| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Dental Anesthesiology, Division of Clinical Medical Science, Programs for Applied Biomedicine, and Department of Dental Pharmacology, Division of Integrated Medical Science, Programs for Biomedical Research, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan; Department of Applied Pharmacology, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan.
Address correspondence and reprint requests to Masahiro Irifune, DDS, PhD, Department of Dental Anesthesiology, Division of Clinical Medical Science, Programs for Applied Biomedicine, Graduate School of Biomedical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan. Address e-mail to mirifun{at}hiroshima-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Riluzole was administered intraperitoneally in adult male ddY mice. To assess the general anesthetic components, three end-points were used: 1) loss of righting reflex (LORR; as a measure of unconsciousness), 2) loss of movement in response to noxious stimulation (as a measure of immobility), and 3) loss of nociceptive response (as a measure of analgesia).
RESULTS: The intraperitoneal administration of riluzole induced LORR in a dose-dependent fashion with a 50% effective dose value of 27.4 (23.3–32.2; 95% confidence limits) mg/kg. The behavioral and microdialysis studies revealed that time-course changes in impairment and LORR induced by riluzole corresponded with decreased glutamate levels in the mouse brain. This suggests that riluzole-induced LORR (unconsciousness) could result, at least in part, from its ability to decrease brain glutamate concentrations. Riluzole dose-dependently produced not only LORR, but also loss of movement in response to painful stimulation (immobility), and loss of nociceptive response (analgesia) with 50% effective dose values of 43.0 (37.1–49.9), and 10.0 (7.4–13.5) mg/kg, respectively. These three dose–response curves were parallel, suggesting that the behavioral effects of riluzole may be mediated through a common site of action.
CONCLUSIONS: These findings suggest that riluzole-induced LORR, immobility, and antinociception appear to be associated with its ability to inhibit glutamatergic neurotransmission in the central nervous system.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The amino acid glutamate plays a major role in the neurotransmission of excitatory signals via long axonal projections of neurons in the central nervous system (CNS). In fact, most of the billions of long-axon neurons in the CNS use glutamate as their principal transmitter, as do excitatory intrinsic neurons; >50% of all CNS synapses may be glutamatergic (4). In vitro, neurochemical studies have revealed that glutamatergic neurons are important targets for anesthetic action at the cellular and microcircuits levels (5). The volatile anesthetics isoflurane, enflurane, and halothane reduce glutamate release from rat hippocampal (6) and cortical (7) brain slices, and from rat cerebral synaptosomes, but the IV anesthetic, propofol, does not (8,9). Other anesthetics (ketamine, nitrous oxide, xenon) have been shown to inhibit the ligand-gated ion channel, N-methyl-d-aspartate (NMDA) receptor—-a subtype of the excitatory amino acid receptors. In electrophysiological studies, ketamine (10), nitrous oxide (11,12), and xenon (13,14) are potent inhibitors of NMDA-activated currents. In vivo, however, the functional relevance of blocking glutamatergic neurotransmission in mediating essential components of the general anesthetic state (unconsciousness, analgesia, immobility, etc.) is unknown.
Riluzole (2-amino-6-[trifluoromethoxy]benzothiazole) is a neuroprotective drug that blocks glutamatergic neurotransmission in the CNS and inhibits the release of glutamate from cultured neurons, from brain slices, and from corticostriatal neurons in vitro. It is thought that these effects may be partly due to inactivation of voltage-dependent Na+ channels on glutamatergic nerve terminals, as well as activation of a G-protein-dependent signal transduction process. Riluzole also blocks some of the postsynaptic effects of glutamate by noncompetitive blockade of NMDA receptors (15). It has been shown that intraperitoneal (IP) administration of large doses of riluzole in rats induces loss of righting reflex (LORR), and that a small, subanesthetic dose produces a significant reduction in minimum alveolar (anesthetic) concentration of halothane (16). In this study, we further examined the relevance of riluzole-induced LORR to its ability to decrease brain glutamate levels using behavioral and microdialysis studies. In addition, we investigated whether riluzole alone could induce other components of anesthesia, such as immobility and analgesia. To assess the general anesthetic components, three end-points were used: 1) LORR (as a measure of unconsciousness), 2) loss of movement in response to noxious stimulation (as a measure of immobility), and 3) loss of nociceptive response (as a measure of analgesia).
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Riluzole was purchased from Sigma-Aldrich Inc. (St. Louis, MO). Chloral hydrate was from Nacalai Tesque Inc. (Kyoto, Japan). The riluzole was dissolved in dimethylsulfoxide (DMSO; Sigma-Aldrich Inc.) and diluted with distilled water. Final concentrations of DMSO were <40%. Chloral hydrate was dissolved in 0.9% saline solution. Each drug was freshly prepared on the day of the experiment. All drugs were administered IP in a volume of 10 mL/kg.
Behavioral Study
General anesthetic states were evaluated using three end-points: LORR (as a measure of unconsciousness), loss of movement in response to noxious stimulation (as a measure of immobility), and loss of nociceptive response (as a measure of analgesia).
The mice were examined individually in a circular glass beaker (13.5 cm diameter x 19 cm high). To examine the righting reflex, we tilted the beaker by hand to an angle of approximately 45° from a horizontal plane. The beaker was tilted three times at each recording time after IP administration of riluzole. Righting reflex was assessed and recorded every 2 min for 5 h after administration by a blinded observer. Righting reflex scores were evaluated according to the rating scale of Irifune et al. (17): a score of 0 indicated a normal righting reflex; +1 indicated that the mouse righted itself within 2 s on all three trials (slightly impaired righting reflex); +2 indicated that the latency to righting was >2 s, but <10 s at the best response in three trials (moderately or severely impaired righting reflex); +3 corresponded to absence of righting reflex (no righting within 10 s on all three trials).
To determine immobility, a tail pinch was applied under LORR with arterial forceps close to the base of the tail for 30 s or until the animal moved at the time when the drug produced a peak effect on the righting reflex. Purposeful movements of head and/or legs after tail-pinch stimulation were considered a response. Purposeless movements, such as coughing or hyperventilation, were excluded. The anesthetized animals were kept warm with an overhead heat lamp.
To examine analgesia, Haffner's method was used for "conscious" mice (18). The arterial forceps were applied close to the base of the mouse tail for 30 s at 10 min postinjection. The animal's continuous attempts to remove the noxious stimulation by turning its head to the tail, biting the forceps and/or crying were considered responses. In a preliminary study, we examined the analgesic effect of morphine using Haffner's method. The 50% effective dose (ED50) of morphine for loss of nociceptive response was 6.0 (3.4–10.7; 95% confidence limits) mg/kg, which was consistent with the data of Bianchi and Franceschini (18).
Because the response is all-or-none, the number of animals losing the righting reflex (scored +3), the movement response, or the nociceptive response out of the total that received a specific treatment (5–7 mice were used per dose) was used to calculate the percent effect of loss of response. The ED50 for LORR (righting- reflex ED50), for loss of movement in response to tail-pinch stimulation (tail-pinch ED50), and for loss of nociceptive response (antinociceptive ED50) with 95% confidence limits and the parallelism between the dose–response curves were determined according to the method of Litchfield and Wilcoxon (19).
Microdialysis Study
Mice were anesthetized with chloral hydrate (400 mg/kg, IP) and placed in a stereotaxic frame (KOPF model 900; David Kopf Instruments, CA). A microdialysis guide cannula (CUP/7; CMA/microdialysis AB, Solna, Sweden) was implanted into the hippocampus (AP: –2.6 mm, L: +2.0 mm, V: –1.2 mm from bregma and top of the skull) according to the atlas of Franklin and Paxinos (20). The guide cannula was secured to the skull using a bone anchor screw and dental acrylic cement. The animals were allowed to recover from the surgery for at least 7 days before perfusion experiments.
All dialysis experiments were performed on conscious, freely moving mice. On the day of experiment, the inlet of the microdialysis probe (CUP/7; 0.24 mm membrane diameter x 2.0 mm membrane length; CMA/Microdialysis AB) was connected to a microinfusion pump (CMA/102; CMA/Microdialysis AB), and its outlet was connected to a fraction collector (CMA/170; CMA/Microdialysis AB). Then, the probe was filled with an artificial cerebrospinal fluid (ACSF). The ACSF consisted of 120 mM NaCl, 5 mM KCl, 1.2 mM MgCl2, 1.8 mM CaCl2, 1.2 mM KH2PO4, and 1.6 mM Na2HPO4 (pH 7.4). After preparation of the probe, the mouse was lightly anesthetized with chloral hydrate (200 mg/kg, IP). The guide obturator was removed and the probe was introduced. Then, the mouse's hippocampus was perfused with the ACFS. Thereafter, the mouse was placed in a large hemispherical bowl (40 cm diameter). A 5-h equilibration period was used before sampling to allow for stabilization of glutamate and aspartate concentrations after probe implantation. Dialysate samples were then collected and stored at 4°C in plastic vials. The flow rate of the perfusion medium (ACFS) was maintained at 2.0 µL/min using the microperfusion pump, and samples were collected every 20 min. Riluzole or vehicle DMSO solution was administered IP 2 h after the start of sampling, and the alterations of glutamate and aspartate were measured for 260 min. The microdialysis samples were stored at –20°C until the amino acids assays were performed.
At the end of each microdialysis experiment, the brains of the mice were quickly removed after decapitation under deep chloral hydrate anesthesia. The brains were flash-frozen in isopentane. Coronal sections (10 µm) were cut using cryostat at –20°C, mounted on gelatin-coated glass slides, stained using a 0.5% solution of cresyl violet acetate (Sigma-Aldrich Inc.), and examined by microscope. The location of the microdialysis probe was determined histologically on serial coronal sections, and only data obtained from mice with properly implanted probes were used in the results.
Glutamate and aspartate in the dialysate were measured by reverse-phase high-performance liquid chromatography with fluorescence detection after precolumn derivatization with orthophthaldialdehyde reagent. The derivatization was performed using a refrigerated autoinjector (SIL-10AD VP; Shimadzu, Kyoto, Japan) with computer (SCL-10A VP; Shimadzu). A 20-µL dialysate sample was mixed with 20-µL of the orthophthaldialdehyde regent. After 150 s of reaction time, a 20-µL aliquot of the mixture was injected into the performance liquid chromatography apparatus, which consisted of a delivery pump (LC-100; BAS Japan, Tokyo, Japan), a degasser (LC-27A; BAS Japan), a reverse-phase column (CMA/Asp plus Glu analysis column, 1.0 mm internal diameter x 100 mm length; CMA/Microdialysis), a fluorescence detector (CMA/280; CMA/Microdialysis) with "amplify" set to 10 and "rise time" to 1, and a computing integrator-printer (C-R6A Chromatopac; Shimadzu). The analytical column temperature was controlled at 40°C (CTO-10AS VP; Shimadzu). The mobile phase consisted of 0.1 M acetate buffer and 10% acetonitrile adjusted pH to 6.0 with concentrated phosphoric acid. It was pumped through the column at a rate of 60 µL/min. The basal level of glutamate or aspartate in each animal was calculated as the mean of the last five consecutive samples immediately before drug treatment. Subsequent results were expressed as a percentage of the basal glutamate or aspartate level in each individual animal. The data are expressed as the mean ± sem.
The data were analyzed by two-factor analysis of variance (ANOVA) with repeated measures, followed by Fisher least significant difference (LSD) test for multiple comparisons. The results were considered statistically significant when P < 0.05.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
In a microdialysis study, the basal levels of glutamate in the hippocampus were 0.10 ± 0.006 (n = 25) and 0.11 ± 0.009 (n = 25) µM in the control and riluzole-treated groups, respectively, at a flow rate of 2.0 µL/min. The basal levels of aspartate were 0.06 ± 0.004 (n = 25) and 0.06 ± 0.008 (n = 25) µM in the control and riluzole groups, respectively. The IP administration of vehicle DMSO produced no significant change from the basal glutamate and aspartate levels (P > 0.05). Riluzole at a dose of 43.0 mg/kg (95% effective dose [ED95] for LORR) produced a gradual decrease in glutamate levels in the hippocampus during the first 40 min after the IP injection. Extracellular glutamate levels were significantly lower 40–60 min postinjection and reached 63% of the control at 120 min postinjection, followed by an inverse plateau (F[1,13] = 5.682, P < 0.05). This inverse plateau continued for 100 min, returning to the control level at 4 h postinjection (Fig. 3). These time-course changes in decreased glutamate levels in mouse brain corresponded with those in increased righting reflex scores except for the onset of action (Figs. 1 and 3). In contrast, the large dose of riluzole did not affect extracellular levels of aspartate (F[1,13] = 0.010, P > 0.05) (data not shown).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
To examine the effect of riluzole on brain glutamate and aspartate levels, we used an in vivo microdialysis technique—-the only technique that can collect virtually any substance from remote brain regions with minimal tissue trauma. The major advantage of the in vivo microdialysis method is that it can be used in the freely moving animal, allowing one to make correlations between changes in extracellular neurotransmitter levels and behavior (21). It has been reported that in awake, freely moving rats, even at 3-h postimplantation of the microdialysis probe both the basal and high K+-evoked release of glutamate are decreased in a Ca2+-dependent manner (22). In addition, it has recently been shown that the blockade of glutamate release by tetrodotoxin, or a Ca2+-free high Mg2+ buffer perfusion is proportional to the release of this neurotransmitter, indicating that the corresponding increase or decrease of glutamate in the brain reflects neuronal activity. In a microdialysis study, approximately 40%–60% of glutamate in the dialysate is believed to be of neuronal origin, although the remainder may be from glial metabolism or reversal of uptake sites (23).
An in vitro synaptosomal study has shown that riluzole exerts presynaptic inhibition through a reduction in the calcium influx mediated by P/Q-type calcium channels, and thereby inhibits glutamate release from nerve terminals. This inhibition of glutamate release may involve a pertusis toxin-sensitive G-protein signaling pathway (24). These findings indicate that the sites of riluzole's action for regulation of release are presynaptic terminals of glutamatergic neurons. However, an in vivo microdialysis study has revealed that only 40%–60% of glutamate is derived from nerve terminals (23). In fact, the present study showed that a large dose of riluzole decreased glutamate levels by approximately 40% at most. These findings suggest that it may take >40 min for even an effective drug to significantly affect extracellular levels of glutamate released from nerve terminals in a microdialysis study. Furthermore, the NMDA receptor blockade action by riluzole might neutralize its ability to inhibit glutamate release early after injection, because the selective NMDA receptor antagonist, MK-801, increases glutamate release (25). This may explain the slow onset of riluzole's action on glutamate release in the current study. The rapid onset of riluzole's action on righting reflex may be due to its ability to noncompetitively block postsynaptic NMDA receptors (15) because the MK-801 enhances LORR induced by volatile and IV anesthetics (26). However, the present study could not completely clarify the cause-effect relationship between the decrease in brain glutamate levels and some components of the behavioral responses. Other mechanisms may be involved in the riluzole-induced behaviors.
Riluzole acts not only on glutamate release but also on other neurotransmitter systems, receptors, and channels, which may contribute to the anesthetic state. Of these targets, as shown in in vitro studies, activation of two-P domain K+ channels (27), blockade of voltage-operated Na+ channels (28), decrease in glutamate- induced dopamine release (29), and blockade of postsynaptic NMDA receptors (30) all may be the putative targets of anesthetic action of riluzole. However, no behavioral finding indicates that a selective K+ channel activator, a Na+ channel blocker, or a dopamine receptor antagonist, on its own, induces LORR. In contrast, the specific, competitive NMDA receptor antagonists, 2-amino-5-phosphonopentanoic acid and CGS-19755, produce anesthetic effects (31). Therefore, riluzole-induced LORR appears to involve both inhibition of glutamate release and blockade of NMDA receptors rather than other effects.
-Aminobutyric acidergic (GABAergic) inhibitory neurotransmissions as well as glutamatergic neurotransmissions are important targets for anesthetic action (32). It has been shown that riluzole blocks synaptosomal uptake of [3H]GABA in the striatum of rats (33). Subsequently, the increased endogenous GABA levels in the synaptic clefts enhance GABA activity. In a microdialysis study, GABA inhibited glutamate release in the hippocampus of rats (34). Thus, another possibility is that the ability of riluzole to block GABA uptake would cause the decreased brain glutamate levels and, therefore, modulate the anesthetic behaviors.
In this study, the effect of riluzole on extracellular glutamate levels was examined in the hippocampus of mice. The primary function of the hippocampus is disputed, but it is generally associated with the coding of visual and auditory inputs, arousal and attention, voluntary movements, exploratory behavior, and especially formation of memory and retention of learned behavior. A particularly important hippocampal function with respect to anesthesia may be the association of this structure with memory formation. The general anesthetic state comprises many components, including loss of sensation, absence of awareness of surroundings, and unconsciousness. These manifestations may result from the loss of short-term memory traces (1,35). The major excitatory neurotransmitter in the hippocampus is glutamate. Therefore, the link between amnesia and anesthesia suggests that alterations in hippocampal function via glutamatergic neurons may contribute to the state of anesthesia. In fact, in an electrophysiological study, riluzole potently blocked excitatory synaptic transmission via depression of presynaptic conduction in hippocampal glutamatergic nerve fibers (36).
Riluzole induced LORR, loss of movement in response to noxious stimulation (immobility), and antinociceptive effect (analgesia) in a dose-dependent manner. The dose–response curves for loss of righting, movement, and nociceptive responses were parallel (Fig. 2). Thus, these behavioral effects of riluzole may be mediated through a common site of action; that is, inhibition of release from glutamatergic nerve terminals and/or blockade of NMDA receptors. To explore each anesthetic component separately, the present study used the three end-points described above. However, the behavioral findings of this study may have to be interpreted more carefully. The three different tests might measure the same phenomenon because these behaviors could all be induced by a muscle relaxant effect. To prove this, however, the muscle relaxant effect of riluzole and the role of glutamatergic neurons in the effect must be clarified. Further studies will be required.
In vivo, behavioral and neurophysiological studies have shown that immobility induced by inhaled anesthetics is mediated by glycine and NMDA receptors and Na+ channels in the spinal cord, but not by GABAA, nicotinic acetylcholine, serotonin, opioid and 
2-adrenergic receptors and K+ channels [see Sonner et al. (37) for review]. However, some recent studies have pointed out that glycine and NMDA receptors may contribute only partially to inhaled anesthetics-induced immobility (38,39). Systemic administration of riluzole produces neuroprotective effects during transient spinal cord ischemia in rabbits (40). This finding indicates that riluzole easily penetrates the blood–brain barrier and affects the spinal cord. Thus, it is likely that the blockade by riluzole of glutamatergic neurotransmission in the spinal cord plays an important role in inducing immobility.
Intrathecal administration of the metabotropic glutamate receptor agonist, (RS)-3,5-dihydroxyphenylglycine, induces nociceptive behaviors in rats. Systemic administration of small doses of riluzole, insufficient to affect the righting reflex, produces dose-dependent reductions in the metabotropic glutamate receptor agonist-induced nociceptive behaviors (41). These findings support the present evidence that riluzole has an antinociceptive effect via inhibition of glutamatergic neurotransmission.
As shown in Figure 2, while even small doses of riluzole induced antinociception, large doses were required to produce immobility in response to noxious stimulation. Similarly, intrathecal administration of ketamine, an NMDA channel blocker, concentration-dependently inhibits the response to a pinch applied to the tail (Haffner's method) in conscious mice (42). We have recently confirmed that IP injection of large, anesthetic doses of ketamine induce immobility. Larger doses of either riluzole or ketamine may be required to produce immobility rather than antinociception, because immobility involves both a reduction in sensory processing and an inhibition of motor neuron excitation (32).
In conclusion, riluzole-induced LORR, antinociception, and immobility appear to be associated with its ability to inhibit glutamatergic neurotransmission.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Supported, in part, by Ministry of Education, Culture, Sports, Science, and Technology of Japan Grant 10671740 and 13672097.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
2B adrenoceptors. J Neurosci 2000;20:9242–51.-aminobutyric acid-A and excitatory amino acid receptors. Anesth Analg 2003;97:424–9.-hydroxybutyric acid (GHB) and -aminobutyric acidB (GABAB) receptor-mediated release of GABA and glutamate (GLU) in rat thalamic ventrobasal nucleus (VB): a possible mechanism for the generation of absence-like seizures induced by GHB. J Pharmacol Exp Ther 1995;273:1534–43. subunit by the neuroprotective drug riluzole. Mol Pharmacol 1994;45:1055–6.[Abstract]-2-adrenergic agonists, and local anesthetics in mice and rats. Anesth Analg 1988;67:616–24.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
