Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

There is increasing evidence of interaction between mu and delta opiate receptors (see Traynor and Elliot, 1993 for review). These receptors can coexist on the same neuron, as proposed for receptor populations in the neostriatum (Schoffelmeer et al., 1990), or interact at different sites in a common pathway (Rossi et al., 1994). Biochemical evidence also supports the existence of independent and interacting mu and delta receptors. On the basis of radioligand binding studies, Rothman et al. (1988) suggest that delta opiate receptors may exist in two different states: complexed with mu receptor (d_cx) or independent of mu receptors (d_ncx).

Stimulation of delta opiate receptors modulates mu-based antinociception. Two populations of delta receptors (delta₁ and delta₂) have been postulated on the basis of pharmacologic evidence (Jiang et al., 1991). Both delta-1 agonists, such as DPDPE, and delta-2 agonists, such as [D-Ala², Glu⁴] deltorphin, have been demonstrated to interact with morphine (Jiang et al., 1991; Porreca et al., 1992), perhaps in a synergistic way (Malmberg and Yaksh, 1992). However 5-NTII, an antagonist specific for the delta-2 receptor, blocked both effects (Porreca et al., 1992). The agonist data suggest that mu and delta receptor populations can interact to produce antinociception. However, the failure of delta antagonists to block morphine-induced antinociception suggests that delta receptor function is not mandatory for mu-mediated antinociception (Calcagnetti and Holtzman, 1991; Jackson et al., 1989; Jiang et al., 1991).

There is also evidence that delta receptor activation contributes to the development of morphine-induced tolerance and physical dependence. Coadministration of the delta-2 antagonist 5-NTII with either 100 mg/kg of morphine or morphine pellets over 3 days prevented the normal development of tolerance and dependence in mice, whereas the delta-1-specific antagonist DALCE did not prevent the development of physical dependence in mice (Abdelhammid et al., 1991; Miyamoto et al., 1993, 1994). BW373U86, a putative delta opioid agonist, attenuated abstinence behaviors in rats when co-administered with morphine (Lee et al., 1993). Surprisingly, administration of the delta antagonist naltrindole (Portoghese et al., 1988) with morphine did not significantly block physical dependence in this same model system (Lee et al., 1993). Although these findings support a role for delta receptors in opiate dependence, the effectiveness of a delta agonist in this regard appears to contradict the reported effects of delta antagonists. This discrepancy creates controversy as to the specific role of delta receptors in the development of tolerance.

Both endogenous and exogenous opioids are known to induce respiratory depression. One postulated mechanism is diminished sensitivity of the neurons of the brain stem to the stimulatory effects of carbon dioxide (for review see Shook et al., 1990). However, the role of delta receptors in this effect is unclear. Delta agonists have been demonstrated to induce respiratory depression in rats whether they are administered centrally (Pazos and Florez, 1984) or peripherally (Morin-Surun et al., 1984). The experiments of Ling et al. (1985) suggest that the mu-2 receptor is crucial in opiate-induced respiratory depression and that delta receptors are less important. However, there have been no studies investigating the role of delta receptor antagonists in the development of tolerance to opiate-induced respiratory depression.

The present study was conducted to investigate the effect of delta receptor blockade by naltrindole on the development of tolerance to morphine-induced antinociception, on respiratory depression and on the development of physical dependence during chronic morphine treatment of rats. The purposes of this study were to extend previous studies into rats and to investigate delta receptor effects on tolerance to respiratory depression as a system in which the acute effects of delta receptor function are known but chronic effects have not been documented. The results of this study suggest that naltrindole partially attenuates the development of tolerance to morphine-induced antinociception and physical dependence, while not affecting tolerance to morphine-induced respiratory depression.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Adult Sprague-Dawley rats (Charles River Laboratories, Raleigh, NC) were used in all experiments. The rats had access to food and water ad libitum and were on a 12-hr light/dark cycle, with the lights on at 6:00 A.M.

Materials. Naltrindole hydrochloride was purchased from Research Biochemicals International (Natick, MA). Morphine and the morphine pellets were generously supplied by the National Institute on Drug Abuse. Morphine and naltrindole were dissolved in saline.

Chronic morphine injection paradigm. The chronic morphine regimen used was the paradigm shown previously to lead to significant tolerance and measurable withdrawal behaviors (Windh et al., 1995). Morphine was administered s.c. at 12-hr intervals for 5 days. The dosage began at 5 mg/kg and was increased 5 mg/day up to 25 mg/kg on Day 5. The animals received no morphine on Day 6 and were challenged with morphine (10 mg/kg) on Day 7. Control animals were treated with saline. Subcutaneous naltrindole (1 mg/kg) or saline was given 1 hr before the morning morphine doses on Days 1 to 5. The naltrindole dose and daily treatment schedule are based on the long half-life of the drug and on effective blockade of the delta receptor by peripheral administration (Portoghese et al., 1988).

Morphine pellet paradigm. One or two 75-mg pellets were implanted s.c. between the scapulas in animals anesthetized with isoflurane. Control animals received placebo pellets of identical size and weight.

Naltrindole administration. Naltrindole (10 µg) was administered via an i.c.v. route 90 min before pellet implantation. The i.c.v. naltrindole dose and regimen were based on experiments by Eisenberg (1993). For this procedure, animals were anesthetized with isoflurane and injected through the foramen magnum into the fourth ventricle. Control animals received i.c.v. saline. This procedure was repeated 24 hr after the initial naltrindole administration. Animals received an acute morphine challenge of 10 mg/kg at 48 hr.

Antinociceptive assay. The warm-water tail-flick assay was used to measure antinociception (Janssen et al., 1963; Negus et al., 1993). Each rat was placed on a table with the tail hanging freely over the edge. Each rat was gently restrained as the tail was immersed in a beaker of water heated to 55°C. A foot pedal timer was activated upon immersion of the tail and stopped when the tail was withdrawn. During the chronic injection protocol, animals were tested 2 min before their acute morphine challenge and then 40 min after the acute challenge. A maximum withdrawal time of 15 sec was employed. Pelleted-implanted animals were tested at 4, 16, 24 and 48 hr after pelleting and 40 min after an acute challenge with morphine. Pilot studies suggested that tissue damage and subsequent hyperalgesia occurred in rats subjected to serial testing when latencies exceeded 7 sec. Because serial testing was employed for the pellet experiments, the cut-off time for all pellet studies was set at 7 sec.

Withdrawal. After the completion of antinociception testing, withdrawal was precipitated in animals that received two morphine or placebo pellets. Animals were placed in individual cages approximately 2 hr after the morphine treatment. They were allowed to acclimate to these cages for 10 min. Naloxone (5 mg/kg) s.c. was administered, and then the animals were observed for 10 min (Windh et al., 1995). Each animal was observed for 20 sec of each minute. The following behaviors were recorded as present or absent at each minute for 10 min: ptosis, salivation and forepaw treading. Wet dog shakes were counted for total episodes in the 10-min period. At the end of 10 min, the rats were checked for additional withdrawal signs: diarrhea, sensitivity to touch and chromodacryorrhea. The rats were weighed before the morphine challenge and then 1 hr after the naloxone challenge.

Respiratory depression. Respiratory depression was measured 48 hr after pellet implantation and an acute challenge with morphine. A pressure-sensitive chamber was utilized to measure respiratory depression (Weese-Mayer et al., 1992). The machine was initially calibrated for atmospheric pressure, animal temperature and animal weight. The chamber pressure deflection (proportional to tidal volume with appropriate calculation) was computer-sampled at 200 Hz in resting animals with no gross movements. Respiration was sampled for 15 sec in every 30-sec period. The rats were placed in the chamber and allowed to acclimate for 10 min. Respiration was measured over 5 min, and an average basal value was obtained. A 10-mg/kg morphine challenge was administered to the rat, and respiration was measured between 25 and 35 min after the injection, when pilot studies suggested that morphine effects on respiration were most evident. These values were averaged and compared to the basal average for that animal to calculate a percentage difference. Values are expressed as 100(minute ventilation after acute morphine challenge)/minute ventilation at base line.

Statistics. Two-way ANOVA was used to determine the effect of naltrindole vs. saline pretreatment on the development of tolerance to antinociception. Repeated-measures ANOVA was utilized for the antinociception experiments in which serial testing was employed. Scheffé's post-hoc test was performed when ANOVA indicated a significant interaction between groups. The Mann-Whitney U test was utilized for withdrawal data in which the response was quantitated. The -square test was applied to the withdrawal data that measured the presence or absence of a sign in the animal at 10 min after the naloxone challenge. The level of significance for all applicable statistical tests was set at P < .05. Paired t tests were used to compare respiratory rate before and after morphine challenge.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Blockade of tolerance to morphine-induced antinociception by cotreatment with naltrindole. Naltrindole partially blocked the development of tolerance to the antinociceptive effects of chronic morphine administration after all three morphine treatment paradigms. As shown in figure 1, rats given the morphine injection regimen showed a significantly smaller antinociceptive response to morphine than rats injected with saline. Although the antinociceptive response of rats treated daily with naltrindole alone was no different from control values, animals cotreated with morphine and naltrindole showed significantly greater antinociceptive responses (less tolerance) than animals treated with morphine alone.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effects of naltrindole treatment on antinociceptive tolerance after the morphine injection paradigm. Antinociceptive response in rats treated with the increasing dose regimen of morphine for 5 days and cotreated with saline or naltrindole (s.c.) as described in "Materials and Methods." Results are expressed as mean ± S.E.M. of the difference between pre- and post-treatment latency for individual animals. N = 6/experimental condition. * indicates P < .05 or better relative to morphine/saline.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effects of naltrindole treatment on antinociceptive tolerance after implantation of a single morphine pellet. Antinociceptive response in rats implanted with one morphine or placebo pellet and cotreated with saline or naltrindole. Animals were treated with morphine and naltrindole as described in "Materials and Methods." Results are expressed as mean ± S.E.M. of the difference between pre- and post-treatment latency for individual animals. N = 6/experimental condition. * indicates P < .05 or better relative to morphine/saline.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Effects of naltrindole treatment on antinociceptive tolerance after two morphine pellets. Animals were treated with morphine and naltrindole as described in "Materials and Methods." Results are expressed as mean ± S.E.M. of the difference between pre- and post-treatment latency for individual animals. N = 8/experimental condition. * indicates P < .05 or better relative to morphine/saline.

Blockade of withdrawal by cotreatment with naltrindole. Cotreatment with naltrindole attenuated abstinence symptoms in the adult rats treated with the two-pellet paradigm. In animals exposed to morphine chronically, naloxone initiated significant withdrawal signs that were substantially more frequent than such signs in animals receiving placebo pellets (figs. 4 and 5). Seven measurements for withdrawal indicated a difference between the morphine/saline and morphine/naltrindole groups, including quantitated signs of forepaw treading, salivation, wet dog shakes (see fig. 4) and weight loss (see table 1), as well as the checked signs of diarrhea, vocalization on touch and chromodacryorrhea/rhinorrhea (see fig. 5). These symptoms were present among the animals that received morphine and naltrindole, but the severity was diminished relative to the rats that received morphine alone. The degree of ptosis was not different between the two groups.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effects of naltrindole treatment on counted withdrawal signs. Animals were treated with two morphine or placebo pellets followed by naloxone as described in "Materials and Methods." Results are expressed as mean ± S.E.M. of incidences in 10 min. N = 10/experimental condition. * indicates P < .05 or better relative to matched placebo condition. # indicates P < .05 or better relative to morphine/saline treatment.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effects of naltrindole on checked withdrawal signs. Animals were treated with morphine or placebo pellets and with naltrindole or vehicle followed by precipitation of withdrawal as described in "Materials and Methods." Results are expressed as % incidence in the 10-min observation period. N = 10/experimental condition. * indicates P < .05 or better relative to matched placebo condition. # indicates P < .05 or better relative to morphine/saline treatment.

Lack of naltrindole effects on tolerance to morphine-induced inhibition of respiration. The respiratory depression experiments were conducted on rats that were exposed to the two-pellet chronic morphine regimen (fig. 6). Base-line rates were obtained at 48 hr after the initial implantation. The similarities in the base-line values between the morphine-implanted animals and the placebo-implanted animals suggests that rats develop rapid tolerance to the respiratory effects of morphine (see legend of fig. 6). The morphine-naive animals exhibited acute respiratory depression after the morphine challenge, whereas animals that received morphine pellets were not affected by the acute challenge.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Effects of naltrindole on tolerance to respiratory depression. Animals were implanted with one morphine or placebo pellet and cotreated with naltrindole or vehicle as described in "Materials and Methods." Results are presented as mean ± S.E.M. of % change in respiratory rate for individual animals before and after a challenge dose of morphine. Basal respiratory rate = 117 ± 3 for placebo/saline, 106 ± 7 for morphine/saline, 115 ± 5 for placebo/naltrindole and 109 ± 5 for morphine/naltrindole. N = 5/experimental group. * indicates P < .05 or better relative to paired placebo condition.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The findings of the present study suggest that occupation of delta opioid receptors contributes to the development of morphine-induced tolerance to antinociception and physical dependence in rats, whereas tolerance to morphine-induced respiratory depression is not influenced by delta receptor occupation. The findings of these experiments support the results of earlier studies in mice in which development of tolerance was prevented by delta receptor antagonists (Abdelhammid et al., 1991). In addition, the present study extends previous findings of naltrindole blockade of naloxone-induced jumping to demonstrate blockade of additional opioid withdrawal signs. Finally, in contrast to the findings with antinociception, naltrindole cotreatment did not prevent the development of tolerance to morphine-induced respiratory depression.

The significant antinociception induced by morphine injection or pellet implantation was not influenced by coadministration of naltrindole. This is consistent with the majority of published studies, which indicate that although morphine-induced antinociception can be enhanced by administration of subantinociceptive doses of either of the delta-specific agonists DPDPE and [D-Ala², Glu⁴] deltorphin (Heyman et al., 1989; Porreca et al., 1992; Malmberg and Yaksh 1992), administration of delta antagonists typically does not prevent morphine-mediated antinociception (Calcagnetti and Holtzmann, 1991; Abdelhammid et al., 1991; Sofuoglu et al., 1991).

The reason why delta receptor stimulation can augment mu agonist antinociception but delta antagonists do not diminish mu-mediated antinociception is not clear. Coupling of receptors has been demonstrated both at the level of the single cell and at the level of neural pathways. Rothman et al. (1988) have provided both in vivo and in vitro evidence of a coupling of mu and delta receptors. Similarly, Schoffelmeer et al. (1990, 1993) used biochemical techniques to demonstrate interacting mu and delta binding sites that inhibit dopamine-sensitive adenylate cyclase in the rat neostriatum. Alternatively, a mu/delta pathway interaction involving the periaqueductal gray and the rostral ventral medulla has been shown to augment morphine-induced antinociception (Rossi, 1994). However, the inability of delta antagonists to influence mu-mediated antinociception is not well understood.

Although naltrindole did not affect the acute antinociceptive response, it substantially decreased the development of tolerance after either peripheral injection of both morphine and naltrindole or i.c.v. administration of naltrindole before implantation of morphine pellets. These results provide support for the idea of mu/delta cooperativity in the development of morphine tolerance that has been previously demonstrated in mice (Abdelhammid et al., 1991). These authors compared ED₅₀ values for mice that had received morphine alone vs. those that received morphine and naltrindole or morphine and 5NTII. They concluded that antagonism of delta receptors substantially prevented the development of tolerance.

The greater effectiveness of naltrindole administered after higher chronic morphine dose regimens in the present study was surprising. Although the blockade of tolerance after injection paradigms was quite modest, substantial blockade was observed after a single morphine pellet, and even better blockade occurred after the administration of two pellets. One reason might have been the different route used for naltrindole delivery, which probably delivered a significantly higher dose than that obtained after peripheral administration. However, the different blockade observed after one or two pellets is not easily explained pharmacokinetically. Whatever the mechanism, this quality offers potential clinical utility for improvement in analgesia in situations where increasing opiate doses are needed to deal with increasing pain.

Physical dependence is another aspect of chronic morphine administration that is susceptible to modification by delta receptor agonists or antagonists. In the present study, we found that naltrindole pretreatment significantly attenuated seven different withdrawal symptoms. These findings confirm and extend the report of Abdelhammid et al. (1991) that an increase in the amount of naloxone was needed to precipitate withdrawal jumping in morphine-dependent mice that were pretreated with naltrindole. This effect has been attributed to actions of the delta-2 receptor (Miyamoto et al., 1993; Miyamoto et al., 1994). It should be noted that contrasting conclusions have been drawn on the basis of studies with the putative delta agonist BW373U86. Lee et al. (1993) demonstrated that BW373U86 blocked the development of physical dependence, wherein coinfusion of naltrindole with morphine without BW373U86 did not prevent the development of tolerance to physical dependence in rats. However, the blockade of physical dependence by BW373U58 may be due to competitive antagonism at the delta receptor site, because this compound has been demonstrated to be a partial mu and delta agonist (Wild et al., 1993).

The activity of naltrindole against a profile of withdrawal behaviors extends the previous findings with naloxone-induced jumping and suggests that interacting mu/delta receptors that are important for the development of opiate dependence exist at multiple sites in the CNS. Sites responsible for various opiate withdrawal behaviors are located both centrally and peripherally (Koob et al., 1992). The withdrawal signs of diarrhea (Bianchetti et al., 1986), salivation, lacrimation and rhinorrhea may be mediated by peripheral receptors (Maldonado et al., 1992), whereas the locus ceruleus and the periaqueductal gray matter have also been implicated as sites that are active during opiate withdrawal (Maldonado et al., 1992). The anterior preoptic and raphe magnus may be particularly important for the induction of wet dog shakes (Maldonado et al., 1992). Our results suggest that cooperative mu and delta receptors may mediate withdrawal at several of these sites. However, the absence of actions on ptosis was a little surprising, because symptoms such as diarrhea that are thought to have a similar mediation by noradrenergic hyperactivity were blocked (Taylor et al., 1988).

The conclusions of the present study regarding the role of delta opioid receptors rely in part on the reported specificities of morphine and naltrindole for mu and delta receptors, respectively. Morphine is reported to show 10-fold specificity for mu over delta receptors in vitro (Pasternak 1986, Change et al., 1979), whereas naltrindole specificity ranges from 20-fold to 100- to 500-fold in different reports (Portoghese et al., 1988; Ayers et al., 1990; Rogers et al., 1990). In vivo, a number of studies have shown that the present dose regimen for naltrindole fails to block analgesia by mu agonists while effectively blocking that induced by delta agonists (Portoghese et al., 1988; Calcagnetti and Holtzmann, 1991; Drower et al., 1991; Improta and Broccardo, 1992; Malmberg and Yaksh, 1992; Craft et al., 1995; Yaksh et al., 1995). The most relevant report for the present study is the recent demonstration that naltrindole and the more selective delta antagonist TIPP (H-Tyr-Tic-Phe-Phe-OH) similarly block the development of morphine dependence during antagonist and morphine coinfusion into rats (Fundytus et al., 1995). In this study, TIPP but not naltrindole prevented the development of tolerance to opiate-induced analgesia, but differences in the analgesic test conducted might explain this difference. Several factors in the present study support the validity of these assumptions. First, naltrindole did not influence morphine analgesia after the first 24 hr after morphine pellet implantation, when morphine brain levels are higher (Yoburn et al., 1985). Therefore, morphine seems to have retained mu selectivity up to the highest levels observed in the present study. Similarly, naltrindole did not prevent morphine analgesia after an acute challenge of 10 mg/kg. The finding that a similar dose of naltrindole failed to block either morphine- or DAMGO-induced ACTH secretion supports the specificity of this in vivo naltrindole dose (C.M. Kuhn and R. Francis, unpublished observations). Nevertheless, a complete dose-response study for naltrindole-induced blockade of tolerance and the investigation of the ability of other delta antagonists to prevent the development of tolerance would provide additional evidence bearing on this hypothesis.

Our experiments have demonstrated a significant tolerance to the morphine-induced respiratory depression in rats treated with two 75-mg morphine pelletsa tolerance that that was not prevented by coadministration of naltrindole. The tolerance observed resembles that previously reported in mice (Roerig et al., 1987). The lack of naltrindole blockade suggests that the sites at which opiate agonists suppress respiration in the pons and medulla (Taviera da Silva et al., 1983; Hurle et al., 1982, 1985) may not possess interacting mu and delta receptor populations, although both mu and delta agonists are known to produce respiratory depression in rodents (Morin-Surun et al., 1984; Pazos et al., 1984). This conclusion is consistent with the putative receptor mediation of respiratory vs. antinociceptive effects, because agonists specific for the delta-1 receptor have been implicated in control of respiration (Mayfield and D'Alecy, 1994a; Mayfield and D'Alecy, 1994b), whereas delta-2 receptors have been implicated in tolerance and dependence (Miyamoto et al., 1993; 1994). These results imply that independent mu and delta receptors regulate the development of tolerance to morphine-induced respiratory depression.

The finding of blockade of tolerance to antinociception but not respiratory depression has potential clinical significance, regardless of the relative role of mu and delta receptors in mediating this effect. Our findings suggest that coadministration of effective mu agonists in conjunction with delta antagonists could enhance the effectiveness of long-term therapy in which development of tolerance can impair the clinical effectiveness of drugs. The failure to block tolerance to respiratory depression would permit beneficial tolerance to a limiting side effect to develop, while maintaining clinical effectiveness. However, it should be emphasized that more studies using models of chronic pain would be required to demonstrate the utility of this treatment. The effectiveness of antagonists in blocking responses to an acute noxious stimulus are not necessarily predictive of responses in chronic pain models.

In summary, the present findings have demonstrated that development of tolerance to certain effects during chronic morphine administration relies on mu/delta cooperativity, whereas tolerance to other effects develops independently of delta receptors. This difference may create a window of opportunity for drug development. A delta antagonist may be useful in preventing the development of tolerance to morphine-induced analgesia, while minimizing withdrawal symptoms and avoiding the deleterious consequences of respiratory depression.