- Email: [email protected]
The role of spinal nitric oxide and glutamate in nociceptive behaviour evoked by high-dose intrathecal morphine in rats
Pain 106 (2003) 269–283 www.elsevier.com/locate/pain
The role of spinal nitric oxide and glutamate in nociceptive behaviour evoked by high-dose intrathecal morphine in rats Chizuko Watanabea, Tsukasa Sakuradab,*, Kazuhiro Okudac, Chikai Sakuradab, Ryuichiro Andoa, Shinobu Sakuradad a
Center for Laboratory Animal Science, Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan Department of Biochemistry, Daiichi College of Pharmaceutical Sciences, 22-1 Tamagawa-cho, Minami-ku, Fukuoka 815-8511, Japan c Department of Internal Medicine, Mogi Hospital, 6-1-20 Meinohama, Nishi-ku, Fukuoka 819-0002, Japan d Department of Physiology and Anatomy, Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan
b
Received 23 January 2003; received in revised form 24 June 2003; accepted 9 July 2003
Abstract Injection of high-dose of morphine into the spinal lumbar intrathecal (i.t.) space of rats elicits a nociceptive behavioural syndrome characterized by periodic bouts of spontaneous agitation and severe vocalization. The induced behavioural response such as vocalization and agitation was observed dose-dependently by i.t. administration of morphine (125 – 500 nmol). Pretreatment with naloxone (s.c. and i.t.), an opioid receptor antagonist, failed to reverse the morphine-induced behavioural response. The excitatory effect of morphine was inhibited dose-dependently by pretreatment with 3-((þ )2-carboxy-piperazin-4-yl)-propyl-1-phosphonic acid (CPP), a competitive N-methyl-D aspartate (NMDA) receptor antagonist and MK-801, a non-competitive NMDA receptor antagonist. The non-selective nitric oxide (NO) synthase inhibitor N G-nitro L -arginine methyl ester (L -NAME) inhibited dose-dependently the behavioural response to high-dose i.t. morphine (500 nmol), whereas D -NAME was without affecting the response to high-dose i.t. morphine. In the present study, we measured NO metabolites (nitrite/nitrate) in the extracellular fluid of rat dorsal spinal cord using in vivo microdialysis. The i.t. injection of morphine (500 nmol) evoked significant increases in NO metabolites and glutamate from the spinal cord. Not only NO metabolites but also glutamate released by high-dose morphine were reduced significantly by pretreatment with L -NAME (400 nmol). Pretreatment with CPP and MK-801 showed a significant reduction of the NO metabolites and glutamate levels elevated by high-dose i.t. morphine. These results suggest that the excitatory action of high-dose i.t. morphine may be mediated by an NMDA– NO cascade in the spinal cord. q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: High-dose morphine; Intrathecal injection; Nitric oxide; Glutamate; Excitatory behavioural response (vocalization and agitation); Spinal microdialysis
1. Introduction There is ample evidence to suggest that the dorsal spinal cord is an important site for the mediation of the antinociceptive effect of morphine and opioids (Yaksh, 1981; Akil et al., 1984; Basbaum and Fields, 1984). It is well documented that in the formalin-induced flinching test intrathecal (i.t.) administration of morphine as well as other opioid drugs into rats over a dose range of 1.0 –20 nmol produces a reliable antinociceptive effect by interacting with opioid receptors in the spinal cord. In contrast, previous studies have demonstrated that morphine at doses (150 mg * Corresponding author. Tel.: þ 81-92-541-0161; fax: þ81-92-553-5698. E-mail address: [email protected] (T. Sakurada).
or 520 nmol) far higher than those required for antinociception, injected i.t. into the spinal subarachnoid space, produces a profound spontaneous agitation and squeaking in rats (Tang and Schoenfelt, 1978; Woolf, 1981; Yaksh et al., 1986; Alvarez-Vega et al., 1998). High-doses of spinally injected morphine in the mouse also elicit a nociceptive-related behaviour such as scratching, biting and licking in mice resembling that of i.t.-injected substance P or N-methyl-D -aspartate (NMDA) (Sakurada et al., 1996c). High-dose i.t. morphine exhibits an allodynic response to innocuous tactile stimuli in rats and mice (Hara et al., 1997; Yaksh and Harty, 1988). Additionally, i.t. injections of highdoses of morphine are found to elicit myoclonic seizures in the rat (Shohami and Evron, 1985) or clonic seizure-like
0304-3959/$20.00 q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/S0304-3959(03)00296-3
270
C. Watanabe et al. / Pain 106 (2003) 269–283
behaviours in mice (Lutfy et al., 1994). A hyperalgesic action has also been known to develop after chronically administered morphine in rats (Kayan and Mitchell, 1968; Mule et al., 1968). A similar phenomenon has been recognized in patients who developed myoclonus and hyperalgesia following administration of high-dose subarachnoid morphine (Wert and MacDonald, 1982; Krames et al., 1985; Ali, 1986; Penn and Paice, 1987; Arner et al., 1988). Recently, we have reported that nociceptive behavioural responses consisting of a severe hindlimb scratching followed by biting/licking of the hindpaw could be elicited by i.t. injection of high-dose morphine in mice, which are considered to be mediated by the spinal NMDA receptors (Sakurada et al., 2002). However, linkage between excitatory behavioural responses to high-dose i.t. morphine and changes in concentrations of spinal glutamate and NO needs to be demonstrated. Thus we conducted the present experiments to provide the possible role played by the glutamate receptor–NO system in nociceptive behaviour evoked by high-dose i.t. morphine in rats. An important objective was to explore whether there is an alteration of glutamate and NO metabolites (nitrite/nitrate) in the extracellular fluid of rat dorsal spinal cord using in vivo microdialysis following high-dose i.t. morphine. We also examined the effects of NMDA receptor antagonists and an inhibitor of NO synthase, on spinally mediated behavioural response, and on the release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine in rats.
2. Materials and methods 2.1. Animals Experimental procedures were conducted according to protocols approved by the Committee of Animal Experimentation of Tohoku Pharmaceutical University, Sendai, Japan. Male Sprague– Dawley (SD) rats (SLC, Japan), weighing 250 –260 g, were housed with free access to food and water. They were maintained in a controlled environment (12 h light/dark cycle, room temperature 23 8C, 50 –60% relative humidity). All experiments took place during the light period between 10:00 and 17:00 hours in a quiet room. The animals belonging to the various treatment groups (n ¼ 7 each group) were tested in randomized order. 2.2. Intrathecal injection To permit application of morphine, N G-nitro L -arginine methyl ester (L -NAME), D -NAME, 3-((þ )2-carboxy-piperazin-4-yl)-propyl-1-phosphonic acid (CPP) and MK-801, directly into the spinal subarachnoid space, rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and chronically implanted with i.t. catheters according to the previously reported method (Sakurada et al., 1999).
Briefly, this involved inserting a length of polyethylene tubing (PE-10) following laminectomy between L3 and L4 and carefully placing the catheter tip in the subarachnoid space of L5 and L6. The animals were allowed to recover over 7 days following implantation of the catheter. The catheter was filled with sterile artificial cerebrospinal fluid (ACSF), containing (in g/L) NaCl 7.4, KCl 0.19, MgCl2 0.19, and CaCl2 0.14. Drugs were administered in volumes of 10 ml followed by a flush of 15 ml of ACSF to ensure that each compound reached the spinal cord. L -NAME, D -NAME, CPP and MK-801 were administered i.t. 10 min prior to high-dose i.t. morphine. Rats were pretreated subcutaneously (s.c.) with naloxone (4.0 mg/kg) 10 min before i.t. administration of morphine (500 nmol). Naloxone (150 nmol) was also injected i.t. 2 min prior to spinal administration of morphine. 2.3. Behavioural observation To perform the behavioural observation after high-dose i.t. morphine, the rats were placed individually in an open Plexiglas chamber (34 £ 30 £ 17 cm), which also served as an observation chamber after injection. The animals were habituated to the observation chamber 1 h before actual experimental sessions. Immediately after the high-dose morphine was injected i.t. using a microsyringe with a 26gauge needle, each rat was returned to the observation chamber. High-dose morphine, injected i.t., produced a striking behavioural syndrome consisting of spontaneous agitation and vocalization. The magnitudes of two behavioural responses were quantified: agitation and vocalization. Spontaneous agitation was ranked visually every 1 min after i.t. administration of high-dose morphine as: 0, no sign of excitation; 1, restlessness, scratching and biting at the flank or tail; 2, mild vocalization with restlessness, scratching and biting at the flank or tail; 3, vocalization with spontaneous running and circling; and 4, vigorous vocalization with running, circling, rolling and jumping. The score of agitation was expressed in each animal during each 5-min interval for a 40-min period after high-dose i.t. morphine. The second parameter studied was the appearance of spontaneous vocalization. The total duration of vocalization was recorded by the stop-watch during the 5-min interval for the 40-min period. The latency to induce the first vocalization after i.t. administration of morphine was also recorded. In the behavioural study, one person prepared the drugs and another, blind to the drugs and dosage, ran the behavioural observation. The blind observer was identical throughout this study. 2.4. Microdialysis and implantation An in vivo i.t. microdialysis technique was performed according to the modified method of Marsala et al. (1995). Rats were implanted with a U-shaped microdialysis cannula through a laminectomy between L1 and L2 under anesthesia
C. Watanabe et al. / Pain 106 (2003) 269–283
with sodium pentobarbital (50 mg/kg, i.p.). The tip of the cannula was positioned in the subarachnoid space of L5 and L6. The dialysis catheters were then externalized on the back of the neck in rats. The microdialysis cannula consisted of a 6 cm length of microdialysis fiber (200 mm inner diameter, 300 mm outer diameter, 50 kDa molecular weight cut-off; Filtral, AN69-HF). The fiber was coated with a thin layer of epoxy (Araldite, CIBA GEIGY Inc.) except for a 4 cm portion in the middle. In order to make the fiber firm enough for implantation, a Nichrome – Formvar wire (0.0026 inch inner diameter; A-M systems, Inc., Everret, WA) was then passed through the fiber and the ends of the fiber were attached to a PE-10 (5.0 cm long) with cyanoacrylate. The fiber was then bent so that a U-shaped loop was formed in the middle of the uncoated portion. Following each experiment, all animals were deeply anesthetized and the localization of the microdialysis cannula was verified visually. Only animals with the cannula placed at the L5 and L6 spinal segments were included in this study. 2.5. Sample collection and chemical analysis After allowing 3– 5 days for recovery from i.t. implantation of microdialysis catheters, rats were re-anesthetized with isoflurane (3.3 –3.5% for induction, 1.5– 2.0% for surgery and 1.0 –1.1% for the subsequent dialysis experiment in a 50:50 air/O2 mixture) and the dialysis experiment was started. The dialysis tubing was connected to the syringe pump (EO-60, EICOM) and perfused with ACSF 30 min prior to sample collection to establish a diffusion equilibrium. The flow rate was 10 ml/min. Baseline concentrations were defined as the mean of two 10 min control samples taken after the initial 30 min washout period. Following baseline periods, samples of fluid were collected in polypropylene tubes after i.t. administration of each compound for the duration of the experiment. Dialysate samples were separated for measuring nitrite/nitrate (aliquot 50 ml) and glutamate (residue 50 ml), and maintained at 2 70 8C until chemical analysis. NO was measured as its break products, nitrite (NO2 2 ) and nitrate (NO2 ). The measurement of nitrite/nitrate was 3 made using a commercially available NO analyser (MODEL-280NOA, Sievers, Inc.) and integrator (HO3396, Sievers, Inc.). External standards were run at the beginning of sample analysis. The limit of detection of nitrite/nitrate was 1 –2 pmol/30 ml. The amino acid glutamate in the spinal perfusate was detected by reversed-phase high-performance liquid chromatography (HPLC) (ECD-300, EICOM) with fluorimetric detection following pre-column derivatization with ophthaldialdehyde (OPA). Chromatography was performed on a reversed-phase C-18 column (2:1 £ 150 mm, EICOM PACKED COLUMN) using a pH sodium acetate methanol gradient. A known concentration of glutamate was used as an internal standard. External standards were run at least
271
every five to six samples to ensure accuracy and for calculations of concentrations. The minimal detectable concentration of glutamate was 5– 10 pmol/30 ml. 2.6. Drugs Drugs used and their sources were: morphine hydrochloride (Sankyo, Tokyo, Japan), naloxone hydrochloride (Sigma Chemical Co., St. Louis, MO), L -NAME, D -NAME (Nakalai tesque, Kyoto, Japan), CPP, (5R,10S)-(þ )-5methyl-10,11-dihydro-5H-dibenzo [a,d ] cyclohepten-5, and 10-imine maleate (MK-801) (Research Biochemical Incorporated, Natic, MA). All drugs for spinal administration were dissolved in ACSF. Dosages reflect the salt form of the drugs. 2.7. Data analysis and statistics The time –response data of the excitatory behaviours (agitation and vocalization) induced by different doses of morphine or in combination with compounds are presented as the mean ^ SEM of agitation scores and accumulated vocalization time (s). The spinal release of nitrite/nitrate and glutamate is presented as the mean ^ SEM percentage changes from baseline. Statistical analysis of behavioural and biochemical data was carried out with Bonferroni’s and Dunnett’s test following two-way or two-factor analysis of variance (ANOVA) for repeated measures. Differences are considered to be significant when the critical value reaches a level of P , 0:05.
3. Results 3.1. Spontaneous agitation and vocalization evoked by high-dose i.t. morphine The i.t. administration of morphine at doses ranging from 125 to 500 nmol into the spinal lumbar space evoked a spontaneous agitation and vocalization. A dose-dependent response was seen by measuring these two behavioural items (Fig. 1). The highest dose (500 nmol) of morphine evoked a spontaneous agitation such as vigorous vocalization accompanied by hindlimb scratching toward the flanks and biting of hindpaw or the base of the tail, running and jumping, and occasionally myoclonic seizures. Agitation evoked by i.t. morphine (500 nmol) peaked at 5 –20 min and lasted for approximately 40 min post-injection, whereas 125 and 250 nmol of morphine exhibited a shorter duration of agitation with lower scores (Fig. 1A). Morphine at doses of 125 and 250 nmol caused a weak agitation, which peaked at 5 –10 min and decayed at 30 – 35 min. As shown in Fig. 1B, time-course of morphine (500 nmol)-evoked vocalization was similar to that of agitation by i.t. morphine at a dose of 250 nmol; vocalization evoked by morphine was maximum at 5 –10 min, and disappeared at 30 –35 min. Accumulated
272
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 1. Time-course effect of agitation (A) and vocalization (B) following intrathecal (i.t.) administration of morphine (125–500 nmol) in rats. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats.
vocalization time for the 40-min observation after i.t. administration of 500 nmol morphine was 435:0 ^ 10:2 s (Fig. 1B). As shown in Table 1, the vocalization response was late in onset; even at a dose of 500 nmol the latency to elicit the first vocalization after i.t. injection was over 60 s. No agitation and vocalization responses were observed in rats receiving an antinociceptive dose of 40 nmol morphine or i.t. ACSF (data not shown).
In order to examine the non-specific effect from a high concentration of hydrochloride salt containing in 500 nmol morphine, animals were treated with 0.5% hydrochloride (n ¼ 3) as a control with a comparable concentration to 500 nmol morphine and 2.5% hydrochloride (n ¼ 3) in a fivefold higher concentration, respectively. Treatment with these concentrations of hydrochloride produced no agitation and vocalization like high-dose morphine.
C. Watanabe et al. / Pain 106 (2003) 269–283 Table 1 The latency to first vocalization following intrathecal administration of high-dose morphine or in combination with naloxone, L -NAME, D -NAME, MK-801 and CPP in rats Treatment Morphine 125 nmol 250 nmol 500 nmol ACSF plus morphine 500 nmol Naloxone (4.0 mg/kg) plus morphine Naloxone (150 nmol) plus morphine L -NAME (100 nmol) plus morphine L -NAME (200 nmol) plus morphine L -NAME (400 nmol) plus morphine D -NAME (400 nmol) plus morphine CPP (1.0 nmol) plus morphine CPP (2.0 nmol) plus morphine CPP (4.0 nmol) plus morphine MK-801 (10 nmol) plus morphine MK-801 (20 nmol) plus morphine MK-801 (40 nmol) plus morphine
Administration route
Latency (s)
i.t. i.t. i.t. i.t.
233.8 ^ 35.6 146.9 ^ 11.8 69.3 ^ 5.5 67.2 ^ 5.5
s.c. þ i.t.
69.8 ^ 11.9
i.t. þ i.t.
60.5 ^ 2.9
i.t. þ i.t.
63.6 ^ 8.1
i.t. þ i.t.
64.1 ^ 8.7
i.t. þ i.t.
61.9 ^ 5.0
i.t. þ i.t.
62.8 ^ 5.6
i.t. þ i.t.
125.7 ^ 14.7*
i.t. þ i.t.
143.4 ^ 13.8**
i.t. þ i.t.
160.0 ^ 14.6**
i.t. þ i.t.
64.3 ^ 5.5
i.t. þ i.t.
79.1 ^ 3.9
i.t. þ i.t.
113.6 ^ 8.4**
Rats were pretreated with naloxone at a dose of 4.0 mg/kg (s.c.) or 150 nmol (i.t.) 10 min before i.t. administration of morphine (500 nmol). L NAME, D -NAME, MK-801 or CPP was administered i.t. 10 min prior to i.t. morphine (500 nmol). Asterisks indicate a significant increase compared to i.t. morphine (500 nmol) alone (Dunnett’s test; * P , 0:01, * * P , 0:05).
3.2. Spinal release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine Basal levels of NO breakdown products, nitrite/nitrate and glutamate were detectable in all animals treated under isoflulane throughout the measurement of nitrite/nitrate and glutamate. The time-courses of the mean basal level of nitrite/nitrate are presented in Fig. 2 (upper panel). The means of baseline samples collected before i.t. administration of ACSF were used for determination of nitrite/ nitrate and glutamate in the dorsal spinal cord extracellular fluid. The absolute amounts of nitrite/nitrate collected during the baseline period were 2:90 ^ 0:22 nmol/30 ml (n ¼ 7) and 2:55 ^ 0:15 nmol/30 ml (n ¼ 7) in rats injected with 250 nmol and 500 nmol morphine, respectively. The baseline control levels of nitrite/nitrate were not altered by i.t. administration of ACSF (Table 2). A high-dose of morphine (500 nmol) evoked a significant increase of nitrite/nitrate release in the time interval of 10– 30 min
273
post-i.t. administration, whereas 250 nmol morphine was without effect (Fig. 2, upper panel). The baseline concentrations of glutamate examined were 0:62 ^ 0:12 nmol/30 ml (n ¼ 7) and 0:54 ^ 0:07 nmol/30 ml (n ¼ 7) in rats injected with 250 nmol and 500 nmol morphine, respectively. The baseline control levels of glutamate were not altered by i.t. administration of ACSF (Table 2). Morphine at a dose of 500 nmol, injected i.t., produced over a 2.0-fold increase of glutamate during the first 20 min when compared to ACSF injection (Fig. 2, lower panel). A significant increase of glutamate was also observed from 20 to 40 min following injection of 500 nmol morphine, whereas the increased level of glutamate was elevated only to a smaller degree. The concentration of glutamate increased significantly during the first 10 min after i.t. administration of morphine at a dose of 250 nmol, though there was no significant increase of the glutamate level in the time interval of 10 –40 min. An antinociceptive dose (40 nmol) of morphine gave no effect on the release of nitrite/nitrate and glutamate. 3.3. Effect of naloxone, an opioid receptor antagonist, on the spontaneous behaviour and on the spinal release of nitrite/nitrate and glutamate induced by high-dose i.t. morphine Peripheral and spinal pretreatment with naloxone (4.0 mg/kg, s.c.; 150 nmol, i.t.), an opioid receptor antagonist, gave no reversed effect on the excitatory responses, vocalization and agitation evoked by i.t. morphine (500 nmol), and the latency to the first vocalization (Fig. 3, Table 1). Biochemical analysis revealed that i.t. pretreatment with naloxone (150 nmol, i.t.) 10 min prior to morphine (500 nmol, i.t.) produced no significant effect on morphine-evoked release of nitrite/nitrate and glutamate (Fig. 2). 3.4. Effect of NO synthase inhibitor and NMDA receptor antagonists on the spontaneous behaviour induced by high-dose i.t. morphine Pretreatment with L -NAME (100 –400 nmol, i.t.) caused a dose-related inhibition of the agitation and vocalization response to i.t. morphine (Fig. 4A and B). The highest dose (400 nmol) of L -NAME used in these experiments was most effective in inhibiting these behavioural responses. The excitatory behavioural response was not affected by pretreatment with D -NAME (400 nmol, i.t.). None of the doses of L -NAME (100, 200 and 400 nmol) or D -NAME (400 nmol) tested altered the latency to first vocalization response evoked by i.t. high-dose morphine (Table 1). Morphine (500 nmol)-evoked agitation and vocalization were inhibited dose-dependently by pretreatment with CPP (1.0 –4.0 nmol), a competitive NMDA antagonist (Fig. 5A and B). Similarly, a dose – response inhibition of MK-801 (10 –40 nmol), a non-competitive NMDA antagonist, was
274
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 2. Time-course of spinal release of nitrite/nitrate (A) and glutamate (B) following intrathecal (i.t.) administration of morphine and in combination with naloxone, an opioid receptor antagonist in rats. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline after i.t. injection of ACSF or morphine (500, 250 and 40 nmol) or in combination with naloxone (150 nmol). Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant increase compared to time-matched ACSF (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
275
Table 2 Effect of L -NAME, CPP and MK-801 on resting release of nitrite/nitrate and glutamate in the lumbar intrathecal space as measured by microdialysis in rats Substances measured
Time period (min)
Mean % of baseline ACSF control
L -NAME
(400 nmol)
Nitrite/nitrate
0– 10 10– 20 20– 30 30– 40
87.5 ^ 5.2 88.8 ^ 1.4 89.8 ^ 3.0 86.8 ^ 5.4
82.3 ^ 13.3 98.3 ^ 2.7 88.0 ^ 8.1 100.3 ^ 3.2
Glutamate
0– 10 10– 20 20– 30 30– 40
95.3 ^ 2.2 92.5 ^ 1.7 88.8 ^ 5.6 76.5 ^ 2.7
95.0 ^ 1.0 79.7 ^ 9.0 81.7 ^ 3.3 82.0 ^ 2.6
CPP (4.0 nmol)
MK-801 (40 nmol)
99.7 ^ 9.8 97.0 ^ 6.1 93.7 ^ 2.7 101.0 ^ 5.7
99.0 ^ 7.1 99.3 ^ 9.3 102.8 ^ 12.6 86.5 ^ 8.7
85.8 ^ 7.8 86.5 ^ 6.2 81.3 ^ 4.9 84.8 ^ 12.1
95.3 ^ 4.2 82.5 ^ 12.7 79.8 ^ 11.5 86.3 ^ 12.8
Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage of baseline after i.t. injection of ACSF, L -NAME (400 nmol), CPP (4.0 nmol) and MK-801 (40 nmol).
constructed in morphine-evoked agitation and vocalization (Fig. 6A and B). The appearance of the first vocalization response was prolonged by both NMDA antagonists (Table 1). General behaviour was not recognized in rats pretreated with L -NAME, CPP or MK-801 at doses employed in the experiment. 3.5. Effect of NO synthase inhibitor and NMDA receptor antagonists on the spinal release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine Pretreatment with L -NAME (400 nmol), a non-selective NO synthase inhibitor, 10 min prior to i.t. administration of morphine (500 nmol), prevented a significant increase of nitrite/nitrate in samples collected in the time interval of 10 –40 min following i.t. administration of morphine (Fig. 7, upper panel). Morphine (500 nmol)-evoked release of nitrite/nitrate was not blocked by pretreatment with 200 nmol L -NAME and 400 nmol D -NAME, an inactive stereoisomer of L -NAME. L -NAME (400 nmol) pretreatment completely blocked the morphine-evoked release of glutamate during 0– 40 min period (Fig. 7, lower panel). A significant block was also observed in the time interval of 20 –30 min by pretreatment with 200 nmol L -NAME but not 400 nmol D -NAME. There were no significant effects of L -NAME on the resting level of nitrite/nitrate and glutamate (Table 2). The i.t. administration of CPP (2.0 and 4.0 nmol) and MK-801 (20 and 40 nmol) prior to i.t. morphine (500 nmol) prevented significantly the increased concentrations of nitrite/nitrate evoked by i.t. morphine (Figs. 8 and 9, upper panels). Pretreatment with 4.0 nmol CPP and 40 nmol MK-801 produced a significant effect on the increased concentrations of glutamate evoked by i.t. morphine (Figs. 8 and 9, lower panels), whereas 2.0 nmol CPP and 20 nmol MK-801 were ineffective in reducing morphine-induced increased concentrations of glutamate. The resting level of
nitrite/nitrate and glutamate was not altered by CPP (2.0 nmol) or MK-801 (40 nmol) (Table 2).
4. Discussion High-dose morphine, injected into the spinal lumbar i.t. space, has been shown to produce an excitatory reaction such as spontaneous agitation and vocalization accompanied by scratching, biting and licking, and touchevoked allodynia in rats and mice (Sakurada et al.,1996c; Hara et al., 1997; Alvarez-Vega et al., 1998; Yaksh and Harty, 1988), as opposed to an antinociceptive action at lower doses ranging from 1.0 to 20 nmol (Yaksh, 1981; Akil et al., 1984; Basbaum and Fields, 1984). Consistent with these previous studies, the present investigation demonstrated that injection of morphine at a high-dose (500 nmol) into the spinal lumbar i.t. space elicited a spontaneous nociceptive syndrome mainly consisting of agitation and vocalization in rats. High-doses of spinally administered morphine have also been reported to produce hyperalgesia in response to noxious stimuli in rats (Woolf, 1981). The morphine-induced hyperalgesia is of clinical interest since hyperalgesic states have been reported to occur in humans following administration of high-dose subarachnoid morphine or systemic administration of opioid (Seymour et al., 1982; Ali, 1986; Arner et al., 1988; De Conno et al., 1991; Sjogtrn et al., 1993). In the present study, we first confirmed that i.t. administrations of 0.5% hydrochloride salt with a comparable concentration to 500 nmol morphine or even 2.5-fold higher concentration produced no observable changes in spontaneous behaviours, vocalization and agitation. The result indicates that high-dose i.t. morphine-induced behavioural responses may not be due to non-specific activation of highly concentrated salt. Behavioural activation evoked by high-dose i.t. morphine are irreversible by the opioid receptor antagonist, naloxone or naltrexone,
276
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 3. Effect of naloxone on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. Naloxone in doses of 4.0 mg/kg (s.c.) and 150 nmol (i.t.) was administered 10 min and 2 min prior to i.t. morphine, respectively. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats.
C. Watanabe et al. / Pain 106 (2003) 269–283
suggesting a non-opioid receptor mechanism (Woolf, 1981; Yaksh et al., 1986; Yaksh and Harty, 1988; Okuda et al., 2001b). The seizure-like motor effects of high-dose i.t. morphine are also irreversible by pretreatment with naloxone in mice (Lutfy et al., 1994). The present study also shows that naloxone administered peripherally and spinally neither antagonized the excitatory behavioural response nor reversed the increased concentrations of nitrite/nitrate and glutamate to high-dose i.t. morphine. It therefore reinforced the concept that the effects of high-dose i.t. morphine are brought about by mechanisms that are not dependent of opioid receptor activity in the spinal cord. It has to be noted that i.t. dynorphin A, a proposed endogenous ligand for the kappa opioid receptor, evokes long-lasting allodynia, and scratching, licking and biting behaviour through NMDA receptors rather than nonopioid mechanisms in rats (Vanderah et al., 1996) and mice (Laughlin et al., 1997; Tan-No et al., 2002). In the present study, CPP, a competitive NMDA receptor antagonist, and MK-801, a non-competitive NMDA receptor antagonist, both dose-dependently inhibited agitation and vocalization induced by high-dose i.t. morphine, suggesting that activation of NMDA receptors in the spinal cord may mediate the excitatory effects of i.t. morphine. This suggestion is supported by the results that high-dose i.t. morphine evoked a marked increase of glutamate in the extracellular levels, which was inhibited by pretreatment with CPP and MK-801. There is a difference of nociceptive behavioural response to high-dose i.t. morphine in rats when compared with mice; in mice high-dose (60 and 90 nmol) i.t. morphine produces a severe hindlimb scratching followed by biting and licking toward the caudal part of the body without squeaking (Sakurada et al., 1996c). Recently, we have shown that the onset of high-dose morphine-induced scratching response was late after i.t. injection in mice. The data observed in mice are compatible with the present study that the latency of vocalization was late (more than 60 s) after 500 nmol morphine in rats. In addition, the characteristic behavioural response, scratching, biting and licking in mice has been found to inhibit dose-dependently by: (1) the NK1 receptor antagonists, sendide and CP-96,345; (2) pretreatment with the non-selective substance P depletor capsaicin; and (3) pretreatment with i.t.-injected antiserum against substance P (Sakurada et al., 1996c). It is therefore inferred that the spinally mediated behavioural response induced by high-dose i.t. morphine may occur indirectly, possibly through the release of substance P in the dorsal horn of the spinal cord. If a similar phenomenon may occur in the case of high-dose morphine injected i.t. into rats, it is reasonable to speculate that the excitatory response evoked by highdose i.t. morphine in rats may be mediated in part through spinal NK1 receptors. This is a reason why morphine-induced behavioural response was not
277
eliminated completely by 4 nmol CPP and 40 nmol MK-801, NMDA receptor antagonists, and 400 nmol L -NAME, an NO synthase inhibitor in the present study. It appears evident that the excitatory effects of high-dose i.t. morphine might be a nociceptive event clearly independent of spinal opioid receptors. Taken together, the behavioural activation of glutamate receptors by high-dose i.t. morphine may, at least, in part be responsible for enhancement of glutamate release in the dorsal horn. There is considerable evidence that NO could possibly contribute to synaptic transmission between primary afferent neurones and dorsal horn cells in spinal nociceptive processing. In the beginning of NO studies in the transmission of pain signals, these studies have indirectly assessed the physiological role of NO using an inhibitor of NO synthase such as L -NAME (Moore et al., 1991; Haley et al., 1992; Sakurada et al., 1996a,b, 2001). Recently, it has been shown that injection of capsaicin into the hindfoot increases NO production in the dialysate of spinal cord (Wu et al., 1998). Studies on the formalin model of pain have indicated that formalin injection into the hindpaw evoked a biphasic increase of NO in the extracellular levels that paralleled largely the early and late phase flinching responses (Okuda et al., 2001a), which was inhibited by pretreatment with L -NAME (unpublished data). The ED 50 value of L -NAME obtained in the late phase of formalin (5.0%)-induced flinching response was 360.0 (83.0 – 1560.6 nmol) nmol. Judging from the ED50 value of L -NAME with 210.0 (129.1 – 341.6 nmol) nmol in vocalization response to i.t. morphine (500 nmol), there is no statistically significant difference of the ED50 values for L -NAME between the two assays. These results support the contention that vocalization evoked by high-dose i.t. morphine may be closely related to nociceptive events in the spinal cord. A possible interaction between NO and opioid systems in the spinal cord has been proposed; the inhibition of NO synthase potentiates antinociception induced by spinal administration of morphine and opioid receptor agonist (Przewlocki et al., 1993; Machelska et al., 1997), and diminishes tolerance to morphine in the antinociceptive action of morphine (Kolesnikov et al., 1992; Majeed et al., 1994; Bhargava and Zhao, 1996). In marked contrast with the above reports, the current study showed that high-dose i.t. morphine produced an excitatory behavioural response (agitation and vocalization) and a significant increase of NO release in the spinal cord. Considering that NO functions as an intermediary of NMDA receptors in the spinal cord, it seems clearly that high-dose i.t. morphine can act as a nociceptive input through activation of NMDA receptors and NO production. This concept is in line with the behavioural data that vocalization response evoked by high-dose i.t. morphine was inhibited more potently than
278
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 4. Effect of L -NAME and D -NAME on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. L -NAME or D -NAME was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
279
Fig. 5. Effect of CPP, a competitive NMDA receptor antagonist, on agitation (A) and vocalization evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. CPP was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
280
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 6. Effect of MK-801, a non-competitive NMDA receptor antagonist, on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. MK-801 was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
281
Fig. 7. Effect of L -NAME and D -NAME on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. L -NAME or D -NAME was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
Fig. 8. Effect of CPP on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of high-dose morphine (500 nmol) in rats. CPP was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
morphine-induced agitation. Previously reported results suggest that NMDA receptor-mediated central sensitization and hyperalgesia involves subsequent production of NO (Meller et al., 1992) and augmented release of glutamate induced by NMDA receptor activation occurs secondary to NO production (Sorkin, 1993). This assumption is supported by the present microdialysis data that morphine-evoked releases of nitrite/nitrate and glutamate were inhibited by pretreatment with
the competitive and non-competitive NMDA antagonists, CPP and MK-801. In summary, the present study demonstrates that highdose i.t. morphine evokes a significant increase of nitrite/nitrate and glutamate in spinal dialysates. Increased release of nitrite/nitrate and glutamate largely corresponded with the appearance of agitation and vocalization. These observations suggest that the excitatory state evoked by high-dose i.t. morphine may be
282
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 9. Effect of MK-801 on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. MK-801 was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
mediated by NMDA-NO cascade, since agitation and vocalization could be reduced by pretreatment with either NMDA receptor antagonists (CPP and MK-801) or an inhibitor of NO synthase (L -NAME).
References Akil H, Watson SJ, Young E, Lewis ME, Khachaturian H, Walker JM. Endogenous opioids: biology and function. Annu Rev Neurosci 1984;7: 223– 55.
Ali NMK. Hyperalgesic response in a patient receiving high concentrations of spinal morphine. Anesthesiology 1986;65:443 –9. Alvarez-Vega M, Baamonde A, Gutierrez M, Hidalgo A, Menendez L. Comparison of the effects of calmidazolium and bupivacaine on N-methyl-D -aspartate- and septide-induced nociceptive behaviour. Naunyn Schmiedebergs Arch Pharmacol 1998;358:628–34. Arner S, Rawal N, Gustafsson LL. Clinical experience of long-term treatment with epidural opioids—a nationwide survey. Acta Anaesthesiol Scand 1988;32:253–9. Basbaum AL, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7: 309 –38. Bhargava HN, Zhao GM. Effects of nitric oxide synthase inhibition on tolerance to the analgesic action of D-Pen2, D-Pen5 enkephalin and morphine in the mouse. Neuropeptides 1996;30:219–23. De Conno F, Caraceni A, Martini C, Spoldi E, Salvetti M, Ventafridda V. Hyperalgesia and myoclonus with intrathecal infusion of high-dose morphine. Pain 1991;47:337–9. Haley JE, Dickenson AH, Schachter M. Electrophysiological evidence for a role of nitric oxide in prolonged chemical nociception in the rat. Neuropharmacology 1992;31:251–8. Hara N, Minami T, Okuda-Ashitaka E, Sugimoto T, Sakai M, Onaka M, Mori H, Imanishi T, Shingu K, Ito S. Characterization of nociception hyperalgesia and allodynia in conscious mice. Br J Pharmacol 1997;121:401–8. Kayan S, Mitchell CL. The effects of chronic morphine administration of tooth pulp threshold in dogs and cats. Proc Soc Exp Biol Med 1968;128: 755 –60. Kolesnikov YA, Pick CG, Pasternak GW. N G-nitro-L -arginine prevents morphine tolerance. Eur J Pharmacol 1992;221:399– 400. Krames ES, Gershow J, Glassberg A, Kenefick T, Lyons A, Taylor P, Wilkie D. Continuous infusion of spinally administered narcotics for pain due to malignant disorder. Cancer 1985;56:696–702. Laughlin TM, Vanderah T, Lashbrook JM, Nichols ML, Ossipov MH, Porreca F, Wilcox GL. Spinally administered dynorphin A produces long-lasting allodynia: involvement of NMDA but not opioid receptors. Pain 1997;72:253–60. Lutfy K, Woodward RM, Keana JFW, Weber E. Inhibition of clonic seizure-like excitatory effects induced by intrathecal morphine using two NMDA receptor antagonists: MK-801 and ACEA-1011. Eur J Pharmacol 1994;252:261–6. Machelska H, Labuz D, Przewlocki R, Przewlocka B. Inhibition of nitric oxide synthase enhances antinociception mediated by mu, delta and kappa opioid receptors in acute and prolonged pain in the rat spinal cord. J Pharmacol Exp Ther 1997;282:977– 84. Majeed NH, Przeweocka B, Machelska H, Przewlocki R. Inhibition of nitric oxide synthase attenuates the development of morphine tolerance and dependence in mice. Neuropharmacology 1994;33: 189 –92. Marsala M, Malmberg AB, Yaksh TL. The spinal loop dialysis catheter: characterization of use in the unanesthetized rat. J Neurosci Methods 1995;62:43–53. Meller ST, Dykstra C, Gebhart GF. Production of endogenous nitric oxide and activation of soluble guanylate cyclase are required for N-methylD -aspartate-produced facilitation of the nociceptive tail-flick reflex. Eur J Pharmacol 1992;214:93–6. Moore PK, Oluyomi AO, Babbedge RC, Wallace P, Hart SL. L -N Gnitroarginine methyl ester exhibits antinociceptive activity in the mouse. Br J Pharmacol 1991;102:198–202. Mule SJ, Clements TH, Layson RC, Haertzen CA. Analgesia in guinea pigs: a measure of tolerance development. Arch Int Pharmacodyn Ther 1968;173:210–2. Okuda K, Sakurada C, Takahashi M, Yamada T, Sakurada T. Characterization of nociceptive responses and spinal releases of nitric oxide metabolites and glutamate evoked by different concentrations of formalin in rats. Pain 2001a;92:107 –15. Okuda K, Watanabe C, Ando R, Sakurada C, Takahashi M, Yamada T, Sakurada T. The role of spinal nitric oxide in nociceptive behavior
C. Watanabe et al. / Pain 106 (2003) 269–283 evoked by high dose intrathecal morphine in rats. Jpn J Pharmacol 2001b;85:218. Penn R, Paice J. Chronic intrathecal morphine for intractable pain. J Neurosurg 1987;67:182– 6. Przewlocki R, Machelska H, Przewlocka B. Inhibition of nitric oxide synthase enhances morphine antinociception in the rat spinal cord. Life Sci 1993;53:PL1 –PL5. Sakurada C, Sugiyama A, Nakayama M, Yonezawa A, Sakurada S, Tan-No K, Kisara K, Sakurada T. Antinociceptive effect of spinally injected L -NAME on the acute nociceptive response induced by low concentrations of formalin. Neurochem Int 2001;38:417–23. Sakurada T, Sugiyama A, Sakurada C, Tan-No K, Sakurada S, Kisara K, Hara A, Abiko Y. Involvement of nitric oxide in spinally mediated capsaicin- and glutamate-induced behavioural responses in the mouse. Neurochem Int 1996a;29:271 –8. Sakurada T, Sugiyama A, Sakurada C, Tan-No K, Yonezawa A, Sakurada S, Kisara K. Effect of spinal nitric oxide inhibition on capsaicin-induced nociceptive response. Life Sci 1996b;59:921 –30. Sakurada T, Wako K, Sakurada C, Manome Y, Tan-No K, Sakurada S, Kisara K. Spinally-mediated behavioural responses evoked by intrathecal high-dose morphine: possible involvement of substance P in the mouse spinal cord. Brain Res 1996c;724:213– 21. Sakurada T, Watanabe C, Okuda K, Sugiyama A, Moriyama T, Sakurada C, Tan-No K, Sakurada S. Intrathecal high-dose morphine induces spinally-mediated behavioural responses through NMDA receptors. Mol Brain Res 2002;98:111 –8. Sakurada C, Watanabe C, Sakurada S, Tan-No K, Sakurada T. Major metabolism of substance P degraded by spinal synaptic membranes antagonize the behavioral response to substance P in rats. J Pharm Sci 1999;88:1127 –32. Seymour RA, Rawlins MD, Rowell FJ. Dihydrocodeine-induced hyperalgesia in postoperative dental pain. Lancet 1982;26:1425 –6. Shohami E, Evron S. Intrathecal morphine induces myoclonic seizures in the rat. Acta Pharmacol Toxicol 1985;56:50–4. Sjogtrn P, Jonsson T, Jenssen N-H, Drenck N-E, Jenssen TS. Hyperalgesia
283
and myoclonus in terminal cancer patients treated with continuous intravenous morphine. Pain 1993;55:93– 7. Sorkin LS. NMDA evokes on L -NAME sensitive spinal release of glutamate and citrulline. NeuroReport 1993;4:479– 82. Tang AH, Schoenfeld HJ. Comparison of subcutaneous and spinal subarachnoid injections of morphine and naloxone in analgesic tests in rat. Eur J Pharmacol 1978;52:215–23. Tan-No K, Esashi A, Nakagawasai O, Niijima F, Tadano T, Sakurada C, Sakurada T, Bakalkin G, Terenius T, Kisara K. Intrathecally administered big dynorphin, a prodynorphin-derived peptide, produces nociceptive behavior through an N-methyl-D -aspartate receptor mechanism. Brain Res 2002;952:7–14. Vanderah TW, Laughlin T, Lashbrook JM, Nichols ML, Wilcox GL, Ossipov MH, Malan Jr. TP, Porreca F. Single intrathecal injections of dynorphin A or des-Tyr-dynorphins produce long-lasting allodynia in rats: blockade by MK-801 but not naloxone. Pain 1996;68:275–81. Wert MA, MacDonald RL. Opiate alkaloids antagonize postsynaptic glycine and GABA responses: correlation with convulsant action. Brain Res 1982;236:107– 19. Woolf CJ. Intrathecal high dose morphine produces hyperalgesia in the rat. Brain Res 1981;209:491–5. Wu J, Lin Q, McAdoo DJ, Willis WD. Nitric oxide contributes to central sensitization following intradermal injection of capsaicin. NeuroReport 1998;9:589–92. Yaksh TL. Spinal opiate analgesia: characteristics and principle of action. Pain 1981;11:293 –346. Yaksh TL, Harty GL. Pharmacology of the allodynia in rats evoked by high dose intrathecal morphine. J Pharmacol Exp Ther 1988;244: 501– 7. Yaksh TL, Harty GJ, Burton M, Onofrio BM. High doses of spinal morphine produce a nonopiate receptor-mediated hyperesthesia: clinical and theoretical implications. Anesthesiology 1986;64: 590– 7.
The role of spinal nitric oxide and glutamate in nociceptive behaviour evoked by high-dose intrathecal morphine in rats Chizuko Watanabea, Tsukasa Sakuradab,*, Kazuhiro Okudac, Chikai Sakuradab, Ryuichiro Andoa, Shinobu Sakuradad a
Center for Laboratory Animal Science, Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan Department of Biochemistry, Daiichi College of Pharmaceutical Sciences, 22-1 Tamagawa-cho, Minami-ku, Fukuoka 815-8511, Japan c Department of Internal Medicine, Mogi Hospital, 6-1-20 Meinohama, Nishi-ku, Fukuoka 819-0002, Japan d Department of Physiology and Anatomy, Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan
b
Received 23 January 2003; received in revised form 24 June 2003; accepted 9 July 2003
Abstract Injection of high-dose of morphine into the spinal lumbar intrathecal (i.t.) space of rats elicits a nociceptive behavioural syndrome characterized by periodic bouts of spontaneous agitation and severe vocalization. The induced behavioural response such as vocalization and agitation was observed dose-dependently by i.t. administration of morphine (125 – 500 nmol). Pretreatment with naloxone (s.c. and i.t.), an opioid receptor antagonist, failed to reverse the morphine-induced behavioural response. The excitatory effect of morphine was inhibited dose-dependently by pretreatment with 3-((þ )2-carboxy-piperazin-4-yl)-propyl-1-phosphonic acid (CPP), a competitive N-methyl-D aspartate (NMDA) receptor antagonist and MK-801, a non-competitive NMDA receptor antagonist. The non-selective nitric oxide (NO) synthase inhibitor N G-nitro L -arginine methyl ester (L -NAME) inhibited dose-dependently the behavioural response to high-dose i.t. morphine (500 nmol), whereas D -NAME was without affecting the response to high-dose i.t. morphine. In the present study, we measured NO metabolites (nitrite/nitrate) in the extracellular fluid of rat dorsal spinal cord using in vivo microdialysis. The i.t. injection of morphine (500 nmol) evoked significant increases in NO metabolites and glutamate from the spinal cord. Not only NO metabolites but also glutamate released by high-dose morphine were reduced significantly by pretreatment with L -NAME (400 nmol). Pretreatment with CPP and MK-801 showed a significant reduction of the NO metabolites and glutamate levels elevated by high-dose i.t. morphine. These results suggest that the excitatory action of high-dose i.t. morphine may be mediated by an NMDA– NO cascade in the spinal cord. q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: High-dose morphine; Intrathecal injection; Nitric oxide; Glutamate; Excitatory behavioural response (vocalization and agitation); Spinal microdialysis
1. Introduction There is ample evidence to suggest that the dorsal spinal cord is an important site for the mediation of the antinociceptive effect of morphine and opioids (Yaksh, 1981; Akil et al., 1984; Basbaum and Fields, 1984). It is well documented that in the formalin-induced flinching test intrathecal (i.t.) administration of morphine as well as other opioid drugs into rats over a dose range of 1.0 –20 nmol produces a reliable antinociceptive effect by interacting with opioid receptors in the spinal cord. In contrast, previous studies have demonstrated that morphine at doses (150 mg * Corresponding author. Tel.: þ 81-92-541-0161; fax: þ81-92-553-5698. E-mail address: [email protected] (T. Sakurada).
or 520 nmol) far higher than those required for antinociception, injected i.t. into the spinal subarachnoid space, produces a profound spontaneous agitation and squeaking in rats (Tang and Schoenfelt, 1978; Woolf, 1981; Yaksh et al., 1986; Alvarez-Vega et al., 1998). High-doses of spinally injected morphine in the mouse also elicit a nociceptive-related behaviour such as scratching, biting and licking in mice resembling that of i.t.-injected substance P or N-methyl-D -aspartate (NMDA) (Sakurada et al., 1996c). High-dose i.t. morphine exhibits an allodynic response to innocuous tactile stimuli in rats and mice (Hara et al., 1997; Yaksh and Harty, 1988). Additionally, i.t. injections of highdoses of morphine are found to elicit myoclonic seizures in the rat (Shohami and Evron, 1985) or clonic seizure-like
0304-3959/$20.00 q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/S0304-3959(03)00296-3
270
C. Watanabe et al. / Pain 106 (2003) 269–283
behaviours in mice (Lutfy et al., 1994). A hyperalgesic action has also been known to develop after chronically administered morphine in rats (Kayan and Mitchell, 1968; Mule et al., 1968). A similar phenomenon has been recognized in patients who developed myoclonus and hyperalgesia following administration of high-dose subarachnoid morphine (Wert and MacDonald, 1982; Krames et al., 1985; Ali, 1986; Penn and Paice, 1987; Arner et al., 1988). Recently, we have reported that nociceptive behavioural responses consisting of a severe hindlimb scratching followed by biting/licking of the hindpaw could be elicited by i.t. injection of high-dose morphine in mice, which are considered to be mediated by the spinal NMDA receptors (Sakurada et al., 2002). However, linkage between excitatory behavioural responses to high-dose i.t. morphine and changes in concentrations of spinal glutamate and NO needs to be demonstrated. Thus we conducted the present experiments to provide the possible role played by the glutamate receptor–NO system in nociceptive behaviour evoked by high-dose i.t. morphine in rats. An important objective was to explore whether there is an alteration of glutamate and NO metabolites (nitrite/nitrate) in the extracellular fluid of rat dorsal spinal cord using in vivo microdialysis following high-dose i.t. morphine. We also examined the effects of NMDA receptor antagonists and an inhibitor of NO synthase, on spinally mediated behavioural response, and on the release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine in rats.
2. Materials and methods 2.1. Animals Experimental procedures were conducted according to protocols approved by the Committee of Animal Experimentation of Tohoku Pharmaceutical University, Sendai, Japan. Male Sprague– Dawley (SD) rats (SLC, Japan), weighing 250 –260 g, were housed with free access to food and water. They were maintained in a controlled environment (12 h light/dark cycle, room temperature 23 8C, 50 –60% relative humidity). All experiments took place during the light period between 10:00 and 17:00 hours in a quiet room. The animals belonging to the various treatment groups (n ¼ 7 each group) were tested in randomized order. 2.2. Intrathecal injection To permit application of morphine, N G-nitro L -arginine methyl ester (L -NAME), D -NAME, 3-((þ )2-carboxy-piperazin-4-yl)-propyl-1-phosphonic acid (CPP) and MK-801, directly into the spinal subarachnoid space, rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and chronically implanted with i.t. catheters according to the previously reported method (Sakurada et al., 1999).
Briefly, this involved inserting a length of polyethylene tubing (PE-10) following laminectomy between L3 and L4 and carefully placing the catheter tip in the subarachnoid space of L5 and L6. The animals were allowed to recover over 7 days following implantation of the catheter. The catheter was filled with sterile artificial cerebrospinal fluid (ACSF), containing (in g/L) NaCl 7.4, KCl 0.19, MgCl2 0.19, and CaCl2 0.14. Drugs were administered in volumes of 10 ml followed by a flush of 15 ml of ACSF to ensure that each compound reached the spinal cord. L -NAME, D -NAME, CPP and MK-801 were administered i.t. 10 min prior to high-dose i.t. morphine. Rats were pretreated subcutaneously (s.c.) with naloxone (4.0 mg/kg) 10 min before i.t. administration of morphine (500 nmol). Naloxone (150 nmol) was also injected i.t. 2 min prior to spinal administration of morphine. 2.3. Behavioural observation To perform the behavioural observation after high-dose i.t. morphine, the rats were placed individually in an open Plexiglas chamber (34 £ 30 £ 17 cm), which also served as an observation chamber after injection. The animals were habituated to the observation chamber 1 h before actual experimental sessions. Immediately after the high-dose morphine was injected i.t. using a microsyringe with a 26gauge needle, each rat was returned to the observation chamber. High-dose morphine, injected i.t., produced a striking behavioural syndrome consisting of spontaneous agitation and vocalization. The magnitudes of two behavioural responses were quantified: agitation and vocalization. Spontaneous agitation was ranked visually every 1 min after i.t. administration of high-dose morphine as: 0, no sign of excitation; 1, restlessness, scratching and biting at the flank or tail; 2, mild vocalization with restlessness, scratching and biting at the flank or tail; 3, vocalization with spontaneous running and circling; and 4, vigorous vocalization with running, circling, rolling and jumping. The score of agitation was expressed in each animal during each 5-min interval for a 40-min period after high-dose i.t. morphine. The second parameter studied was the appearance of spontaneous vocalization. The total duration of vocalization was recorded by the stop-watch during the 5-min interval for the 40-min period. The latency to induce the first vocalization after i.t. administration of morphine was also recorded. In the behavioural study, one person prepared the drugs and another, blind to the drugs and dosage, ran the behavioural observation. The blind observer was identical throughout this study. 2.4. Microdialysis and implantation An in vivo i.t. microdialysis technique was performed according to the modified method of Marsala et al. (1995). Rats were implanted with a U-shaped microdialysis cannula through a laminectomy between L1 and L2 under anesthesia
C. Watanabe et al. / Pain 106 (2003) 269–283
with sodium pentobarbital (50 mg/kg, i.p.). The tip of the cannula was positioned in the subarachnoid space of L5 and L6. The dialysis catheters were then externalized on the back of the neck in rats. The microdialysis cannula consisted of a 6 cm length of microdialysis fiber (200 mm inner diameter, 300 mm outer diameter, 50 kDa molecular weight cut-off; Filtral, AN69-HF). The fiber was coated with a thin layer of epoxy (Araldite, CIBA GEIGY Inc.) except for a 4 cm portion in the middle. In order to make the fiber firm enough for implantation, a Nichrome – Formvar wire (0.0026 inch inner diameter; A-M systems, Inc., Everret, WA) was then passed through the fiber and the ends of the fiber were attached to a PE-10 (5.0 cm long) with cyanoacrylate. The fiber was then bent so that a U-shaped loop was formed in the middle of the uncoated portion. Following each experiment, all animals were deeply anesthetized and the localization of the microdialysis cannula was verified visually. Only animals with the cannula placed at the L5 and L6 spinal segments were included in this study. 2.5. Sample collection and chemical analysis After allowing 3– 5 days for recovery from i.t. implantation of microdialysis catheters, rats were re-anesthetized with isoflurane (3.3 –3.5% for induction, 1.5– 2.0% for surgery and 1.0 –1.1% for the subsequent dialysis experiment in a 50:50 air/O2 mixture) and the dialysis experiment was started. The dialysis tubing was connected to the syringe pump (EO-60, EICOM) and perfused with ACSF 30 min prior to sample collection to establish a diffusion equilibrium. The flow rate was 10 ml/min. Baseline concentrations were defined as the mean of two 10 min control samples taken after the initial 30 min washout period. Following baseline periods, samples of fluid were collected in polypropylene tubes after i.t. administration of each compound for the duration of the experiment. Dialysate samples were separated for measuring nitrite/nitrate (aliquot 50 ml) and glutamate (residue 50 ml), and maintained at 2 70 8C until chemical analysis. NO was measured as its break products, nitrite (NO2 2 ) and nitrate (NO2 ). The measurement of nitrite/nitrate was 3 made using a commercially available NO analyser (MODEL-280NOA, Sievers, Inc.) and integrator (HO3396, Sievers, Inc.). External standards were run at the beginning of sample analysis. The limit of detection of nitrite/nitrate was 1 –2 pmol/30 ml. The amino acid glutamate in the spinal perfusate was detected by reversed-phase high-performance liquid chromatography (HPLC) (ECD-300, EICOM) with fluorimetric detection following pre-column derivatization with ophthaldialdehyde (OPA). Chromatography was performed on a reversed-phase C-18 column (2:1 £ 150 mm, EICOM PACKED COLUMN) using a pH sodium acetate methanol gradient. A known concentration of glutamate was used as an internal standard. External standards were run at least
271
every five to six samples to ensure accuracy and for calculations of concentrations. The minimal detectable concentration of glutamate was 5– 10 pmol/30 ml. 2.6. Drugs Drugs used and their sources were: morphine hydrochloride (Sankyo, Tokyo, Japan), naloxone hydrochloride (Sigma Chemical Co., St. Louis, MO), L -NAME, D -NAME (Nakalai tesque, Kyoto, Japan), CPP, (5R,10S)-(þ )-5methyl-10,11-dihydro-5H-dibenzo [a,d ] cyclohepten-5, and 10-imine maleate (MK-801) (Research Biochemical Incorporated, Natic, MA). All drugs for spinal administration were dissolved in ACSF. Dosages reflect the salt form of the drugs. 2.7. Data analysis and statistics The time –response data of the excitatory behaviours (agitation and vocalization) induced by different doses of morphine or in combination with compounds are presented as the mean ^ SEM of agitation scores and accumulated vocalization time (s). The spinal release of nitrite/nitrate and glutamate is presented as the mean ^ SEM percentage changes from baseline. Statistical analysis of behavioural and biochemical data was carried out with Bonferroni’s and Dunnett’s test following two-way or two-factor analysis of variance (ANOVA) for repeated measures. Differences are considered to be significant when the critical value reaches a level of P , 0:05.
3. Results 3.1. Spontaneous agitation and vocalization evoked by high-dose i.t. morphine The i.t. administration of morphine at doses ranging from 125 to 500 nmol into the spinal lumbar space evoked a spontaneous agitation and vocalization. A dose-dependent response was seen by measuring these two behavioural items (Fig. 1). The highest dose (500 nmol) of morphine evoked a spontaneous agitation such as vigorous vocalization accompanied by hindlimb scratching toward the flanks and biting of hindpaw or the base of the tail, running and jumping, and occasionally myoclonic seizures. Agitation evoked by i.t. morphine (500 nmol) peaked at 5 –20 min and lasted for approximately 40 min post-injection, whereas 125 and 250 nmol of morphine exhibited a shorter duration of agitation with lower scores (Fig. 1A). Morphine at doses of 125 and 250 nmol caused a weak agitation, which peaked at 5 –10 min and decayed at 30 – 35 min. As shown in Fig. 1B, time-course of morphine (500 nmol)-evoked vocalization was similar to that of agitation by i.t. morphine at a dose of 250 nmol; vocalization evoked by morphine was maximum at 5 –10 min, and disappeared at 30 –35 min. Accumulated
272
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 1. Time-course effect of agitation (A) and vocalization (B) following intrathecal (i.t.) administration of morphine (125–500 nmol) in rats. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats.
vocalization time for the 40-min observation after i.t. administration of 500 nmol morphine was 435:0 ^ 10:2 s (Fig. 1B). As shown in Table 1, the vocalization response was late in onset; even at a dose of 500 nmol the latency to elicit the first vocalization after i.t. injection was over 60 s. No agitation and vocalization responses were observed in rats receiving an antinociceptive dose of 40 nmol morphine or i.t. ACSF (data not shown).
In order to examine the non-specific effect from a high concentration of hydrochloride salt containing in 500 nmol morphine, animals were treated with 0.5% hydrochloride (n ¼ 3) as a control with a comparable concentration to 500 nmol morphine and 2.5% hydrochloride (n ¼ 3) in a fivefold higher concentration, respectively. Treatment with these concentrations of hydrochloride produced no agitation and vocalization like high-dose morphine.
C. Watanabe et al. / Pain 106 (2003) 269–283 Table 1 The latency to first vocalization following intrathecal administration of high-dose morphine or in combination with naloxone, L -NAME, D -NAME, MK-801 and CPP in rats Treatment Morphine 125 nmol 250 nmol 500 nmol ACSF plus morphine 500 nmol Naloxone (4.0 mg/kg) plus morphine Naloxone (150 nmol) plus morphine L -NAME (100 nmol) plus morphine L -NAME (200 nmol) plus morphine L -NAME (400 nmol) plus morphine D -NAME (400 nmol) plus morphine CPP (1.0 nmol) plus morphine CPP (2.0 nmol) plus morphine CPP (4.0 nmol) plus morphine MK-801 (10 nmol) plus morphine MK-801 (20 nmol) plus morphine MK-801 (40 nmol) plus morphine
Administration route
Latency (s)
i.t. i.t. i.t. i.t.
233.8 ^ 35.6 146.9 ^ 11.8 69.3 ^ 5.5 67.2 ^ 5.5
s.c. þ i.t.
69.8 ^ 11.9
i.t. þ i.t.
60.5 ^ 2.9
i.t. þ i.t.
63.6 ^ 8.1
i.t. þ i.t.
64.1 ^ 8.7
i.t. þ i.t.
61.9 ^ 5.0
i.t. þ i.t.
62.8 ^ 5.6
i.t. þ i.t.
125.7 ^ 14.7*
i.t. þ i.t.
143.4 ^ 13.8**
i.t. þ i.t.
160.0 ^ 14.6**
i.t. þ i.t.
64.3 ^ 5.5
i.t. þ i.t.
79.1 ^ 3.9
i.t. þ i.t.
113.6 ^ 8.4**
Rats were pretreated with naloxone at a dose of 4.0 mg/kg (s.c.) or 150 nmol (i.t.) 10 min before i.t. administration of morphine (500 nmol). L NAME, D -NAME, MK-801 or CPP was administered i.t. 10 min prior to i.t. morphine (500 nmol). Asterisks indicate a significant increase compared to i.t. morphine (500 nmol) alone (Dunnett’s test; * P , 0:01, * * P , 0:05).
3.2. Spinal release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine Basal levels of NO breakdown products, nitrite/nitrate and glutamate were detectable in all animals treated under isoflulane throughout the measurement of nitrite/nitrate and glutamate. The time-courses of the mean basal level of nitrite/nitrate are presented in Fig. 2 (upper panel). The means of baseline samples collected before i.t. administration of ACSF were used for determination of nitrite/ nitrate and glutamate in the dorsal spinal cord extracellular fluid. The absolute amounts of nitrite/nitrate collected during the baseline period were 2:90 ^ 0:22 nmol/30 ml (n ¼ 7) and 2:55 ^ 0:15 nmol/30 ml (n ¼ 7) in rats injected with 250 nmol and 500 nmol morphine, respectively. The baseline control levels of nitrite/nitrate were not altered by i.t. administration of ACSF (Table 2). A high-dose of morphine (500 nmol) evoked a significant increase of nitrite/nitrate release in the time interval of 10– 30 min
273
post-i.t. administration, whereas 250 nmol morphine was without effect (Fig. 2, upper panel). The baseline concentrations of glutamate examined were 0:62 ^ 0:12 nmol/30 ml (n ¼ 7) and 0:54 ^ 0:07 nmol/30 ml (n ¼ 7) in rats injected with 250 nmol and 500 nmol morphine, respectively. The baseline control levels of glutamate were not altered by i.t. administration of ACSF (Table 2). Morphine at a dose of 500 nmol, injected i.t., produced over a 2.0-fold increase of glutamate during the first 20 min when compared to ACSF injection (Fig. 2, lower panel). A significant increase of glutamate was also observed from 20 to 40 min following injection of 500 nmol morphine, whereas the increased level of glutamate was elevated only to a smaller degree. The concentration of glutamate increased significantly during the first 10 min after i.t. administration of morphine at a dose of 250 nmol, though there was no significant increase of the glutamate level in the time interval of 10 –40 min. An antinociceptive dose (40 nmol) of morphine gave no effect on the release of nitrite/nitrate and glutamate. 3.3. Effect of naloxone, an opioid receptor antagonist, on the spontaneous behaviour and on the spinal release of nitrite/nitrate and glutamate induced by high-dose i.t. morphine Peripheral and spinal pretreatment with naloxone (4.0 mg/kg, s.c.; 150 nmol, i.t.), an opioid receptor antagonist, gave no reversed effect on the excitatory responses, vocalization and agitation evoked by i.t. morphine (500 nmol), and the latency to the first vocalization (Fig. 3, Table 1). Biochemical analysis revealed that i.t. pretreatment with naloxone (150 nmol, i.t.) 10 min prior to morphine (500 nmol, i.t.) produced no significant effect on morphine-evoked release of nitrite/nitrate and glutamate (Fig. 2). 3.4. Effect of NO synthase inhibitor and NMDA receptor antagonists on the spontaneous behaviour induced by high-dose i.t. morphine Pretreatment with L -NAME (100 –400 nmol, i.t.) caused a dose-related inhibition of the agitation and vocalization response to i.t. morphine (Fig. 4A and B). The highest dose (400 nmol) of L -NAME used in these experiments was most effective in inhibiting these behavioural responses. The excitatory behavioural response was not affected by pretreatment with D -NAME (400 nmol, i.t.). None of the doses of L -NAME (100, 200 and 400 nmol) or D -NAME (400 nmol) tested altered the latency to first vocalization response evoked by i.t. high-dose morphine (Table 1). Morphine (500 nmol)-evoked agitation and vocalization were inhibited dose-dependently by pretreatment with CPP (1.0 –4.0 nmol), a competitive NMDA antagonist (Fig. 5A and B). Similarly, a dose – response inhibition of MK-801 (10 –40 nmol), a non-competitive NMDA antagonist, was
274
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 2. Time-course of spinal release of nitrite/nitrate (A) and glutamate (B) following intrathecal (i.t.) administration of morphine and in combination with naloxone, an opioid receptor antagonist in rats. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline after i.t. injection of ACSF or morphine (500, 250 and 40 nmol) or in combination with naloxone (150 nmol). Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant increase compared to time-matched ACSF (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
275
Table 2 Effect of L -NAME, CPP and MK-801 on resting release of nitrite/nitrate and glutamate in the lumbar intrathecal space as measured by microdialysis in rats Substances measured
Time period (min)
Mean % of baseline ACSF control
L -NAME
(400 nmol)
Nitrite/nitrate
0– 10 10– 20 20– 30 30– 40
87.5 ^ 5.2 88.8 ^ 1.4 89.8 ^ 3.0 86.8 ^ 5.4
82.3 ^ 13.3 98.3 ^ 2.7 88.0 ^ 8.1 100.3 ^ 3.2
Glutamate
0– 10 10– 20 20– 30 30– 40
95.3 ^ 2.2 92.5 ^ 1.7 88.8 ^ 5.6 76.5 ^ 2.7
95.0 ^ 1.0 79.7 ^ 9.0 81.7 ^ 3.3 82.0 ^ 2.6
CPP (4.0 nmol)
MK-801 (40 nmol)
99.7 ^ 9.8 97.0 ^ 6.1 93.7 ^ 2.7 101.0 ^ 5.7
99.0 ^ 7.1 99.3 ^ 9.3 102.8 ^ 12.6 86.5 ^ 8.7
85.8 ^ 7.8 86.5 ^ 6.2 81.3 ^ 4.9 84.8 ^ 12.1
95.3 ^ 4.2 82.5 ^ 12.7 79.8 ^ 11.5 86.3 ^ 12.8
Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage of baseline after i.t. injection of ACSF, L -NAME (400 nmol), CPP (4.0 nmol) and MK-801 (40 nmol).
constructed in morphine-evoked agitation and vocalization (Fig. 6A and B). The appearance of the first vocalization response was prolonged by both NMDA antagonists (Table 1). General behaviour was not recognized in rats pretreated with L -NAME, CPP or MK-801 at doses employed in the experiment. 3.5. Effect of NO synthase inhibitor and NMDA receptor antagonists on the spinal release of nitrite/nitrate and glutamate evoked by high-dose i.t. morphine Pretreatment with L -NAME (400 nmol), a non-selective NO synthase inhibitor, 10 min prior to i.t. administration of morphine (500 nmol), prevented a significant increase of nitrite/nitrate in samples collected in the time interval of 10 –40 min following i.t. administration of morphine (Fig. 7, upper panel). Morphine (500 nmol)-evoked release of nitrite/nitrate was not blocked by pretreatment with 200 nmol L -NAME and 400 nmol D -NAME, an inactive stereoisomer of L -NAME. L -NAME (400 nmol) pretreatment completely blocked the morphine-evoked release of glutamate during 0– 40 min period (Fig. 7, lower panel). A significant block was also observed in the time interval of 20 –30 min by pretreatment with 200 nmol L -NAME but not 400 nmol D -NAME. There were no significant effects of L -NAME on the resting level of nitrite/nitrate and glutamate (Table 2). The i.t. administration of CPP (2.0 and 4.0 nmol) and MK-801 (20 and 40 nmol) prior to i.t. morphine (500 nmol) prevented significantly the increased concentrations of nitrite/nitrate evoked by i.t. morphine (Figs. 8 and 9, upper panels). Pretreatment with 4.0 nmol CPP and 40 nmol MK-801 produced a significant effect on the increased concentrations of glutamate evoked by i.t. morphine (Figs. 8 and 9, lower panels), whereas 2.0 nmol CPP and 20 nmol MK-801 were ineffective in reducing morphine-induced increased concentrations of glutamate. The resting level of
nitrite/nitrate and glutamate was not altered by CPP (2.0 nmol) or MK-801 (40 nmol) (Table 2).
4. Discussion High-dose morphine, injected into the spinal lumbar i.t. space, has been shown to produce an excitatory reaction such as spontaneous agitation and vocalization accompanied by scratching, biting and licking, and touchevoked allodynia in rats and mice (Sakurada et al.,1996c; Hara et al., 1997; Alvarez-Vega et al., 1998; Yaksh and Harty, 1988), as opposed to an antinociceptive action at lower doses ranging from 1.0 to 20 nmol (Yaksh, 1981; Akil et al., 1984; Basbaum and Fields, 1984). Consistent with these previous studies, the present investigation demonstrated that injection of morphine at a high-dose (500 nmol) into the spinal lumbar i.t. space elicited a spontaneous nociceptive syndrome mainly consisting of agitation and vocalization in rats. High-doses of spinally administered morphine have also been reported to produce hyperalgesia in response to noxious stimuli in rats (Woolf, 1981). The morphine-induced hyperalgesia is of clinical interest since hyperalgesic states have been reported to occur in humans following administration of high-dose subarachnoid morphine or systemic administration of opioid (Seymour et al., 1982; Ali, 1986; Arner et al., 1988; De Conno et al., 1991; Sjogtrn et al., 1993). In the present study, we first confirmed that i.t. administrations of 0.5% hydrochloride salt with a comparable concentration to 500 nmol morphine or even 2.5-fold higher concentration produced no observable changes in spontaneous behaviours, vocalization and agitation. The result indicates that high-dose i.t. morphine-induced behavioural responses may not be due to non-specific activation of highly concentrated salt. Behavioural activation evoked by high-dose i.t. morphine are irreversible by the opioid receptor antagonist, naloxone or naltrexone,
276
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 3. Effect of naloxone on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. Naloxone in doses of 4.0 mg/kg (s.c.) and 150 nmol (i.t.) was administered 10 min and 2 min prior to i.t. morphine, respectively. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats.
C. Watanabe et al. / Pain 106 (2003) 269–283
suggesting a non-opioid receptor mechanism (Woolf, 1981; Yaksh et al., 1986; Yaksh and Harty, 1988; Okuda et al., 2001b). The seizure-like motor effects of high-dose i.t. morphine are also irreversible by pretreatment with naloxone in mice (Lutfy et al., 1994). The present study also shows that naloxone administered peripherally and spinally neither antagonized the excitatory behavioural response nor reversed the increased concentrations of nitrite/nitrate and glutamate to high-dose i.t. morphine. It therefore reinforced the concept that the effects of high-dose i.t. morphine are brought about by mechanisms that are not dependent of opioid receptor activity in the spinal cord. It has to be noted that i.t. dynorphin A, a proposed endogenous ligand for the kappa opioid receptor, evokes long-lasting allodynia, and scratching, licking and biting behaviour through NMDA receptors rather than nonopioid mechanisms in rats (Vanderah et al., 1996) and mice (Laughlin et al., 1997; Tan-No et al., 2002). In the present study, CPP, a competitive NMDA receptor antagonist, and MK-801, a non-competitive NMDA receptor antagonist, both dose-dependently inhibited agitation and vocalization induced by high-dose i.t. morphine, suggesting that activation of NMDA receptors in the spinal cord may mediate the excitatory effects of i.t. morphine. This suggestion is supported by the results that high-dose i.t. morphine evoked a marked increase of glutamate in the extracellular levels, which was inhibited by pretreatment with CPP and MK-801. There is a difference of nociceptive behavioural response to high-dose i.t. morphine in rats when compared with mice; in mice high-dose (60 and 90 nmol) i.t. morphine produces a severe hindlimb scratching followed by biting and licking toward the caudal part of the body without squeaking (Sakurada et al., 1996c). Recently, we have shown that the onset of high-dose morphine-induced scratching response was late after i.t. injection in mice. The data observed in mice are compatible with the present study that the latency of vocalization was late (more than 60 s) after 500 nmol morphine in rats. In addition, the characteristic behavioural response, scratching, biting and licking in mice has been found to inhibit dose-dependently by: (1) the NK1 receptor antagonists, sendide and CP-96,345; (2) pretreatment with the non-selective substance P depletor capsaicin; and (3) pretreatment with i.t.-injected antiserum against substance P (Sakurada et al., 1996c). It is therefore inferred that the spinally mediated behavioural response induced by high-dose i.t. morphine may occur indirectly, possibly through the release of substance P in the dorsal horn of the spinal cord. If a similar phenomenon may occur in the case of high-dose morphine injected i.t. into rats, it is reasonable to speculate that the excitatory response evoked by highdose i.t. morphine in rats may be mediated in part through spinal NK1 receptors. This is a reason why morphine-induced behavioural response was not
277
eliminated completely by 4 nmol CPP and 40 nmol MK-801, NMDA receptor antagonists, and 400 nmol L -NAME, an NO synthase inhibitor in the present study. It appears evident that the excitatory effects of high-dose i.t. morphine might be a nociceptive event clearly independent of spinal opioid receptors. Taken together, the behavioural activation of glutamate receptors by high-dose i.t. morphine may, at least, in part be responsible for enhancement of glutamate release in the dorsal horn. There is considerable evidence that NO could possibly contribute to synaptic transmission between primary afferent neurones and dorsal horn cells in spinal nociceptive processing. In the beginning of NO studies in the transmission of pain signals, these studies have indirectly assessed the physiological role of NO using an inhibitor of NO synthase such as L -NAME (Moore et al., 1991; Haley et al., 1992; Sakurada et al., 1996a,b, 2001). Recently, it has been shown that injection of capsaicin into the hindfoot increases NO production in the dialysate of spinal cord (Wu et al., 1998). Studies on the formalin model of pain have indicated that formalin injection into the hindpaw evoked a biphasic increase of NO in the extracellular levels that paralleled largely the early and late phase flinching responses (Okuda et al., 2001a), which was inhibited by pretreatment with L -NAME (unpublished data). The ED 50 value of L -NAME obtained in the late phase of formalin (5.0%)-induced flinching response was 360.0 (83.0 – 1560.6 nmol) nmol. Judging from the ED50 value of L -NAME with 210.0 (129.1 – 341.6 nmol) nmol in vocalization response to i.t. morphine (500 nmol), there is no statistically significant difference of the ED50 values for L -NAME between the two assays. These results support the contention that vocalization evoked by high-dose i.t. morphine may be closely related to nociceptive events in the spinal cord. A possible interaction between NO and opioid systems in the spinal cord has been proposed; the inhibition of NO synthase potentiates antinociception induced by spinal administration of morphine and opioid receptor agonist (Przewlocki et al., 1993; Machelska et al., 1997), and diminishes tolerance to morphine in the antinociceptive action of morphine (Kolesnikov et al., 1992; Majeed et al., 1994; Bhargava and Zhao, 1996). In marked contrast with the above reports, the current study showed that high-dose i.t. morphine produced an excitatory behavioural response (agitation and vocalization) and a significant increase of NO release in the spinal cord. Considering that NO functions as an intermediary of NMDA receptors in the spinal cord, it seems clearly that high-dose i.t. morphine can act as a nociceptive input through activation of NMDA receptors and NO production. This concept is in line with the behavioural data that vocalization response evoked by high-dose i.t. morphine was inhibited more potently than
278
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 4. Effect of L -NAME and D -NAME on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. L -NAME or D -NAME was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
279
Fig. 5. Effect of CPP, a competitive NMDA receptor antagonist, on agitation (A) and vocalization evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. CPP was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
280
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 6. Effect of MK-801, a non-competitive NMDA receptor antagonist, on agitation (A) and vocalization (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. MK-801 was administered i.t. 10 min prior to i.t. morphine. (A) Agitation was scored every 1 min for the 40-min period. The accumulated score of agitation during each 5-min interval is represented (left panel). Each bar represents accumulated score for 40 min (right panel). (B) Duration of vocalization time was observed during each 5-min interval for the 40-min period (left panel). Each bar represents accumulated vocalization time for 40 min (right panel). Data are shown as the mean ^ SEM of seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
C. Watanabe et al. / Pain 106 (2003) 269–283
281
Fig. 7. Effect of L -NAME and D -NAME on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. L -NAME or D -NAME was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
Fig. 8. Effect of CPP on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of high-dose morphine (500 nmol) in rats. CPP was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
morphine-induced agitation. Previously reported results suggest that NMDA receptor-mediated central sensitization and hyperalgesia involves subsequent production of NO (Meller et al., 1992) and augmented release of glutamate induced by NMDA receptor activation occurs secondary to NO production (Sorkin, 1993). This assumption is supported by the present microdialysis data that morphine-evoked releases of nitrite/nitrate and glutamate were inhibited by pretreatment with
the competitive and non-competitive NMDA antagonists, CPP and MK-801. In summary, the present study demonstrates that highdose i.t. morphine evokes a significant increase of nitrite/nitrate and glutamate in spinal dialysates. Increased release of nitrite/nitrate and glutamate largely corresponded with the appearance of agitation and vocalization. These observations suggest that the excitatory state evoked by high-dose i.t. morphine may be
282
C. Watanabe et al. / Pain 106 (2003) 269–283
Fig. 9. Effect of MK-801 on the increased spinal release of nitrite/nitrate (A) and glutamate (B) evoked by intrathecal (i.t.) administration of morphine (500 nmol) in rats. MK-801 was administered i.t. 10 min prior to i.t. morphine. Concentrations of nitrite/nitrate and glutamate in spinal microdialysate are expressed as the mean ^ SEM percentage change from baseline. Each value on the graph represents the mean ^ SEM for seven rats. Asterisks indicate a significant decrease compared to 500 nmol morphine (* * P , 0:01, * P , 0:05).
mediated by NMDA-NO cascade, since agitation and vocalization could be reduced by pretreatment with either NMDA receptor antagonists (CPP and MK-801) or an inhibitor of NO synthase (L -NAME).
References Akil H, Watson SJ, Young E, Lewis ME, Khachaturian H, Walker JM. Endogenous opioids: biology and function. Annu Rev Neurosci 1984;7: 223– 55.
Ali NMK. Hyperalgesic response in a patient receiving high concentrations of spinal morphine. Anesthesiology 1986;65:443 –9. Alvarez-Vega M, Baamonde A, Gutierrez M, Hidalgo A, Menendez L. Comparison of the effects of calmidazolium and bupivacaine on N-methyl-D -aspartate- and septide-induced nociceptive behaviour. Naunyn Schmiedebergs Arch Pharmacol 1998;358:628–34. Arner S, Rawal N, Gustafsson LL. Clinical experience of long-term treatment with epidural opioids—a nationwide survey. Acta Anaesthesiol Scand 1988;32:253–9. Basbaum AL, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7: 309 –38. Bhargava HN, Zhao GM. Effects of nitric oxide synthase inhibition on tolerance to the analgesic action of D-Pen2, D-Pen5 enkephalin and morphine in the mouse. Neuropeptides 1996;30:219–23. De Conno F, Caraceni A, Martini C, Spoldi E, Salvetti M, Ventafridda V. Hyperalgesia and myoclonus with intrathecal infusion of high-dose morphine. Pain 1991;47:337–9. Haley JE, Dickenson AH, Schachter M. Electrophysiological evidence for a role of nitric oxide in prolonged chemical nociception in the rat. Neuropharmacology 1992;31:251–8. Hara N, Minami T, Okuda-Ashitaka E, Sugimoto T, Sakai M, Onaka M, Mori H, Imanishi T, Shingu K, Ito S. Characterization of nociception hyperalgesia and allodynia in conscious mice. Br J Pharmacol 1997;121:401–8. Kayan S, Mitchell CL. The effects of chronic morphine administration of tooth pulp threshold in dogs and cats. Proc Soc Exp Biol Med 1968;128: 755 –60. Kolesnikov YA, Pick CG, Pasternak GW. N G-nitro-L -arginine prevents morphine tolerance. Eur J Pharmacol 1992;221:399– 400. Krames ES, Gershow J, Glassberg A, Kenefick T, Lyons A, Taylor P, Wilkie D. Continuous infusion of spinally administered narcotics for pain due to malignant disorder. Cancer 1985;56:696–702. Laughlin TM, Vanderah T, Lashbrook JM, Nichols ML, Ossipov MH, Porreca F, Wilcox GL. Spinally administered dynorphin A produces long-lasting allodynia: involvement of NMDA but not opioid receptors. Pain 1997;72:253–60. Lutfy K, Woodward RM, Keana JFW, Weber E. Inhibition of clonic seizure-like excitatory effects induced by intrathecal morphine using two NMDA receptor antagonists: MK-801 and ACEA-1011. Eur J Pharmacol 1994;252:261–6. Machelska H, Labuz D, Przewlocki R, Przewlocka B. Inhibition of nitric oxide synthase enhances antinociception mediated by mu, delta and kappa opioid receptors in acute and prolonged pain in the rat spinal cord. J Pharmacol Exp Ther 1997;282:977– 84. Majeed NH, Przeweocka B, Machelska H, Przewlocki R. Inhibition of nitric oxide synthase attenuates the development of morphine tolerance and dependence in mice. Neuropharmacology 1994;33: 189 –92. Marsala M, Malmberg AB, Yaksh TL. The spinal loop dialysis catheter: characterization of use in the unanesthetized rat. J Neurosci Methods 1995;62:43–53. Meller ST, Dykstra C, Gebhart GF. Production of endogenous nitric oxide and activation of soluble guanylate cyclase are required for N-methylD -aspartate-produced facilitation of the nociceptive tail-flick reflex. Eur J Pharmacol 1992;214:93–6. Moore PK, Oluyomi AO, Babbedge RC, Wallace P, Hart SL. L -N Gnitroarginine methyl ester exhibits antinociceptive activity in the mouse. Br J Pharmacol 1991;102:198–202. Mule SJ, Clements TH, Layson RC, Haertzen CA. Analgesia in guinea pigs: a measure of tolerance development. Arch Int Pharmacodyn Ther 1968;173:210–2. Okuda K, Sakurada C, Takahashi M, Yamada T, Sakurada T. Characterization of nociceptive responses and spinal releases of nitric oxide metabolites and glutamate evoked by different concentrations of formalin in rats. Pain 2001a;92:107 –15. Okuda K, Watanabe C, Ando R, Sakurada C, Takahashi M, Yamada T, Sakurada T. The role of spinal nitric oxide in nociceptive behavior
C. Watanabe et al. / Pain 106 (2003) 269–283 evoked by high dose intrathecal morphine in rats. Jpn J Pharmacol 2001b;85:218. Penn R, Paice J. Chronic intrathecal morphine for intractable pain. J Neurosurg 1987;67:182– 6. Przewlocki R, Machelska H, Przewlocka B. Inhibition of nitric oxide synthase enhances morphine antinociception in the rat spinal cord. Life Sci 1993;53:PL1 –PL5. Sakurada C, Sugiyama A, Nakayama M, Yonezawa A, Sakurada S, Tan-No K, Kisara K, Sakurada T. Antinociceptive effect of spinally injected L -NAME on the acute nociceptive response induced by low concentrations of formalin. Neurochem Int 2001;38:417–23. Sakurada T, Sugiyama A, Sakurada C, Tan-No K, Sakurada S, Kisara K, Hara A, Abiko Y. Involvement of nitric oxide in spinally mediated capsaicin- and glutamate-induced behavioural responses in the mouse. Neurochem Int 1996a;29:271 –8. Sakurada T, Sugiyama A, Sakurada C, Tan-No K, Yonezawa A, Sakurada S, Kisara K. Effect of spinal nitric oxide inhibition on capsaicin-induced nociceptive response. Life Sci 1996b;59:921 –30. Sakurada T, Wako K, Sakurada C, Manome Y, Tan-No K, Sakurada S, Kisara K. Spinally-mediated behavioural responses evoked by intrathecal high-dose morphine: possible involvement of substance P in the mouse spinal cord. Brain Res 1996c;724:213– 21. Sakurada T, Watanabe C, Okuda K, Sugiyama A, Moriyama T, Sakurada C, Tan-No K, Sakurada S. Intrathecal high-dose morphine induces spinally-mediated behavioural responses through NMDA receptors. Mol Brain Res 2002;98:111 –8. Sakurada C, Watanabe C, Sakurada S, Tan-No K, Sakurada T. Major metabolism of substance P degraded by spinal synaptic membranes antagonize the behavioral response to substance P in rats. J Pharm Sci 1999;88:1127 –32. Seymour RA, Rawlins MD, Rowell FJ. Dihydrocodeine-induced hyperalgesia in postoperative dental pain. Lancet 1982;26:1425 –6. Shohami E, Evron S. Intrathecal morphine induces myoclonic seizures in the rat. Acta Pharmacol Toxicol 1985;56:50–4. Sjogtrn P, Jonsson T, Jenssen N-H, Drenck N-E, Jenssen TS. Hyperalgesia
283
and myoclonus in terminal cancer patients treated with continuous intravenous morphine. Pain 1993;55:93– 7. Sorkin LS. NMDA evokes on L -NAME sensitive spinal release of glutamate and citrulline. NeuroReport 1993;4:479– 82. Tang AH, Schoenfeld HJ. Comparison of subcutaneous and spinal subarachnoid injections of morphine and naloxone in analgesic tests in rat. Eur J Pharmacol 1978;52:215–23. Tan-No K, Esashi A, Nakagawasai O, Niijima F, Tadano T, Sakurada C, Sakurada T, Bakalkin G, Terenius T, Kisara K. Intrathecally administered big dynorphin, a prodynorphin-derived peptide, produces nociceptive behavior through an N-methyl-D -aspartate receptor mechanism. Brain Res 2002;952:7–14. Vanderah TW, Laughlin T, Lashbrook JM, Nichols ML, Wilcox GL, Ossipov MH, Malan Jr. TP, Porreca F. Single intrathecal injections of dynorphin A or des-Tyr-dynorphins produce long-lasting allodynia in rats: blockade by MK-801 but not naloxone. Pain 1996;68:275–81. Wert MA, MacDonald RL. Opiate alkaloids antagonize postsynaptic glycine and GABA responses: correlation with convulsant action. Brain Res 1982;236:107– 19. Woolf CJ. Intrathecal high dose morphine produces hyperalgesia in the rat. Brain Res 1981;209:491–5. Wu J, Lin Q, McAdoo DJ, Willis WD. Nitric oxide contributes to central sensitization following intradermal injection of capsaicin. NeuroReport 1998;9:589–92. Yaksh TL. Spinal opiate analgesia: characteristics and principle of action. Pain 1981;11:293 –346. Yaksh TL, Harty GL. Pharmacology of the allodynia in rats evoked by high dose intrathecal morphine. J Pharmacol Exp Ther 1988;244: 501– 7. Yaksh TL, Harty GJ, Burton M, Onofrio BM. High doses of spinal morphine produce a nonopiate receptor-mediated hyperesthesia: clinical and theoretical implications. Anesthesiology 1986;64: 590– 7.