Free
Education  |   May 2003
GABAergic Interneurons at Supraspinal and Spinal Levels Differentially Modulate the Antinociceptive Effect of Nitrous Oxide in Fischer Rats
Author Affiliations & Notes
  • Ryo Orii, M.D., D.M.Sc.
    *
  • Yoko Ohashi, M.D.
    *
  • Sunil Halder, B.Sc.
  • Mariangela Giombini, M.D.
    *
  • Mervyn Maze, M.B., Ch.B.
  • Masahiko Fujinaga, M.D.
    §
  • *Postdoctoral Fellow. †Medical Student. ‡Professor. §Senior Lecturer.
  • Received from the Department of Anaesthetics and Intensive Care, Imperial College of Science, Technology and Medicine, University of London, and Magill Department of Anaesthesia, Intensive Care and Pain Management, Chelsea & Westminster Hospital, Chelsea & Westminster Healthcare National Healthcare Service Trust, London, United Kingdom.
Article Information
Education
Education   |   May 2003
GABAergic Interneurons at Supraspinal and Spinal Levels Differentially Modulate the Antinociceptive Effect of Nitrous Oxide in Fischer Rats
Anesthesiology 5 2003, Vol.98, 1223-1230. doi:
Anesthesiology 5 2003, Vol.98, 1223-1230. doi:
STUDIES over the last several decades have revealed, in part, the underlying mechanisms of nitrous oxide (N2O)-induced analgesic/antinociceptive effect. The initiating action of N2O appears to be the release of opioid peptides in the brain stem, which activates the descending noradrenergic inhibitory neurons, leading to the modulation of nociceptive processing in the spinal cord. 1 However, precise neuronal pathways that mediate these effects of N2O are not yet fully understood.
Descending noradrenergic inhibitory neurons are important components of the endogenous pain suppression system and it is thought that they are tonically inhibited under resting conditions by inhibitory γ-amino butyric acid–mediated (GABAergic) neurons. 2 Because activation of opioid receptors by endogenous opioid peptides generally mediates inhibitory neurotransmission, we speculate that it is the GABAergic interneurons in the brain stem that are inhibited by N2O, which leads to disinhibition (activation) of the descending noradrenergic inhibitory neurons (fig. 1). In the spinal cord, activation of the descending noradrenergic inhibitory neurons by N2O leads to the release of norepinephrine in the dorsal horn, 3 which in turn modulates nociceptive processing by at least two neuronal pathways (fig. 1). One of these neuronal pathways involves the activation of inhibitory GABAergic neurons through α1 adrenoceptors. 4 If this theory is correct, GABAergic neurons modulate the antinociceptive properties of N2O in diametrically opposite direction; activation of GABAergic neurons at the supraspinal level will be pronociceptive, while at the spinal level activation of GABAergic neurons will be antinociceptive.
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e.  , systemically).
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e. 
	, systemically).
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e.  , systemically).
×
To test this hypothesis, we examined the effects of muscimol (a GABAAreceptor agonist) at the supraspinal level and gabazine (a GABAAreceptor antagonist) at the spinal level on N2O-induced antinociceptive effects using the plantar test in rats. Because of the diametrically opposite effects, we next tested whether and in which direction a systemically administered GABA-mimetic compound will affect the antinociceptive properties of N2O. Thus, we studied the effects of midazolam (a benzodiazepine receptor agonist exhibiting GABA-mimetic effects at GABAAreceptors) on the N2O-induced antinociceptive effect behaviorally and immunohistochemically.
Materials and Methods
Animals
Adult male Fischer rats weighing between 250 and 320 g were used throughout the study (B&K Universal, Grimston Aldbrough, Hull, UK). This strain of rat exhibits strong antinociceptive effect in the presence of N2O. 5 All animal procedures were carried out in accordance with the United Kingdom (Scientific Procedures) Act of 1986, and the study protocol was approved by the Home Office of the United Kingdom (London, UK). Rats were provided with access to food and water ad libitum  and kept in a 12-h light and dark cycle. All efforts were made to minimize animal suffering and reduce the number of animals used. No animal was used for more than one experiment.
Study I: Effect of Supraspinally Administered GABAAReceptor Agonist on N2O-induced Antinociceptive Effect
Intracerebroventricular Cannulation.
Rats were anesthetized with 1.5–2.0% isoflurane/oxygen. After stereotactic apparatus was applied to the head, a hole was trephined overlying the left ventricle. Through the hole, a 22G stainless steel cannula was threaded into the lateral ventricle using the atlas of Paxinos and Watson 6 with the following coordinates: with bregma as a reference, 4.0 mm posterior, 1.0 mm lateral, and 1.0 mm ventral to the surface. The guide cannula (a BD-7 stainless steel, 11.0 mm in length) was then inserted 3.0 mm into the lateral ventricle and fixed in position with methylmethacrylate resin. After the surgery, 0.6 ml of 0.5% bupivacaine (10 mg/kg) was injected locally around the incision. A minimum period of 4 days elapsed between cannula implantation and experiment being performed. At the conclusion of the experiment, the site of cannulation was examined by injecting pontamine blue dye, of which location within the cerebroventricular was confirmed by autopsy; success rate was 100%.
Drug Administration and Gas Exposure.
A 100-μl Hamilton microsyringe was attached to the chronically implanted cannula by PE10 tubing. The microsyringe was placed on the infusion pump (Pump 22, Harvard Apparatus, South Natick, MA), which was set at a constant flow rate of 5 μl/min for a period of 2 min for drug delivery. Rats were administered 10 μl of either 0.9% sodium chloride (saline) or muscimol (a GABAAreceptor agonist; 3 μg/10 μl). Dosage of muscimol (3 μg) and injection volume (10 μl) were chosen from the data in a study by Hough et al.  7 Fifteen minutes after completing drug administration, gas exposure was performed in an exposure chamber (18 inches long, 9 inches wide, and 8 inches high) made of plastic. The chamber and holding area of rats were maintained between 23 and 24°C. Humidity within the chamber was between 35 and 45%. Either air or 75% N2O/25% O2at a flow rate of 4 l/min was continuously delivered into the exposure chamber from an anesthetic machine via  an inflow port and exhausted via  an outflow port flowing through the Anesthetic Gas Scavenging System (Boyle 2000, Ohmeda, Hatfield, Herz, UK). Gas concentrations, including those for N2O, O2, and CO2, in the chamber were measured continuously by infrared gas spectrometry (Ohmeda 5250 RGM, Ohmeda). Wash-in of the gas mixture to achieve a steady-state concentration within the chamber was achieved in 30 min, after which animals were placed into the chamber through the side door; this procedure did not significantly change the gas concentrations within the exposure chamber. Initially, paper towels were laid down on the floor of the chamber to soak up excrement produced by animals. The paper towels and all excrement collected were removed from the chamber before the plantar test was started. The side door was also used to clean the chamber throughout experimentation, because urine and feces may affect measurements obtained by the plantar test. After 30 min of gas exposure, which coincides with the peak antinociceptive effect of N2O by the tail flick test, 5 animals were subjected to the plantar test.
Plantar Test.
The plantar test, i.e.  , thermal nociceptive testing, was performed using the plantar test device (Plantar test 7370, Ugo Basile, Comerio, Italy). The assessor was blinded to drug or vehicle administration. Radiant heat was applied on the plantar surface of hind paws through the floor of the exposure chamber and the paw withdrawal latency (PWL), defined as the time between the activation of the heat source and hind paw withdrawal, was automatically recorded. Approximately 30–60 min before gas exposure, heat intensity was adjusted such that the baseline PWL was approximately 4 s. To avoid tissue damage, a predetermined cut-off time of 10 sec was imposed. Each PWL data consisted of a mean of three trials for each animal. From the PWL, the percentage of maximal possible effect (%MPE) was calculated.
Study II: Effect of Spinally Administered GABAAReceptor Antagonist on N2O-induced Antinociceptive Effect
Intrathecal (Intrathecal) Injection, Gas Exposure, and Plantar Test.
The rat was anesthetized with 1.5% isoflurane/33% O2/66% N2O during the following procedures, which lasted approximately 2 min. Immediately after induction of anesthesia, the rat was shaved around the lumbar area to facilitate identification of the midline of the spine. The animal was then braced over a 50-ml centrifuge tube to splay the intervertebral spaces of the lumbar (L) spine. While firmly holding the rat's vertebral column in one hand, a 27G needle attached to a 50-μl Hamilton syringe was inserted into the intervertebral space between L5 and L6 as described by Hylden and Wilcox. 8 Entry into the subarachnoid space was accompanied by a sudden lateral movement of the tail as described by Mestre et al.  9 A total volume of 10 μl vehicle with saline (as a control) or vehicle containing gabazine (a GABAAreceptor antagonist; 1 μg/10 μl) was injected, and the needle was removed. The vehicle contained a mixture of 2% pontamine (blue dye)/6% dextrose which was slightly hyperbaric with respect to the cerebrospinal fluid. A dosage of gabazine, 1 μg/10 μl, was chosen because, in preliminary experiments, this proved to be the lowest efficacious blocking dose at the receptor, which does not exhibit its own behavioral effect. (At higher dosages of gabazine administration, animals showed signs of toxicity, e.g.  , prominent repetitive tail flexing.) As soon as these procedures were completed, anesthesia was discontinued and the rat awakened within 10 min. Fifteen minutes following intrathecal injection, the rat was exposed to air or 75% N2O/25% O2and subjected to the plantar test 30 min later as described in Study I. After experimental procedures had been completed, the site of injection was confirmed by autopsy. Successful intrathecal administration (13 of 17, 76.5%) was evidenced by the presence of the blue dye within the subarachnoid space; data from unsuccessful injections (4 of 17) were not further considered.
Study III: Effect of Systemically Administered Midazolam on N2O-induced Antinociceptive Effects
Intraperitoneal (Intraperitoneal) Drug Administration, Gas Exposure, and Plantar Test.
Rats were injected intraperitoneally with either saline or midazolam (a benzodiazepine receptor agonist; 1, 5, and 10 mg/kg). These doses are less than those that produce a hypnotic effect that can confound interpretation of the plantar test. Fifteen minutes after injection, rats were exposed to air or 75% N2O/25% O2for 30 min and were subjected to the plantar test as described in Study I.
Immunohistochemical Examination of Spinal Cord and Brain.
In some animals, immunohistochemical studies were performed instead of subjecting animals to nociceptive testing. Rats were given midazolam (1, 5, and 10 mg/kg) intraperitoneally. Fifteen minutes later they were subjected to 90 min of gas exposure to air or 75% N2O/25% O2, which was the time to the peak effect of N2O exposure on c-Fos induction in the spinal cord. 4 Animals were then injected with sodium pentobarbital (100 mg/kg) intraperitoneally for perfusion. A thoracotomy was performed by transverse incision at the level of the diaphragm and midline sternotomy. The animals were then perfused with 0.1 m phosphate-buffered saline (PBS) followed by 4% paraformaldehyde in 0.1 m phosphate buffer (PFA) via  a 16G cannula inserted through the left ventricle into the ascending aorta. An incision was made in the right ventricle for drainage. Following decapitation, the brain was extracted and the spinal cord was expelled by rapid injection of PBS into the spinal canal at the sacral level. Collected tissues were stored in PFA overnight then transferred into 30% sucrose in PBS for at least 48 h at 4°C. A 5-mm portion of the spinal cord at the lumbar enlargement was cut by a razor blade and was freeze-mounted in embedding matrix. Transverse sections of 30 μm were cut at −15°C, which were collected in PBS; every third section was collected (approximately 40–50 sections per sample). A 2-mm portion of the brain, containing the A7 nuclei region, was cut by a razor blade using specifications from the atlas of Paxinos and Watson 6 and freeze-mounted in embedding matrix. Transverse sections of 30 μm were cut at −15°C and were collected in PBS.
Fluorescent Double Staining of c-Fos and Dopamine β-Hydroxylase and Colocalization Quantification.
Free-floating brain sections were first incubated for 1 h in blocking solution consisting of 3% donkey serum (Chemicon, Temecula, CA) in 0.3% Triton X in PBS (PBT). They were then incubated overnight at 4°C on an orbital shaker at 75 r.p.m. with goat antic-Fos antibody (1:2000, Cat. No. sc-52-G, Santa Cruz Biotechnology, Santa Cruz, CA) and rabbit anti-dopamine β-hydroxylase (anti-DBH) antibody (1:2000, Cat. No. DZ1020, Affiniti Research Products, Mamhed, UK) in 1% donkey serum in PBT. The use of c-Fos has been validated as a biochemical marker of neuronal activation in brainstem catecholaminergic neurons using various stimuli. 10–12 DBH is a norepinephrine synthesizing enzyme, which is commonly used as an immunohistochemical marker of noradrenergic neurons. Sections were rinsed with PBT, incubated for 1 h in darkness with a mixture of Cy3-conjugated donkey antigoat secondary antibody (1:200, Cat. No. 705–166–147, Jackson Immuno Research Laboratories, West Grove, PA) and fluorescein isothiocyanate (FITC)-conjugated donkey antirabbit secondary antibody (1:200, cat No. AP182F, Chemicon) in 1% donkey serum in PBT, then rinsed with PBS, floated in water, and mounted on glass slides. After drying, the slides were fitted with cover slips with one drop of VectaShield (Vector Laboratories, Burlingame, CA) and were mounted. The four well-preserved undamaged sections were randomly selected from each animal and examined under a fluorescent microscope (Leica DMR microscope, Leica, Wetzlar, Germany). Red staining of the nucleus identified the c-Fos positive neurons. Green staining of the cytoplasm identified the DBH-positive neurons. The A7 region, located according to a rat brain atlas, 6 was easily identified by DBH staining of the cytoplasm. Results from four animals for each group were collated and the prevalence of c-Fos/DBH colocalization was calculated. The investigator performing the c-Fos and DBH counting was blinded to the origin of the sections.
Diaminobenzidine Staining and c-Fos Quantification.
Free-floating spinal cord sections were first incubated at room temperature for 30 min in 0.3% hydrogen peroxide in 70% methanol/PBS. Then they were incubated for 1 h in blocking solution consisting of 3% rabbit serum in PBT, followed by overnight incubation with goat anti–c-Fos antibody (1:10000, Cat. No. sc-52-G, Santa Cruz Biotechnology) in blocking solution (1% normal rabbit serum in PBT) at 4°C on an orbital shaker at 75 rpm. The use of c-Fos has been validated as a marker of neuronal activation in the spinal cord. 13,14 Sections were then rinsed with PBT, incubated for 1 h with biotinylated rabbit antigoat immunoglobulin (1:200, Vector Laboratories) in 1% normal rabbit serum, rinsed with PBT, and incubated for 1 h with avidin-biotin-peroxidase complex (Vector Laboratories) in PBT at room temperature on an orbital shaker at 75 rpm. Visualization of the immunohistochemical reaction was achieved by incubation with diaminobenzidine with nickel-ammonium sulfate (DAB kit, Vector Laboratories). After the staining procedure was completed, the sections were rinsed in PBS and then rinsed in distilled water, mounted on glass slides (which were dehydrated in 100% ethanol), cleared in 100% xylene, and fitted with cover slips. By diaminobenzidine staining with nickel enhancement, c-Fos positive cells were identified by dense black nuclear staining under a bright field microscope (Olympus Model BX50 Research Photomicroscope, Olympus Optical, Southall, Middlesex, UK). Randomly selected undamaged sections were photographed using a digital camera (Olympus Digital Camera Model C2020Z, Olympus Optical). The number of c-Fos positive cells was counted for each area of the spinal cord, i.e.  , laminae I-II (superficial area), laminae III-IV (nucleus proprius area), laminae V-VI (neck area), and laminae VII-X (ventral area), according to the method by Presley et al.  15 Five photographs were taken from each rat and the mean number of c-Fos positive cells per section was calculated. Each group was composed of at least four animals, and the number of c-Fos positive cells in each group was calculated as mean ± SD. The investigator was blinded to the treatment group during the counting process.
Statistical Analyses
Plantar Test.
Data of the plantar test results were expressed as %MPE. Comparisons at the supraspinal and spinal levels were conducted in the following groups: saline/air, drug/air, saline/75% N2O, and drug/75% N2O. Data were analyzed by one-way analysis of variance and Bonferroni correction as a post hoc  test. A P  value less than 0.05 was considered to be statistically significant.
Immunohistochemistry.
Results from c-Fos and DBH double staining in the A7 nuclei were compared in the following groups: saline/air, drug/air, saline/75% N2O and drug/75% N2O. Data were analyzed using a contingency table, and Fisher exact test was used as post hoc  test. Results from c-Fos single-staining were compared in each laminal scheme in the following groups: saline/air, drug/air, saline/75% N2O, and drug/75% N2O. Data were analyzed by one-way analysis of variance and Bonferroni correction as a post hoc  test. A P  value less than 0.05 was considered to be statistically significant.
Results
Rats exposed to an air environment were awake and active during the experiment, while those exposed to a N2O environment were excited for the first 5–10 min of exposure, followed by a relatively calm state. No animal exhibited motor dysfunction, as assessed by loss of righting reflex, after administration of agonists or antagonists by any of the three routes of administration.
Study I: Effect of Supraspinally Administered GABAAReceptor Agonist on N2O-induced Antinociceptive Effect
Results from the plantar test are shown in table 1. N2O significantly increased the PWL from a %MPE of 3.6 ± 4.3% (mean ± SD, n = 6) with air to 30.6 ± 7.3% (n = 7). Muscimol had no effect on the %MPE with air (1.2 ± 6.2%, n = 6) but completely attenuated the N2O-induced enhancement of PWL (%MPE of 3.6 ± 9.0%, n = 7).
Table 1. Effects of Supraspinally Administered Muscimol on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 1. Effects of Supraspinally Administered Muscimol on N2O-induced Antinociceptive Effect by the Plantar Test
×
Study II: Effect of Spinally Administered GABAAReceptor Antagonist on N2O-induced Antinociceptive Effect
Results from the plantar test are shown in table 2. N2O significantly increased the PWL from a %MPE of 3.5 ± 3.4% (mean ± SD, n = 4) to 38.3 ± 10.5% (n = 4). Gabazine had no effect on the %MPE with air (0.9 ± 4.8, n = 4) but significantly lowered the %MPE (15.0 ± 3.7%, n = 4) when compared with the saline/N2O group.
Table 2. Effects of Intrathecally Administered Gabazine on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 2. Effects of Intrathecally Administered Gabazine on N2O-induced Antinociceptive Effect by the Plantar Test
×
Study III: Effect of Systemically Administered Midazolam on N2O-induced Antinociceptive Effect
Plantar Test.
Results from the plantar test are shown in table 3. N2O significantly increased the PWL from a %MPE of 2.6 ± 1.3% (mean ± SD, n = 6) to 38.2 ± 6.9% (n = 6). The midazolam/air group showed no difference in %MPE at 1, 5, and 10 mg/kg (n = 6 each) when compared with the saline/air group. Midazolam significantly lowered antinociceptive effect of N2O in a dose-dependent fashion, showing a significant difference in the %MPE at 5 mg/kg (15.5 ± 5.4%, n = 6) and 10 mg/kg (5.8 ± 7.0%, n = 6); no difference at 1 mg/kg (37.8 ± 7.8%, n = 6).
Table 3. Effects of Systemically Administered Midazolam on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 3. Effects of Systemically Administered Midazolam on N2O-induced Antinociceptive Effect by the Plantar Test
×
Immunostaining in the A7.
A total of four animals were examined in each group and the results are shown in table 4. In saline/air and midazolam/air groups, the prevalence of c-Fos/DBH colocalization was very low (1.0 and 4.8%, respectively). The saline/N2O group showed significantly higher incidence of colocalization (53.1%); this effect was significantly attenuated by midazolam pretreatment (7.3%).
Table 4. Effects of Systemically Administered Midazolam on N2O-induced c-Fos Expression in Noradrnergic Cells (DBH Positive Cells) in the A7 Nuclei
Image not available
Table 4. Effects of Systemically Administered Midazolam on N2O-induced c-Fos Expression in Noradrnergic Cells (DBH Positive Cells) in the A7 Nuclei
×
Immunostaining in the Spinal Cord.
A total of four animals were examined in each group and the results are shown in figure 2. The number of c-Fos positive cells in the entire gray area of the spinal cord section in the saline/air group was 76.2 ± 3.5 (mean ± SD). Exposure to 75% N2O increased the number of c-Fos positive cells to 121.8 ± 11.5. The increase in c-Fos positive cells was significant in laminae III-IV (63.0 ± 4.9 vs.  32.8 ± 1.5, saline/air group) and laminae VII-X (22.0 ± 4.1 vs.  16.0 ± 2.9, saline/air group). The midazolam (10 mg/kg)/air group showed no difference when compared with the saline/air group. In the midazolam/N2O group, c-Fos expression was attenuated when compared with the saline/N2O group; at lower concentration of midazolam (1 mg/kg), a partial reduction occurred while at higher concentrations (5 and 10 mg/kg), almost complete suppression of N2O-induced c-Fos expression was noted.
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P  < 0.05 versus  saline/air group. #P  < 0.05 versus  saline/N2O group.
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P 
	< 0.05 versus 
	saline/air group. #P 
	< 0.05 versus 
	saline/N2O group.
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P  < 0.05 versus  saline/air group. #P  < 0.05 versus  saline/N2O group.
×
Discussion
Descending noradrenergic inhibitory neurons are tonically inhibited under the resting conditions, although the precise neuronal pathways involved have yet to be clarified. 2 Evidence to date indicate that N2O induces opioid peptide release in the periaqueductal gray area of the midbrain. We propose that the released opioid peptide inhibits GABAergic interneurons, thereby disinhibiting the descending noradrenergic inhibitory neurons, resulting in modulation of nociceptive processing in the spinal cord (fig. 1). 1 Our hypothesis is supported by the finding that the antinociceptive properties of N2O are mitigated when GABAergic inhibition is restored by supraspinal administration of a GABAAreceptor agonist, muscimol (table 1). In the spinal cord, activation of descending noradrenergic inhibitory neurons by N2O leads to the release of norepinephrine, 3 which modulates nociceptive processing by at least two different neuronal pathways (fig. 1). One mechanism involves norepinephrine-induced activation of α2 adrenoceptors, 16,17 which is thought to decrease neurotransmission in the primary afferent neuron by presynaptic inhibition of calcitonin gene-related peptide and substance P release 18 as well as inhibiting firing of the ascending second order neurons, 19 thereby interrupting the transmission of nociceptive impulses. Recently, we provided evidence for a second mechanism that involves activation of inhibitory GABAergic neurons 4 through α1 adrenoceptors. 20 Now we show that spinally administered gabazine, a GABAAreceptor antagonist, significantly attenuates the antinociceptive effect of N2O (table 2), thereby establishing the mechanistic role that activation of spinal GABAergic interneurons plays in transducing the antinociceptive effect of N2O at the spinal level.
Taken together, activation of GABAergic interneurons differentially modulates the antinociceptive effect of N2O depending on whether interneurons are located supraspinally (where they inhibit) or spinally (where they facilitate). In clinical anesthetic practice, drugs that either directly activate or enhance GABA-mediated chloride influx GABAAreceptor are commonly used; therefore, we next inquired whether a systemically administered GABA-mimetic agent would increase (through a spinal site) or decrease (through a supraspinal site) the antinociceptive effect of N2O. As a model GABA-mimetic compound, we used midazolam, which is commonly used as a sedative agent during the perioperative period. Midazolam does not directly “gate” the GABAAreceptor; rather, midazolam exhibits GABA-mimetic properties by enhancing the effect of endogenous GABA within the synapse. 21 Because midazolam's sedative and muscle relaxant properties may confound interpretation of behavioral tests, we first established that, at the midazolam doses used (i.e.  , 1, 5, 10 mg/kg), there was neither a detectable reduction in motor response nor loss of righting reflex. At these doses, midazolam attenuated N2O-induced antinociceptive effects in a dose-dependent fashion (table 3).
These data suggested that, when administered systemically, GABA-mimetic agents are more likely to enhance tonic GABAergic inhibition of the descending noradrenergic neurons and thus prevent the spinal modulatory effects of N2O. To test this further, we examined the effect of intraperitoneal midazolam on N2O-induced neuronal activation of the pontine A7 nuclei. The N2O-induced activation of the noradrenergic neurons in the A7 (as reflected by increased c-Fos expression in DBH positive cells) was inhibited by systemically administered midazolam (table 4). While N2O activates all three major noradrenergic nuclei in the brain stem (i.e.  , A5, locus caeruleus, and A7), 17 we have focused our attention on the A7 because only neurons originating from this nucleus terminates on relevant pain processing neurons in the spinal cord. 22–27 In experimental settings, disruption of the descending noradrenergic pathway blocks the spinal modulatory and antinociceptive effect of N2O 3,17; our finding that midazolam prevents N2O-induced activation of descending noradrenergic pathway (as evidenced by c-Fos expression in the A7) and thereby its antinociceptive effect corroborates our earlier studies (fig. 2). 3,17 
Our data provide a plausible mechanism whereby midazolam blocks N2O-induced activation of descending noradrenergic pathways. At the supraspinal level, midazolam's GABA-mimetic action is expressed because of the presence of endogenously released GABA 2; at this site, it has the effect of tonically inhibiting the descending noradrenergic neurons, thereby mitigating the actions of N2O (table 4). No norepinephrine is released at spinal cord terminals of the descending noradrenergic neurons; therefore, α1 adrenoceptors are not activated and hence GABA interneurons remain quiescent. In the absence of endogenous GABA, midazolam cannot exert its GABA-mimetic antinociceptive action at the spinal cord level. Earlier, Dirig and Yaksh 28 reported that intrathecal administration of muscimol (a direct activator of the GABAAreceptor), but not midazolam (an enhancer of activation of GABAAreceptor activation by endogenous GABA), exhibited antinociceptive effect in rats. These data are in agreement with our interpretation. In a series of studies with the combination of midazolam and morphine, Luger et al.  also showed that it depended on the site of administration whether midazolam inhibited (icv administration) or potentiated (intrathecal administration) the analgesic of morphine in rats. 29,30 
In clinical practice, midazolam is frequently used to sedate patients in preparation for surgery in which N2O and/or opioids are used as part of the anesthetic regimen. If our results can be extrapolated to the clinical settings, then it could mean that the analgesic effects of N2O and opioids may be attenuated by midazolam (and possibly by other indirectly acting GABA-mimetic agents, including barbiturates and propofol). Because volatile anesthetics exhibit both direct and indirect GABA-mimetic properties, 31 their effect on N2O and opioid-induced analgesic may be difficult to predict. However, Goto et al.  reported that halothane and isoflurane inhibited N2O-induced antinociceptive effect in rats. 32,33 More recently, Janieszewski et al.  reported that sevoflurane attenuated the analgesic effect of N2O in humans. 34 While barbiturates and propofol alone have been demonstrated to have a hyperalgesic action, 35–38 we did not observe a hyperalgesic effect of icv muscimol (3 μg). Apart from the fact that these compounds differ in their ability to directly gate chloride conductance through the GABAAreceptor, it is possible that our experimental paradigm is less stressful than that used in the other reports. 35–38 Thus the basal state in those reports may in fact be one of stress-induced analgesia and the administration of propofol, barbiturates, and midazolam may interrupt descending noradrenergic inhibitory pathways and reduce  the latency to withdrawal from the noxious stimulus, providing the interpretation that the drugs are hyperalgesic. 35–38 We posit that, under our experimental conditions, the basal state is stress-free because the subject is unrestrained; in this basal state, the descending noradrenergic inhibitory neurons are tonically inhibited by GABAergic neurons at the supraspinal level. Under these circumstances, the addition of an exogenous GABA agonist, such as muscimol, does not contribute further to what is already being effected by endogenous GABA.
In summary, we have demonstrated that GABAergic neurons differentially modulate the antinociceptive effects of N2O at supraspinal and spinal levels in Fischer rats. Systemic administration of midazolam, the GABA-mimetic agent, attenuates the N2O-induced antinociceptive effects, assessed both behaviorally and immunohistochemically. If these findings are extrapolated to humans, it suggests that the analgesic effect of N2O may be attenuated by midazolam.
References
Fujinaga M, Maze M: Neurobiology of nitrous oxide-induced antinociceptive effects. Mol Neurobiol 2002; 25: 167–89Fujinaga, M Maze, M
Basbaum AL, Fields HL: Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984; 7: 309–38Basbaum, AL Fields, HL
Zhang C, Davies MF, Guo TZ, Maze M: The analgesic action of nitrous oxide is dependent on the release of norepinephrine in the dorsal horn of the spinal cord. A nesthesiology 1999; 91: 1401–7Zhang, C Davies, MF Guo, TZ Maze, M
Hashimoto T, Maze M, Ohashi Y, Fujinaga M: Nitrous oxide activates GABAergic neurons in the spinal cord in Fischer rats. A nesthesiology 2001; 95: 463–9Hashimoto, T Maze, M Ohashi, Y Fujinaga, M
Fender C, Fujinaga M, Maze M: Strain differences in the antinociceptive effect of nitrous oxide on the tail flick test in rats. Anesth Analg 2000; 90: 195–9Fender, C Fujinaga, M Maze, M
Paxinos G, Watson C: The rat brain in stereotaxic co-ordinates, 2nd edition. Sydney, Academic Press, 1986
Hough LB, Nalwalk JW, Leurs R, Menge WM, Timmerman H: Significance of GABAergic systems in the action of improgan, a non-opioid analgesic. Life Sci 2001; 68: 2751–7Hough, LB Nalwalk, JW Leurs, R Menge, WM Timmerman, H
Hylden JL, Wilcox GL: Intrathecal morphine in mice: a new technique. Eur J Pharm 1980; 67: 313–6Hylden, JL Wilcox, GL
Mestre C, Pelissier T, Fialip J, Wilcox G, Eschalier A: A method to perform direct transcutaneous intrathecal injection in rats. J Pharm Toxicol Meth 1994; 32: 197–200Mestre, C Pelissier, T Fialip, J Wilcox, G Eschalier, A
Monnikes H, Lauer G, Bauer C, Tebbe J, Zittel TT, Arnold R: Pathways of Fos expression in locus ceruleus, dorsal vagal complex, and PVN in response to intestinal lipid. Am J Physiol 1997; 273: 2059–71Monnikes, H Lauer, G Bauer, C Tebbe, J Zittel, TT Arnold, R
Hahn MK, Bannon MJ: Stress-induced c-fos  expression in the rat locus coeruleus is dependent on neurokinin 1 receptor activation. Neuroscience 1999; 94: 1183–8Hahn, MK Bannon, MJ
Li C, Chen P, Smith MS: Corticotropin releasing hormone neurons in the paraventricular nucleus are direct targets for neuropeptide Y neurons in the arcuate nucleus: an anterograde tracing study. Brain Res 2000; 854: 122–9Li, C Chen, P Smith, MS
Hunt SP, Pini A, Evan G: Induction of c-fos  -like protein in spinal cord neurons following sensory stimulation. Nature 1987; 328: 632–4Hunt, SP Pini, A Evan, G
Harris JA: Using c-fos  as a neural marker of pain. Brain Res Bull 1998; 45: 1–8Harris, JA
Presley RW, Menetrey D, Levine JD, Basbaum AI: Systemic morphine suppresses noxious stimulus-evoked Fos protein-like immunoreactivity in the rat spinal cord. J Neurosci 1990; 10: 323–35Presley, RW Menetrey, D Levine, JD Basbaum, AI
Guo TZ, Poree L, Golden W, Stein J, Fujinaga M, Maze M: Antinociceptive response to nitrous oxide is mediated by supraspinal opiate and spinal alpha 2 adrenergic receptors in the rat. A nesthesiology 1996; 854: 846–52Guo, TZ Poree, L Golden, W Stein, J Fujinaga, M Maze, M
Sawamura S, Kingery WS, Davies MF, Agashe GS, Clark JD, Kobilka BK, Hashimoto T, Maze M: Antinociceptive action of nitrous oxide is mediated by stimulation of noradrenergic neurons in the brainstem and activation of α2Badrenoceptors. J Neurosci 2000; 20: 9242–51Sawamura, S Kingery, WS Davies, MF Agashe, GS Clark, JD Kobilka, BK Hashimoto, T Maze, M
Takano M, Takano Y, Yaksh TL: Release of calcitonin gene-related peptide (CGRP), substance P (SP), and vasoactive intestinal polypeptide (VIP) from rat spinal cord: modulation by alpha 2 agonists. Peptides 1993; 14: 371–8Takano, M Takano, Y Yaksh, TL
Murata K, Nakagawa I, Kumeta Y, Kitahata LM, Collins JG: Intrathecal clonidine suppresses noxiously evoked activity of spinal wide dynamic range neurons in cats. Anesth Analg 1989; 69: 185–91Murata, K Nakagawa, I Kumeta, Y Kitahata, LM Collins, JG
Orii R, Ohashi Y, Guo T, Laura E. Nelson, Hashimoto T, Maze M, Fujinaga M: Evidence for the involvement of spinal cord α1 adrenoceptors in nitrous oxide-induced antinociceptive effects of in Fischer rats. A nesthesiology 2002; 97: 1458–65Orii, R Ohashi, Y Guo, T Laura E, Nelson Hashimoto, T Maze, M Fujinaga, M
Charney DS, Mihic SJ, Harris RA: Hypnotics and sedatives, Goodman & Gilman's The Pharmacological Basis of Therapeutics, 10th Ed. Edited by Hardman JG, Limbird LE. New York, McGraw-Hill, 2001, pp. 399–427
Clark FM, Proudfit HK: The projection of noradrenergic neurons in the A7 catecholamine cell group to the spinal cord in the rat demonstrated by anterograde tracing combined with immunocytochemistry. Brain Res 1991; 547: 279–88Clark, FM Proudfit, HK
Proudfit HK, Clark FM: The projections of locus coeruleus neurons to the spinal cord. Prog Brain Res 1991; 88: 123–41Proudfit, HK Clark, FM
Fritsky JM, Grazanna R: Demonstration of two separate descending noradrenergic pathways to the rat spinal cord: evidence for an intragriseal trajectory of locus coeruleus axons in the superficial layers of the dorsal horn. J Comp Neurol 1990; 291: 553–82Fritsky, JM Grazanna, R
Clark FM, Proudfit HK: Anatomical evidence for genetic differences in the innervation of the rat spinal cord by noradrenergic locus coeruleus neurons. Brain Res 1992; 59: 44–53Clark, FM Proudfit, HK
Clark FM, Proudfit HK: The projections of noradrenergic neurons in the A5 catecholamine cell group to the spinal cord in the rat: anatomical evidence that A5 neurons modulate nociception. Brain Res 1993; 616: 200–10Clark, FM Proudfit, HK
Sluka KA, Westlund KN: Spinal projections of the locus coeruleus and the nucleus subcoeruleus in the Harlan and the Sasco Sprague-Dawley rat. Brain Res 1992; 579: 67–73Sluka, KA Westlund, KN
Dirig DM, Yaksh TL: Intrathecal baclofen and muscimol, but not midazolam, are antinociceptive using the rat-formalin model. J Pharm Exp Ther 1995; 275: 219–27Dirig, DM Yaksh, TL
Luger TJ, Hayashi T, Lorenz IH, Hill HF: Mechanisms of the influence of midazolam on morphine antinociception at spinal and supraspinal levels in rats. Eur J Pharm 1994; 271: 421–31Luger, TJ Hayashi, T Lorenz, IH Hill, HF
Luger TJ, Hayashi T, Weiss CG, Hill HF: The spinal potentiating effect and the supraspinal inhibitory effect of midazolam on opioid-induced analgesia in rats. Eur J Pharm 1995; 275: 153–62Luger, TJ Hayashi, T Weiss, CG Hill, HF
Daniels S, Smith EB: Effects of general anaeshetics on ligand-gated ion channels. Br J Anaesth 1993; 71: 59–64Daniels, S Smith, EB
Goto T, Marota JJ, Crosby G: Nitrous oxide induces preemptive analgesia in the rat that is antagonized by halothane. A nesthesiology 1994; 80: 409–16Goto, T Marota, JJ Crosby, G
Goto T, Marota JJ, Crosby G: Volatile anaesthetics antagonize nitrous oxide and morphine-induced analgesia in the rat. Br J Anaesth 1996; 76: 702–6Goto, T Marota, JJ Crosby, G
Janiszewski DJ, Galinkin JL, Klock PA, Coalson DW, Pardo H, Zacny JP: The effects of subanesthetic concentrations of sevoflurane and nitrous oxide, alone and in combination, on analgesia, mood, and psychomotor performance in healthy volunteers. Anesth Analg 1999; 88: 1149–54Janiszewski, DJ Galinkin, JL Klock, PA Coalson, DW Pardo, H Zacny, JP
Tatsuo MA, Salgado JV, Yokoro CM, Duarte ID, Francischi JN: Midazolam-induced hyperalgesia in rats: modulation via GABAAreceptors at supraspinal level. Eur J Pharm 1999; 370: 9–15Tatsuo, MA Salgado, JV Yokoro, CM Duarte, ID Francischi, JN
Hugo S, Lake APJ, Cooledge BA: Are barbiturates hyperalgesic? Anesthesiology 1994; 80: 1413–4Hugo, S Lake, APJ Cooledge, BA
Ewen A, Archer DP, Samanani N, Roth SH: Hyperalgesia during sedation: effects of barbiturates and propofol in the rat. Can J Anaesth 1995; 42: 532–40Ewen, A Archer, DP Samanani, N Roth, SH
Tatsuo MA, Yokoro CM, Salgado JV, Pesquero SM, Santana MA, Francischi JN: Hyperalgesic effect induced by barbiturates, midazolam and ethanol: pharmacological evidence for GABAAreceptor involvement. Braz J Med Biol Res 1997; 30: 251–6 GABAergic neurons modulate the antinociceptive effect of N2O differentially at supraspinal and spinal levels. Midazolam attenuates the antinociceptive effect of N2O at the supraspinal level.Tatsuo, MA Yokoro, CM Salgado, JV Pesquero, SM Santana, MA Francischi, JN
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e.  , systemically).
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e. 
	, systemically).
Fig. 1. Putative neuronal pathways involved in the antinociceptive effect of N2O in rats. Closed triangles indicate excitatory synapses, and open triangles indicate inhibitory synapses. Small closed circles indicate the nuclei of cells activated by N2O exposure or nociception, and small open circles indicate the nuclei of cells inactivated by N2O exposure. In the present study, muscimol, a GABAAreceptor agonist, was administered intracerebroventricularlly (supraspinally), gabazine, a GABAAreceptor antagonist, was administered intrathecally (spinally), and midazolam, a benzodiazepine receptor agonist with GABA-mimetic properties, was administered intraperitoneally (i.e.  , systemically).
×
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P  < 0.05 versus  saline/air group. #P  < 0.05 versus  saline/N2O group.
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P 
	< 0.05 versus 
	saline/air group. #P 
	< 0.05 versus 
	saline/N2O group.
Fig. 2. The effect of midazolam on N2O-induced c-Fos positive cells in the spinal cord. Rats were administered either saline (0.5 ml) or midazolam (1, 5, and 10 mg/kg/0.5 ml) intraperitoneally and 15 min later were exposed to either air or 75% N2O/25% O2for 90 min. The spinal cord was harvested and the lumbar spine was stained for c-Fos and counted in the laminae. The numbers in the parentheses indicate the dosages of midazolam (mg/kg) injection. Each bar indicates mean ± SD (n = 4 each). *P  < 0.05 versus  saline/air group. #P  < 0.05 versus  saline/N2O group.
×
Table 1. Effects of Supraspinally Administered Muscimol on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 1. Effects of Supraspinally Administered Muscimol on N2O-induced Antinociceptive Effect by the Plantar Test
×
Table 2. Effects of Intrathecally Administered Gabazine on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 2. Effects of Intrathecally Administered Gabazine on N2O-induced Antinociceptive Effect by the Plantar Test
×
Table 3. Effects of Systemically Administered Midazolam on N2O-induced Antinociceptive Effect by the Plantar Test
Image not available
Table 3. Effects of Systemically Administered Midazolam on N2O-induced Antinociceptive Effect by the Plantar Test
×
Table 4. Effects of Systemically Administered Midazolam on N2O-induced c-Fos Expression in Noradrnergic Cells (DBH Positive Cells) in the A7 Nuclei
Image not available
Table 4. Effects of Systemically Administered Midazolam on N2O-induced c-Fos Expression in Noradrnergic Cells (DBH Positive Cells) in the A7 Nuclei
×