2016 年 39 巻 2 号 p. 174-180
Many anti-epileptic drugs (AEDs) that mainly target ion channels or post-synaptic receptors are in clinical use, but a proportion of patients are resistant to these traditional AEDs and experience repeated severe break-out seizures. Given its involvement in the etiology of epilepsy, the neurotransmitter glycine may serve as a novel target for epilepsy treatment. Increasing evidence suggests that inhibitors of glycine transporter 1 (GlyT1) exhibit anti-seizure properties in mouse models and show potential as anti-convulsions drugs. In the present study, we investigated the effect of a highly selective GlyT1 inhibitor (named M22) on glycine transport kinetics using a radioactive substrate uptake assay and investigated the anti-seizure effects of M22 on the maximal electroshock seizure threshold (MEST) test and the timed intravenous (i.v.) pentylenetetrazole (PTZ) intravenous test. Our results demonstrate that M22 was capable of elevating the seizure threshold in the MEST test but did not alter the seizure threshold in the PTZ i.v. test. Strychnine, an inhibitor of glycine receptor activity, reversed the threshold elevation at a subconvulsive dosage (0.1 mg/kg subcutaneously) in the MEST test and did not affect M22 plasma levels in mice, suggesting that the anti-seizure effect in this model may be mediated by increased glycine receptor activity. Moreover, M22 administration did not influence motor function and coordination in mice. In combination with the previously reported excellent pharmacokinetic features of M22, our present results suggested that M22 has the potential to serve as a new anti-convulsive drug or as a lead compound for the development of AEDs.
Epilepsy is a common neurological disease that occurs in 3% of the population. Regarding the pathogenesis and treatment of this disease, an imbalance of excitable and inhibitory neurotransmitters remains an attractive hypothesis.1,2) Various anti-epileptic drugs (AEDs) that restore the imbalance between the glutamate and gamma-aminobutyric acid (GABA) systems, such as diazepam and vigabatrin, have been used in clinical practice.3–5) Despite this progress, however, a considerable number of epileptic patients still exhibit resistance to these types of therapies. Moreover, cognitive dysfunction and hyperactivity disorder are frequently observed in epileptic patients, which may be related to treatment with AEDs.6) However, current AEDs do little to improve cognitive or affective disorders, and some of these agents even aggravate the above mental symptoms.7,8) Thus, new types of AEDs with fewer side-effects need to be explored.
Glycine acts as an inhibitory neurotransmitter when binding to the strychnine-sensitive site/glycine receptor.9) However, upon binding to the strychnine-insensitive site in the N-methyl-D-aspartate (NMDA) receptor complex, it also co-agonizes the NMDA receptor and participates in glutamate-mediated neurotransmission. Glycine transporters (GlyTs) catalyze the re-uptake of excess glycine into the nerve terminals or the surrounding glia, thus terminating the signal.10) GlyTs are divided into two groups based on their reaction to N-methyl glycine (sarcosine) : GlyT1 proteins are inhibited by sarcosine, whereas GlyT2 proteins are not.11) Furthermore, GlyT1 can co-localize with NMDA receptors and modulate their function.12–14) Additionally, GlyT1 has been reported to be a target for the treatment of schizophrenia.10,15–17) Despite the fact that some GlyT1 inhibitors activate the NMDA receptor, potentially increasing the risk of seizure, a large number of studies indicate that GlyT1 inhibitors can be used as anticonvulsants for a variety of seizures.18,19) Kalinichev et al. indicated that several GlyT1 inhibitors exhibit anticonvulsive effects in a model of epilepsy.20) Socala et al. demonstrated that sarcosine displays anticonvulsant properties in the timed intravenous (i.v.) pentylenetetrazole (PTZ) infusion test and the maximal electroshock seizure threshold (MEST) test,21) suggesting GlyT1 as a potential target for antiepileptic drugs. However, few studies have examined GlyT1 inhibitors for their anti-seizure properties.
In the present study, a highly selective inhibitor of glycine transport, 2,4-dichloro-N-{[4-(cyclopropylmethyl)-1-(ethylsulfonyl)piperidin-4-yl]methyl}benzamide (M22) (Fig. 1), which has been reported as a potential GlyT1 inhibitor in the development of anti-schizophrenia drugs and which exhibits excellent pharmacokinetic properties (Oral bioavailability 38%, t1/2: 11.8 h),22–24) was evaluated for GlyT1 inhibition and anti-seizure activity in a mouse model. Our results showed that M22 elevated the seizure threshold in the MEST test and did not impair locomotion or motor coordination in mice.
Three-month-old male C57BL/6J mice (weight: 22–25 mg) were housed in plastic cages with metal covers and were fed a standard rodent diet and water ad libitum. The animals were maintained on a 12 h/12 h light–dark schedule with lights on at 07:00, and they were tested during the diurnal phase, namely, between 09:00 and 18:00. Neonatal mice were purchased from Guangdong Medical University Laboratory Animal Center. All experiments were approved by the Institutional Animal Care and Use Committee of Guangdong Medical University.
Drugs and AdministrationM22 was provided by Chengyuan Chemical Co., Shenzhen, China. The drug was suspended in 0.1% (w/v) carboxymethylcellulose and administered intragastrically (i.g.) at 1 mL/100 g body weight. Strychnine, sarcosine and N-[3-(4′-fluorophenyl)-3-(4′-phenylphenoxy)propyl]sarcosine (NFPS) were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.) and dissolved in normal saline prior to use. 3H-Glycine was purchased from PerkinElmer, Inc. (Waltham, MA, U.S.A.).
The Kinetics of GlyT1 and the Effect M22 on GlyT1 KineticsGlyT1 function and kinetics were examined as described previously.25) Briefly, rat glioma C6 cells (preserved in our laboratory) that stably expressed GlyT1 were plated into 24-well culture plates (1×106 per well). After 18 h, N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES) buffer solution was used to replace the culture medium. Thereafter, M22 (20 µL, final concentration ranges: 10−10–10−4 M) and 3H-glycine (20 µL; final concentration: 5, 10 or 50 nM) were added. The total volume was 200 µL. In the total uptake wells, the tested compounds were replaced with Hank’s balanced salt solution (HBSS; 20 mL). NFPS (10 mM) was used to determine nonspecific uptake. After incubation for 30 min at 37°C, HBSS was discarded, and the plate was washed twice with phosphate buffered saline (PBS) (500 mL). The cells were then lysed with NaOH (100 mL, 2 M). The lysates were collected over Whatman GF/B glass fiber filters. Radioactivity was measured with a liquid scintillation spectrometer (PerkinElmer, Inc.).
The Kinetics of GlyT2 and the Effects of M22 on GlyT2 KineticsInhibitory activity against GlyT2 was examined as described previously.26) Briefly, brain stems were obtained from rat pups and digested with a mixture of 0.05% trypsin (Sigma) and 0.01% DNase I (Sigma) at 37°C for 10 min. Then, the digested tissues were incubated with Dulbecco’s modified Eagle’s medium (DMEM)/F12 (3 mL; 1 : 1; Gibco BRL, Gaithersburg, MD, U.S.A.) and mechanically dissociated into single-cell suspensions. After centrifugation for 5 min at 1000 rpm, the dissociated cells were re-suspended in DMEM containing F12 and then plated into 24 plates pre-coated with polylysine (Sigma). Two days later, cytosine arabinoside (5 mM, Sigma) was added to kill the glial cells. After the cells were cultured for 10 d, the inhibition of GlyT2 was tested. Sarcosine (10−3 M) was employed to block GlyT1 activity, and glycine (10−4 M) was used to determine nonspecific uptake action. The remaining procedures were the same as those described above for GlyT1.
The Maximal Electroshock Seizure Threshold Test in MiceThe electroconvulsive instrument was purchased from Academy of Military Medical Sciences. Prior to conducting the MEST test, mice were randomly divided into several groups (n=10/group). To explore dose–response parameters, individual groups were given different doses of M22 (10, 20 and 40 mg/kg) i.g. over 30 min. When time–course experiments were conducted, thresholds were evaluated 15, 30, 60 and 90 min after commencement of M22 treatment. Strychnine (0.05, 0.1, 0.2 and 0.4 mg/kg) was administered subcutaneously (s.c.) 15 min before treatment with M22 (or vehicle control). Each group of mice was tested once; thus, multiple groups were employed to obtain time-points.
All experimental mice received electrical stimulation (0.250 s in duration; a 50-Hz sine wave) via saline-moistened corneal electrodes. Successful induction of generalized tonic seizure was signified by tonic hindlimb extension at a 180° angle to the torso for one minute after stimulation. The threshold of maximal electroconvulsion was determined using the “up and down” method.27) Briefly, the first mouse was shocked by the expected/estimated CC50 current value, the next mouse received a shock 1 mA higher than that of the previous mouse if tonic seizure was not induced. Otherwise, the next mouse received a shock 1 mA lower than formerly administered. This procedure was followed for all mice within a given treatment group. The CC50 value (the threshold current level inducing hindlimb extension in 50% of animals) was then calculated.
Pentylenetetrazole-Induced ConvulsionWhen timed i.v. PTZ infusion tests were conducted, the mice were administered either M22 or vehicle. After 30 min, each animal was placed in a restrainer, and a needle was inserted into the lateral tail vein. A plastic syringe was attached to the needle via polyethylene tubing. PTZ solution (1% (w/v)) was dispensed from the syringe into the vein at a constant rate of 0.2 mL/min. The time interval from the commencement of PTZ infusion to the onset of each of three separate events (the first myoclonic twitch, generalized clonus with loss of the righting reflex, and forelimb tonus) was recorded.
Determination of Plasma M22 LevelsAfter the MEST test, mice were decapitated, and blood aliquots approximately 1 mL in volume were collected into Eppendorf tubes and mixed with 300 µL methanol. Each tube was vortexed for approximately 1 min and left to stand for 30 min prior to precipitate removal by centrifugation at 10000×g for 20 min. Next, each supernatant sample was placed into a fresh tube and dried under vacuum. Each pellet was re-dissolved in 100 µL of the HPLC mobile phase, and a 60-µL sample was injected into an HPLC system employing a stainless-steel octadecylsilyl (ODS) column (200×4.6 mm) operating at the ambient temperature of 22°C. The mobile phase was methanol–water–diethylamide 40 : 60 : 0.05 (v/v/v), and the flow rate was 1 mL/min. A stock solution of M22 served as an external standard. M22 levels (µg/mL of blood) were determined by comparing sample peak areas with those of the standards.
Locomotion and Motor CoordinationThe test apparatus was constructed by Shanghai Jiliang Company (Shanghai, China). The locomotion apparatus contained four chambers, each of that was sound-attenuated and illuminated by diffuse white light at the center (10 lux). Each mouse was placed into the central quadrant of the open field and allowed to freely explore the arena for 120 min. After each trial, the entire apparatus was cleaned with 30% (v/v) ethanol and thoroughly dried. Cameras located above each chamber recorded all activity, and an automated video tracking system (Jiliang Software, Shanghai, China) was used to quantify activity. Motor coordination was assessed using an accelerating Rotarod system. The rotating rod was equipped with a surface-striated axis (with a diameter of 3.5 cm) and was elevated by 10 cm. For each trial, one mouse was placed on the inactive rod with its head pointed in the direction opposite to that of rod motion. The rod was accelerated to 50 rotations per min (rpm) over 1 min in a stepwise manner. Latency time (the time from commencement of rotation to the time when the mouse fell from the rod) was recorded and used to evaluate motor coordination.
StatisticsData are expressed as the mean±standard error of the mean (S.E.M.). In the experiments investigating the combined effect between strychnine and M22, two-way ANOVA was employed to determine statistical significance, followed by Bonferroni’s post-hoc test for multiple comparisons. In other experiments, one-way ANOVA was used, and Dunnett’s post-hoc test was also applied. p<0.05 was considered statistically significant. Data were analyzed using GraphPad Prism software (Version 5.0).
As shown in Fig. 2A, M22 inhibited the uptake function of GlyT1. The maximal inhibition ratios were 93.35, 94.71 and 90.38% when the concentration of 3H-glycine was 5, 10 and 50 nM, respectively. The 95% confidence intervals (95% CI) were 81.06–105.6, 81.54–107.9 and 79.48–101.3%. There were no significant differences in the maximal inhibition ratios. However, the IC50 value increased with the increasing doses of 3H-glycine: 18.11 nM (95% CI 12.02–26.30 nM), 37.30 nM (95% CI 28.84–47.86 nM), and 114.0 nM (95% CI 53.70–234.0 nM) at 5, 10 and 50 nM 3H-glycine, respectively. These results indicated that M22 is a competitive inhibitor of GlyT1.
(A) M22 (10−10–10−4 M) can inhibit the uptake function of GlyT1; (B) M22 promote the saturating curve to shift in the right direction while there were no differences in the maximal up-taking velocities (dpm/min 106 cells); (C) The Lineweaver–Burk transformation indicated that the slopes were raised with the increase of M22.
To confirm that M22 is a competitive inhibitor of GlyT1, we examined the kinetics of GlyT1 and the effect of M22 on the kinetics of GlyT1. As shown in Fig. 2B, M22 shifted the saturation curve to the right. However, there were no differences in the maximal uptake velocities (dpm/min 106 cells). The Lineweaver–Burk transformation (Fig. 2C) also indicated that the slopes were higher with increasing doses of M22. Thus, the kinetics assays confirmed that M22 is a competitive inhibitor of GlyT1. Using the fitting equation for a competitive inhibitor,26,28) we established that the Ki value of M22 was 129.2 nM (95% CI 8.44–249.9 nM).
Moreover, we also tested the inhibitory activity of M22 on GlyT2, and the results showed that the inhibition ratio of M22 for GlyT2 was 17.3±4.6% at 10 µM, with an IC50 value or 583 µM (95% CI 371–1124 µM), indicating no inhibitory effect on GlyT2 compared with GlyT1.
M22 Administration Significantly Elevated Generalized Tonic Seizure Thresholds in the MEST TestWe first evaluated the anti-seizure effect of M22 in the MEST test in mice. As shown in Fig. 3A, M22 significantly elevated the seizure threshold in the MEST test at all doses tested (F(3, 8)=9.224, p=0.001). Furthermore, in the time–response curve (Fig. 3B), M22 demonstrated an anti-convulsive effect at 15 min (12.16±0.63 vs. 9.14±0.71 mA at 0 min); this effect culminated at 30 min (15.35±1.17 mA), diminished starting at 60 min (13.89±1.29 mA), and was lost at 90 min (11.55±0.77 mA). The ratio of test to control group data was calculated in this experiment (controls were normalized to 100%).
(A) The effect of M22 on the MEST threshold. M22 (10, 20, and 40 mg/kg) was given i.g. 30 min prior to testing. * p<0.05 vs. the vehicle control (0 mg/kg); (B) A time–response curve revealing the anti-convulsive effect of M22 (40 mg/kg i.g.) at 15, 30, 60 and 90 min after commencement of M22 treatment; * p<0.05 vs. the observed commencement threshold (0 min).
In contrast to the MEST test results, M22 did not change the convulsive threshold in the timed i.v. PTZ infusion test at any dose tested (Fig. 4). The control PTZ threshold for myoclonic twitch onset was 33.16±5.21 mg/kg. After treatment with M22, the thresholds were 35.12±5.32, 34.08±4.27, and 36.51±3.29 mg/kg at 10, 20, and 40 mg/kg, respectively (F(3, 36)=0.097, p=0.96) (Fig. 4A). The same effect was observed with the thresholds for generalized clonus (Fig. 4B) (F(3, 36)=0.11, p=0.95) and forelimb tonus (Fig. 4C) (F(3, 36)=0.38, p=0.76).
(A) Onset thresholds of the first myoclonic twitches; (B) Thresholds for generalized clonic seizures; (C) Thresholds for forelimb tonic seizures.
We further analyzed whether M22 is sensitive to strychnine. As shown in Fig. 5A, the seizure threshold was not changed when the strychnine dose was under 0.2 mg/kg in the MEST test. However, when strychnine was co-administered with M22, it reversed the effect of M22 (Fig. 5B). When M22 was combined with strychnine (0.1 mg/kg) the threshold was significantly reduced compared with M22 alone (two-way ANOVA, F(4, 18)=12.69, p<0.001). M22 plasma levels were further measured by HPLC analysis. As shown in Fig. 5C, the level of M22 was 5.53±1.41 µg/mL after M22 administration for 30 min (20 mg/kg i.g.). Administration of strychnine at either 0.1 mg/kg or 0.2 mg/kg did not influence the M22 plasma levels achieved following the administration of either 20 mg/kg M22 (F(2, 12) value=0.01, p-value=0.98) or 40 mg/kg M22 (F(2, 12) value=0.02, p-value=0.94).
(A) The effect of strychnine on the seizure threshold. Strychnine (0.05, 0.1, 0.2 and 0.4 mg/kg) was administered s.c. 45 min prior to testing. * p<0.05 vs. the vehicle control (0 mg/kg). (B) The effect of strychnine on the M22 CC50 value. Strychnine was given s.c. 45 min before testing. M22 (20 and 40 mg/kg) was administered i.g. 30 min prior to testing. # p<0.05 vs. the strychnine-treated group. (C) The influence of strychnine on M22 plasma levels in mice.
We last evaluated the possible influence of M22 administration on motor function in mice. Compared with the spontaneous activity of vehicle-treated mice (70.0±11.8 min), M22 had no effect (75.0±9.8, 77.18±10.5, and 81.3±16.0 min with 10, 20, and 40 mg/kg M22, respectively) (F(3, 36)=0.359, p=0.783) (Fig. 6A). Furthermore, the Rotarod test did not reveal any impairment of motor performance in the range of M22 doses tested (43±5.7 s; F(3, 36)=0.8841, p=0.480) (Fig. 6B), and observing caged animals also revealed no grossly apparent behavioral abnormalities in mice administered M22 at any dose.
(A) The free-moving distance was measured 2 h after treatment with M22; (B) Latency was evaluated as the time from commencement of rod rotation to the time when a mouse fell from the rod.
In the present study, we report for the first time that M22, an effective inhibitor of GlyT1 but not GlyT2, elevated the tonic seizure threshold in the mouse MEST model and did not impair motor function. Given that current epilepsy treatment is limited by poor responses to available AEDs and limited tolerance due to major cognitive side effects, we believe that this new target for epilepsy, the GlyT1 inhibitor M22, has potential as a new anti-convulsive drug or as the lead compound for AEDs development.
Glycine receptors are anion-selective transmitter-gated ion channels that promote Cl− influx and decrease neuronal excitability when activated and that can be specifically blocked by strychnine, which underscores the important role of glycine receptor in the treatment of seizures.29) Proposed strategies for the treatment of anti-epilepsy via the direct activation of glycine receptors remain controversial but are in progress.30) GlyT1 inhibitors have been proposed for the treatment of schizophrenia because they can also increase NMDA receptor function in the prefrontal cortex.31) Sarcosine, another GlyT1 inhibitor, can increase the concentration of extracellular glycine in vivo.32) Moreover, recent studies demonstrated that GlyT1 is also present in regions adjacent to glycinergic synapses, suggesting the potential efficacy of these inhibitors for the treatment of related neurological diseases.11,33) From the above progresses, we can speculate that the anti-seizure effect of GlyT1 inhibitors may be attributable to increased extracellular glycine levels and enhanced glycine receptor activity, thus contributing to the suppression of seizures in epilepsy.34)
Our results are consistent with previous work by Socala et al., who found that sarcosine exhibits anti-convulsive activity as a traditional GlyT1 inhibitor.21) Moreover, it is worth noting that in addition to agonizing the glycine receptor, sarcosine also directly potentiates NMDA receptor function as a co-agonist.35,36) However, due to the high doses of sarcosine (800–1000 mg/kg) required for effective activity,21) the possibility of taking sarcosine as a potential antiepileptic drug may remain elusive. In the present study, we employed M22 as a probe and further confirmed the role of GlyT1 in anti-convulsive action. Compared with other GlyT1 inhibitors such as sarcosine, M22 showed a more effective antiepileptic activity (10–40 mg/kg). Considering that M22 exhibits more desirable pharmacokinetic properties (oral bioavailability 38%, t1/2 11.8 h),12,24) we believe that the anti-seizure properties of M22 are more clinically relevant.
In this study, our results showed that M22 did not prevent myoclonic, clonic, or tonic seizures in the i.v. PTZ test in a mouse model. The discrepancy between the MEST and PTZ data may be due to the different brain areas involved in the different types of seizures. Tonic seizure follows widespread activation of the brainstem, whereas clonic seizure often originates from the forebrain cortex.37,38) Glycine receptors are prominently distributed in the brain stem, cerebellum and retina but are scarcely distributed in the forebrain cortex. Hence, the uneven distribution of this receptor may contribute to these results. Nevertheless, the exact mechanism underlying this phenomenon requires further investigation.
Inhibiting GlyT1 activity could in turn increase the concentration of glycine in the synapse and promote glycine receptor activity, finally leading to inhibitory neurotransmission in the central nervous system (CNS). Strychnine is an inhibitor of glycine receptor activity; thus, given that strychnine counteracted the effects of M22, strychnine must be blocking glycine receptor activity independent of the increased concentration of glycine caused by the inhibition of GlyT1 by M22. In this study, we further confirmed that strychnine was able to reverse the effect of M22, suggesting that M22 likely exhibits anti-seizure activity through the inhibition of GlyT1, as expected. However, we cannot exclude the possibility that strychnine-insensitive binding sites in the NMDA receptor are also involved in the anti-convulsive effect of M22. Orthodox NMDA receptor agonists have been reported to potentiate the occurrence of seizures while some NMDA receptor antagonists exhibit anti-seizure activity.39) By contrast, co-agonists that act on the strychnine-insensitive binding site mediate both pro- and anti-convulsive activity. For example, D-serine potentiates NMDA-induced convulsions but antagonizes electroshock-induced tonic–clinic seizures in mice.40) Regardless of whether M22 affects NMDA receptors, however, our results showed that M22 ultimately elevated the seizure threshold in the mouse maximal electroshock model. Additional evidence is required to establish the role of the strychnine-insensitive binding site in the pathogenesis of seizures.
In conclusion, the present study confirmed that M22 exhibits an anti-tonic seizure effect in MEST testing and did not impair locomotion or motor coordination in mice. As M22 exhibits excellent pharmacokinetic features,22) our results suggested that this compound is a promising candidate for use as an anti-epilepsy drug.
This work was supported by funding from the National Natural Science Foundation of China (Grant number 81271214) and the Medical Scientific Research Foundation of Guangdong Province (B2013306). In addition, we sincerely thank Prof. Xuechu Zhen and Dr. Lin Guo at the Soochow University for providing research assistance and for reviewing the manuscript.
The authors declare no conflict of interest.
