Neuroscience Letters 368 (2004) 82–86
Cannabinoids blocks tactile allodynia in diabetic mice without attenuation of its antinociceptive effect Ahmet Do˘grula,∗ , H¨usamettin G¨ula , O˘guzhan Yıldıza , Ferruh Bilginb , M. Erdal G¨uzeldemirb a b
Department of Pharmacology, G¨ulhane Academy of Medicine, 06018 Etlik, Ankara, Turkey Department of Anesthesiology, G¨ulhane Academy of Medicine, 06018 Etlik, Ankara, Turkey Received 25 May 2004; received in revised form 23 June 2004; accepted 23 June 2004
Abstract Diabetic neuropathic pain is one of the most commonly encountered neuropathic pain syndromes. However, the treatment of diabetic neuropathic pain is challenging because of partial effectiveness of currently available pain relievers. It is well known that diabetic animals are less sensitive to the analgesic effect of morphine, and opioids are found to be ineffective in the treatment of diabetic neuropathic pain. Cannabinoids are promising drugs and they share a similar pharmacological properties with opioids. It has been reported that cannabinoid analgesia remained intact and to be effective in some models of nerve injury. Thus, we investigated antinociceptive efficacy and the effects of cannabinoids on behavioral sign of diabetic neuropathic pain in diabetic mice by using WIN 55, 212-2, a cannabinoid receptor agonist. Diabetes was induced by streptozotocin (STZ) (200 mg/kg) and animals were tested between 45 and 60 days after onset of diabetes. Antinociception was assessed using the radiant tail-flick test. Mechanical and thermal sensitivities were measured by Von Frey filaments and hot-plate test, respectively. Tactile allodynia, but not thermal hyperalgesia developed in diabetic mice. Systemic WIN 55, 212-2 (1, 5 and 10 mg/kg) produced a dose-dependent antinociception both in diabetic and control mice. WIN 55, 212-2-induced antinociception were found to be similar in diabetic mice when compared to controls suggesting efficacy of cannabinoid antinociception was not diminished in diabetic mice. WIN 55, 212-2 also produced a dose-dependent antiallodynic effect in diabetic mice. This study suggests that cannabinoids have a potential beneficial effect on experimental diabetic neuropathic pain. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cannabinoid; Diabetes mellitus; Neuropathy; Analgesia; Allodynia; Hyperalgesia
Painful neuropathy is one of the most common long-term complications of diabetes mellitus and often proves difficult to relieve [13,25,27,38]. Diabetic patients frequently exhibit one or more types of stimulus-evoked pain, including increased responsiveness to noxious stimuli (hyperalgesia) as well as hyper-responsiveness to normally innocuous stimuli (allodynia) [25,35]. Neuropathic pain has previously been regarded as less responsive to opioids than nociceptive pain [5,38]. Clinical and animal studies show that morphine has a limited effect in the treatment of painful diabetic neuropathy [3,4,9,29,30]. It is also well established that morphine shows less antinociceptive effects in diabetic animals when com∗
Corresponding author. Tel.: +90 312 3044765; fax: +90 312 3042150. E-mail address:
[email protected] (A. Do˘grul).
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.06.060
pared to naive controls [16,29,30,36]. It has been suggested that factors triggering pathological pain status in diabetes may also activate some intracellular cascades and contribute to reduced effectiveness of opioid analgesia in diabetic neuropathic pain [15,28,30]. Cannabinoids are promising analgesic drugs and they share a similar pharmacological properties with opioids. The ability of cannabinoids to inhibit acute nociception are well known and acute cannabinoid analgesia are comparable with opiates in potency and efficacy [9,14,39,40]. There are some studies showing that cannabinoids are potential analgesics in chronic pain status and cannabinoids have a beneficial effect in various neuropathic pain models [6,7,12,18,34,33]. In contrary to reduced opioid analgesia in pathological pain status [3,28,32] 9 -tetrahydrocannabinol-induced antinoci-
A. Do˘grul et al. / Neuroscience Letters 368 (2004) 82–86
ception remained intact in rat model of neuropathic pain resulting from chronic constriction nerve injury [12,28]. It has been reported that endogenous cannabinoids regulates sensory threshold and attenuating of this inhibitory tonus may result in enhanced excitatory input and sensory hypersensitivity [8,14]. To our knowledge, there is no study to explore the antinociceptive efficacy of cannabinoids in diabetes and the contribution of cannabinoid system in diabetic neuropathic pain. Therefore, we aimed to compare the antinociceptive efficacy of cannabinoids in diabetic and control mice using WIN 55, 212-2, a cannabinoid receptor agonist. We also investigated the effects of WIN 55, 212-2 on behavioral sign of diabetic neuropathic pain. Adult female Bulb/c mice weighing 25–30 g were used. Animals were maintained on a 12 h/12 h light–dark cycle with food and water available ad libitum. Mice were handled in accordance with guidelines for the care of laboratory animals and the ethical guidelines for investigations of experimental pain in conscious animals [41]. R-(+)-WIN 55, 212-2, a mixed CB1 /CB2 receptor agonist was obtained from Sigma-RBI (St. Louis, USA). R-(+)-WIN 55, 212-2 was dissolved 50% DMSO in saline. Diabetes was induced by a single intraperitoneal (i.p.) injection of streptozotocin (STZ) (200 mg/kg body weight) after an overnight fast. STZ was dissolved in 3 mM citric acid buffer (pH 4.5) immediately before injection. Age matched non-diabetic mice received similar volumes of vehicle alone by the same route. Mice with serum glucose levels above 250 mg/dl were considered diabetic and used in this study. Mice were left for 45–60 days following STZ injection to allow for the development of neuropathic changes in diabetic mice. At the end of this period, control and diabetic mice were divided into different groups, each of which containing six to eight animals. Antinociception was assessed using the radiant heat tailflick test (Columbus, OH, USA; Type 812). Baseline tailflick latency (BL) for each mouse typically ranged from 2.5 to 3 s. Cut-off time was set at 7 s to prevent tissue damage in mice. Mice were tested at 15, 30, 45 and 60 min after WIN 55, 212-2 injections to measure test latency (TL). In order to generate a dose–response curve against antinociception, data were converted to percent maximal possible effect (MPE) by using the equation: %MPE = (TL − BL) × 100/(7 − BL). ED50 values with 95% confidence intervals (CI) of WIN 55, 212-2 in control and diabetic mice were calculated from regression analysis of the linear portion of the dose–response curves by means of the customized Visual Basic program FlashCalc (Michael H. Ossipov, personal communication) at the time of the peak effect. If CI of WIN55, 212-2 induced antinociception in diabetic groups did not overlap CI of WIN-55, 212-2-induced antinoception in control mice, a difference of efficacy of WIN-55, 212-2-induced antinociception was considered to be present in control and diabetic mice. The assessment of tactile allodynia was determined by the application of calibrated von Frey filaments to the plantar as-
83
pect of the hindpaw of mice which were kept in suspended wire-mesh cages. A response was indicated by lifting of the hindpaw. Withdrawal threshold was determined by sequentially increasing and decreasing the stimulus strength (‘up and down’ method), analyzed with a Dixon nonparametric test [2,7]. Tactile allodynia was indicated by a significant reduction (P < 0.05; Student’s t-test) in the paw withdrawal threshold when compared to that obtained before any manipulations. Mice were tested at 0, 15, 30, 45 and 60 min after injections of WIN-55, 212-2 or vehicle. The hot-plate test was used to assess paw withdrawal latency to a thermal nociceptive stimulus. Each mouse was placed individually on a hot plate and the reaction time was measured starting from time the mouse was placed on the plate until the mouse either demonstrated hind paw licking or jumping. To allow for the detection of thermal hyperalgesia, baseline temperature was maintained at 52 ◦ C. The cut-off latency used was 60 s to prevent tissue damage. A non-parametric method of statistical analysis was used. Statistical significance of more than two groups were evaluated by ANOVA (P < 0.05), followed by Dunnett’s multiple test for individual comparisons. Student’s t-test was used to compare two groups (P < 0.05). After 14 days following STZ injection, STZ-induced diabetic mice exhibited significantly increased blood glucose level (303 ± 21.2 mg/dl) compared to the vehicle-treated control mice (137 ± 8.4 mg/dl; P < 0.01). Baseline tail-flick latencies were found to be 2.85 ± 0.33 s and 2.35 ± 0.31 s in control and diabetic groups, respectively, that were not significantly different. We assessed the dose–response curves of WIN 55, 212-2 administered by i.p. in control and diabetic mice. WIN 55, 212-2 (1, 5 and 10 mg/kg) significantly prolonged the tail-flick latencies in a dose-dependent manner in control (Fig. 1a) and diabetic mice (Fig. 1b). Time action curves revealed that antinociceptive response of highest dose of WIN 55, 212-2 (10 mg/kg) began at 15 min and peaked to 6.8 ± 0.12 s and 6.59 ± 0.27 s at 30 min in control and diabetic mice, respectively. There was no significant difference in antinociceptive responses between the control and diabetic groups at the corresponding dose and time matched. The A50 (95% CI) value for WIN 55, 212-2-induced antinociception was 3.42 mg/kg (2.44–5.07) and 2.69 mg/kg (1.38–3.75) in control and diabetic mice at 30 min, respectively (Fig. 1c). The CIs of WIN 55, 212-2 in control and diabetic mice did overlap suggesting that WIN 55, 212-2-induced antinociception were not significantly different between control and diabetic mice. Vehicle-treated control mice displayed paw withdrawal thresholds to probing with von Frey filaments of 2.42 ± 0.14 g. The hindpaw response to probing with von Frey filaments before STZ injection was 2.69 ± 0.11 g. The sensory threshold was significantly reduced in diabetic group to 0.55 ± 0.17 g indicating the development of tactile allodynia in diabetes (P < 0.05). The i.p. injection of 1, 5 and 10 mg/kg of WIN 55, 212-2 dose dependently and significantly reversed tactile allodynia (P < 0.05) (Fig. 2), while injection of WIN
84
A. Do˘grul et al. / Neuroscience Letters 368 (2004) 82–86
Fig. 2. The effects of WIN 55, 212-2 administration by i.p. route on mechanical allodynia in diabetic mice. Paw withdrawal thresholds to probing with von Frey filaments were determined prior to and at select time points after several doses of WIN 55, 212-2. WIN 55, 212-2 produced dose-dependent increases in paw withdrawal thresholds to tactile stimuli in diabetic mice, n = 6–8 mice per group.
Fig. 1. Groups of mice received WIN 55, 212-2 (1, 5 and 10 mg/kg, i.p.) and tested in tail-flick latencies in control (a) and diabetic (b) mice at the indicated time over 60 min. Tail-flick latencies at the time of peak effect (30 min) were converted to %MPE in order to generate dose–response curves (c); n = 6–8 animals per group.
55, 212-2 i.p. at these doses did not effect the paw withdrawal thresholds in control mice (data not shown). A peak antiallodynic effect was evident at 60 min after the highest dose of WIN 55, 212-2 (10 mg/kg) in diabetic mice (Fig. 2). Thermal sensitivity of animals in hot-plate test were found to be 32.07 ± 3.63 s and 28.53 ± 1.65 s in control and diabetic mice, respectively, that were not significantly different from each other suggesting that thermal hyperalgesia did not develop in diabetic mice in hot-plate test.
In the present study, we showed that WIN 55, 212-2 had antinociceptive effects both in control and diabetic mice. The antinociceptive potency of WIN 55, 212-2 was found to be similar in diabetic and control mice suggesting that the antinociceptive effect of cannabinoids was preserved in diabetes. In addition, WIN 55, 212-2 had an antiallodynic effects in diabetic mice by restoring the decreased paw withdrawal thresholds to baseline control levels. It has been reported that STZ-induced diabetic mice are significantly less responsive to the antinociceptive effects of opioids in tail-flick test [11,24,36,29]. In contrast to opioids, our study demonstrates that cannabinoid antinociception is not changed in STZ diabetic mice when compared to respective controls. It has been well known that both opioids and cannabinoids are effective antinociceptive drugs in normal animals without pathological pain [28]. However, there are some reports that the antinociceptive effects of opioids and cannabinoids differ in pathological pain states [28]. In pathological pain states, such as painful peripheral neuropathy, opioid antinociception was reduced, but cannabinoid antinociception was not changed [28]. In support of efficacy of cannabinoids in pathological pain, our findings showed that the efficacy of WIN 55, 212-2, a cannabinoid agonist, remains unaffected in a streptozotocin model of painful diabetic neuropathy. The mechanism underlying differences between the efficacy of opioid and cannabinoid-induced antinociception in diabetic mice is not clear and remains to be determined. However, anatomical dissociation between cannabinoid and mu-opioid modulation of sensory transmission at the level of the primary afferent inputs to the spinal cord may explain these differences. It has been reported that the majority of mu-opioid receptors reside presynaptically on the central terminals of thin primary afferent fibers, but only a subpopulation of CB1 receptors situated on the central terminals of primary afferent C-fibers [21]. The reduced efficacy of morphine in diabetic status may be due to the pe-
A. Do˘grul et al. / Neuroscience Letters 368 (2004) 82–86
ripheral nerve lesions and deafferentation has been attributed to loss of presynaptic receptors that mediate the analgesic action of morphine. But, the majority of cannabinoid receptors are not localized presynaptically on thin primary afferent fibers [21], so the antinociceptive efficacy of cannabinoids may remain unchanged in diabetes. Alternatively, the differences in the efficacy of opioids and cannabinoids in diabetic status may be caused by disruption of different type of G protein coupling, alterations in second/third messenger systems and/or ion channels which affects the efficacy of opioids [3,31], but not cannabinoids. Mechanical allodynia, assessed by von Frey filaments, developed within 45–60 days in diabetic mice after STZ injection is consistent with a lot previous studies. However, we did not observe significant changes in the withdrawal latency to noxious heat stimulation into paw in the hot-plate test in diabetic mice. The failure to detect thermal hyperalgesia in hot-plate test may reflect the rapid response latency in these test endpoints. However, although mechanical allodynia are fairly consistent with reduction in mechanical threshold as demonstrated in STZ diabetic models, reports of changes in thermal nociceptive thresholds have been highly variable, with thermal hyperalgesia observed in some studies and others observing no changes or decreased thermal sensitivity [11,37]. Consistent with our study, there are some reports that tactile allodynia but not thermal hyperalgesia developed in STZ diabetic rat [10,24]. It is not clear why the mechanical threshold is preferentially altered in diabetic mice. It has been proposed that mechanical allodynia is predominantly mediated by myelinated afferents, while thermal hyperalgesia is mainly transmitted through unmyelinated C-fiber in neuropathic pain [31]. Khan et al. [26] showed that in STZ diabetic rats, unlike C-fiber afferents which mediating thermal nociception, myelinated A-delta and A-beta fiber afferents had a lower threshold for activation and their responses to von Frey filaments were significantly augmented, which compared to the same type of afferent fibers recorded from non-diabetic rats suggesting that mechanical allodynia in STZ diabetic model nociception is transmitted by myelinated A-fibers. The findings in the present study demonstrate that systemically administered WIN 55, 212-2 blocks tactile allodynia in diabetic mice. However, the mechanism by which WIN 55, 212-2 blocked tactile allodynia in diabetic mice, remains to be clarified. Systemically administered cannabinoid agonists have been shown to be effective in reducing mechanical allodynia in various nerve injury models of neuropathic pain in animals [1,12,18] and the antiallodynic effects of cannabinoids are mediated by activation of CB1 receptors in those models [12,28]. It has been demonstrated that A-fibers play an important role in the allodynia associated with peripheral nerve injury [19,22] and abnormal sensory input due to aberrant activity in A-delta and A-beta fiber underlies mechanical allodynia in diabetes [22]. Interestingly, Bridges et al. [1] have showed that cannabinoid CB1 receptors predominantly located in myelinated A-fibers on DRG neurons. Thus, taken together, our results would support that the action of WIN
85
55, 212-2 via the CB1 receptors on A-fiber afferents which might account for its anti-allodynic effects observed in STZ diabetic mice. However, diabetes cause nerve injury at whole of the nervous system, namely, from the peripheral nerves to the brain [23], and CB1 receptors has been found throughout the central nervous system [17,20]. So, it is possible that changes in the expression or sensitivity of CB1 receptors not only at the primary afferent fibers but in whole of the nervous system due to the diabetic status may affect the pharmacology of WIN 55, 212-2 over the CB1 receptors. To address this issue, further studies are needed to compare the activity and the expression of CB1 receptors in nerve tissue in diabetic and non-diabetic control animals. In conclusion, cannabinoid agonist WIN 55, 212-2 has clearly important antinociceptive effect both in control and diabetic mice. The antinociceptive potency of WIN 55, 212-2 is preserved in diabetes. WIN 55, 212-2 has also antiallodynic effect in diabetic neuropathic pain. Since painful neuropathy is an important diabetic complication, which severely affects quality of life, our study support a role of cannabinoid system in diabetic neuropathic pain
Acknowledgements This work was supported by TUBITAK (SBAG-AYD407). The authors would like to thank Professor Frank Porreca for helpful comments and suggestions and Michael H. Ossipov for providing Visual Basic program FlashCalc. The ¨ ur YES˙ILYURT for authors also thank Tayfun Ide and Ozg¨ helpful suggestions.
References [1] D. Bridges, A.S.C. Rice, M. Egertov´ac, M.R. Elphick, J. Winter, G.J. Michael, Localisation of cannabinoid receptor 1 in rat dorsal root ganglion using in situ hybridisation and immunohistochemistry, Neuroscience 119 (2003) 803–812. [2] 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. [3] S.R. Chen, H.L. Pan, Hypersensitivity of spinothalamic tract neurons associated with diabetic neuropathic pain in rats, J. Neurophysiol. 87 (2002) 2726–2733. [4] S.R. Chen, H.L. Pan, Antinociceptive effect of morphine, but not mu opioid receptor number, is attenuated in the spinal cord of diabetic rats, Anesthesiology 99 (2003) 1409–1414. [5] N.I. Cherny, H.T. Thaler, H. Friedlander-Klar, J. Lapin, K.M. Foley, R. Houde, R.K. Portenoy, Opioid responsiveness of cancer pain syndromes caused by neuropathic or nociceptive mechanisms: a combined analysis of controlled, single-dose studies, Neurology 44 (1994) 857–861. [6] B. Costa, M. Colleoni, S. Conti, A.E. Trovato, M. Bianchi, M.L. Sotgiu, G. Giagnoni, Repeated treatment with the synthetic cannabinoid WIN 55, 212-2 reduces both hyperalgesia and production of pronociceptive mediators in a rat model of neuropathic pain, Br. J. Pharmacol. 141 (2004) 4–8. [7] W.J. Dixon, Efficient analysis of experimental observations, Annu. Rev. Pharmacol. Toxicol. 20 (1980) 441–462.
86
A. Do˘grul et al. / Neuroscience Letters 368 (2004) 82–86
[8] A. Dogrul, L.R. Gardell, S. Ma, M.H. Ossipov, F. Porreca, J. Lai, ‘Knock-down’ of spinal CB1 receptors produces abnormal pain and elevates spinal dynorphin content in mice, Pain 100 (2002) 203– 209. [9] A. Dogrul, H. Gul, A. Akar, O. Yildiz, F. Bilgin, E. Guzeldemir, Topical cannabinoid antinociception: synergy with spinal sites, Pain 105 (2003) 11–16. [10] M.J. Field, S. McCleary, J. Hughes, L. Singh, Gabapentin and pregabalin, but not morphine and amitriptyline, block both static and dynamic components of mechanical allodynia induced by streptozocin in the rat, Pain 80 (1999) 391–398. [11] A. Fox, C. Eastwood, C. Gentry, D. Manning, L. Urban, Critical evaluation of the streptozotocin model of painful diabetic neuropathy in the rat, Pain 81 (1999) 307–316. [12] A. Fox, A. Kesingland, C. Gentry, K. McNair, S. Patel, L. Urban, I. James, The role of central and peripheral Cannabinoid1 receptors in the antihyperalgesic activity of cannabinoids in a model of neuropathic pain, Pain 92 (2001) 91–100. [13] B.S. Galer, A. Gianas, M.P. Jensen, Painful diabetic polyneuropathy: epidemiology, pain description, and quality of life, Diabetes Res. Clin. Pract. 47 (2000) 123–128. [14] L.R. Gardell, S.E. Burgess, A. Dogrul, M.H. Ossipov, T.P. Malan, J. Lai, F. Porreca, Pronociceptive effects of spinal dynorphin promote cannabinoid-induced pain and antinociceptive tolerance, Pain 98 (2002) 79–88. [15] M.S. Gold, Spinal nerve ligation: what to blame for the pain and why, Pain 84 (2000) 117–120. [16] H. Gul, O. Yildiz, A. Dogrul, O. Yesilyurt, A. Isimer, The interaction between IL-1beta and morphine: possible mechanism of the deficiency of morphine-induced analgesia in diabetic mice, Pain 89 (2000) 39–45. [17] M. Herkenham, A.B. Lynn, M.D. Little, M.R. Johnson, L.S. Melvin, B.R. de Costa, K.C. Rice, Cannabinoid receptor localization in brain, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 1932–1936. [18] U. Herzberg, E. Eliav, G.J. Bennett, I.J. Kopin, The analgesic effect of R-(+) WIN 55, 212-2 mesylate, a high affinity cannabinoid agonist, in a rat model of neuropathic pain, Neurosci. Lett. 221 (1997) 157–160. [19] H.S. Huang, Q.H. Zhang, F.X. Bao, L. Lu, Y.J.C. Guo, L. Yang, W.J. Huang, G. Fu, S.H. Xu, X.P. Cheng, Q. Yan, Z.D. Zhu, X. Zhang, Z. Chen, Z.G. Han, X. Zhang, Identification of gene expression profile of dorsal root ganglion in the rat peripheral axotomy model of neuropathic pain, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 8360–8365. [20] A.G. Hohmann, E.M. Briley, M. Herkenham, Pre- and postsynaptic distribution of cannabinoid and mu opioid receptors in rat spinal cord, Brain Res. 822 (1999) 17–25. [21] A.G. Hohmann, M. Herkenham, Regulation of cannabinoid and mu opioid receptors in rat lumbar spinal cord following neonatal capsaicin treatment, Neurosci. Lett. 252 (1998) 13–16. [22] L.J. Hudson, S. Bevan, G. Wotherspoon, C. Gentry, A. Fox, J. Winter, VR1 protein expression increases in undamaged DRG neurons after partial nerve injury, Eur. J. Neurosci. 13 (2001) 2105– 2114. [23] A.R. Jadat, D. Carroll, C.J. Glynn, R.A. Moore, H.J. McQuay, Morphine responsiveness of chronic pain: double-blind randomized crossover study with patient-controlled analgesia, Lancet 339 (1992) 1367–1371.
[24] J. Kamei, Y. Ohhashi, T. Aoki, N. Kawasima, Y. Kasuya, Streptozotocin-induced diabetes selectively alters the potency of analgesia produced by mu-opioid agonists, but not by delta- and kappa-opioid agonists, Brain Res. 571 (1992) 199–203. [25] D. Kapur, Neuropathic pain and diabetes, Diabetes Met. Res. Rev. 19 (2003) S9–S15. [26] G.M. Khan, S.R. Chen, H.L. Pan, Role of primary afferent nerves in allodynia caused by diabetic neuropathy in rats, Neuroscience 114 (2002) 291–299. [27] S.T. Krishnan, G. Rayman, New treatments for diabetic neuropathy: symptomatic treatments, Curr. Diab. Rep. 6 (2003) 3459–3467. [28] J. Mao, D.D. Price, J. Lu, L. Keniston, D.J. Mayer, Two distinctive antinociceptive systems in rats with pathological pain, Neurosci. Lett. 280 (2000) 13–16. [29] M. Ohsawa, H. Mizoguchi, M. Narita, J. Kamei, H. Nagase, L.F. Tseng, Effects of a mu-opioid receptor agonist on G-protein activation in streptozotocin-induced diabetic mice, Eur. J. Pharmacol. 401 (2000) 55–58. [30] M. Ohsawa, H. Nagase, J. Kamei, Role of intracellular calcium in modification of mu and delta opioid receptor-mediated antinociception by diabetes in mice, J. Pharmacol. Exp. Ther. 286 (1998) 780–787. [31] M.H. Ossipov, D. Bian, T.P. Malan, J. Lai, F. Porreca, Lack of involvement of capcaicin-sensitive primary afferents in nerve-ligation injury induced tactile allodynia in rats, Pain 79 (1999) 127–1333. [32] M.H. Ossipov, Y. Lopez, M.L. Nichols, D. Bian, F. Porreca, The loss of antinociceptive efficacy of spinal morphine in rats with nerve ligation injury is prevented by reducing spinal afferent drive, Neurosci. Lett. 199 (1995) 87–90. [33] Z. Rudich, J. Stinson, M. Jeavons, S.C. Brown, Treatment of chronic intractable neuropathic pain with dronabinol: case report of two adolescents, Pain Res. Manage. 8 (2003) 221–224. [34] D.A. Scott, C.E. Wright, J.A. Angus, Evidence that CB-1 and CB-2 cannabinoid receptors mediate antinociception in neuropathic pain in the rat, Pain 109 (2004) 124–131. [35] H. Shuangsong, T.J. Morrow, P.E. Paulson, L.L. Isom, J.W. Wiley, Early diabetic neuropathy is associated with differential changes in tetrodotoxin-sensitive and -resistant sodium channels in dorsal root ganglion neurons in the rat, J. Biol. Chem. (2004) in press. [36] G.S. Simon, W.L. Dewey, Narcotics and diabetes. I. The effects of streptozotocin-induced diabetes on the antinociceptive potency of morphine, J. Pharmacol. Exp. Ther. 218 (1981) 318–323. [37] A. Ulugol, H.C. Karadag, M. Tamer, Z. Firat, A. Aslantas, I. Dokmeci, Involvement of adenosine in the anti-allodynic effect of amitriptyline in streptozotocin-induced diabetic rats, Neurosci. Lett. 328 (2002) 129–132. [38] C.P. Watson, D. Moulin, J. Watt-Watson, A. Gordon, J. Eisenhoffer, Controlled-release oxycodone relieves neuropathic pain: a randomized controlled trial in painful diabetic neuropathy, Pain 105 (2003) 71–78. [39] S.P. Welch, M. Eads, Synergistic interaction of endogenous opioids and cannabinoid systems, Brain Res. 848 (1999) 183–190. [40] O. Yesilyurt, A. Dogrul, H. Gul, M. Seyrek, O. Kusmez, Y. Ozkan, O. Yildiz, Topical cannabinoid enhances topical morphine antinociception, Pain 105 (2003) 303–308. [41] M. Zimmerman, Ethical guidelines for investigation of experimental pain on conscious animals, Pain 16 (1983) 109–110.