Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Nociceptin/orphinan FQ is an endogenous nonopioid peptide ligand for the opioid receptor-like ORL-1 (LC123) receptor, which is involved in pain mediation (Meunier et al., 1995; Reinscheid et al., 1995). Administration of nociceptin produces allodynia, hyperalgesia, and antianalgesia (Henderson and McKnight, 1997; Meunier, 1997). These effects are produced by different mechanisms (Grisel et al., 1996; Mogil et al., 1996a,b; Okuda-Ashitaka et al., 1996; Hara et al., 1997; Minami et al., 1997; Morgan et al., 1997). The purpose of the present study was to determine a mode of the antianalgesic action for nociceptin given i.c.v. into the brain to attenuate the antinociceptive action of intrathecally (i.t.) administered morphine using the mouse tail-flick test. Attenuation of i.c.v. morphine-induced analgesia by i.c.v. nociceptin is well established in the mouse (Grisel et al., 1996; Mogil et al., 1996a,b). Even though in rats nociceptin interferes with the analgesic action of morphine elicited from the periaqueductal gray area, one of the sites of action of morphine (Morgan et al., 1997; Pan et al., 2000), it is not known whether nociceptin can act in the brain through a descending neuronal system to antagonize i.t. administered morphine. Our approach to study this latter interaction takes advantage of the mouse model in which i.t. morphine acts in the spinal cord to increase the latency of the tail-flick response and the antianalgesic action of nociceptin administered i.c.v. indicates the presence of a descending neuronal influence of nociceptin on spinal morphine action. It is likely that such a descending neuronal influence occurs through spinal release of an endogenous mediator that has antianalgesic properties. The focus was narrowed to the systems already known to be involved in producing either hyperalgesia or allodynia when nociceptin is given i.t.

Minami et al. (1997) have shown that allodynia produced in mice by i.t. administration of nociceptin is inhibited by PGD₂ given i.t. In contrast to this, the hyperalgesic action of i.t. nociceptin is not inhibited by i.t. PGD_2. Similarly, i.t PGE₂ produces allodynia that is inhibited by i.t. PGD₂ and hyperalgesia that is not (Minami et al., 1996, 1997). Thus, PGD₂ differentiates the allodynic from the hyperalgesic actions of both i.t. nociceptin and PGE₂. In spite of the parallelism between i.t. nociceptin and i.t. PGE₂, nociceptin allodynia is not inhibited by i.t. indomethacin (Minami et al., 1997), indicating that i.t. nociceptin may not produce allodynia through the release of spinal PGE₂. Thus, there is no connection between the allodynic actions of i.t. nociceptin and PGE₂. However, the present investigation suggests that the antianalgesic action of i.c.v. nociceptin originating from the brain (unlike that for the allodynic action of i.t. nociceptin) is mediated by spinal PGE₂ and thereby involves a descending system to the spinal cord. We have previously used the mouse model to demonstrate descending systems for analgesia (Rady et al., 2000) and antianalgesia (Rady et al., 1998, 1999; Holmes et al., 1999).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Antinociceptive Response. Male CD-1 mice weighing 25 to 30 g were obtained from Charles River Laboratories (Wilmington, MA). The radiant heat tail-flick test was used with the lamp intensity set to provide a predrug tail-flick latency response time of 2 to 4 s. An automatic cut-off time of 10 s was used to prevent tissue injury. In single dose experiments the percentage of maximum possible effect (% MPE) was calculated according to Dewey et al. (1970):
Where dose-response relationships were established for morphine antinociception as affected by various treatments, the quantal response to morphine was determined. Tail-flick latencies greater than the mean predrug time plus three standard deviations were considered to be antinociceptive. At each dose of morphine the mice showing antinociception were expressed as a percentage of the total number of animals (usually eight but in a few cases seven) in that group. Three or more dose levels were used to construct each of the dose-response curves. The probit values were plotted against the morphine dose and the parallelism of the curves and potency ratio comparisons were made by the method of Litchfield and Wilcoxon (1949).

Drug Administration. The basic protocol involved administration of opioid agonists i.t. (Hylden and Wilcox, 1980) to produce analgesia and nociceptin i.c.v. into the right ventricle [under light halothane anesthesia by the method of Haley and McCormick (1957)] to produce an antianalgesic action. The modulatory effect of i.c.v. nociceptin to the spinal cord was investigated by the i.t. administration of various agents to either produce or eliminate antianalgesia. The time for peak action of i.c.v.- and i.t.-administered agents was taken to be at 10 and 5 min, respectively, based on previous work done in this laboratory (see the Introduction) and others (Minami et al., 1996, 1997; Mogil et al., 1996a,b, 1999; Hara et al., 1997). Times, doses, and routes of administration are indicated under Results presented with each experiment. All studies were done in compliance with the Institutional Animal Care and Use Committee (Animal Studies Subcommittee).

Statistical Analysis. Student's t test (comparison of two groups) and ANOVA followed by either Dunnett's test (comparison of each group to a control group) or Newman-Keuls test (comparison of all groups to each other) were used for determining significant differences as indicated by a P  0.05. The ED₅₀ values, 95% confidence intervals, and significance of shifts in the dose-response curves for morphine produced by various treatments as indicated by the ED₅₀ potency ratio (P  0.05) were derived according to the Litchfield and Wilcoxon (1949) method.

Source of Drugs and Drug Solutions. Morphine sulfate (Mallinckrodt Chemical Works, St. Louis, MO) was dissolved in a 0.9% sodium chloride solution. Nociceptin (Bachem, Torrence, CA), [D-Pen^2,5]-enkephalin (DPDPE) (Sigma, St. Louis, MO), and [D-Ser²,Leu⁵]-enkephalin-Thr (DSLET) (Peninsula Laboratories, Belmont, CA) were dissolved in a 0.01% Triton X-100 in 0.9% sodium chloride solution. PGE₂ and PGD₂ (Sigma) and the prostanoid agonists sulprostone, 17-phenyl--trinor PGE₂, and 11-deoxyPGE₁ (Cayman Chemical, Ann Arbor, MI) were dissolved in a 10% (v/v) ethanol solution. Indomethacin (Sigma) was prepared in ethanol with a final concentration of 30%. The substance P antiserum (obtained from Dr. J. S. Hong, National Institutes of Environmental Health, Research Triangle Park, NC) and control serum were used as described in a previous study (Arts et al., 1992). The doses of the drugs given for the experiments were for the above-stated forms. The drug vehicle (designated in each of the experiments presented) was administered to appropriate control groups in the volumes used for the i.c.v. and i.t. drug administrations.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Antianalgesic Action of i.c.v. Nociceptin. In Fig. 1A i.t. morphine at a 1-µg dose produced an analgesic response of about 80% MPE. This response was inhibited by i.c.v. administration of nociceptin. Figure 1B demonstrated the duration of action of a 5.5-µg dose of nociceptin. The nociceptin was given 10 to 120 min before the tail-flick test, whereas the i.t. morphine was given at a fixed dose 5 min before the tail-flick test. The duration of action of the i.c.v. nociceptin was about 1 h. Nociceptin at this dose did not produce an analgesic response (Table 1).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Antianalgesic action of i.c.v. nociceptin against the analgesic action of i.t. morphine in the tail-flick test in mice. A, nociceptin designated in the rectangle for the main parameter evaluated was given i.c.v. at the doses stated at 10 min before the tail-flick test; the v indicates that the vehicle, 0.01% Triton X-100 in saline, was given in place of nociceptin. Morphine was given i.t., 1 µg, at 5 min before the tail-flick test. The columns represent the mean % MPE, the vertical line indicates the S.E., and the number next to it is the number of mice within that group. The asterisk indicates a significant difference (P  0.05) compared with the control group by Dunnett's test. B, duration of action of i.c.v. nociceptin was determined by its administration at the various times indicated before the tail-flick test. Morphine was administered at a fixed time and dose. The asterisk (*) indicates a significant difference from the paired control group (P  0.05) by Student's t test.

Elimination of the Antianalgesic Action of i.c.v. Nociceptin by Indomethacin. The next consideration was whether the antianalgesic action of i.c.v. nociceptin was mediated by spinal prostaglandin. The result in Fig. 2A indicated that the i.c.v. nociceptin-induced antianalgesic action was inhibited by treatment with i.t. indomethacin, a cyclooxygenase 1 and 2 inhibitor (Yamamoto and Nozaki-Taguchi, 1996). The combination of nociceptin and indomethacin gave a response no different from indomethacin by itself; neither treatment produced much analgesia. The antianalgesic action of i.t. nociceptin was not inhibited by the 0.7-µg dose of i.t. indomethacin (Fig. 2B). The next step was to show that the prostaglandin involved in the i.c.v. nociceptin-induced antianalgesia might be PGE₂.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Ability of indomethacin to inhibit the antianalgesic action of i.c.v. nociceptin (A) but not i.t. nociceptin (B). Designations for nociceptin administration are given in the rectangle where v represents vehicle and + indicates nociceptin administration. A, indomethacin treatment, indicated in the top rectangle, was given 15 min before the tail-flick test and the vehicle was 30% ethanol. Note that saline was given i.t. to the two last groups instead of morphine. The asterisk (*) indicates these means were significantly different from those that have no asterisk (P  0.05) by Newman-Keuls test. B, indomethacin treatment given 10 min before the tail-flick test and nociceptin was given i.t. instead of i.c.v. Indomethacin did not have a significant effect. The asterisk (*) indicates significant difference from nonasterisk group (P  0.05) by Newman-Keuls test.

View larger version (36K): [in this window] [in a new window]   Fig. 3.   Antianalgesic action of i.t. PGE2 against i.t. morphine. A, various doses of PGE2 were evaluated against morphine-induced analgesia. The vehicle (v) was 10% ethanol in saline solution. The asterisk (*) indicates significant difference (P  0.05) compared with the control group by Dunnett's test. B, PGE2 was given at a fixed dose at different times before the tail-flick test, whereas the i.t. morphine dose and time were kept constant. The asterisk (*) indicates significant difference (P  0.05) from the paired control group by Student's t test.

View larger version (53K): [in this window] [in a new window]   Fig. 4.   Ability of PGD2 to eliminate the antianalgesic action of i.t. PGE2 and i.c.v. nociceptin. A, vehicle (v), 10% ethanol in saline solution, or PGD2 was given in the same solution with PGE2 at the designated doses and time. Morphine was given as indicated. The asterisk (*) indicates a significant difference (P  0.05) from the other groups by Newman-Keuls test. B, nociceptin or vehicle (Triton X-100 in saline) was followed in 5 min by morphine along with either vehicle (10% ethanol in saline) or PGD2, i.t. The asterisk (*) indicates significant difference (P  0.05) from all other groups by Newman-Keuls test. C, duration of action of PGD2 was determined by its administration at various times before the tail-flick test, whereas the morphine and nociceptin dose and time were kept fixed. Single (*) and double (**) asterisks indicate significant difference (P  0.05) from groups designated differently by Newman-Keuls test.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Dose-response curves for i.t. morphine at 5 min before the tail-flick test and groups given additional treatments. A, groups were given i.c.v. vehicle (0.01% Triton X-100 in saline) or nociceptin and i.t. morphine in 10% ethanol in saline or morphine with PGD2 as stated. The quantal response for groups of seven to eight mice were expressed as % MPE and transformed to probit units. B, control group for morphine is the same data as stated above. All i.t. agents were given in the same solution.

Lack of Antianalgesic Action of i.c.v. Nociceptin and i.t. PGE₂ against the Antinociceptive Action of DPDPE and DSLET. Neither i.t. PGE₂ nor i.c.v. nociceptin at doses that were antianalgesic against i.t. morphine were effective against i.t. DPDPE- and DSLET-induced antinociception (Fig. 6). These results confer further matching selectivity to the antianalgesic actions produced by i.t. PGE₂ and i.c.v. nociceptin.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Neither i.c.v. nociceptin (A) nor i.t. PGE2 (B) produced antianalgesia against i.t. 1 or 2 agonist- (DPDPE or DSLET, respectively) induced analgesia (P > .05 by Student's t test).

Evaluation of Desensitization and Cross-Desensitization between i.c.v. Nociceptin and i.t. PGE₂ Antianalgesic Actions. Figure 7A provides the results in which 3-h pretreatment with i.c.v. nociceptin produced desensitization to the antianalgesic action of i.c.v. nociceptin; the 3-h pretreatment significantly reduced the antianalgesic action of i.c.v. nociceptin. The nociceptin pretreatment did not produce desensitization to the antianalgesic action of i.t. PGE₂. In Fig. 7B, desensitization was produced to the antianalgesic action of i.t. PGE₂ by 3-h pretreatment with PGE_2. This pretreatment effectively desensitized to the antianalgesic action of i.c.v. nociceptin.

View larger version (35K): [in this window] [in a new window]   Fig. 7.   Desensitization to the antianalgesic action of i.c.v. nociceptin and i.t PGE2 by 3-h pretreatment with i.c.v. nociceptin (A) or i.t. PGE2 (B). A, nociceptin pretreatment desensitized to nociceptin but not to PGE2. The single (*) and double (**) asterisks indicate groups were significantly different from other groups not similarly marked (P  0.05) by Newman-Keuls test. B, PGE2 pretreatment desensitized to both i.c.v. nociceptin and i.t. PGE2. The asterisk (*) indicates significant difference (P  0.05) from groups not marked similarly by Newman-Keuls test.

Determination of the Prostaglandin Receptor Involved in the Antianalgesic Action of PGE₂. Because PGE₂ could be acting on several different prostaglandin receptors (Narumiya et al., 1999), an attempt was made to determine the prostanoid E (EP) receptor subtype involved by use of compounds with different relative selectivity for the receptor subtypes. The receptor agonists were chosen based on our previous experience (Csukas et al., 1998) and are the same as those used by Minami et al. (1994). In Fig. 8, the antinociceptive action of i.t. morphine was unaffected by i.t. administration of a 100-ng dose of 11-deoxyPGE₁, which has equal activity for EP₂ and EP₃ receptors. Sulprostone with EP₃ > EP₁ receptor activity effectively produced antianalgesia at a 1- and 10-ng dose. A 10-ng dose of 17-phenyl--trinorPGE₂ that has EP₁  EP₃ activity was also antianalgesic.

View larger version (39K): [in this window] [in a new window]   Fig. 8.   Examination of different prostaglandins with various prostaglandin E (EP) receptor selectivity (EP1, EP2, and EP3) for antianalgesic activity. Both sulprostone (selectivity, EP3 > EP1) and 17-phenyl--trinorPGE2 (EP1 > EP3) possessed antianalgesic activity against i.t. morphine. 11-DeoxyPGE1 (EP2 = EP3) was not effective. The asterisk (*) indicates significant difference from control value (P  0.05) by Dunnett's test or Student's t test. Vehicle (v) was 10% ethanol in saline.

Mediation of the Antianalgesic Action of i.t. Nociceptin by Substance P (SP). In contrast to the antianalgesic effect of i.c.v. nociceptin, the antianalgesia produced by i.t. nociceptin was not inhibited by i.t. PGD₂ (data not shown). Instead, the results in Fig. 9A indicated that the antianalgesic action of i.t. nociceptin was eliminated by the i.t. administration of SP antiserum 1 h before the tail-flick test. As expected and demonstrated previously (Arts et al., 1992) the antianalgesic action of SP was eliminated by the SP antiserum treatment. However, the antianalgesic action of i.c.v. nociceptin was not affected by i.t. SP antiserum treatment (Fig. 9B).

View larger version (30K): [in this window] [in a new window]   Fig. 9.   Effect of i.t. substance P antiserum on the antianalgesic action of i.t. (A) and i.c.v. (B) nociceptin-induced antianalgesia against i.t. morphine. In A and B, the substance P antiserum or control antiserum was given 1 h before the tail-flick test. Designations for nociceptin are v for vehicle (0.01% Triton X-100 in saline) and + for nociceptin administration. Note that the antiserum was effective in eliminating the antianalgesic action of i.t. SP and i.t. but not i.c.v. nociceptin-induced antianalgesia. The asterisk (*) indicates significant difference from all other groups not similarly marked (P  0.05) by Newman-Keuls test.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Administration of nociceptin i.c.v. in Swiss-Webster mice produces antianalgesia against i.c.v. µ-, -, and -agonists (Mogil et al., 1996b). Nociceptin also functions as an antiopioid peptide (Grisel et al., 1996; Mogil et al., 1996a). Carrying this antiopioid peptide theme further, Mogil et al. (1999) postulate that i.c.v. nociceptin acts against the opioid component of stress-induced analgesia to produce hyperalgesia. The i.c.v. drug administration procedure produces stress-induced analgesia only in certain strains of mice, such as Swiss-Webster, and it is the antagonism of this stress-induced analgesia that leads to hyperalgesia. This association found in several different strains of mice seemed to explain the variability observed in eliciting the hyperalgesic response (Henderson and McKnight, 1997; Meunier, 1997). In CD-1 mice (used in the present study), the i.c.v. injection procedure does not produce stress-induced analgesia (Rossi et al., 1997; Mogil et al., 1999). One group found no hyperalgesia after i.c.v. nociceptin administration (Mogil et al., 1999) yet the other group (Rossi et al., 1997) found a biphasic response consisting of hyperalgesia followed by analgesia in CD-1 mice. In spite of the variability in the hyperalgesic response, an antianalgesic effect of i.c.v. nociceptin occurs in both Swiss-Webster (Grisel et al., 1996; Mogil et al., 1996a,b, 1999) and CD-1 mice (the present study). The previous work on Swiss-Webster mice (Grisel et al., 1996; Mogil et al., 1996a,b, 1999) examines the interaction between i.c.v. nociceptin and i.c.v. opioids, whereas our work in CD-1 mice focuses on the antianalgesic effect of i.c.v. nociceptin elicited against morphine given i.t. at a site remote from where the nociceptin was given and therefore provided a way to characterize the neuronal mediator involved at the spinal site.

The finding that the antianalgesic action of i.c.v. nociceptin was inhibited by the i.t. treatment with indomethacin suggested the involvement of a prostaglandin at the spinal level. These results with indomethacin were consistent with indomethacin acting in the spinal cord to inhibit the cyclooxygenase, which converts arachidonic acid to prostaglandins (Yamamoto and Nozaki-Taguchi, 1996). An alternative argument might be that an analgesic action of indomethacin adds to the analgesic action of morphine wherein the outcome might look like indomethacin was inhibiting nociceptin action. There was no hint of indomethacin analgesia adding to that of morphine nor indomethacin having an analgesic action of its own (Fig. 6). However, the response to i.t. morphine was near maximal so the conditions might not have been suitable for showing a possible synergistic interaction between morphine and indomethacin. In the other experiment, i.t. indomethacin did not affect the antianalgesic action of i.t. nociceptin against i.t. morphine. If indomethacin were interacting with morphine to produce a greater analgesic response, then elimination of the antianalgesic action of i.t. nociceptin should have been obtained in this experiment. The outcome was not the same as for i.c.v. nociceptin. Thus, the alternative explanation could be discounted. The finding that i.t. nociceptin antianalgesia was not inhibited by i.t. indomethacin also eliminated the possibility that i.c.v. nociceptin-induced antianalgesia was produced through spinal nociceptin, either endogenously released or diffused down from the brain. Preliminary results suggested that the antianalgesic action of i.t. nociceptin was mediated by spinal SP because SP antiserum eliminated this antianalgesic action. However, further work is necessary to establish this mechanism. A connection between nociceptin and SP is not completely new because nociceptin is known to release SP from peripheral nerves (Inoue et al., 1998). In contrast to i.t. nociceptin-induced antianalgesia, the antianalgesic action of i.c.v. nociceptin was not affected by i.t. administration of SP antiserum. These results provided further evidence that i.c.v. nociceptin did not release spinal nociceptin. The differences between i.c.v. and i.t. nociceptin-induced antianalgesia also suggested that the allodynic and hyperalgesic effects of spinal nociceptin (Hara et al., 1997; Minami et al., 1997) were not involved in the antianalgesic action of i.c.v. nociceptin.

The results suggest a link between a descending neuronal system activated in the brain by nociceptin and a prostaglandin-mediated system in the spinal cord. It should be noted that the dose of i.c.v. nociceptin is high (5.5 µg) so that even though activation is produced pharmacologically, whether such a nociceptin-induced activation occurs physiologically needs to be established. In rats in which the hind paw is immersed in hot (50°C) water, PGE₂ is released into spinal perfusates (Coderre et al., 1990), suggesting a physiological function related to pain. Also, PGE₂ is released from the spinal cord when formalin is injected into the hind paw of the rat (Malmberg and Yaksh, 1995). Thus, PGE₂ seems to be released by several different means, including brain nociceptin action. The allodynic and hyperalgesic actions of spinal PGE₂ are produced by different mechanisms (Minami et al., 1996; Hara et al., 1997). The allodynic but not the hyperalgesic action of spinal PGE₂ is antagonized by PGD₂; thus, the hyperalgesic action of spinal PGE₂ is not related to the allodynic action. The relationship between the allodynic and antianalgesic actions of PGE₂ cannot be resolved without further studies.

Minami et al. (1994) found that the allodynic effect of PGE₂ involved the EP₁ receptor in the spinal cord of mice. Their conclusion was based on the finding that the prostaglandin analog 17-phenyl--trinorPGE₂ was more potent than sulprostone in producing allodynia. The 17-phenyl--trinorPGE₂ is relatively more selective for EP₁ than EP₃ receptors and sulprostone is more selective for EP₃ than EP₁ receptors (Minami et al., 1996; Csukas et al., 1998). The present results suggested that 17-phenyl--trinorPGE₂ and sulprostone appeared to have about the same potency. However, 11-deoxyPGE₁, which has equal selectivity for EP₂ and EP₃, was not antianalgesic. By process of elimination it would seem that the EP₁ receptor is the one involved in producing antianalgesia. Thus, the receptor involved in the antianalgesic and allodynic action might be the same. However, the present data is far from robust and the receptor assignment based on agonist potencies is tenuous. No selective antagonists are available (Narumiya et al., 1999).

Our finding that i.t. morphine antinociception was inhibited by i.t. nociceptin is at odds with that of Grisel et al. (1996) where i.t. nociceptin did not affect the antinociceptive action of i.t. morphine. One explanation may be that they used Swiss-Webster and we used CD-1 mice. Swiss-Webster mice have a higher threshold in the tail withdrawal test (Mogil et al., 1999). Other laboratories find results similar to those presented here. Nociceptin given i.t. produces allodynia and hyperalgesia in ddY mice (Minami et al., 1997) and antianalgesia against i.t. morphine in the rat (Jhamandas et al., 1998).

As mentioned in the Introduction, Morgan et al. (1997) show that the intracerebral injection of nociceptin into the ventral periaqueductal gray of the rat antagonizes the analgesic action of morphine elicited from the same site. They infer that nociceptin inhibits the activation of periaqueductal gray neurons by morphine. Pan et al. (2000) show that the analgesic action of morphine in the periaqueductal gray area of the rat is attenuated by administration of nociceptin into the nucleus raphe magnus. The analgesic action of morphine from the periaqueductal gray is obtained by disinhibition of primary neurons in the nucleus raphe magnus and nociceptin inhibits this disinhibition by hyperpolarizing the neuron through an increase in K⁺ conductance, a direct interaction within the primary neuron. Thus, it seems unlikely that the effect on the primary neuron in the nucleus raphe magnus would explain our results with i.c.v. nociceptin, which requires activating a descending antianalgesic PGE₂ system and inhibition of the analgesic action of morphine at spinal sites. Further work is needed to more clearly separate these modes of action.