 0.5 Hz. The possible mechanisms of LFS effects on paired‐pulse indices have been discussed elsewhere (Mohammad‐Zadeh et al., 2007, 2009).
Recently, we showed that activation of adenosine A1 but not A2A receptors may have a role in mediating the inhibitory effects of LFS on perforant path kindling rate (Mohammad‐Zadeh et al., 2009). The inhibitory effect of LFS on kindling acquisition was significantly reduced when the animals received an adenosine A1 receptor antagonist (Mohammad‐Zadeh et al., 2009). Results obtained in the present study also confirmed the role of adenosine receptors in LFS anticonvulsant effects.
Our RT‐PCR data showed a significant reduction in the A1AR mRNA level following stage 5 kindled seizures. This effect was prevented by application of LFS at the frequencies of 1 and 5 Hz, so that no difference was observed in this parameter between control and KLFS groups. Changes in adenosine A1 receptors have been investigated during short‐term (less than 24 h) and long‐term (after 48 h) periods after seizure induction. There is an increase in the density of adenosine A1 receptors during the short‐term period (Daval & Sarfati, 1987; Angelatou et al., 1990; Daval & Werck, 1991; Pagonopoulou et al., 1993; Vanore et al., 2001). Of course, some investigators showed no short‐term changes (Newman et al., 1984; Gleiter et al., 1989; Ekonomou et al., 2000) or even a reduction in the density of brain adenosine A1 receptors upon seizure induction (Ochiishi et al., 1999). In contrast, except for one study that reported a maintenance of the density and increased affinity of A1 receptors in the hippocampus of amygdala kindled rats (Simonato et al., 1994), most studies seem to find a long‐term decease in the density of adenosine A1 receptors (Glass et al., 1996; Ochiishi et al., 1999; Ekonomou et al., 2000; Rebola et al., 2003).
The reduction in adenosine A1 receptor gene expression observed 24 h after the last seizure in the present study is opposite to those evaluated in these receptors at short‐term periods after seizure induction. Several factors may be considered to be account for this discrepancy: (1) The electrically kindled seizures result in region‐specific changes in expression of A1 (and A2A) receptor gene expression (Aden et al., 2004). (2) The experimental model of seizure used in this study (rapid kindling) was different from those used in previous studies. (3) In most of the mentioned reports, changes in amount of A1 receptor density had been investigated. However, we measured the mRNA level (as an index of gene expression) of adenosine receptors.
Although our previous study showed no role for adenosine A2A receptors in antiepileptogenic effects of LFS (Mohammad‐Zadeh et al., 2009), in the present study application of LFS during kindling strongly reduced A2AAR mRNA level. It may explain why inhibition of adenosine A2A receptors in our previous study could not exert any effect on suppressing action of LFS during kindling acquisition. There was not any significant change in the A2AAR mRNA level following kindled seizure, which is consist with previous studies (Newman et al., 1984).The role of A2A receptors in seizures has been less well studied. There are reasons to believe that a reduction in A2A receptor activity may play a role in the control of seizure (El Yacoubi et al., 2001; Zeraati et al., 2006; Hosseinmardi et al., 2007). Meanwhile, the effect of LFS on long‐term depression induction is facilitated by A2A receptor inhibition (Fujii et al., 1992, 2000). Therefore, a reduction in A2AAR following LFS induction can improve the anticonvulsant effect of LFS.
The effect of LFS on adenosine receptor gene expression was changed following changes in its frequency. Again, LFS was more effective at the frequency of 5 Hz. Of course, one must keep in mind that in all of our experiments the duration of LFS application was equal. Consequently, when LFS was applied at the frequency of 5Hz, the number of pulses was more than when applied at the frequency of 1 or 0.5 Hz.
In conclusion, the observed preventing effect of LFS on post seizure A1 receptor reduction, and the parallel decreasing effect of LFS on A2A receptors provide a preliminary rationale for the development of novel anticonvulsive strategies based on the use of the LFS together with the correct manipulation of the adenosine neuromodulatory system. It has to be emphasized that the observed effects of LFS were obtained during kindling acquisition and it remains to be determined whether similar changes would be observed in fully kindled animals.
Acknowledgement
This study was supported by a grant from Tarbiat Modares University. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
None of the authors has any conflict of interest to disclose.
Number of times cited: 11
- Simin Namvar, Yaghoub Fathollahi, Mohammad Javan, Maryam Zeraati, Mohammad Mohammad-Zadeh, Amir Shojaei and Javad Mirnajafi-Zadeh, The antiepileptogenic effect of low-frequency stimulation on perforant path kindling involves changes in regulators of G-protein signaling in rat, Journal of the Neurological Sciences, 375, (450), (2017).
- Khadijeh Esmaeilpour, Vahid Sheibani, Mohammad Shabani and Javad Mirnajafi-Zadeh, Low frequency electrical stimulation has time dependent improving effect on kindling-induced impairment in long-term potentiation in rats, Brain Research, 1668, (20), (2017).
- Khadijeh Esmaeilpour, Vahid Sheibani, Mohammad Shabani and Javad Mirnajafi-Zadeh, Effect of low frequency electrical stimulation on seizure-induced short- and long-term impairments in learning and memory in rats, Physiology & Behavior, 168, (112), (2017).
- Samireh Ghafouri, Yaghoub Fathollahi, Mohammad Javan, Amir Shojaei, Azam Asgari and Javad Mirnajafi-Zadeh, Effect of low frequency stimulation on impaired spontaneous alternation behavior of kindled rats in Y-maze test, Epilepsy Research, 126, (37), (2016).
- Razieh Rohani, Abbas Piryaei, Ali Jahanshahi, Yousef Sadeghi and Javad Mirnajafi-Zadeh, Effect of low-frequency stimulation on kindling induced changes in rat dentate gyrus: an ultrastructural study, Acta Neurologica Belgica, 114, 1, (47), (2014).
- Nealen G. Laxpati, Willard S. Kasoff and Robert E. Gross, Deep Brain Stimulation for the Treatment of Epilepsy: Circuits, Targets, and Trials, Neurotherapeutics, 11, 3, (508), (2014).
- Adi Cymerblit-Sabba, Maya Schiller and Yitzhak Schiller, Termination of Chemoconvulsant-Induced Seizures by Synchronous and Asynchronous Electrical Stimulation of the Hippocampus In-Vivo, Brain Stimulation, 6, 5, (727), (2013).
- Zhenghao Xu, Yi Wang, Miaomiao Jin, Jiaxing Yue, Cenglin Xu, Xiaoying Ying, Dengchang Wu, Shihong Zhang and Zhong Chen, Polarity-dependent effect of low-frequency stimulation on amygdaloid kindling in rats, Brain Stimulation, 6, 2, (190), (2013).
- Mahnaz Davoudi, Amir Shojaei, Mohammad Reza Palizvan, Mohammad Javan and Javad Mirnajafi-Zadeh, Comparison between standard protocol and a novel window protocol for induction of pentylenetetrazol kindled seizures in the rat, Epilepsy Research, 106, 1-2, (54), (2013).
- Marzieh Shahpari, Javad Mirnajafi-Zadeh, Seyed Mohammad P. Firoozabadi and Ali Yadollahpour, Effect of low-frequency electrical stimulation parameters on its anticonvulsant action during rapid perforant path kindling in rat, Epilepsy Research, 99, 1-2, (69), (2012).
- Geoffrey Burnstock, Ute Krügel, Maria P. Abbracchio and Peter Illes, Purinergic signalling: From normal behaviour to pathological brain function, Progress in Neurobiology, 10.1016/j.pneurobio.2011.08.006, 95, 2, (229-274), (2011).
