Amiloride and SN‐6 Suppress Audiogenic Seizure Susceptibility in Genetically Epilepsy‐Prone Rats

Summary

Aims

We have recently reported that amiloride, a potent and nonselective blocker of acid‐sensing ion channels, prevents the development of pilocarpine‐induced seizures and status epilepticus. Amiloride is also known to suppress the activity of Na⁺/Ca²⁺ and Na⁺/H⁺ exchangers that have been implicated in the pathophysiology of seizures. Here, we evaluated the effects of amiloride, SN‐6 (a potent blocker of Na⁺/Ca²⁺ exchangers) and zoniporide (a potent blocker of Na⁺/H⁺ exchangers) on acoustically evoked seizures (audiogenic seizures, AGS) in genetically epilepsy‐prone rats (GEPR‐3s), a model of inherited generalized epilepsy.

Methods

Male, six‐week‐old GEPR‐3s were used. The GEPR‐3s were tested for AGS susceptibility before and after treatment with various doses of amiloride, SN‐6, and zoniporide (1, 3, 10, and 30 mg/kg; per os).

Results

We found that pretreatment with amiloride and SN‐6 markedly reduced the incidence and severity of AGS in the GEPR‐3s. In contrast, administration of zoniporide only minimally reduced the incidence and severity of AGS in the GEPR‐3s. A combination of noneffective doses of SN‐6 and zoniporide also suppressed AGS susceptibility in the GEPR‐3s.

Conclusions

These findings suggest acid‐sensing ion channels and the Na⁺/Ca²⁺ exchanger may play an important role in the pathophysiology of inherited AGS susceptibility in the GEPR‐3s.

Introduction

Acidosis is a significant decrease in extracellular pH, and it plays a major role in the pathophysiology of many neuronal disorders, such as ischemic brain injury and seizures 1, 2, 3. The fundamental mechanism of action of acidosis lies in its ability to activate acid‐sensing (or proton‐gated) ion channels (ASICs) that belong to the degenerin/epithelial Na⁺ channel superfamily 1. Upon activation, these proton‐gated channels allow for a rapid influx of Na⁺ and Ca²⁺ into neurons; furthermore, a disturbance in this Na⁺ and Ca²⁺ homeostasis may contribute to cell injury and cell death, as well as a neuronal hyperexcitability that can lead to seizures 2, 3. Evidence indicates that both acute extracellular and intracellular acidification can suppress seizure activity 4, 5. Alternatively, chronic extracellular and/or intracellular acidification may play a role in the generation of seizures. Consistent with this hypothesis, amiloride, a potent blocker of ASICs, has proven to be effective in suppressing acute seizures induced by pentylenetetrazole, electroshock, and pilocarpine 6, 7. Amiloride was also reported to delay the development of pentylenetetrazole kindling 8. In the pilocarpine model of acute seizures, the mRNA expression of the ASIC2b and ASIC1a subunits were decreased in the hippocampus 24 h after the onset of seizures 9. These findings suggest the involvement of ASICs in the mechanism underlying pilocarpine‐induced seizures. Recent studies demonstrated that amiloride reduced the incidence of acoustically evoked seizures (audiogenic seizures, AGS) in rats with cardiac arrest‐induced global cerebral hypoxia 10. The role of ASICs in inherited seizure susceptibility in genetically epilepsy‐prone rats (GEPR‐3s) remains unknown. In addition to ASICs, amiloride is also known to inhibit membrane exchangers, including the Na⁺/H⁺ and Na⁺/Ca²⁺ exchangers, and various ligand‐ and voltage‐gated ion channels 11, 12. It is unclear as to whether the potential anticonvulsant effect of amiloride is attributable to its effect on the Na⁺/H⁺ and/or Na⁺/Ca²⁺ exchanger. Here, we report the effects of blocking ASICs, Na⁺/Ca²⁺ exchangers and Na⁺/H⁺ exchangers with amiloride, SN‐6, and zoniporide, respectively, on AGS susceptibility in the validated GEPR‐3 model of inherited generalized epilepsy.

Methods

Six‐week‐old, male GEPR‐3s were used. These rats were obtained from our animal colony that is maintained at Georgetown University Medical Center. The GEPR‐3s were maintained in a temperature/humidity‐controlled room on a 12 h/12 h light/dark cycle with free access to food and water. All possible efforts were made to minimize the number of animals used in experiments and their discomfort. All experimental procedures were approved by the Georgetown University Animal Care and Use Committee. We used amiloride (3,5‐Diamino‐N‐(aminoiminomethyl)‐6‐chloropyrazinecarboxamide hydrochloride, Tocris Bioscience, Ellisville, MO, USA) to evaluate the role of ASICs in AGS susceptibility in the GEPR‐3s. To assess the role of the Na⁺/Ca²⁺ exchanger, we antagonized its reverse mode activity by using SN‐6 (2‐[[4‐[(4‐nitrophenyl)methoxy]phenyl]methyl]‐4‐thiazoli dinecarboxylic acid ethyl ester, Tocris Bioscience) 13, 14. The role of the Na⁺/H⁺ exchanger in AGS susceptibility was evaluated using zoniporide ([1‐(Quinolin‐5‐yl)‐5‐cyclopropyl‐1H‐pyrazole‐4‐carbonyl]guanidine dihydrochloride, Tocris Bioscience) 15. For each pharmacological agent, the GEPR‐3s were randomly separated into groups of n = 8 and were used as their own controls. The GEPR‐3s were first tested for AGS susceptibility 90 min after administration of vehicle (0.3 mL of sterile water). Those GEPR‐3s exhibiting seizures were subsequently used for pharmacological studies 1 h later and referred as controls. Amiloride (1, 3, 10, and 30 mg/kg), SN‐6 (1, 3, 10, and 30 mg/kg) and zoniporide (1, 3, 10, and 30 mg/kg) were dissolved in sterile water using sonication (80 kHz, 100% power), filtered, and administered 90 min before seizure testing. Amiloride, SN‐6, and zoniporide were given per os (p.o.) by gastric intubation with a volume of 0.2 mL/100 g body weight using an 18‐gauge stainless steel feeding needle with a round tip (ball diameter 3 mm). An initial dose of 1 mg/kg was chosen based on published in vivo pharmacological studies and preliminary data 7, 16, 17, 18. Based on our previous study, a 90 min timeframe is the most effective pretreatment window against seizure activity 7, 16, 17. Following administration of amiloride, SN‐6 or zoniporide, the GEPR‐3s were placed in an acoustic chamber (Med Associates, ST Albans, VT) and tested for AGS susceptibility. To evaluate the long‐lasting effect of amiloride, SN‐6, and zoniporide, the GEPR‐3s were again tested for AGS susceptibility 1 and 24 h after the first seizure testing. To induce AGS, an acoustic stimulus that consisted of pure tones at a 100–105 decibels sound pressure level (Med Associated, St Albans, VT) was presented until either seizures were elicited or 60 seconds passed with no seizure activity. The GEPR‐3s were closely monitored following amiloride, SN‐6 or zoniporide administration as well as during and following seizure testing. Convulsive seizure behavior was classified into stages: stage 0, no seizures in response to acoustic stimulus; stage 1, wild running seizures (WRS); stage 2, two or more episodes of WRS; stage 3, WRS followed by tonic‐clonic seizures characterized by tonic dorsiflexion of the neck, tonic flexion of shoulder, and bouncing clonic seizures (or clonus, i.e., tonic‐clonic seizures while the animal is lying on its belly) 19. Wild running seizures and clonus correspond to a preconvulsive phase and convulsive phase of AGS, respectively. In another set of experiments, GEPR‐3s (n = 4) were subjected to an assessment of general behavior (up to 24 h) following administration of amiloride, SN‐6, and zoniporide. The occurrence of the following abnormalities was recorded including: lethargy, ataxia, tremor, Straub's tail, and spontaneous seizures. At the end of the experiment, animals were euthanized with a lethal dose of Nembutal (100 mg/kg, i.p.).

Statistical Analysis

Following pharmacological pretreatment and seizure testing, GEPR‐3s that did not display seizures within the 60‐seconds observation period were considered to be protected from seizure activity. Only data obtained in control conditions and on the first seizure testing following administration of amiloride, SN‐6 or zoniporide was included in the analysis. For each group, the incidence of WRS and clonus components of AGS in the GEPR‐3s were recorded. The time interval from the start of acoustic stimulus to the appearance of the first episode of WRS was recorded as the seizure latency. The incidence of WRS and clonus was analyzed using the McNemar's test; this test compares categorical data when subjects serve as their own controls. The seizure severity was analyzed using the Wilcoxon signed‐rank test; this test is used ordinal data in subjects before and after treatment. The seizure latency was analyzed using a paired t‐test. The cutoff for statistical significance was P < 0.05 and no multiple comparisons and adjustment were performed. Data are presented as the mean ± SEM for seizure latency, median seizure score ± SEM for seizure severity and percentages (%) for the incidence of WRS and clonus.

Results

Administration of amiloride, SN‐6, and zoniporide at the tested doses did not induce abnormal behaviors (i.e., reduced exploratory behavior, lethargy, ataxia, Straub's tail, and spontaneous seizures) of the GEPR‐3s. All GEPR‐3s (n = 8) tested under control conditions exhibited WRS that progressed into clonus. Pretreatment with amiloride at 1 mg/kg significantly reduced the incidence of WRS and clonus by 37.5% and 50%, respectively, when compared with controls (P < 0.001, Figure 1A,B). In addition, amiloride significantly increased the seizure latency to 41 ± 6 seconds (n = 8, t = −3.2, P < 0.01, Figure 1C) when compared with controls (28 ± 4 seconds, n = 8). The seizure severity was nonsignificantly reduced to 2 (n = 8) following amiloride pretreatment when compared with controls (3, n = 8, Figure 1D). At 3 mg/kg, the incidence of WRS and clonus was significantly reduced by 62.5% and 75%, respectively, when compared with controls (P < 0.001, Figure 1A,B). Amiloride significantly delayed the onset of seizures to 46 ± 7 seconds (n = 8, t = −4, P < 0.01, Figure 1C) when compared with controls (25 ± 2 seconds, n = 8). The seizure severity was significantly reduced to 0 (n = 8) when compared with a control group score of 3 (n = 8, P < 0.05, Figure 1D). The time course evaluation of the anticonvulsant effect revealed that AGS susceptibility returned to control levels 24 h later (Figure 2A). At a dose of 10 mg/kg, the incidence of WRS and clonus was significantly reduced by 25% and 62.5%, respectively, when compared with controls (Figure 1A,B). Amiloride significantly increased the seizure latency to 32 ± 7 seconds (n = 8, t = −2.5, P < 0.05, Figure 1C) when compared with controls (22 ± 3 seconds, n = 8). The seizure severity was nonsignificantly reduced to 1 (n = 8) when compared with a control group score of 3 (n = 8, Figure 1D). At a dose of 30 mg/kg, the incidence of WRS and clonus was significantly (P < 0.001) reduced by 25% and 62.5%, respectively, when compared with controls (n = 8, Figure 1A,B). Amiloride significantly delayed the onset of seizures (40 ± 7 seconds, n = 8, t = −2.9, P < 0.01, Figure 1C) when compared with controls (22 ± 2 seconds). The seizure severity was nonsignificantly reduced to 1 (n = 8) when compared with controls (3, n = 8, Figure 1D).

Figure 1.

Amiloride suppresses AGS susceptibility in the GEPR‐3s. The effects of various doses of amiloride were evaluated on the incidence and severity of AGS in the GEPR‐3s. Pretreatment with amiloride at doses of 1, 3, 10, and 30 mg/kg (p.o.) markedly reduced the incidence of WRS (A) and clonus (B) components of AGS in the GEPR‐3s. At a dose of 3 mg/kg (p.o.), amiloride completely suppressed AGS susceptibility in the GEPR‐3s (C). Amiloride delayed the onset of seizures in the GEPR‐3s (D). Data represent the mean ± SEM. for seizure latency and seizure severity, and percentages for the incidences of WRS and clonus. McNemar's test was used to compare seizure incidence, whereas paired t‐test and Wilcoxon signed‐rank‐test were used for seizure latency and seizure severity analysis, respectively. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2.

Time course of the effects of amiloride (3 mg/kg; A), SN‐ (10 mg/kg, B), zoniporide (3 mg/kg, C), and a combination of noneffective dose of SN‐6 and zoniporide (3 mg/kg, D). Complete seizure blockade was observed following amiloride pretreatment, whereas the seizure severity was markedly reduced by SN‐6, zoniporide and a combination of noneffective dose of SN‐6 and zoniporide. Note that AGS susceptibility in the GEPR‐3s returned to control levels 24 h later.

Pretreatment of SN‐6 at 1 mg/kg significantly reduced the incidence of WRS and clonus by 25% and 37.5%, respectively, when compared with controls (P < 0.001, Figure 3A,B). No significant change was found in the seizure latency (SN‐6: 33 ± 6 seconds, n = 8, Figure 3C) when compared with controls (24 ± 2 seconds, n = 8). SN‐6 did not alter the seizure severity (Figure 3D). At 3 mg/kg, the incidence of WRS and clonus was significantly reduced by 37.5% and 50%, respectively, when compared with controls (P < 0.001, Figure 1A,B). SN‐6 significantly delayed the onset of seizures to 35 ± 7 seconds (n = 8, t = −2.3, P < 0.05, Figure 3C) when compared with controls (24 ± 5 seconds, n = 8). The seizure severity was nonsignificantly reduced to 2 (n = 8) when compared with controls (3, n = 8, Figure 3D). At a dose of 10 mg/kg, the incidence of WRS and clonus was significantly reduced by 50% and 75%, respectively, when compared with controls (P < 0.001, Figure 3A,B). SN‐6 significantly increased the seizure latency to 43 ± 6 seconds (n = 8, t = −4.0, P < 0.01, Figure 3C) compared to controls (22 ± 3 seconds, n = 8). The seizure severity was significantly reduced to 0.5 (n = 8) when compared with controls (3, n = 8, Figure 3D), but returned to control levels at 24 h (Figure 2B). At a dose of 30 mg/kg, the incidence of WRS and clonus was significantly (P < 0.001) reduced by 25% and 62.5%, respectively, when compared with controls (n = 8, Figure 3A,B). SN‐6 significantly delayed to onset of seizures (47 ± 7 seconds, n = 8, t = −2.3, P < 0.01, Figure 3C) when compared with controls (28 ± 3 seconds). The seizure severity was nonsignificantly reduced to 1 (n = 8) when compared with controls (3, n = 8, Figure 3D).

Figure 3.

SN‐6 alters the expression of AGS in the GEPR‐3s. The effects of various doses of SN‐6 were evaluated based on the incidence and severity of AGS in the GEPR‐3s. SN‐6 pretreatment (1, 3, 10 and 30 mg/kg; p.o.) significantly reduced the incidence of WRS (A) and clonus (B) in the GEPR‐3s. At a dose of 10 mg/kg (p.o.), SN‐6 significantly reduced AGS severity in the GEPR‐3s (C). SN‐6 pretreatment delayed the seizure onset in the GEPR‐3s. Data represent the mean ± SEM. for seizure latency and seizure severity, and percentages for the incidences of WRS and clonus. McNemar's test was used to compare seizure incidence, whereas paired t‐test and Wilcoxon signed‐rank‐test were used for seizure latency and seizure severity analysis, respectively. *P < 0.05, **P < 0.01, ***P < 0.001.

Pretreatment of zoniporide at a dose of 1 mg/kg did not affect the incidence of WRS (Figure 4A) and nonsignificantly reduced the incidence of clonus when compared with controls (Figure 4B). No significant change was found in the seizure latency (controls: 27 ± 1 seconds, n = 8, zoniporide: 29 ± 4 seconds, n = 8, Figure 4C). Zoniporide nonsignificantly reduced the seizure severity to 2 when compared with a control group score of 3 (n = 8, Figure 4D). At a dose of 3 mg/kg, zoniporide significantly reduced the incidence of both WRS and clonus by 25% and 62.5%, respectively, when compared with controls (P < 0.001, Figure 4A,B). Zoniporide significantly delayed the onset of seizures to 42 ± 4 seconds (n = 8, t = −2.4, P < 0.05, Figure 4C) when compared with controls (28 ± 1 seconds, n = 8). The seizure severity was nonsignificantly reduced to 1 (n = 8) when compared with controls (3, n = 8, Figure 4D), but return to control levels 24 h later (Figure 2C). At a dose of 10 mg/kg, zoniporide significantly (P < 0.001) reduced the incidence of WRS and clonus by 12.5% and 25%, respectively, when compared with controls (P < 0.001, Figure 3A,B). No significant change was found in the seizure latency following zoniporide administration (31 ± 4 seconds, n = 8) when compared with controls (22 ± 2 seconds, n = 8, t = −2.2, P < 0.06, Figure 4C). The seizure severity was not altered following zoniporide administration (Figure 4D). At a dose of 30 mg/kg, zoniporide significantly (P < 0.001) reduced the incidence of both WRS and clonus by 25% when compared with controls(P < 0.001, Figure 4A,B), but it failed to alter the seizure latency (zoniporide: 35 ± 6 seconds, n = 8, controls: 22 ± 2 seconds, n = 8, t = −2, P < 0.08, Figure 4C). The seizure severity was nonsignificantly reduced to 1 (n = 8) when compared with controls (3, n = 8, Figure 4D).

Figure 4.

The effects of zoniporide on AGS susceptibility in the GEPR‐3s. Pretreatment with zoniporide at doses of 3, 10, and 30 mg/kg (p.o.) mildly reduced the incidence of WRS (A) and clonus (B) components of AGS in the GEPR‐3s. Zoniporide at a dose of 1 and 3 mg/kg (p.o.) did not significantly reduce AGS severity in the GEPR‐3s (C). Zoniporide at the dose of 3 mg/kg delayed the onset of seizures (D). Data represent the mean ± SEM for seizure latency and seizure severity, and percentages for the incidences of WRS and clonus. McNemar's test was used to compare seizure incidence, whereas paired t‐test and Wilcoxon signed‐rank‐test were used for seizure latency and seizure severity analysis, respectively. *P < 0.05, ***P < 0.001.

We also evaluate the extent to which a combination of noneffective doses of SN‐6 and zoniporide affects the occurrence of AGS susceptibility in the GEPR‐3s. Coadministration of SN‐6 (1 mg/kg; p.o.) and zoniporide (1 mg/kg; p.o.) reduced the incidence of WRS and clonus by 12.5% and 50%, respectively, when compared to controls (P < 0.001, Figure 5A,B). This treatment nonsignificantly increased the seizure latency and reduced the seizure severity (Figure 5C,D). At a dose of 3 mg/kg (p.o.), coadministration of SN‐6 and zoniporide suppressed the incidence of WRS and clonus by 50% and 87.5%, respectively, when compared to controls (Figure 5A,B). The reduced AGS incidence was accompanied by significant delay of the seizure onset and reduction in seizure severity (P < 0.01, Figure 5C,D). The reduced AGS susceptibility returns to control levels 24 h later (Figure 2D).

Figure 5.

Coadministration of noneffective dose of SN‐6 and zoniporide suppresses AGS susceptibility in the GEPR‐3s. SN‐6 and zoniporide at the dose of 3 mg/kg markedly reduced the incidence of WRS (A) and clonus (B) component of AGS, delayed the seizure onset (C), and reduced the seizure severity (D). Data represent the mean ± SEM for seizure latency and seizure severity, and percentages for the incidences of WRS and clonus. McNemar's test was used to compare seizure incidence, whereas paired t‐test and Wilcoxon signed‐rank‐test were used for seizure latency and seizure severity analysis, respectively.**P < 0.01, ***P < 0.001.

Discussion

This study demonstrates that amiloride and SN‐6 pretreatment preferentially suppressed the occurrence of the clonus component of AGS in the GEPR‐3s. In contrast, zoniporide administration was relatively ineffective against AGS in the GEPR‐3s. A combination of noneffective doses of SN‐6 and zoniporide also suppressed clonus component of AGS in the GEPR‐3s. Amiloride, SN‐6, and zoniporide are known to preferentially inhibit ASICs, Na⁺/Ca²⁺ exchangers and Na⁺/H⁺ exchangers, respectively, suggesting that ASICs and the Na⁺/Ca²⁺ exchanger, but not the Na⁺/H⁺ exchanger, may play an important role in the pathogenesis of inherited AGS susceptibility in the GEPR‐3s. Amiloride is also known to inhibit the activity of the Na⁺/Ca²⁺ and Na⁺/H⁺ exchangers 11, 12. Thus, it is tempting to suggest that amiloride may exert its anticonvulsant action via inhibition of the Na⁺/Ca²⁺ exchanger in the GEPR‐3 model.

Multiple lines of evidence indicate that amiloride has both antiepileptogenic and anticonvulsant effects. In particular, amiloride pretreatment has been shown to suppress generalized seizures induced by maximal electroshock, pentylenetetrazole, and pilocarpine 6, 7. Similarly, we now report that amiloride suppresses AGS susceptibility in the GEPR‐3s. The mechanisms underlying the potential anticonvulsant effects of amiloride are not yet fully understood. Amiloride is known to cause a biphasic alteration of neuronal activity that consists of an initial excitation followed by inhibition 20, 21. The amiloride‐induced inhibition of neuronal activity is consistent with its anticonvulsant effect. Amiloride also inhibits various ligand‐gated channels, voltage‐gated ion channels, and membrane transporters 22. Of interest, amiloride inhibits channels that have been implicated in the pathophysiology of seizures, including ASICs, voltage‐gated Na⁺ channels, and low threshold (T‐type) voltage‐gated Ca²⁺ channels. Evidence suggests that ethosuximide, a clinically used anticonvulsant and potent blocker of T‐type Ca²⁺ channels, suppresses AGS susceptibility in the GEPR‐3s 23, 24. Similarly, phenytoin and carbamazepine are clinically used anticonvulsants thought to block Na⁺ channels, and they have been suggested to reduce AGS susceptibility in the GEPR‐3s 24, 25. Unlike the action of amiloride, the anticonvulsant effects of ethosuximide, phenytoin, and carbamazepine were dose‐dependent in the GEPR‐3s. Thus, the anticonvulsant effect of amiloride in the GEPR‐3s may not result from the blockade of voltage‐gated Na⁺ and/or T‐type Ca²⁺ channels. Amiloride is thought to exert its anticonvulsant effect via blockade of ASICs 6, 7. However, Ziemann et al. 26 reported that disrupting ASIC1a subtype gene or pharmacologically inhibiting ASIC1a subtype increased seizure severity in the kainate and pentylenetetrazole model of acute seizures, whereas overexpressing ASIC1a had the opposite effect. One potential mechanism underlying seizure suppression following activation of ASICs would be their activation of inhibitory interneurons 26. Interestingly, inhibitory interneurons had large proton‐gated currents than excitatory brain neurons 26, 27. Quantitation shows that amiloride was the most effective at reducing the incidence and severity of AGS at a dose of 3 mg/kg. A dose‐dependent effect was found with lower doses (1–3 mg/kg), but higher doses were associated with less efficacious anticonvulsant properties. These findings suggest that the anticonvulsant effect of amiloride in the GEPR‐3 model is complex and cannot be explained by a conventional dose‐relationship. A potential mechanism that may explain the anticonvulsant effect of amiloride lies in its action on membrane transporters. One molecular target of interest is the Na⁺/Ca²⁺ exchanger, an electrogenic membrane transporter that allows Ca²⁺ influx into the cells under a reverse mode of activity. Thus, inhibition of the reverse mode activity of the Na⁺/Ca²⁺ exchanger would prevent massive cell Ca²⁺ overload and the resulting neuronal hyperexcitability that may lead to seizures. The Na⁺/Ca²⁺ exchanger has been implicated in the pathophysiology of acute generalized tonic‐clonic seizures and inherited generalized tonic‐clonic epilepsy 16, 17, 28, 29. In normal physiological conditions, the Na⁺/Ca²⁺ exchanger couples the export of one Ca²⁺ with the import of three Na⁺, which is known as the Ca²⁺ exit mode or forward mode of Na⁺/Ca²⁺ activity 30, 31. However, the Na⁺/Ca²⁺ operates in the reverse mode of activity known as Ca²⁺ entry mode in pathophysiological conditions. The electrogenic effect of the reverse mode activity of the Na⁺/Ca²⁺ exchanger is thought to contribute to the anticonvulsant effects of blockers of this exchanger 16. We have recently reported that blockade of the Na⁺/Ca²⁺ exchanger suppresses pentylenetrazole‐induced seizures; moreover, this effect was not dose‐dependent 16. Similarly, in the present study, we found that blockade of the reverse mode activity of Na⁺/Ca²⁺ exchangers suppressed AGS in the GEPR‐3s in a manner that was not dose‐dependent. Interestingly, the anticonvulsant effect of SN‐6 is reminiscent of those observed for amiloride in the GEPR‐3s. Thus, it is tempting to suggest that amiloride may suppress AGS via inhibition of the Na⁺/Ca²⁺ exchanger in the GEPR‐3s. Evidence indicates that amiloride can inhibit both the forward mode and the reverse mode activity of Na⁺/Ca²⁺ exchangers 32. Three isoforms of the Na⁺/Ca²⁺ exchanger (i.e., NCX1, NCX2, and NCX3) have been identified 33. SN‐6 predominantly inhibits NCX1 suggesting that this isoform of the Na⁺/Ca²⁺ exchanger may be a potential molecular target for seizure suppression in the GEPR‐3s.

The Na⁺/H⁺ exchanger is also a molecular target of amiloride 20, 21. Activation of the Na⁺/H⁺ exchanger results in a transport of protons into the extracellular space via a 1:1 exchange for Na⁺. This transporter plays an important role in the regulation of pH homeostasis and affects intracellular Na⁺ levels, which in turn can alter the function of various ion channels and neuronal excitability. The Na⁺/H⁺ exchanger has been implicated in the pathogenesis of seizures because mice lacking isoform 1 of the Na⁺/H⁺ exchanger exhibit an absence slow‐wave epilepsy and generalized tonic‐clonic seizures 34, 35. Evidence indicates that amiloride blocks subtypes 1, 2, and 4 of the Na⁺/H⁺ exchanger, all of which are abundant in the brain 36, 37. Pharmacological studies suggest that Na⁺/H⁺ inhibition mediates the anticonvulsant and antiepileptogenic effects of amiloride in acute models of seizures and a model of kindling epileptogenesis 6, 8. However, the effects of a selective blocker of the Na⁺/H⁺ exchanger, such as zoniporide, have not yet been tested in these models of acute generalized and epileptogenesis. In the present study, we found that selective blockade of the Na⁺/H⁺ exchanger by zoniporide reduced the incidence of AGS in the GEPR‐3s; in addition, this effect was associated with a nonsignificant reduction in seizure severity. Similarly, zoniporide affected AGS expression in rats with cardiac arrest‐induced global cerebral hypoxia 10. These findings suggest that the Na⁺/H⁺ exchanger may not play an important role in the pathophysiology of acoustically evoked generalized seizures.

In this study, we found that coadministration of noneffective doses of both SN‐6, and zoniporide suppresses AGS susceptibility in the GEPR‐3s. SN‐6 and zoniporide suppress Na⁺/Ca²⁺ exchanger and Na⁺/H⁺ exchanger, respectively. The Na⁺/H⁺ exchanger contributes in the rise of intracellular Na⁺, while the levels of intracellular Ca²⁺ is closely regulated by the Na⁺/Ca²⁺ exchanger, and Ca²⁺ efflux is dependent on the levels of intracellular Na⁺. Both elevated Na⁺ and Ca²⁺ influxes have been implicated in the pathophysiology of seizures. Thus, inhibition of both Na⁺/H⁺ and Na⁺/Ca²⁺ exchanger prevent for Na⁺ and Ca²⁺ influx has anticonvulsant potential.

In this study, a relatively moderate to small sample size was used. The lack of study power as function of a small sample size is usually associated with type II error (false negatives; i.e., lack of zoniporide's anticonvulsant effect) rather than type I error (false positives). Thus, it is possible that the small sample size could lead to type II statistical error in this study.

Here, we report that blockade of the reverse mode activity of the Na⁺/Ca²⁺ exchanger allows for Ca²⁺ influx and results in anticonvulsant effects in the GEPR‐3 model. Understanding how the Na⁺/Ca²⁺ exchanger contribute to the neuronal hyperexcitability that leads to seizures may provide new insights into the pathophysiology of seizures and will help in the development of therapeutic strategies for the treatment of generalized seizures.

Conflict of Interest

The authors declare no conflict of interest.