A murine model of opioid-induced hyperalgesia

Molecular Brain Research 86 (2001) 56–62 www.elsevier.com / locate / bres

Research report

A murine model of opioid-induced hyperalgesia Xiangqi Li, Martin S. Angst, J. David Clark* Veterans Affairs Palo Alto Health Care System and Stanford University Department of Anesthesiology, 3801 Miranda Ave., Palo Alto, CA 94304, USA Accepted 10 October 2000

Abstract Controversies surround the possible long-term physiological and psychological consequences of opioid use. Analgesic tolerance and addiction are commonly at the center of these controversies, but other concerns exist as well. A growing body of evidence suggests that hyperalgesia caused by the chronic administration of opioids can occur in laboratory animals and in humans. In these studies we describe a murine model of opioid-induced hyperalgesia (OIH). After the treatment of mice for 6 days with implanted morphine pellets followed by their removal, both thermal hyperalgesia and mechanical allodynia were documented. Additional experiments demonstrated that prior morphine treatment also increased formalin-induced licking behavior. These effects were intensified by intermittent abstinence accomplished through administration of naloxone during morphine treatment. Experiments designed to determine if the m-opioid receptor mediated OLH in our model revealed that the relatively-selective m-opioid receptor agonist fentanyl induced the thermal hyperalgesia and mechanical allodynia characteristic of OIH when administered in intermittent boluses over 6 days. In complimentary experiments we found that CXBK mice which have reduced m-opioid receptor binding displayed no significant OIH after morphine treatment. Finally, we explored the pharmacological sensitivities of OIH. We found that the N-methyl-D-aspartate (NMDA) receptor antagonist MK-801, the nitric oxide synthase (NOS) inhibitor N G -nitro-L-arginine methyl ester ( L-NAME) and the heme oxygenase (HO) inhibitor tin protoporphyrin (Sn-P) dose-dependently reduced OIH in this model while the NSAID indomethacin had no effect. Thus we have characterized a murine model of OIH which will be useful in the pursuit of the molecular mechanisms underlying this phenomenon.  2001 Elsevier Science B.V. All rights reserved. Theme: Sensory systems Topic: Pain modulation: pharmacology Keywords: Opioid; Mouse; CXBK; Hyperalgesia; Morphine; Fentanyl; Opponent-Process Theory

1. Introduction Though opioids occupy a position of unsurpassed clinical utility for the treatment of many types of pain, some practitioners harbor concerns about the long term consequences of their use. These concerns can generally be classified as either psychological, e.g. psychological dependence and addiction, or physiological, e.g. analgesic tolerance and constipation. Another complication which may be observed with the use of opioids is opioid-induced

*Corresponding author. Tel.: 11-650-493-5000; fax: 11-650-8523423. E-mail address: [email protected] (J.D. Clark).

hyperalgesia (OIH). One setting in which OIH has been observed is with the use of very high doses of either intravenous or intrathecal opioids like morphine [2,8,24,27]. Other opioids have been noted to induce hyperalgesia at high doses as well [9]. This condition generally resolves quickly with reductions in opioid dosage [26]. A second setting in which OIH has been observed is with the chronic use of opioids followed by abrupt reductions in dosage. This manifestation of OIH is not limited to the well-described pain complaints of those who abuse drugs an subsequently undergo withdrawal. Reports now exist which document spontaneous pain, allodynia and thermal hyperalgesia in humans after abrupt cessation of opioids administered for therapeutic purposes [19,21].

0169-328X / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0169-328X( 00 )00260-6

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62

Likewise, sudden cessation of intrathecal morphine delivery because of catheter malfunction can lead to a state of allodynia to mechanical stimuli (light touch) which resolves upon resumption of opioid administration [10]. The increasing clinical enthusiasm for the use of opioids for chronic pain leads us to examine OIH as a potential complication of long-term use. Several reports exist in which OIH was modeled in rats. These studies demonstrated that thermal hyperalgesia could be induced in rats after several days of intermittent systemic morphine dosing, or if continuous delivery of morphine was interrupted by brief periods of naloxoneprecipitated abstinence [14,20,31]. Importantly, no signs of acute withdrawal were present when the hyperalgesia was documented in these studies. In fact, as few as four bolus doses of the relatively selective m-opioid receptor agonist fentanyl given within 1 h can lead to hyperalgesia in a paw pressure pain model lasting for several days [4], though it is unclear what the chronic administration of this opioid would do. Also unclear is whether other opioid receptor subtypes are capable of mediating OIH. Evidence exists that spinal cord opioid receptors can mediate OIH as intermittent intrathecal morphine administration also leads to thermal hyperalgesia [11,14]. Though not well understood, the induction of OIH seems dependent on NMDA receptors as the blockade of these receptors during opioid treatment reduces the resulting OIH [3,11]. These studies extend our understanding of OIH by: (1) characterizing a murine model of OIH, (2) assessing the impact of OIH not only on baseline thresholds to noxious mechanical and thermal stimuli, but also on the impact OIH has on pain behaviors manifested in a model of inflammatory pain, (3) providing evidence that m-opioid receptors mediate OIH in our model using the selective m-opioid receptor agonist fentanyl and m-opioid receptor deficient CXBK mice, and (4) implicating the involvement of NMDA receptors and the NOS and HO enzyme systems through additional pharmacological studies.

2. Methods

2.1. Animals

2.1.1. Experimental animals All protocols involving animals were approved by our institution’s Subcommittee on Animal Studies. The mouse strain used for the majority of these experiments was the C57BL / 6J (Jackson Labs, Bar Harbor, MD). Male mice between 16 and 22 wks (25–30 g) were used. Where noted male mice of the same age and size of the CXBK strain (Jackson Labs) were used. All animals were kept 4–6 / cage with a 12 h–12 h light–dark cycle and food and water ad libitum.

57

2.2. Drugs 2.2.1. Morphine treatment The chronic administration of morphine was accomplished by implanting mice with subcutaneous morphine pellets as we have previously reported [15,16]. For these experiments a 75 mg morphine pellet (NIDA) was first coated on one surface with acrylic nail polish which prevented the disintegration of the tablet with time in the moist subcutaneous environment. Mice in the morphine treated group were briefly anesthetized with isoflurane, and a small subcutaneous skin pocket was made on the animal’s back into which a morphine tablet was placed followed by closure with surgical staples. Animals in the control group had a skin pocket made and closed. Animals in the morphine / naloxone group underwent morphine pellet implantation, then intermittent morphine abstinence was accomplished by injecting naloxone 1 mg / kg subcutaneously on days 2, 4 and 6. Behavioral testing always preceded naloxone injection. On day 6, morphine pellets were removed from animals in the morphine and morphine / naloxone groups. Control animals underwent reopening followed by re-closure of the skin pocket on day 6. Thermal hyperalgesia and mechanical allodynia were assayed beginning 24 h later. 2.2.2. Fentanyl treatment In some studies mice were injected daily at 08:00 and 17:00 for 6 days with 0.3 mg / kg fentanyl subcutaneously. Control animals received saline injections of equal volume. No naloxone was administered in this protocol. Behavioral testing began 24 h after the last fentanyl injection which was carried out in the same manner as described for mice after morphine treatment. 2.2.3. Reduction of OIH The analgesic effects of various drugs were assessed in the setting of OIH24 h after the removal of morphine pellets from animals treated according to the morphine / naloxone protocol. The NOS inhibitor N G -nitro-L-arginine methyl ester ( L-NAME), NMDA receptor agonist Nmethyl-D-aspartate and the NSAID indomethacin were purchased form Sigma (St. Louis, MO). The heme oxygenase inhibitor Sn-P was obtained from Porphyrin Products (Ogden, UT). After measuring baseline thermal withdrawal latencies and mechanical withdrawal thresholds, drugs were injected subcutaneously in 50–100 ul 0.9% NaCl followed by retesting after 25–35 min. 2.3. Behavioral testing 2.3.1. Formalin assay The formalin assay was used as a model of inflammatory pain as we have described in previous reports [6,15]. For this assay mice were first taken to the behavioral testing area and placed on a glass surface inside a

58

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62

cylindrical clear plastic enclosure 10 cm in diameter and 30 cm in height. After acclimating for 20 min the mice were injected with 25 ul formalin 0.5 or 5% in 0.9% NaCl on the dorsal surface of one hind paw and placed back in the plastic enclosure. For this a microsyringe (Hamilton, Las Vegas, NV) was used with a 30 ga needle. The number of seconds each animal spent licking or biting the injected hind paw over the first 5 min (phase I) and form 10–40 min post injection (phase II) were recorded.

prevent tissue damage. For experiments in which morphine-induced analgesia was being measured, the light beam intensity was initially adjusted to provide a relatively short 4–5 s baseline thus facilitating the quantification of analgesia. In experiments in which we sought to quantify thermal hyperalgesia the light beam intensity was adjusted to provide a somewhat longer (8–9 s) baseline thus facilitating the detection of thermal hyperalgesia. Two measurements were made per animal per test session.

2.3.2. Mechanical withdrawal thresholds Mechanical allodynia was assayed using nylon von Frey filaments according to the ‘up–down’ algorithm described by Chaplan et al. [5] as we have used previously [18]. In these experiments mice were placed in mesh platforms in plastic enclosures as described above. After 20 min of acclimation, fibers of sequentially increasing stiffness (0.2–2 g, 7 fibers) were applied to the center of the plantar surface of the hind paw just distal to the first set of foot pads and left in place for 5 s. Purposeful withdrawal of the hind paw from the fiber was scored as a response. When no response was obtained the next stiffest fiber in the series was applied; if a response was obtained a less stiff fiber was next applied. Testing proceeded in this manner until four fibers had been applied after the first one causing a withdrawal response allowing the estimation of the mechanical withdrawal threshold.

2.4. Statistical analysis

2.3.3. Thermal withdrawal latency Response latencies to noxious thermal stimulation were measured using the method of Hargreaves [13] as we have modified for use with mice [6]. In this assay mice were placed on a glass platform (23.5–24.08C) in a plastic enclosure as described above. After 20 min of acclimation, a beam of focused light was directed towards the same area of the hind paw as described for the von Frey assay. The time to purposeful withdrawal of the foot from the beam of light was measured. A 15-s cutoff was used to

Analysis of repeated measures was accomplished using ANOVA analysis followed by Dunnett’s testing. Comparison of 2 means, e.g. in the formalin testing paradigm, was carried out using a two-tailed Student’s t-test. Data are presented as the mean6standard error of measurement, S.E.M. *P,0.05, **P,0.01.

3. Results

3.1. Effects of subcutaneous morphine administration The continuous administration of morphine by implantation of morphine pellets lead to transient analgesia in mice. Data in Fig. 1 demonstrate that mice implanted with morphine pellets had prolonged paw flick latencies compared with sham-operated control animals for the first 2 days after surgery, but indistinguishable latencies after that time. At no point was thermal hyperalgesia observed during this phase of the experiments where morphine pellets were in place. There were no differences between the latencies observed in the morphine and morphine / naloxone groups, though paw flick latencies were never measured directly after naloxone injection. We demonstrated in previous studies that male C57B1 / 6 mice treated with subcutaneous morphine pellets in similar protocols

Fig. 1. Male C57BL / 6J mice implanted with subcutaneous morphine pellets exhibit transient analgesia as demonstrated with Hargreaves testing. Sham operated mice do not exhibit significant changes in their paw flick latencies. Withdrawal latencies were measured prior to naloxone injection on days 2, 4 and 6. For each group n56. *P,0.05, **P,0.01.

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62

were tolerant to the analgesic effects of morphine after 6 days of treatment by virtue of a rightward shift of their dose–response relationships in the Hargreaves thermal withdrawal paradigm [15,16].

3.2. Hyperalgesia and allodynia after morphine pellet removal After removal of the morphine pellets, mice exhibited OIH. Fig. 2A and B show data demonstrating thermal hyperalgesia and mechanical allodynia which were maximal at 24 h after pellet removal and which slowly resolved over the following days. Animals in the morphine / naloxone group appeared to have a more profound and longerlived period of OIH than animals receiving morphine alone. Again, sham operated control animals did not show any changes in thermal or mechanical responses during this time. No overt signs of withdrawal were observed during behavioral testing such as shaking, jumping or diarrhea.

3.3. Formalin testing in the setting of OIH In the next series of experiments we wished to examine the interaction of OIH with noxious stimulation of an

Fig. 2. Mice treated 6 days with morphine or morphine / naloxone exhibited profound thermal hyperalgesia and mechanical allodynia after removal of the morphine pellets. Panel A contains data from Hargreaves thermal paw withdrawal latency experiments, while panel B contain data collected using von Frey filaments for mechanical stimulation. Note that Hargreaves assay conditions have been adjusted in these assays so that the baseline paw flick latencies approximate 8 s in Hargreaves experiments thus providing greater sensitivity to hyperalgesic changes. For each group n510. *P,0.05, **P,0.01.

59

inflammatory etiology. For this reason we measured formalin-induced hind paw licking in control animals and ones treated with either the morphine or morphine / naloxone protocols. All mice in these experiments were used 48 h after morphine pellet removal. The results presented in Fig. 3A and B demonstrate that while there were no differences in formalin-induced licking times when high (5%) formalin was used, when low (0.5%) formalin was injected animals having been treated with morphine / naloxone had significantly greater phase I and phase II licking times. Preliminary experiments established 5% formalin to be supra-maximal for the stimulation of licking behavior.

3.4. Opioid receptor subtype modulating OIH We then changed the focus of our experiments to the issue of the type of opioid receptor which might mediate OIH in this murine model. Given that one other group had reported the m-selective agonist fentanyl effective in causing OIH in rats after relatively acute exposure [4], we hypothesized that m-opioid receptors could mediate OIH in our murine model in a more chronic administration paradigm. To test this hypothesis, we used two approaches, the use of a selective m-opioid receptor agonist to attempt to induce OIH, and the use of a strain of mice having severely reduced CNS m-opioid receptor expression, the CXBK strain [22,23]. In the first experiments we injected C57B1 / 6J mice as used in the morphine studies with fentanyl twice daily for 6 days. The data in Fig. 4 demonstrate that the fentanyl protocol did in fact lead to

Fig. 3. Mice treated 6 days with morphine / naloxone exhibited increased formalin-induced licking behavior when sub-maximal concentrations were used (panel A, 0.5%), but not when supra-maximal concentrations were used (panel B, 5%). For panel A n56 / group, for panel B n57 mice / group. *,0.05, **P,0.01.

60

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62 Table 1 The pharmacological sensitivities of opioid-induced hyperalgesia a Treatment

Fig. 4. Mice treated with intermittent fentanyl boluses develop thermal hyperalgesia and mechanical allodynia after 6 days of treatment. In these experiments n56 / group. *P,0.05, **,0.01.

thermal hyperalgesia and mechanical allodynia after cessation of injections. Furthermore, the magnitude of these changes was comparable to those seen in the morphine studies.

Paw flick latency

Paw withdrawal threshold

Control

7.760.4

1.3560.14

Morphine / Naloxone Treated 1MK-801 0.02 mg / kg 1MK-801 0.10 mg / kg 1Sn-P 5 umol / kg 1Sn-P 25 umol / kg 1L-NAME 3 mg / kg 1L-NAME 30 mg / kg 1Indomethacin 3 mg / kg 1Indomethacin 10 mg / kg

4.960.6

0.5560.06

6.1*60.3 7.9*60.7 5.560.8 7.2**60.9 6.1*60.5 7.3**60.9 4.460.5 5.260.7

1.16**60.22 1.41**60.43 0.77*60.12 1.50**60.22 1.04**60.30 1.41**60.35 0.6360.04 0.4760.07

a

After rendering mice hyperalgesic using the morphine / naloxone protocol, various pharmacological agents were administered in attempts to reduce or reverse the thermal hyperalgesia (paw flick latency, s) and mechanical allodynia (withdrawal threshold, g). Data represent mean responses for 5–6 mice per each dose of each drug. *P,0.05, **P,0.01.

In complimentary experiments CXBK mice were treated according to our morphine / naloxone protocol. As opposed to the results presented in Fig. 2, the CXBK mice did not exhibit either thermal hyperalgesia or mechanical allodynia after morphine treatment at the 24 h time point where those phenomena were maximal in the C57B1 / 6J animals. These data are presented in Fig. 5.

3.5. Pharmacological sensitivities of OIH The next stage in our characterization of OIH in C57BL / 6 mice was to test the hypothesis that agents commonly reported to reduce or reverse neuropathic and / or inflammatory pain reduce or reverse OIH. In Table 1 data are presented demonstrating that the NMDA receptor antagonist MK-801, the NOS inhibitor L-NAME and the HO inhibitor Sn-P all dose-dependently reverse the thermal hyperalgesia and mechanical allodynia in mice treated according to the morphine / naloxone protocol. No significant effects of indomethacin at doses up to 10 mg / kg were observed, though profound analgesic effects of this drug at this dose have been noted in the mouse formalin model of inflammatory pain [25].

4. Discussion

Fig. 5. Mice of the CXBK strain do not develop significant thermal hyperalgesia or mechanical allodynia after exposure to morphine / naloxone. Data from CXBK mice are displayed alongside data from C57B1 / 6J mice used as controls. Baseline data are from animals prior to morphine pellet (morphine sulfate, MSO4) implantation. Hyperalgesia (panel A) and mechanical allodynia (panel B) were assessed 24 h after pellet removal. For both strains n59 / group. *P,0.05, **P,0.01.

The principal conclusions we draw from our data are: (1) that C57B1 / 6J mice demonstrate profound and longlasting thermal hyperalgesia and mechanical allodynia after chronic treatment with morphine with periodic abstinence achieved through the injection of the opioid receptor antagonist naloxone, (2) that pain behaviors in response to inflammatory stimuli are likewise increased in the setting of OIH, (3) that m-opioid receptors may mediate 01H since selective m-opioid receptor agonists can cause OIH in

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62

mice, and m-opioid receptor deficient CXBK mice do not develop OIH under the same conditions leading to the manifestation of OIH in the C57B1 / 6J strain, and (4) that NMDA receptors and the NOS and HO enzyme systems may be mechanistically involved in the manifestation of OIH. To this point there has not been a murine model of OIH described in this degree of detail. With this model in hand we are now in a position to use transgenic and knockout strains of mice as well as pharmacological and biochemical techniques to probe the underlying mechanisms responsible for OIH. Though another group has suggested that m-opioid receptors can be involved in OIH after acute opioid administration, ours are the first experiments employing pharmacological tools as well as mice having altered m-opioid receptor signaling to implicate a specific type of opioid receptor in the etiology of OIH due to the chronic administration of opioids. We have not addressed, however, whether or not it is possible that other opioid receptor subtypes could mediate OIH under specific circumstances. It is known that physical dependence to both d and k opioid receptors can occur [7,12,30], So it may be that these opioid receptor subtypes can mediate OIH as well. A second issue our studies did not address is the anatomical location of the opioid receptors mediating OIH after systemic administration; is the OIH observed after systemic morphine administration a peripherally or centrallymediated phenomenon? Data from rat studies suggest that chronic intrathecal opioid administration can lead to a state of OIH [14]. However, persistent activation of peripheral m-opioid receptors followed by naloxone-precipitated abstinence has also been linked to hyperalgesia [1]. Therefore, OIH observed after systemic opioid administration may have both central and peripheral components. Little is understood of the mechanism whereby chronic stimulation of opioid receptors leads to the manifestation of hyperalgesia when opioid administration is discontinued. One behavioral theory which seems to fit our observations is the opponent-process theory of acquired motivation [28,29]. Applied to opioid administration this theory would predict that when mice are exposed to analgesics chronically, that an opposing process is activated to return the pharmacologically elevated pain thresholds towards baseline values (tolerance). Furthermore, this theory would predict that if the analgesics were to be abruptly discontinued that pain thresholds would transiently drop to levels below baseline (hyperalgesia). These predictions are consistent with our data. On a molecular level it seems clear from our data and the data from rat model systems that NMDA receptors are involved in the manifestation of OIH. Others have shown in rats that administering NMDA receptor antagonists at the same time that morphine was administered not only reduced analgesic tolerance, but reduced the signs of OIH after the cessation of opioid administration as well [4,11,14]. Our data demonstrate that NMDA receptor

61

blockers reverse OIH once established. In light of the coexistence of opioid tolerance and OIH in the same animals, and the fact that blockade of NMDA receptors can reduce both tolerance and OIH, it would seem reasonable to hypothesize that tolerance and OIH have significant mechanistic similarities, though this remains to be demonstrated empirically. It has been hypothesized that NMDA receptor, nitric oxide synthase (NOS) and protein kinase C (PKC) activation which occur during chronic exposure to opioids are events common to both the mechanism of opioid tolerance and OIH [20]. Our studies also implicate the HO enzyme system in OIH. The HO enzyme system has been shown to be involved in pain of several other etiologies such as inflammatory [17,32], incisional [18] and neuropathic [18]. In fact we have recently reported that spinal cord HO enzymatic activity is increased in animals chronically treated with morphine, and that an increased expression of the HO-2 isozyme is likely partially responsible [16]. Our observation that both NOS and HO inhibitors reverse OIH is consistent with the hypothesis that these enzyme systems play parallel roles in nociception of various etiologies. The phenomenon under investigation here (OIH) may be relevant to clinical medicine. As presented in the introduction, clinical reports have appeared linking states of hyperalgesia to rapid reductions in opioid doses [10,19,21]. Given that the use of opioids is gaining popularity in the management of chronic pain, it behooves us to understand all of the potential complications of that form of management. In particular, the information to date suggests that OIH might be more likely to occur in situations where intermittent short acting opioids are used in preference to long-acting opioids. Additionally, rapid discontinuation of intrathecal opioids as could occur if an implanted system’s pump or catheter were to fail could lead to a state of OIH. Though our understanding of OIH in humans lacks even our rudimentary understanding of the phenomenon in animals, it seems reasonable that the pursuit of studies in animals could lead to human studies directed as assessing the importance of OIH during or after the administration of opioids for therapeutic reasons. Later, strategies may be developed to limit the severity of OIH generated in clinical populations, or perhaps treatments for combating OIH once present could be designed.

References [1] K.O. Aley, J.D. Levine, Multiple receptors involved in peripheral alpha 2, mu, and Al antmociception, tolerance, and withdrawal, J. Neurosci. 17 (1997) 735–744. [2] N.M. Ali, Hyperalgesic response in a patient receiving high concentrations of spinal morphine [letter], Anesthesiology 65 (1986) 449. [3] E. Celerier, J. Laulin, A. Larcher, M. Le Moal, G. Simonnet, Evidence for opiate-activated NMDA processes masking opiate analgesia in rats, Brain Res. 847 (1999) 18–25.

62

X. Li et al. / Molecular Brain Research 86 (2001) 56 – 62

[4] E. Celerier, C. Rivat, Y. Jun, J.P. Laulin, A. Larcher, P. Reynier, G. Simonnet, Long-lasting hyperalgesia induced by fentanyl in rats: preventive effect of ketamine [see comments], Anesthesiology 92 (2000) 465–472. [5] S.R. Chaplan, F.W. Bach, J.W. Pogrel, J.M. Chung, T.L. Yaksh, Quantitative assessment of tactile allodynia in the rat paw, J. Neurosci. Methods 53 (1994) 55–63. [6] J.D. Clark, B.L. Tempel, Hyperalgesia in mice lacking the Kv1.1 potassium channel gene, Neurosci. Lett. 251 (1998) 121–124. [7] F.C. Dalman, K.L. O’Malley, kappa-Opioid tolerance and dependence in cultures of dopaminergic midbrain neurons, J. Neurosci. 19 (1999) 5750–5757. [8] F. De Conno, A. Caraceni, C. Martini, E. Spoldi, M. Salvetti, V. Ventafridda, Hyperalgesia and myoclonus with intrathecal infusion of high-dose morphine, Pain 47 (1991) 337–339. [9] J. Devulder, Hyperalgesia induced by high-dose intrathecal sufentanil in neuropathic pain, J. Neurosurg. Anesthesiol. 9 (1997) 146–148. [10] J. Devulder, P. Bohyn, F. Castille, M. De Laat, G. Rolly, A case of uncommon withdrawal symptoms after a short period of spinal morphine administration, Pain 64 (1996) 589–591. [11] S.A. Dunbar, I.J. Pulai, Repetitive opioid abstinence causes progressive hyperalgesia sensitive to N-methyl-D-aspartate receptor blockade in the rat, J. Pharmacol. Exp. Ther. 284 (1998) 678–686. [12] Y.Z. Feng, T. Zhang, S. Toknyama, H. Zhu, R.W. Rockhold, L.K. Ho, Mu- and delta-opioid receptor antagonists precipitate similar withdrawal phenomena in butorphanol and morphine dependence, Neurochem. Res. 21 (1996) 63–71. [13] K. Hargreaves, R. Dubner, F. Brown, C. Flores, J. Joris, A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia, Pain 32 (1988) 77–88. [14] T. Ibuki, S.A. Dunbar, T.L. Yaksh, Effect of transient naloxone antagonism on tolerance development in rats receiving continuous spinal morphine infusion, Pain 70 (1997) 125–132. [15] X. Li, J.D. Clark, Changes in transcription factor expression in the spinal cords of morphine tolerant mice, Neurosci. Lett. 272 (1999) 79–82. [16] X. Li, J.D. Clark, Chronic morphine exposure and the expression of heme oxygenase type 2, Molecular Brain Res. 75 (2000) 179–184. [17] X. Li, J.D. Clark, Heme oxygenase type 2 plays a role in formalininduced nociception, Pain 86 (2000) 75–80, [Record as supplied by publisher]. [18] X. Li, J.D. Clark, The role of heme oxygenase in neuropathic and incisional pain, Anesthesia and Analgesia 90 (2000) 677–682. [19] J.J. Lipman, B. Blumenkopf, Comparison of subjective and objec-

[20]

[21]

[22]

[23]

[24]

[25] [26]

[27]

[28]

[29]

[30]

[31]

[32]

tive analgesic effects of intravenous and intrathecal morphine in chronic pain patients by heat beam dolorimetry, Pain 39 (1989) 249–256. D.J. Mayer, J. Mao, J. Holt, D.D. Price, Cellular mechanisms of neuropathic pain, morphine tolerance, and their interactions, Proc. Natl. Acad. Sci. USA 96 (1999) 7731–7736. A.W. Miser, K.J. Chayt, J.T. Sandlund, P.S. Cohen, J.A. Dothage, J.S. Miser, Narcotic withdrawal syndrome in young adults after the therapeutic use of opiates, Am. J. Dis. Child 140 (1986) 603–604. A.S. Moskowitz, R.R. Goodman, Autoradiographic analysis of mu1, mu2, and delta opioid binding in the central nervous system of C57BL / 6BY and CXBK (opioid receptor-deficient) mice, Brain Res. 360 (1985) 108–116. A.S. Moskowitz, G.W. Terman, K.R. Carter, M.J. Morgan, J.C. Liebeskind, Analgesic, locomotor and lethal effects of morphine in the mouse: strain comparisons, Brain Res. 361 (1985) 46–51. J.M. Potter, D.B. Reid, R.J. Shaw, P. Hackett, P.E. Hickman, Myoclonus associated with treatment with high doses of morphine: the role of supplemental drugs [see comments], Br. Med. J. 299 (1989) 150–153. M. Shibata, T. Ohkubo, H. Takahashi, R. Inoki, Modified formalin test: characteristic biphasic pain response, Pain 38 (1989) 347–352. P. Sjogren, N.H. Jensen, T.S. Jensen, Disappearance of morphineinduced hyperalgesia after discontinuing or substituting morphine with other opioid agonists, Pain 59 (1994) 313–316. P. Siogren, T. Jonsson, N.H. Jensen, N.E. Drenck, T.S. Jensen, Hyperalgesia and myoclonus in terminal cancer patients treated with continuous intravenous morphine, Pain 55 (1993) 93–97. R.L. Solomon, The opponent-process theory of acquired motivation: the costs of pleasure and the benefits of pain, Am. Psychol. 35 (1980) 691–712. R.L. Solomon, J.D. Corbit, An opponent-process theory of motivation. I. Temporal dynamics of affect, Psychol. Rev. 81 (1974) 119–145. T. Suzuld, M. Tsuji, T. Mori, M. Misawa, H. Nagase, Involvement of delta 1 and delta 2 opioid receptor subtypes in the development of physical dependence on morphine in mice, Pharmacol. Biochem. Behav. 57 (1997) 293–299. H.A. Tilson, R.H. Rech, S. Stolman, Hyperalgesia during withdrawal as a means of measuring the degree of dependence in morphine dependent rats, Psychopharmacologia 28 (1973) 287–300. T. Yamamoto, N. Nozaki-Taguchi, Zinc protoporphyrin IX, an inhibitor of the enzyme that produces carbon monoxide, blocks spinal nociceptive transmission evoked by formalin injection in the rat, Brain Res. 704 (1995) 256–262.